King’s College London!

Thesis submitted for the degree of

PhD

Andrea Corrado Profeta

BDS Hons

Department of Restorative Dentistry

Biomaterials Science, Biomimetics and Biophotonics (B3) Research Group

King’s College London Dental Institute

at Guy’s, King’s College and St Thomas’ Hospitals

MMXIII

This electronic theses or dissertation has been downloaded from the King’s Research Portal at https://kclpure.kcl.ac.uk/portal/

The copyright of this thesis rests with the author and no quotation from it or information derived from it may be published without proper acknowledgement.

Take down policy

If you believe that this document breaches copyright please contact librarypure@kcl.ac.uk providing details, and we will remove access to the work immediately and investigate your claim.

END USER LICENSE AGREEMENT

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0

Unported License. http://creativecommons.org/licenses/by-nc-nd/3.0/

You are free to:

Share: to copy, distribute and transmit the work Under the following conditions:

Attribution: You must attribute the work in the manner specified by the author (but not in any way that suggests that they endorse you or your use of the work).

Non Commercial: You may not use this work for commercial purposes.

No Derivative Works - You may not alter, transform, or build upon this work.

Any of these conditions can be waived if you receive permission from the author. Your fair dealings

and other rights are in no way affected by the above.

Title:Hybridisation of dental hard tissues with modified adhesive systems: therapeuticimpact of bioactive silicate compounds on bonding to dentine

Author:Andrea Corrado Profeta

Copyright

Copyright © 2013 by Profeta, Andrea Corrado

All rights reserved.

The copyright of this thesis rests with the Author and no quotation from it or

information derived from it may be published without proper acknowledgement.

Copies (by any process) either in full, or of extracts, may be made only in

accordance with instructions given by the Author and lodged in the Maughan

Library of King’s College London. Details may be obtained from the Librarian.

This page must form part of any such copies made.

Further copies (by any process) of copies made in accordance with such

instructions may not be made without the permission (in writing) of the Author.

The ownership of any intellectual property rights which may be described in this

thesis is vested in King’s College London, subject to any prior agreement to the

contrary, and may not be made available for use by third parties without the

written permission of the University, which will prescribe the terms and

conditions of any such agreement.

Further information on the conditions under which disclosures and exploitation

may take place is available online at the College institutional repository:

http://www.kcl.ac.uk/library/visiting/maughan.aspx

Recommended Citation:

Hybridisation of dental hard tissues with modified adhesive systems:

therapeutic impact of bioactive silicate compounds on bonding to dentine.

Profeta AC.

PhD Thesis 2013. King’s College London, Strand, London WC2R 2LS,

England, United Kingdom.

Hybridisation of dental hard tissues

with modified adhesive systems:

therapeutic impact of bioactive silicate

compounds on bonding to dentine

Andrea Corrado Profeta

Bachelor of Dental Surgery BDS Hons

Università Cattolica del Sacro Cuore (UCSC)

Class of 2006

Thesis submitted for the degree of

Doctor of Philosophy PhD in Clinical Dentistry

King’s College London (KCL)

Department of Restorative Dentistry

Biomaterials Science, Biomimetics and Biophotonics (B3) Research Group

KCL Dental Institute

at Guy’s, King’s College and St Thomas’ Hospitals

2013

2

I dedicate this work to the adversities that made it so worthwhile

3

Structure of the thesis, objectives and working plan

The first section of this work is a review of the literature necessary to

understand the objectives of the project; it includes general information about

dental adhesive technology as well as adhesion testing, about dentine

hybridisation and about the drawbacks of contemporary bonding systems.

Several studies revealed excellent immediate and short-term bonding

effectiveness of etch-and-rinse adhesives, yet substantial reductions in resin-

dentine bond strength occur after ageing. Degenerative phenomena involve

hydrolysis of suboptimally polymerised hydrophilic resin components and

degradation of mineral-deprived water-rich resin-sparse collagen matrices by

matrix metalloproteinases and cysteine cathepsins.

Silicate compounds, including calcium/sodium phosphosilicates, such as

commercially available bioactive glass, and calcium-silicate Portland-derived

cements are known to promote the formation of apatite in aqueous

environments that contain calcium and phosphate (e.g. saliva); thus, we have

raised questions about whether their presence at the bonded interface could

increase the in vitro durability of resin-dentine bonds through crystal formation

and self-sealing, in the presence of phosphate buffered saline or simulated

body fluid solutions.

In answering these questions, the objectives were accomplished by employing

Bioglass® 45S5 in etch-and-rinse bonding procedures either (i) included within

the composition of a resin adhesive as a tailored micro-filler, or (ii) applied

directly onto acid-etched wetted dentine. Alternative light-curable methacrylate-

4

based agents containing (iii) three modified calcium-silicates derived from

ordinary Portland cement were also tested.

Confirming the relative success of bioactive materials incorporated in the

dentine bonding procedures required assessment of the potential to reduce

nano-leakage, as well as their effect upon the strength of the bond over time.

In order to explore these possibilities, which have not been previously

investigated, a combination of methods were applied in the second

experimental section. Bond strength variations were quantified using the

microtensile test while scanning electron microscopy, confocal laser scanning

microscopy and Knoop micro-indentation analysis were used to evaluate

optically and mechanically adjustments to mineral and water content within the

resin bonded-dentine interface. Initially, high microtensile values were achieved

in each tested group. All the resin-dentine interfaces created with bonding

agents containing micro-fillers showed an evident reduction of nano-leakage

and mineral deposition after the ageing period. However, only adhesive

systems containing Bioglass and two modified Portland cement-based micro-

fillers were found to reduce nano-leakage with no negative effects on bond

strength. Furthermore, specimens created with the same experimental

adhesives did not restore micro-hardness to the level of sound dentine but were

able to maintain statistically unaltered Knoop values.

The second section is also composed of a set of preliminary studies that

involved the use of up-to-date spectroscopic (attenuated total reflection Fourier

transform infrared spectroscopy) and thermoanalytical (differential scanning

calorimetry) techniques to predict the chemical-physical properties and apatite-

forming ability of the novel ion-leachable hybrid materials. Lastly, the overall

5

conclusions of the present work and directions for future research are

discussed.

6

Acknowledgements

According to Merriam-Webster's dictionary, adversity means “a state, condition, or instance of serious or continued difficulty or adverse fortune” while triumph denotes “a great victory or success.” In any case, it is impossible to experience a sense of triumph over adversity unless you have first stared the possibility of disaster in the face. The taste of success means little unless you have a hint of the flavour of failure to compare it to. Acts of great courage are only taken after terrifying fears have been acknowledged and understood. Against almost everyone’s predictions, this thesis is respectfully submitted to Professor Dianne Rekow, Dean of the Dental Institute at King’s College London (KCL), and to Professor Tim Watson, Head of the Institute’s Biomaterials, Biomimetics and Biophotonics (B3) Research Group. The Dental Institute at KCL is full of talented, masterful and honourable people. I am proud to have been part of the B3 team and lucky to know so many brilliant clinicians and scientists. I wish past and present staff members who interacted with me throughout this project all the best; most especially, I would like to place on record my thanks to Professors Alistair Lax and Gordon Proctor for their direct involvement in bringing it to a successful conclusion. Also, I would like to extend my appreciation to Dr. Richard Foxton for his assistance in the academic and administrative requirements involved in my candidacy. Of course I am grateful to my family for their unconditional support in everything I choose to do and obsess over. Special mention to Agnė for helping me going through all those years, and for so much more... She knows the kind of pandemonium I endured in my life and that completing this work was a pretty big deal for me. Something I am glad I experienced, but would never welcome back again. Should somebody else ask me now, ‘Did you enjoy your PhD?’ ‘Did you use your time wisely?’ I will not hand over a piece of paper with the CV and other achievements on it to use up most of the alphabet after my name, or give an explanation of why I might be better than others. It is not, at least for me, about looking back or looking down, about titles, honorifics and status. I am simply going to stand up and smile a smile which lets people know I have no regrets at all. I was eager to be faced with all this experience had to offer, the intensity and unique opportunity to do things at the highest level, and discover what it might show me about myself. Unexpectedly my world was turned upside-down, my trust tested and my ego crushed. I had to be twice as good, three times as sharp, four times as focused than all the other PhD candidates. I had to prove myself ten times over but I never gave up and I succeeded where others failed. I can look at this record now and think how far I have come, and how far I have grown and also how grateful I am for all those experiences, regardless of how difficult they were at the time. Things I can take with me wherever I go, essential ingredients in a better me which can never be taken away, not just material goods I own briefly. The latin saying NIL DIFFICILE VOLENTI has certainly proved true for me and I am sure it will hold true for anyone who believes it.

7

List of contents

Structure of the thesis, objectives and working plan.............................

Acknowledgements...................................................................................

List of Figures............................................................................................

List.of Tables..............................................................................................

Section I - A review of the literature.........................................................

Chapter 1: Adhesive technology and dentine bonding

limitations................................................................................

1.1 Introduction............................................................................................

1.1.1 Coupling resin monomers to enamel...........................................

1.1.2 Adhesion to dentinal substrates...................................................

1.2 Development of dentine-resin bonding technology................................

1.2.1 Early dentine bonding agents.......................................................

1.2.2 Smear-layer removal and acid conditioning……………………….

1.2.3 Dentine hybridisation and resin-infiltrated smear-layer................

1.3 Physico-mechanical considerations of resin-bonded dentine................

1.3.1 Wettability of dentinal surfaces and contact angle.......................

1.3.2 Solubility of adhesive monomers.................................................

1.3.3 Permeability of the collagen network and

monomers diffusivity....................................................................

1.3.4 Permeability of adhesive resins and water sorption.....................

1.4 Mechanisms responsible for loss of mechanical stability.......................

3 6 14 17

19

20 21 22 23 28 29 31 32 35 36 39 41 44 47

8

1.4.1 Hydrolytic degradation of dental adhesive resins.........................

1.4.2 Endogenous collagenolytic activity..............................................

1.5 Adhesion testing.....................................................................................

1.5.1 Assessment of sealing ability.......................................................

1.5.1.1 Micro-leakage and micro-permeability............................

1.5.1.2 Nano-leakage..................................................................

1.5.2 Bond strength measurement........................................................

1.5.2.1 Macro-bond strength test................................................

1.5.2.2 Micro-bond strength test.................................................

1.6 Classification of contemporary bonding systems...................................

1.6.1 Etch-and-rinse..............................................................................

1.6.2 Self-etch.......................................................................................

1.6.3 Self-adhesive...............................................................................

Chapter 2: Strategies for preventing resin-dentine bond

degradation..............................................................................

2.1 Introduction............................................................................................

2.1.1 Improvement of degree of conversion and esterase

resistance......................................................................................

2.1.2 Inhibition of enzyme-catalysed hydrolytic cleavage

of collagen.....................................................................................

2.1.3 Use of collagen cross-linking agents.............................................

2.1.4 Ethanol-wet bonding technique.....................................................

48 50 56 60 61 62 65 66 68 71 72 75 82 87 88 89 90 96 102

9

2.1.5 Restoring the mineral phase of the collagen

matrix…………………...…………………………………………….

2.1.5.1 Guided tissue remineralisation.........................................

2.1.5.2 Top-down remineralisation via epitaxial growth…….……

2.1.5.3 Key objectives in the design of bioactive dentine

bonding systems..............................................................

2.2 Development of ion-releasing adhesives comprising

bioactive fillers........................................................................................

2.2.1 Calcium/sodium phosphate-phyllosilicates fillers..........................

2.2.2 Filler phase consisting of calcium silicate cements.......................

2.2.3 Dye-assisted confocal microscopy imaging of

remineralised hard tissues............................................................

2.2.4 Aims of the study...........................................................................

Section II - Experimental projects............................................................

Chapter 3: Chemical-physical properties and apatite-forming

ability of experimental dental resin cements

containing bioactive fillers.....................................................

3.1 Introduction............................................................................................

3.2 Materials and methods...........................................................................

3.2.1 Experimental micro-fillers and resin blends

formulation....................................................................................

3.2.2 Specimen preparation...................................................................

105 108 114 122 124 128 133 137 141 143 144 145 147 147 150

10

3.2.3 Water sorption and solubility evaluation........................................

3.2.4 Differential scanning calorimetry (DSC)........................................

3.2.5 Statistics........................................................................................

3.2.6 ATR-FTIR spectroscopy................................................................

3.3 Results...................................................................................................

3.3.1 Water sorption and solubility evaluation.......................................

3.3.2 Differential scanning calorimetry (DSC).......................................

3.3.3 ATR-FTIR spectroscopy...............................................................

3.4 Discussion..............................................................................................

3.5 Conclusion.............................................................................................

Chapter 4: Bioactive effects of a calcium/sodium phosphosilicate

on the resin-dentine interface: a microtensile bond

strength, scanning electron microscopy, and confocal

microscopy study...................................................................

4.1 Introduction............................................................................................

4.2 Materials and methods...........................................................................

4.2.1 Specimen preparation..................................................................

4.2.2 Experimental bonding procedures and formulation

of resin adhesives.........................................................................

4.2.3 μTBS and SEM fractography and failure analysis.........................

4.2.4 Confocal microscopy ultramorphology and nano-leakage

evaluation......................................................................................

4.3 Results...................................................................................................

151 152 153 153 154 154 157 159 164 169 170 171 172 172 173 178 179 182

11

4.3.1 μTBS and SEM fractography and failure analysis……..………….

4.3.2 Confocal microscopy ultramorphology and nano-leakage

evaluation.......................................................................................

4.4 Discussion..............................................................................................

4.5 Conclusion.............................................................................................

Chapter 5: Experimental etch-and-rinse adhesives doped with

calcium silicate-based micro-fillers to generate

therapeutic bioactivity within resin-dentine

interfaces.................................................................................

5.1 Introduction............................................................................................

5.2 Materials and methods...........................................................................

5.2.1 Preparation of the experimental bioactive

resin-base bonding agents............................................................

5.2.2 Specimen preparation and bonding procedures...........................

5.2.3 μTBS and SEM observations of the failed bonds..........................

5.2.4 Dye-assisted CLSM evaluation.....................................................

5.3 Results...................................................................................................

5.3.1 μTBS and SEM observations of the failed bonds..........................

5.3.2 Dye-assisted CLSM evaluation.....................................................

5.4 Discussion..............................................................................................

5.5 Conclusion.............................................................................................

182 186 189 195 196 197 199 199 203 205 206 207 207 211 216 222

12

Chapter 6: In vitro micro-hardness of resin-dentine interfaces

created by etch-and-rinse adhesives comprising

bioactive fillers........................................................................

6.1 Introduction............................................................................................

6.2 Materials and methods...........................................................................

6.2.1 Teeth collection and preparation...................................................

6.2.2 Formulation of the comonomer resin

adhesive blend………………………………………………………..

6.2.3 Bioactive fillers and experimental bonding

systems.........................................................................................

6.2.4 Bonding procedures......................................................................

6.2.5 Knoop micro-hardness (KHN) analysis.........................................

6.3 Results...................................................................................................

6.3.1 Knoop micro-hardness (KHN) analysis.........................................

6.4 Discussion..............................................................................................

6.5 Conclusion.............................................................................................

Chapter 7: General discussion and conclusion......................................

7.1 Summary................................................................................................

7.2 Research contributions..........................................................................

7.3 Recommendations for future research...................................................

Bibliography...............................................................................................

223 224 226 226 226 229 230 231 234 234 237 242 243 244 249 251 254

13

List of publications in international peer-reviewed journals

as a result of this work..............................................................................

List of abstracts in international conferences of dental research

from this work…............…….….….….….….…...........………....................

Appendix.....................................................................................................

325 326 327

14

List of Figures

Figure 1.1 - Crystal structure of biogenic hydroxyapatite.…………................ 24

Figure 3.1 - ATR-FTIR spectra of the unmilled comonomer blend, of Bioglass®

45S5, HOPC, HPCTO and HPCMM powders and of the hybrid experimental

adhesives immediately after curing and following 60 days in DPBS……….. 162

Figure 4.1 - Schematic illustrating the experimental study design................. 176

Figure 4.2 - Schematic illustrating the composite-tooth matchsticks (1 mm)

prepared using a water-cooled diamond saw, stored in PBS for 24 h or 6

months, and then subjected to microtensile bond strength (μTBS) testing and

scanning electron microscopy failure analysis. This schematic also illustrates

how composite-tooth slabs were prepared, stored in PBS for 24 h or 6 months,

and evaluated by confocal laser scanning microscopy................................... 181

Figure 4.3 - Scanning electron microscopy images of failure modes of the resin-

bonded specimens created using the three different bonding approaches

tested.............................................................................................................. 185

Figure 4.4 - Confocal laser scanning microscopy (CLSM) images showing the

interfacial characterisation and nanoleakage, after 24 h of storage in PBS, of

15

the resin-dentine interfaces created using the three different bonding

approaches tested......................................................................................... 187

Figure 4.5 - Confocal laser scanning microscopy (CLSM) images showing the

interfacial characterisation and nanoleakage, after 6 months of storage in PBS,

of the resin-dentine interfaces......................................................................... 188

Figure 5.1 - Chemical structures of the methacrylate monomers used in the

tested resin blends.......................................................................................... 201

Figure 5.2 - Schematic illustrating the resin-dentine match-sticks prepared

using a water-cooled diamond saw, stored in SBS for 24 h or 6 months, and

then subjected to microtensile bond strength (µTBS) testing and scanning

electron microscopy fractography. This schematic also illustrates how

composite-tooth slabs were prepared, stored in SBS for 24 h or 6 months,

immersed in fluorescein (nanoleakage) or Xylenol Orange (Calcium-binding

dye) and finally evaluated by confocal laser scanning microscopy

(CLSM)............................................................................................................ 204

Figure 5.3 - SEM failure analysis of debonded specimens............................ 210

Figure 5.4 - Confocal laser scanning microscopy (CLSM) single-projection

images showing the interfacial characterisation and nanoleakage, after 24 h of

storage in SBS................................................................................................ 213

16

Figure 5.5 - CLSM single-projection images disclosing the fluorescent calcium-

chelators dye xylenol orange.......................................................................... 214

Figure 5.6 - Confocal laser scanning microscopy (CLSM) single-projection

images showing the interfacial characterisation and nanoleakage after 6 months

of SBS storage................................................................................................ 215

Figure 6.1 - Optical images obtained during the micro-hardness test along the

resin-dentine interface.................................................................................... 233

17

List of Tables

Table 3.1 - Chemical structures of the constituent monomers and composition

(wt%) of the experimental adhesives used in this study................................ 149

Table 3.2 - Summary of maximum water uptake, solubility and net water uptake

data................................................................................................................ 156

Table 3.3 - Means and standard deviations for Tg initially, after the ageing

period and percentage change as determined by DSC

analysis.......................................................................................................... 158

Table 4.1 - Composition of the experimental bonding procedures/adhesive

systems used in this study............................................................................. 177

Table 4.2 - Means and standard deviations (SD) of the microtensile bond

strength values (MPa) obtained for the different experimental groups and

percentage distribution of failure modes after microtensile bond strength testing;

total number of beams (tested stick/pre-load failure)..................................... 184

Table 5.1 - Chemical composition (wt%) and application mode of the

experimental adhesive system used in this study.......................................... 202

Table 5.2 - Mean and standard deviation (SD) of the μTBS (MPa) to

dentine........................................................................................................... 209

18

Table 6.1 - Chemical composition (wt%) of the experimental adhesive systems

used in this study........................................................................................... 228

Table 6.2 - The results of the micro-hardness measurements for each bonding

system after 24 hours and 6 months of PBS storage.................................... 236

19

Section I - A review of the literature

20

Chapter 1: Adhesive technology and dentine

bonding limitations

21

1.1 Introduction

Adhesion or bonding is the process of forming an adhesive junction, which

consists of two materials joined together. Any event described as adhesion is

really an assembly involving a substrate (or ‘adherend’) with an applied

‘adhesive’ that creates an intervening ‘interface’. In reparative dentistry (Small,

2008), the adherends are enamel and dentine to which the adhesive is applied.

Dental adhesives are solutions of resin monomers that join a restorative

material with the tooth structure after their polymerisation is completed. While

most adhesive joints involve only two interfaces, dental adhesive joints may be

more complex such as the dentine-adhesive-composite interface of a bonded

composite direct restoration. The aim is to create a close relationship between

the dental substrate and restorative material, reproducing the natural

relationship of the dental tissues, and to protect the pulp. Biomimetics, or

imitating nature, is concerned with not only the natural appearance and

aesthetic aspects of the restorations but the way they work. To copy nature is to

understand the mechanics of the tooth, the way it looks and functions, and the

way every stress is distributed. Ideally, the interface should provide a secure

marginal seal and have the ability to withstand the stresses that have an effect

on the bonding integrity of the adhesives, in order to keep the restoration

adherent to the cavity walls. There are several sequential events that are

necessary to form an effective adhesive joint. Bonding between hard tissues of

the tooth and dental adhesive involves potential contributions from chemical

(e.g., ionic bonds), physical (e.g., van der Waals) and mechanical sources but

primarily relies on micro-mechanical interaction for success. For the

development of strong adhesion, good wetting and intimate contact between the

22

adhesive and substrate, which must be clean and therefore in a high energy

state, are required. Films of water, organic debris, and/or biofilms are always

present in the clinical situation as dental surfaces that are prepared for

restorative procedures, which present contaminants that remain on them.

Consequently the steps for interface formation are the creation of a clean

surface and the generation of a rough surface for interfacial interlocking.

1.1.1 Coupling resin monomers to enamel

In 1955, Michael G. Buonocore reported the use of 85% phosphoric acid for 60

seconds in order to improve the retention of an acrylic resin on enamel

(Buonocore, 1955, Simonsen, 2002). Still today adhesion to enamel is achieved

by means of etching with 32-37% phosphoric acid, ideally in a gel form, for 15-

20 seconds. Acid etching creates microporosities, produces surface roughness,

reduces the surface tension of the prepared surface and forms facets on the

mineral crystals (Baier, 1992); this permits the hydrophilic monomers of the fluid

adhesive resin to penetrate into the micro-retentive spaces in-between or within

the enamel crystals. Accordingly, the micromechanical nature of the interaction

of dental adhesives with enamel is a result of the infiltration of resin monomers

into the microporosities left by the acid dissolution of enamel and subsequent

enveloping of the exposed hydroxyapatite (HAp) crystals with the polymerised

monomers (Swift et al., 1995). This makes it possible to obtain an adhesive-

composite-enamel bond strength able to resist a shearing force of more than 20

MPa, which is clinically remarkably effective (Swift et al., 1998).

23

1.1.2 Adhesion to dentinal substrates

Whilst the adhesion to enamel, thanks to the etching technique, promptly

demonstrated its efficacy, convincing in the following years researchers and

clinicians, the same cannot be said for the adhesion to dentinal surfaces. The

quest for an adequate dentine bonding agent has been longer and even today

there is no confirmation of having attained the effectiveness which the adhesion

to enamel has demonstrated. Enamel is composed of 96% HAp (mineral) by

weight, whereas dentine contains a large percentage of organic material and

water. It has been found that its bulk chemical composition is about 50% in

volume made of mineral substance, 30% in volume formed of organic material,

and the residue 20 vol% represented by water (Marshall et al., 1997). This

tissue can be considered a biological composite that consists of a highly cross-

linked and insoluble in acids collagen matrix filled with mineral crystallites

located both within (intrafibrillar) and between (interfibrillar) collagen fibrils

(Kinney et al., 2003).

The mineral component is primarily a carbonated nanocrystalline hydroxyapatite

whose structure is far different from stoichiometric hydroxyapatite, represented

by the formula:

Me10(XO4)6Y2

Where Me is a divalent metal (Ca2+, Sr2+, Ba2+, Pb2+ …), XO4 is a trivalent anion

(PO43-, AsO4

3-, VO43- …), and Y is a monovalent anion (F-, Cl-, Br-, I-, OH-…).

Given the unique mechanism involved in apatite crystal formation in biology,

biogenic apatite varies in several ways from the corresponding geologically

produced mineral.

24

First, biogenic apatite has a hexagonal lattice structure, having a strong ability

to form solid solutions, and to accept numerous substitutions (Figure 1.1).

Figure 1.1 - Crystal structure of biogenic hydroxyapatite.

These substitutions affect the apatitic lattice parameters: the crystal size is

decreased, and thereby the surface area is increased compared to

stoichiometric HAp, thus permitting additional adsorption of ions and molecules

on the apatite surface (LeGeros, 1991).

Biological apatite contains in fact various trace elements from intrinsic or

extrinsic origins, namely significant carbonate substitutions, OH- deficiencies,

and imperfections in the crystal lattice (Boskey, 2007). This phenomenon

25

provides certain physico-chemical, biological, functional, and chemical features

important in the formation and dissolution of the crystals in dental tissues. For

example, F- ions are readily incorporated into dental apatite, forming

fluoroapatite, a less soluble phase of calcium phosphate as compared to HAp,

confering to enamel its low dissolution properties to resist acidic attacks.

Likewise, trace elements present in extracellular fluids may have a specific role

on mineral quality and condition.

With respect to dentine apatite structure, this is represented by numerous

substitutions (i) by hydrogenophosphate (HPO42-) of XO4 groups and (ii) by

carbonate (CO32-) of Y2 and XO4 groups.

Finally, biological minerals tend to attain high crystallinity and a more organised

structure on the time scale of days or months rather than years (Verdelis et al.,

2007).

The dentine matrix is mainly composed of type-I collagen fibrils with associated

noncollagenous proteins, to form a three-dimensional matrix that is reinforced

by the apatite crystals (Marshall et al., 1997).

Collagen microfibrils are described as those strands of collagen that are 5-10

nm in diameter, collagen fibrils are bundles of microfibrils that are 50-100 nm in

diameter, and collagen fibres are bundles or networks of fibrils that are

approximately 0.5-1 μm thick (Eick et al., 1997).

This mineral-reinforced fibril composite is described by Weiner and Wagner as

containing parallel platelike HAp crystals with their c-axis aligned with the long

axis of the fibril (Weiner and Wagner, 1998). The location of these crystals in

the fibril was demonstrated in a study by Traub and co-workers that showed

26

that mineralised collagen fibrils had the same banded pattern as negatively

stained collagen fibrils (Traub et al., 1989).

This indicated that mineral is concentrated in the hole zones of the fibril. It was

proposed that these mineral platelets were arranged in parallel like a stack of

cards within the interstices of the fibril (Palmer et al., 2008).

The quality of dentine is dependent upon the total sum of characteristics of the

tissue that influence its competence: microstructure, mineral density and

especially the particular location of the mineral with respect to organic

structures of the tissue. From a microstructural perspective, the collagen fibrils

in dentine serve as a scaffold for mineral crystallites that reinforce the matrix,

supporting the surrounding enamel. This microstructure suggests the necessity

of a hierarchical approach to the understanding of its mechanical properties

(Kinney et al., 2003).

The mineral component incorporates oriented tubules that run continuously

from the dentine-enamel junction (DEJ) to the pulp in coronal dentine, and from

the cementum-dentine junction (CEJ) to the pulp canal in the root. Each tubule

is encased in a collar of highly mineralised dentine, called peritubular dentine,

embedded in the intertubular matrix (Marshall, 1993).

These tubes are elongated cones with their largest diameters (ca. 3.0 µm) at

the pulp and their smallest diameters (ca. 0.8 µm) at the DEJ, and are filled with

a liquid that flows inside, with a pressure of about 20 mm of Hg (Van Hassel,

1971). The quantity of the tubules decreases from about 45,000 per mm2 in the

proximity of the pulp to about 20,000 per mm2 near the dentino-enamel junction.

As they converge on the pulp chamber, the surface area of the intertubular

dentine diminishes while the tubule density augments, from about 1.9 x 106

27

tubules/cm2 at the DEJ to between 4.5 x 106 and 6.5 x 106 tubules/cm2 at the

dentine-pulp edge (Garberoglio and Brännström, 1976). This humid and organic

nature of dentine makes it very challenging to bond to and have an effect on the

integrity of the tooth-adhesive side of the interface. A peculiarity of dentine is

the presence of the dentinal fluid in the tubular constitution that couples the pulp

with the enamel-dentinal junction (EDJ). As stated by the hydrodynamic theory

(Neill, 1838, Gysi, 1900, Brannström and Aström, 1972), when the enamel is

lost and the dentine is exposed, external stimuli cause fluid shifts across the

dentine which activate pulpal nerves and cause pain. This fluid flux within

tubules, accountable for dentine sensitivity (Pashley et al., 1993), is also

responsible for the persistent wetness of exposed dentine surfaces due to the

outward fluid movement from the pulp, which may influence the quality of the

adhesive-dentine interface and may decrease the bond strength between resins

and dentine (Sauro et al., 2007). Furthermore, the increasing number of tubules

with depth and, consequently, the increment in dentine wetness, can make

bonding to deeper dentine even more difficult than to superficial dentine.

The fluid movement in the dentinal tubules under the influence of pulpal

pressure may in fact interfere with the penetration of the adhesive into the

conditioned dentine surface (Chersoni et al., 2004), as well as causing

deterioration of the adhesive interface with time. Another characteristic of

dentine is the presence of a coating of debris produced with mechanical

preparation, called smear-layer, consisting of shattered and crushed HAp, as

well as fragmented and denatured collagen that is contaminated by bacteria

and saliva (Brännström et al., 1981). It is revealed by scanning electron

microscopy (SEM) as a 1-2 μm adherent surface with a mainly granular

28

substructure that varies in roughness, density and degree of attachment to the

underlying tooth structure according to the surface preparation (Pashley et al.,

1988). While cutting dentine, the heat and shear forces produced by the rotary

movement of the bur cause this debris to compact and aggregate. The orifices

of the dentinal tubules are obstructed by debris tags, called smear-plugs, that

are contiguous with the smear-layer and may extend into the tubules to a depth

of 1-10 μm (Prati et al., 1993).

The application of acidic agents opens the pathway for the diffusion of

monomers into the collagen network, but it also facilitates the outward seepage

of tubular fluid from the pulp to the dentine surface, leading to a deterioration in

the bonding effectiveness of some of the current adhesives. After the HAp

crystals have been removed, it is quite challenging to also maintain the spaces

created between collagen fibrils to allow monomers to diffuse into the substrate.

The demineralised dentinal matrix can actually easily collapse if the matrix

peptides, including collagen, are denatured during the conditioning, causing a

decrease in the interfibril spacing and a loss of permeability to resin monomers

(Nakabayashi et al., 1982).

1.2 Development of dentine-resin bonding technology

In the developments of dental adhesives several attempts have been made to

provide a stronger and more reliable bond as well as simplifying the clinical

procedures. These attempts have resulted in the introduction of different

generations of bonding systems which are different in chemistry, mechanism,

number of bottles, application techniques and clinical effectiveness.

In general, dentine bonding agents all contain similar ingredients, namely cross-

29

linking agents, bifunctional monomers, organic solvents, curing initiators,

inhibitors or stabilisers, and sometimes inorganic filler particles.

Whereas cross-linkers have two polymerisable groups (vinyl-groups or -CQC-)

or more, functional monomers commonly have only one polymerisable group

and a functional group, which can serve different purposes, such as enhancing

wetting of dentine. Bifunctional monomers have in fact (meth)acrylate functions

at one end, in order to provide covalent bonds with the composite monomers,

and the so-called functional group, usually carboxyl, phosphate, or phosphonate

at the other end which will impart monomer-specific functions (Van Landuyt et

al., 2008).

1.2.1 Early dentine bonding agents

Since calcium is abundant in dentine, the earliest dentine bonding formulas

attempted to chemically bond to dentine by ionic bonds to this alkaline metal.

The first adhesive resin system was created and manufactured at the

Amalgamated Dental Company, England, UK, by a Swiss chemist called Oscar

Hagger: it was composed of glycerophosphoric acid dimethacrylate (GPDM)

and it was made available on the market as Sevriton Cavity Seal (The

Amalgamated Dental Company, Ltd, London, UK) (Haggar, 1951). Kramer and

McLean (1952) were among the first to investigate the bonding ability of this

material to dentine (Kramer and Mc Lean, 1952). The dentine bonded with this

adhesive system was observed by light microscopy: during the histologic

examination they demonstrated altered staining of the bonded subsurface which

took up haematoxylin more readily than did the control surfaces. It was

supposed that the resin-primer had altered the dentine. This study was followed

30

by a work of Buonocore and co-workers (Buonocore and Quigley, 1958), who

etched dentine with 7% hydrochloric acid and then applied GPDM bonding

resin. These attempts were unsuccessful because of limitations in the adhesive

monomer formulations and a general lack of knowledge of dentine as a bonding

substrate. It was demonstrated that there was little evidence for the formation of

chemical bonds between resins and dentine. A few years later, coincident with

an expansion of knowledge in this area, considerable advances were made in

adhesive monomer formulations to improve resin penetration into the tissue

matrix. The development of a cross-linking dimethacrylate 2,2-bis[4(2-hydroxy-

3-methacryloyloxy-propyloxy)-phenyl] propane (Bis-GMA) reinvigorated the

research on adhesion to dentine (Bowen, 1963). Dentine adhesive products

such as Dentin Adhesit (Vivadent, Schaan, Liechtenstein), Scotchbond Dual-

Cure (3M Dental Products, St. Paul, MN, USA), Prisma Universal Bond

(Caulk/Dentsply, Milford, DE, USA) and Bondlite (Kerr, Danburry, CT, USA) did

not remove the smear-layer prior to resin application but were applied directly

to smear-layer covered dentine. The presence of smear-layer on ground

dentinal surfaces greatly reduced the permeability of tubular (Pashley, 1991)

and intertubular dentine (Watanabe et al., 1994); for this reason the resin was

unable to penetrate profoundly enough to establish a bond with intact dentine,

and hence gave very low bond strength values (ca. 3-7 MPa) (Eick et al.).

Examination of both parts of the failed resin-dentine bonds employing scanning

electron microscopy (SEM) revealed smear-layer that split into upper and lower

halves on each side. Thus, the “bond strength” was not really a measure of

bonding, but measured the strength of the cohesive forces holding smear-layer

particles together (Pashley, 1991). The actual interfacial bond strength between

31

the resin and the uppermost part of the smear-layer was higher by an unknown

amount, because it did not fail.

1.2.2 Smear-layer removal and acid conditioning

Despite widespread scepticism among the dental academic community, T.

Fusayama proposed in 1980 to remove the smear-layer along with the

underlying smear-plugs that prevented resin tag formation (Fusayama, 1980).

Acid etching permitted demineralisation of the top 5-10 µm of the underlying

sound dentine allowing dentinal tubules to receive micro-tags of resin and

represented an important innovation which paved the way for the modern

concepts of dentine bonding. However, even smear-layer removal was

insufficient for high resin-dentine bond strength. The technique fell short of

expectations in practice because of the intratubular fluid’s pressure that makes

the dentine extremely humid, especially in deep cavities where a large number

of wider dentinal tubules are exposed. This phenomenon inhibited the

hydrophobic resin to efficiently adhere to the dentinal substrate and water was

regarded as a contaminant. As a result, these bonding agents required severe

air-drying of the dentine surface before application. The outcome of this

manoeuvre was frequently a layer so thin that atmospheric oxygen inhibited

their polymerisation (Erickson, 1989). Air-drying led to the evaporation of water

maintaining the collagen network expanded and its collapse due to surface

tension forces. The spaces between the collapsed collagen fibrils were

therefore greatly reduced and with this the permeability of intertubular dentine to

adhesive resins (Pashley et al., 1995a). What was also required was stripping

the mineral phase from the collagen fibrillar matrix of dentine and keeping it as

32

expanded as possible in order to produce a large increase in surface and

subsurface porosity. If monomers could have infiltrated this mesh-work and

coated the fibrils with polymerised resin to improve micromechanical retention,

they would have produced high bond strength. For the mentioned reasons and

in view of the fact that a water-free environment is unachievable during clinical

procedures, dentine bonding agents were reformulated in a more hydrophilic

blend (Nakabayashi and Takarada, 1992). The introduction of low molecular

weight monomers called primers, such as 4-methacryloxyethyl-trimellitic

anhydride (4-META) and 2-hydroxyethylmethacrylate (HEMA), containing

bifunctional groups - a hydrophilic functional group with a high affinity for the

aqueous dentinal substrate, and a hydrophobic functional group having high

affinity for the bonding adhesive - along with the etch-and-rinse technique

enhanced the strength of the adhesion to dentine and provided reliable

resin/dentine bond strengths (Barkmeier and Cooley, 1992). Furthermore,

leaving the dentine wet made it possible to preserve porosity necessary for

primer penetration in the demineralised spaces (Tay et al., 1995) and led to the

formation of a acid resistant layer consisting of polymerisable hydrophilic

monomers and exposed collagen fibrils (Nakabayashi and Takarada, 1992).

1.2.3 Dentine hybridisation and resin-infiltrated smear-layer

Nakabayashi and his colleagues were the first to use transmission electron

microscopy (TEM) with sufficient resolution to show the penetration of resin

nano-tags into the demineralised dentine matrix to create an entirely new

biomaterial that was half collagen fibrils and half resin. It was neither resin nor

dentine but a hybrid of the two, and so was called hybrid layer (Nakabayashi

33

and Pashley, 1998). Hybridisation thoroughly modified the physico-chemical

properties of tooth surfaces and subsurfaces and was considered a form of

tissue engineering.

With the introduction of the hybrid layer, many clinicians believed that the

mechanism of bonding had been solved. Instead, the complexities of bonding

became apparent when the same adhesive agents produced hybrid layers of

different thicknesses depending on dentine depth and dentine condition. Hybrid

layer formation was the major bonding mechanism in superficial dentine, which

incorporates fewer tubules than deep dentine, due to the amount of intertubular

dentine present in this area with little contribution from resin tags. Whereas, in

deep dentine, resin micro-tag formation remained accountable for the bond

strength, with a reduced contribution of the hybrid layer due to the limited

amount of intertubular dentine available, as the tubules become larger and

closer together.

Nano-tags seemed to be much more important to overall retention (Marshall et

al., 2010) increasing the bond strengths to 32 MPa, concurring for a better

marginal seal and acting as an elastic cushion that, thanks to its elasticity

(modulus of elasticity 3.4 GPa), was able to moderate the polymerisation

shrinkage stress of the restorative composite (Wang and Spencer, 2003).

Although the smear-layer is regarded a limiting factor in achieving high bond

strengths, nowadays it can also be considered as a bonding substrate thanks to

the development of smear-layer incorporating systems called self-etch. This

was realised by raising the amount of acidic monomers and adding 20-30%

acidic methacrylates (pH 1.9-2.8) to 20% water, 20% ethanol, 30% HEMA or

dimethacrylates. Self-etch adhesives contain high concentrations of water and

34

acidic monomers (Watanabe et al., 1994). Water is a necessary ingredient

required to ionise the acidic monomers so that they can etch through smear-

layers into the underlying hard tissues (Tay et al., 2002b). Its presence entailed

the use of water-miscible hydrophilic comonomers (e.g. HEMA) and/or acetone

or ethanol as a solvent to prevent phase changes from occurring (Van Landuyt

et al., 2005). Smear-layer covered dentine is substantially drier than acid-etched

dentine. Smear-layer and smear-plugs being present, the transdentinal

permeability is greatly reduced and no significant wetness is present on the

dentine surface (Pashley, 1989). All the same, self-etching bonding systems are

applied to smear-layer covered dentine under dry conditions, since they contain

their own water. Combining acidic conditioners and resin primers did not require

a separate etch-and-rinse phase and made these agents able to simultaneously

condition and prime enamel and dentine (Chigira et al., 1994). Self-etching

systems interact very superficially with the smear-layer and the underlying

dentine. They can easily penetrate 1-2 μm of smear-layers but their penetration

is restricted just to 0.5 μm into the top portion of the underlying intact dentinal

matrix (Watanabe et al., 1994). This is due, in part, to the fact that the acidity is

partially buffered by the smear-layer during comonomer penetration (Reis et al.,

2004), and because the underlying mineralised dentine is less porous, and

hence less permeable, than smear-layers. Water is also useful for solubilising

the calcium and phosphate ions that are liberated by the etching. These ions,

released from apatite crystallites during self-etching, get incorporated into the

water of the adhesive blend or precipitate as calcium phosphates, which

become dispersed within the comonomers in the interfibrillar spaces. Some of

the calcium ions may also associate with the acidic monomers as calcium salts.

35

This mixture fills the interfibrillar spaces and some free ions may still diffuse up

into the overlying adhesive layer (Bayle et al., 2007). The primed surfaces are

not rinsed with water, leaving the dissolved smear-layer and demineralisation

products to reprecipitate within the diffusion channels created by the acid

primers. Compared with etch-and-rinse adhesives, many advantages have

been attributed to self-etch adhesives. It has been suggested that they improve

the efficiency of clinical procedures by omitting the obligatory rinse phase in

etch-and-rinse adhesives and thus reducing the chairside time. Conditioning,

rinsing and drying steps, which may be critical and difficult to standardise in

clinical conditions, are eliminated in self-etch adhesives. Technique sensitivity

correlated with bonding to dehydrated demineralised dentine is eliminated, as

rinsing and drying phases are no longer needed. Since monomers infiltrate

concomitantly as they demineralise, the collapse of the collagen network is

prevented (Peumans et al., 2005). For the same reason, incomplete resin

infiltration should be avoided. As the smear-layer and smear-plugs are not

removed before the actual bonding procedure, rewetting of dentine by dentinal

fluid should be disallowed too (Van Meerbeek et al., 2005). However, some

leakage observations in the hybrid layer, and especially beyond the hybrid

layer, have shed doubt on the concept that self-etch adhesives guarantee

complete resin infiltration (Carvalho et al., 2005).

1.3 Physico-mechanical considerations of resin-bonded dentine

One factor that could be easily overlooked is the requirement for the bonding

system to act as a means of transferring load from one part of a structure to

another. This generates stresses and strains within the resin-bonded dentine

36

and it is important that the adhesive has the necessary physico-mechanical

properties to withstand these stresses and strains. Thus, the assessment of a

bonding system should be based on its ability to carry load and contribute to the

structural integrity of the whole unit.

The durability of the resin-dentine bond is related to the depth of

demineralisation versus the depth of monomer penetration, and the ability of the

polymer not only to envelope each fibril but also to do so without leaving any

gap or space between the resin and the fibril. That is, the resin-infiltrated layer

must be free of any porosity or defects that can act as stress raisers under

function or permit hydrolysis of collagen fibrils (Nakabayashi et al., 1982).

1.3.1 Wettability of dentinal surfaces and contact angle

Wetting is a general term used to indicate the ability of a liquid to come into

intimate contact with a solid substrate and to maintain contact with it. The

balance between adhesive and cohesive forces dictates the degree of wetting

(wettability). Adhesive forces between the liquid and the solid cause a drop to

spread across, whereas cohesive forces within the liquid cause the drop to ball

up and avoid contact with the surface. If a liquid can spread across a surface, it

is said to “wet the surface”. This wetting ability of a liquid for a surface is usually

characterised by measuring the contact angle (resultant between adhesive and

cohesive forces) of a droplet on the surface.

Resin contact angle measurement on dentine provides information on the

interaction between adhesives and dentine, and it also indicates the affinity of

dentine for the adhesive resin (Rosales-Leal et al., 2001).

37

Low contact angles imply good wetting while a contact angle greater than 90°

usually means that wetting of the surface is unfavourable: the fluid does not

spread over a large area of the surface but tends to minimise contact with it

forming a compact liquid droplet. The tendency of a drop to spread out over a

flat, solid surface hence increases as the contact angle decreases. For water, a

wettable surface may also be termed hydrophilic and a non-wettable surface

hydrophobic. Superhydrophobic surfaces have contact angles greater than

150°, showing almost no contact between the liquid drop and the surface (Feng

et al., 2002).

Wettability of dentine is an important topic to take into consideration as good

spreading of monomers on this tissue is very important for successful bonding.

For a liquid to spread uniformly across a solid surface, the surface tension of

the liquid must be less than the free surface energy of the substrate. Substrates

for bonding may present low or high surface energy. HAp is a high-energy

substrate while collagen has a low-energy surface (Akinmade and Nicholson,

1993). Accordingly, acid etching increases the surface energy of enamel but

decreases that of dentine. Unlike enamel, acid-etched dentine does not

increase its surface energy to facilitate spreading of adhesive resins (Attal et al.,

1994). Thus, for hybridisation of demineralised dentine with resin to occur, it is

necessary to match the surface tension of the primer with that of the

demineralised dentinal surface, depending on whether it is wet or dry.

Commonly used bonding monomers such as HEMA have excellent spreading

properties (Bowen et al., 1996) and could be considered to be surface-active

comonomers (Rosales-Leal et al., 2001). That is, they are considered to

38

improve the ability of the monomers to wet the surface of acid-etched dentinal

substrate.

Wetting of the surface of dentine by monomers is a necessary initial step in

bonding, but it alone is not sufficient to establish a successful bond, because it

does not guarantee monomer penetration into the subsurface. The permeability

of the demineralised intertubular dentinal network to monomers is a critical

variable in dentine bonding (Nakabayashi and Takarada, 1992). To attain

intimate association between resin monomers and collagen fibrils, the primers

and bonding agents must be able to “wet” the collagen fibrils. If the fibril is

enveloped by water, the monomers must be able to successfully compete with

water for the fibril surface.

Barbosa and collaborators found that dentine permeability was also intensified

by the removal of organic materials (Barbosa et al., 1994). Sodium hypochlorite

(NaOCl) is a well-known nonspecific proteolytic agent and its collagen removal

ability after acid conditioning has been evaluated (Wakabayashi et al., 1994).

After NaOCl treatment, the extent to which the primer wets the dentine surface

is increased because the interactions between the primer and the deproteinised

dentine are greater than before (Toledano et al., 2002). Deproteinisation leads

to a hydrophilic surface (Attal et al., 1994) and eliminates the exposed collagen

fibres. Besides, dentine becomes a porous structure with multiple irregularities

which allows good mechanical retention (Vargas et al., 1997). However,

complete removal of the collagen matrix with NaOCl as an adjunctive step of

restorative and adhesive dentistry is still a subject for debate. Sauro et al.

(Sauro et al., 2009a) evaluated the efficacy of a 12% w/v NaOCl solution for

complete removal of exposed collagen matrices from acid-etched dentine

39

surfaces within a maximum clinically possible period of 120 seconds and a

longer period of application (10 minutes) using confocal reflection/immuno-

fluorescence microscopy and ESEM. An extended period (45 minutes) of

NaOCl application was also performed as a negative control. This study

demonstrated that complete removal of the exposed collagen matrix from the

etched dentine surface can be achieved by applying a 12% w/v NaOCl solution,

but at this concentration, it required a far longer reaction time than is clinically

acceptable.

1.3.2 Solubility of adhesive monomers

Solubility is the property of a substance called solute to dissolve in a liquid

solvent to form a homogeneous solution. It is measured as the saturation

concentration where adding more solute does not increase the concentration of

the solution.

The term “solubility parameter” was first used in dentistry by Asmussen

(Asmussen et al., 1991). They regarded demineralised collagen as a porous

solid polymer and reasoned that for primers to penetrate demineralised dentine,

the primer should have a solubility parameter that is similar to the polymeric

substrate, as is generally true in polymer chemistry.

The concept was extended to Hansen’s triple solubility parameters so as to

calculate the relative contribution of dispersive force (δd), polar force (δp),

hydrogen bonding force (δh), and the total cohesive energy density of adhesive

(δt). As Hoy’s triple solubility parameters are more widely used on dentine

bonds, chemical structures modify the calculated Hoy’s triple solubility

parameters for δd, δp, δh and δt (Mai et al., 2009).

40

Solubility parameter calculations have been used to quantify the degrees of

hydrophilicity of polymers, important for the adhesive penetration into exposed

collagen fibrils, and predict dentine-adhesive bond strengths (Asmussen and

Uno, 1993).

When a primer that has a low solubility in water is applied to moist

demineralised dentine, the result is a limited distribution of the monomers into

the water-filled three-dimensional network between the collagen fibrils, with a

consequent low bond strength. Some hydrophilic monomers, such as HEMA,

are very solubile in either water or acetone. Replacing water in the spaces

around collagen fibrils, HEMA acts like a polymerisable solvent for the adhesive

monomers placed thereafter. The uptake of adhesive monomers into these

nano-spaces is contingent on their solubility in the solvent that occupies the

spaces, hence this theory is very useful in predicting how miscible monomers

should be in demineralised matrices saturated with various solvents (Sadek et

al., 2007). Furthermore, the diffusion of the monomers is also determined by the

size of the spaces between collagen fibrils and by the depth that they must

reach from the surface. Wet demineralised dentine exhibits a fully expanded

collagen network that offers maximal volumes between its fibrils. Under similar

conditions, the bonding substrate has high permeability. At the other extreme,

when there are no spaces between the collagen fibrils, as in air-dried, fully

collapsed dentine (Carvalho et al., 1996), the permeability to monomer is

extremely low. The ideal condition exists when there is both high permeability of

the substrate (dentine) and high diffusivity of the solute (resin monomer)

(Nakabayashi and Takarada, 1992).

41

Unfortunately, many adhesive monomers are not very soluble in water. That is

why marketed adhesives are generally solvated in ethanol or acetone. When

solvated adhesives are placed on water-saturated acid-etched dentine, their

solvents attempt to penetrate into the water-filled spaces and some of the water

in these spaces diffuses into the solvent. This culminates in too little solvent

remaining in the infiltrating adhesive with the capacity to keep hydrophobic

dimethacrylates like BisGMA (2,2-bis [4-(2-hydroxy-3-methacryloyloxypropoxy)]

- phenyl propane) in solution. The net result is partial penetration of BisGMA

into water-saturated matrices (Spencer and Wang, 2002). When BisGMA-

HEMA mixtures are placed on water-saturated dentine, the applied

concentrations changes as the much more water-soluble HEMA diffuses to the

base of the demineralised zone. This can result in final molar ratios of BisGMA

and HEMA in the hybrid layer that are very different from the applied molar

ratio.

1.3.3 Permeability of the collagen network and monomers diffusivity

Permeability quantifies the effort with which a substance can penetrate a

membrane or diffusion barrier. The permeability of dentinal substrate to

monomers and their diffusivity are extremely important for the creation of the

hybrid layer.

After the dentinal surface is acid etched and subsequently rinsed, intertubular

spaces are filled with water and are presumed to be still as wide as when they

were occupied by apatite crystallites (Van Meerbeek et al., 1996).

Maintaining the permeability of the substrate as high as possible allows the

achievement of good monomer infiltration because it is through these 15 to 20

42

nm wide diffusion pathways that adhesive monomer must move to fill the

demineralised dentinal matrix and envelop every fibril. As these molecules

diffuse into demineralised dentine, they may encounter some very small or

narrow constrictions within the interfibrillar spaces, especially if the permeability

of the collagen network has not been maintained. This reduces the rate of

inward diffusion of adhesive monomers. If the strength of the bond is

proportional to the sum of the cross-sectional areas of the resin-infiltrated

interfibrillar spaces, then reductions in the size of these spaces should lead to

lower bond strengths. Therefore it is essential to increase monomer

concentration in demineralised dentine and to ensure that it becomes fully

polymerised, to produce strong, durable hybrid layers.

The ability of resins to infiltrate the exposed collagen mesh of dentine and to

create a molecular-level intertwining within the fibril network depends upon their

concentration and uniformity of penetration (Eick et al., 1996), their degree of

polymerisation and cross-linking, and the amount of water that should be

replaced in the demineralised dentinal substrate (Jacobsen and Soderholm,

1995).

The mechanism available for resin infiltration involves the diffusion of the

monomer into the solvent present in the spaces of the substrate and along

collagen fibrils. That is the reason why this zone is also known as the resin

interdiffusion zone (Van Meerbeek et al., 1996). The rate of diffusion depends

on the affinity of the monomer for the substrate and is proportional to the

concentration, temperature and viscosity of the solution (Cussler, 1976). The

intrinsic diffusivity of the molecule, namely, its intrinsic free diffusion coefficient

in the solvent, which is inversely related to its molecular weight or size, is also

43

an important variable. As the diffusion rate is proportional to the square root of

the molecular weight, the smaller molecules diffuse faster and deeper than the

larger ones (Nakabayashi and Pashley, 1998). On this account, whenever a

blend of monomers of widely differing molecular weights is used in a primer or

bonding agent, the rate of diffusion into the underlying substrate may vary to a

considerable extent. This can result in final molar ratios of monomers in the

hybrid layer that are very different from the initial applied concentrations (Eick et

al., 1997).

It has been mentioned how the presence of water during bonding procedures

may come from several sources (i.e. tubular fluid, relative humidity, rinsing

procedures). Post-etching rinsing thoroughly sponged out the dissolved dentine

minerals and left approximately 70% of the demineralised dentine occupied by

water (Nakabayashi et al., 2004). One of the assumptions with the 'wet-bonding'

technique is that exposed collagen is not dried out thoroughly after etching to

prevent its collapse to a thinner less permeable layer and the consequent

restriction of the spaces around fibrils through which resins had to diffuse

(Nakaoki et al., 2000). One way to avoid more than necessary and desirable air

drying of dentine is to add water-miscible solvents in the primer solutions to

chemically remove water from demineralised dentine (Suh, 1991). During the

priming phase, the solvent (which exceeds the water) diffuses through the

spaces between the collagen fibrils to reach the bottom of the demineralised

zone in conjunction with the monomers that therefore have less water to

challenge with (Eick et al., 1996). After evaporation of the solvent, the resin

infiltration is thought to take the place of all the water present between the

collagen fibrils.

44

However, when it was demonstrated that acid-etching lowered the stiffness of

dentine from 18000 MPa to 1-5 MPa (Eddleston et al., 2003), also the

susceptibility of the demineralised matrix to collapse became evident. It was

discovered that even after primer infiltration (35% HEMA in 65% water) into the

matrix, this was still so compliant that evaporation of the solvent was enough to

cause it to collapse and extrude much of the monomers it had taken up

(Eddleston et al., 2003).

Solvents such as ethanol or acetone have much higher vapour pressures and

generate less surface tension forces on the collagen fibrils network compared

with aqueous primers while they evaporate (Maciel et al., 1996). Despite this,

the use of ethanol-solvated primer mixtures also seems to stiffen the matrix

enough to lower, but not to completely prevent, matrix collapse (Agee et al.,

2006).

1.3.4 Permeability of adhesive resins and water sorption

Ideally, polymer networks should be insoluble materials with relatively high

chemical and thermal stability. Unfortunately, very few polymers are absolutely

impermeable to water. Water movement in a polymer system is related to the

availability of molecular-sized pores in its structure, and the affinity of the

polymer components with water (Van Landingham et al., 1999). The availability

of nanopores depends on the polymer microstructure, morphology and cross-

link density, which are functions of degree of cure, relationship between the

relative quantities of substances forming the compound, molecular chain

stiffness and the cohesive energy density of the polymer (Soles and Yee, 2000).

The affinity of the polymer to water is related to the presence of hydrogen

45

bonding sites along the polymer chains which create attractive forces between

the polymer and water molecules (Soles and Yee, 2000). Incorporation of high

concentrations of hydrophilic functional groups and methacrylate-based resin

monomers in contemporary bonding systems, to achieve immediate bond

strength to an intrinsically wet substrate such as dentine, also increased their

attraction of water (Nishitani et al., 2007). The more hydrophilic the polymer, the

greater is also the likelihood of formation of micro-cavities of different sizes in

the polymeric network (Van Landingham et al., 1999). Many in vivo and in vitro

studies have shown that resin-dentine interfaces become much weaker over

time (Hashimoto et al., 2003). Sauro and collaborators (Sauro et al., 2007)

showed that continued water flow under simulated pulpal pressure increased

convective fluid movement through polymerised resins. It was also

demonstrated that the higher is the dentine permeability, the lower is the tensile

bond strengths of simplified adhesives. The presence of hydroxyl, carboxyl and

phosphate groups in monomers and their resultant polymers make them more

hydrophilic and, as a result, more prone to water sorption. In the manners now

being exemplified, when water sorption is sufficiently high, macromolecular

polymer chains undergo a relaxation process as they swell to absorb the water.

Most of the unreacted methacrylate groups trapped in the polymer network

should not be released into aqueous environments, because they are still part

of dimethacrylate molecules that have reacted and therefore are covalently

bonded to the main polymer chain. Despite this, significant amounts of

unreacted monomer or small chain polymer are released to the surrounding

environment at a rate that is controlled by the swelling and relaxation capacities

of the polymer (Santerre et al., 2001).

46

A number of studies have shown that elution for resin-based materials ranged

from 0.05% to 2.0% of the weight of the specimen into aqueous media, with

elution into alcohol and other organic solvents being higher in most cases (2-

6%) (Ferracane, 1994, Hume and Gerzina, 1996, Pelka, 1999, Munksgaard et

al., 2000, Tanaka et al., 1991). It has been demonstrated that the movement of

water from hydrated dentine may cause the formation of water filled channels

within the polymer matrices of contemporary hydrophilic dentine adhesives (Tay

et al., 2004b). More hydrophilic polymer networks permit a faster release of

unreacted monomers through nanovoids in the material (Brazel and Peppas,

1999). Accordingly, these water filled channels may accelerate elution of

unreacted monomers from polymerised resins (Ito et al., 2005), as well as

further the progress of weakening of the polymers by plasticisation (Wang and

Spencer, 2003).

This phenomenon decreases the stiffness of the polymers (Ito et al., 2005),

produces stresses on the interface with the cavity wall and reduces bond

strengths (Carrilho et al., 2005b).

Water sorption/solubility investigations of hydrophilic adhesives in common use

demonstrated that these systems have much higher water sorption than the

more hydrophobic BisGMA/TEGDMA resins employed to seal multi-step

adhesives (Ito et al., 2005). The hybrid layer created by simplified adhesives,

containing high percentages of hydrophilic monomers, resulted in the formation

of a porous interface (Wang and Spencer, 2003). This interface behaved as a

permeable membrane (Tay et al., 2002a) that allowed water sorption, polymer

swelling, resin hydrolysis and elution of unreacted monomers (Malacarne et al.,

2006). When 3-step etch-and-rinse and 2-step self-etch adhesives were

47

challenged with thermomechanical loading between 5 and 55°C and up to 100

000 cycles, their microtensile bond strengths fell 25-30%. Conversely, the

microtensile bond strengths of 1-step self-etch adhesives fell 50-80% after

thermomechanical loading (Frankenberger et al., 2005). When dentine,

respectively bonded with 3-step etch-and-rinse, 2-step etch-and-rinse, 2-step

self-etch and 1-step self-etch adhesives, was directly exposed to water using

miniature specimens that accelerate water sorption, the microtensile bond

strengths of 3-step etch-and-rinse and 2-step self-etch adhesives did not lessen

remarkably after one year of direct water storage. In contrast, the bonding

effectiveness values of the 2-step etch-and-rinse and 1-step self-etch adhesives

were reduced to almost zero after the same period of direct water exposure (De

Munck et al., 2006). Clearly, the more hydrophilic the resins, the more water the

polymers absorb, the more the polymers become plasticised and the more they

lose their mechanical properties. Thus, water plasticisation of resins contributes

to a reduction in resin-dentine bond strength durability.

1.4 Mechanisms responsible for loss of mechanical stability

Despite successful immediate bonding, the longevity of resin-bonded

restorations remains questionable due to physical (occlusal forces, expansion

and contraction stresses related to temperature changes) and chemical factors

challenging the adhesive interface (Breschi et al., 2008). Today, the most

difficult task in adhesive dentistry is to make the adhesive-tooth interface more

resistant against ageing, thereby rendering the restorative treatment more

predictable in terms of clinical performance in the long term. Despite the

enormous advances made in adhesive technology during the last 50 years, the

48

bonded interface itself remains the weakest area of composite restorations and

none of the current adhesives or techniques is able to produce an interface that

is absolutely resistant to degradation (Breschi et al., 2008). The degradation of

the adhesive interface, which may occur in a relatively short term, depends on

the way the adhesive has been manipulated, on the actual adhesive approach

and on the adhesive composition.

Hydrolysis of interface components, such as dentinal collagen and resin, due to

water sorption, potentially enhanced by enzymatic degradation, and subsequent

elution of the break-down products are the major factors thought to destabilise

the adhesive-dentine bond (De Munck et al., 2009).

1.4.1 Hydrolytic degradation of dental adhesive resins

Dental polymer networks have been shown to be susceptible to hygroscopic

and hydrolytic effects to varying extents dependent upon their chemistry and

structure (Ferracane, 2006).

In the evolution of dentine adhesives, manufacturers have incorporated

increasing concentrations of hydrophilic and ionic monomers to make these

adhesives more compatible for bonding to intrinsically moist, acid-etched

dentine (Van Landuyt et al., 2007).

Increasing the hydrophilic nature of the adhesive-dentine interface has several

disadvantages (Tay and Pashley, 2003a) and affects the integrity and durability

of the adhesive/dentine interfacial bond (Spencer et al., 2010).

Hydrophilic and ionic resin monomers are vulnerable to hydrolysis, due to the

presence of ester linkages, typical of all methacrylates (Ferracane, 2006).

49

These ester linkages are theoretically susceptible to several esterases in body

fluids (Soderholm et al., 1984).

Adhesive hydrophilicity, water sorption, and subsequent hydrolytic degradation

have been considered as highly correlative, because hydrolytic degradation

occurs only in the presence of water (Carrilho et al., 2005a).

Several studies have established a direct relationship between the presence of

hydrophilic and acidic resin monomers in adhesive blends with decreased

longevity of resin-dentine bonds (Peumans et al., 2005), owing to the fact that

resin composition and hydrophilicity expedite water sorption in hydrophilic

resins (Malacarne et al., 2006).

Even the inclusion of small amounts of water may culminate in nano-phase

separation of the adhesive components in the form of nanoscopic worm-like

structures between the polymerised hydrophilic and hydrophobic resin phases

(Ye et al., 2009b). Nano-phase separation reduces the dynamic mechanical

properties of the polymerised adhesives (Park et al., 2010) and increases their

susceptibility to esterase-catalysed hydrolysis (Kostoryz et al., 2009).

Esterases known to activate ester hydrolysis include salivary esterase,

cholesterol esterase, pseudocholinesterase, porcine liver esterase, and

acetylcholinesterase. In contrast to HEMA, Bis-GMA has greater susceptibility

to hydrolysis by cholesterol esterase and acetylcholinesterase. Biodegradation

of HEMA/Bis-GMA adhesives in the presence of either enzyme appear to be

more clinically relevant, since they simulate salivary enzyme activity (Yourtee et

al., 2001).

Previous work has shown that human saliva contains sufficient esterase activity

to attack resin composites (Lin et al., 2005).

50

Nevertheless, it is not known whether there are similar esterases in dentinal

fluid and how they could reach resin-dentine interfaces.

Hydrolysis of methacrylate ester bonds caused either by the increase in acidity

of monomer components (Aida et al., 2009) or by salivary esterases (Shokati et

al., 2010) can break covalent bonds between the polymers by the addition of

water to the ester bonds.

Apart from water, the interfibrillar spaces in acid-etched dentine also include

highly hydrated negatively charged proteoglycans that constitute a hydrogel

within that space (Scott and Thomlinson, 1998). If these hydrogels continue to

be hydrated in interfibrillar spaces, they may be responsible for “molecular

sieving” of larger dimethacrylates like BisGMA, allowing only smaller molecules

such as HEMA to infiltrate the base of the hybrid layers. Since HEMA forms a

linear polymer that does not cross-link, HEMA-rich regions of hybrid layers may

undergo large strains during function that prompt further degradation and

compromise the longevity of resin-dentine bonds (Liu et al., 2011c).

1.4.2 Endogenous collagenolytic activity

Collagen serves as a structural barrier between tissues, and thus collagen

catabolism (collagenolysis) is required to be a tightly regulated process in

normal physiology. The turnover of connective tissue and degradation of nearly

all extracellular matrix components has been ascribed to different members of

the matrix metalloproteinase (MMP) family, due to their ability to catalyse the

hydrolysis of type I collagen triple helical structure. MMPs are a group of zinc-

and calcium-dependent enzymes operating in homeostatic and reparative

51

processes, but unregulated catalysis by these extracellular proteinases leads to

the pathological destruction of the tissues to which they are bound.

In soft tissues, these collagenases are either secreted in a latent form or

inhibited by tissue inhibitors or metalloproteinases (TIMPs). In mineralised

tissues, these enzymes may be active, secreted in a latent form or inhibited by

TIMPs as well as being incorporated by apatite crystallites that fossilise them

and enable their activity.

It has been mentioned that resin-dentine bonding could be considered a unique

form of tissue engineering in which dentists utilise the natural collagen fibril

matrix of demineralised dentine, which is continuous with the underlying

mineralised matrix, as a scaffold for resin infiltration. The collagen fibrils of the

hybrid layer, by being anchored into the underlying mineralised matrix, provide

micromechanical retention of adhesive resins that, in turn, retain resin

composites. The only continuity between adhesively retained restorations and

the hybrid layer are the resin tags in the tubules, along with the nanometre-wide

resin extensions that pass around and between collagen fibrils.

Nevertheless, unprotected type I collagen fibrils situated at the bottom of the

hybrid layer are subjected to deterioration over time due to the activation of

endogenous collagenolytic enzymes (Mazzoni et al., 2006).

Several studies reported that mineralised dentine contains in fact bound MMPs

such as MMP-2, -3, -8, -9 and -20 (Toledano et al., 2010). Even though the

quantitative analysis of different MMPs in dentine remains to be completed, the

currently available data indicate that MMP- 2 may be the prevalent MMPs in

human dentine matrix (Mazzoni et al., 2007). Although classified as a gelatinase

52

(gelatinase A), MMP- 2 is also an effective collagenase (Aimes and Quigley,

1995).

These host-derived proteases contribute to the breakdown of collagen matrices

in the pathogenesis of dentinal caries (Chaussain-Miller et al., 2006) and

periodontal disease (Hannas et al., 2007). In addition, non-collagen-bound

MMPs are also present in saliva (Sulkala et al., 2001), in dentinal tubules, and,

presumably, in dentinal fluid (Boushell et al., 2008).

Proof of degenerative modifications in hybrid layers was offered by De Munck

and collaborators (2003) with long-term in vitro TEM studies that indicated loss

of staining and loss of cross-banded collagen after 4-5 years of water storage

(De Munck et al., 2003). The degradation was irregular and variable but also

extensive. The high resolution provided by TEM examination suggested that

collagen had been converted into gelatin. That is, the hybrid layer was not

empty but still contained organic material not pigmented with heavy metal stains

which are typically taken up by native cross-banded collagen fibrils (García-

Godoy et al., 2007).

When normal hybrid layers receive tensile stressing, the collagen fibrils share

the stress with the resin network by being loaded in parallel. Subsequent to

cleavage of collagen and its conversion to weaker gelatin (i.e. loss of cross-

banded collagen), the stresses applied to the weakened hybrid layer are carried

only by the stiffest surviving material. In this way the resin meshworks pull out of

the "gelatinised" hybrid layer, producing lower bond strengths (De Munck et al.,

2003).

To demonstrate the degradation of dentine matrices by endogenous MMPs,

Pashley and collaborators (2004) acid-etched disks of dentine with 37%

53

phosphoric acid for 15 s, then placed them in buffered calcium- and phosphate-

containing media with or without four protease inhibitors, normally utilised in

biochemistry to prevent MMPs during collagen extraction and purification

(Pashley et al., 2004). Since MMPs are technically hydrolases, that is to say

they catalyse specific peptide bonds in presence of water, half of the etched

specimens were incubated in mineral oil. Specimens were removed and

processed for TEM observation of the quality of the collagen after 24h, 90 days

and 250 days. The naked collagen fibrils had degraded down to the mineralised

base after the period of incubation in the absence of protease inhibitors. By

contrast, in specimens incubated in the presence of protease inhibitors, the

collagen fibrils appeared normal. Similarly, specimens incubated in oil looked

normal over the 250 days, as in the absence of water MMPs could not cleave

collagen.

Mazzoni and collaborators (2006) reported that when etch-and-rinse systems

were applied on dentine their intrinsic acidity (i.e. pHs between 2.6 and 4.7) was

enough to demineralise dentine but not to denature the collagenases. Hence,

the pH of the adhesives was sufficient to expose and set in motion dentinal

MMPs, initiating autolytic phenomena that ultimately affected the hybrid layer

(Mazzoni et al., 2006).

Such results were consistent with a previous study showing that exposure of

MMPs to an acidic pH (c. pH 4.5) activates MMPs in carious dentine

(Tjäderhane et al., 1998).

Furthermore, when normal mineralised human dentine powder was mixed with

different self-etch adhesives with pHs between 1.5 and 2.7, the gelatinolytic and

collagenolytic activity of dentine increased more than 10-fold (Nishitani et al.,

54

2006). Following application of self-etching primers, increases of collagenolytic

activity were also reported for root canal dentine shavings produced during

rotary instrumentation with Gates-Glidden burs (Tay et al., 2006).

This body of increasing evidence indicates that endogenous MMPs are

uncovered and/or activated by many, if not all, dentine bonding procedures.

It was also suggested that mildly acidic resin monomers can activate MMPs by

inhibiting TIMPs (Ishiguro et al., 1994) in TIMP-MMP complexes, thereby

producing active MMPs (Tjäderhane et al., 1998, Sulkala et al., 2001).

Alternatively, acidic resin monomers may set in motion latent forms of MMPs

(pro-MMPs) via the cysteine-switch mechanism that uncovers the catalytic

domain of these enzymes that were blocked by propeptides (Tallant et al.,

2010).

Cysteine cathepsins are papain-like endopeptidases having a vital role in

mammalian cellular turnover, e.g. bone remodelling and resorption.

Most of these peptidases become activated at the low pH found in lysosomes.

Thus, their activities occur almost entirely within those organelles, playing a part

in intracellular proteolysis within the lysosomal compartments of living cells

(Dickinson, 2002).

However, they also exist as exopeptidases and participate in extracellular

matrix degradation through the breakdown of type I collagen and proteoglycans

(Obermajer et al., 2008). For example, cathepsin K, highly expressed in type I

collagen degradation, works extracellularly after secretion by osteoclasts during

bone homeostasis.

The different members of this family of proteases are distinguished by their

structure, catalytic mechanism, and which proteins they cleave. Cathepsins B,

55

L, and S cleave the non-helical telopeptide extensions of collagen molecules,

while cathepsin K cleaves the collagen molecules along their triple helix region

(Liu et al., 2011c).

Unlike the collagenolytic MMPs (MMP-1, -2, -8, and -13) that cleave type I

collagen into a ¾ N-terminal fragment and ¼ C-terminal fragment at a single

site within the triple helix (between amino acids 775 and 776 from the first GXY

triplet of the triple helix domain), cathepsin K cleaves collagen molecules at

multiple sites within the triple helix, thereby giving rise to fragments of various

sizes (Garnero et al., 1998).

Tersariol et al. reported for the first time the presence of cysteine cathepsins in

dentine demonstrating their expression by mature human odontoblasts

(Tersariol et al., 2010). However, these collagen-degrading enzymes are

thought to be more abundant (approximately 10-fold) in carious dentine (Liu et

al., 2011c).

Like MMPs, cysteine cathepsins may be activated in mildly acidic environments.

Acid activation of dentine-bound cathepsins may also coincide with the

conversion of matrix-bound MMPs into their reactive form. On top of that,

glycosaminoglycans (GAGs) can promote further conversion of the latent forms

of the cathepsin enzyme family into their mature forms at neutral pH (Obermajer

et al., 2008). Consequently, GAG-cathepsin activation allows active cathepsins

to be functional even in neutral pH environments.

The existence of cysteine cathepsins in dentinal tubules (Tersariol et al., 2010)

indicates that they are derived from the dental pulp via the dentinal fluid and

may be activated by mildly acidic resin monomers. They may subsequently

56

interact with GAGs and assist salivary MMPs in the degradation of incompletely

infiltrated collagen fibrils within the hybrid layer.

1.5 Adhesion testing

Several aspects should be considered when testing the strength and durability

of the bond to dentine. These include the heterogeneity of its structure and

composition, the features of the dentinal surface exposed after cavity

preparation, and the characteristics of the adhesive itself, such as its strategy of

interaction and basic physicochemical properties. Laboratory experiments

conducted on dental adhesives can be classified into two types, namely

behavioural tests and structural integrity tests. In the behavioural tests the focus

is on understanding how the material behaves and how one might be able to

change the properties of the material by changing such things as its

composition. These experiments are not designed to assess the clinical

performance of the material used. Examples of the sorts of things one might

measure are tensile/shear bond-strength, thus enabling bond strength to be

measured as a material property, elastic modulus, fracture toughness,

coefficient of thermal expansion and translucency. However, all sorts of

chemical and mechanical challenges that are inherent to the oral environment

should also be taken into account, such as moisture, masticatory stresses,

changes in temperature and pH, and dietary and chewing related habits (Mjör

and Gordan, 2002). Structural integrity tests aim to provide an experimental

arrangement that mimic the performance of the material during function. In

other words, the material is being applied in a situation in an attempt to provide

some insight into how the material might respond to a clinical environment and

57

to learn what makes the structure fail. This will be a complex interaction

between material, design and environment. Thus, the structural integrity test is

seeking to establish a link between the material and its performance in a clinical

situation. Typical examples of such tests are represented by fatigue tests.

Besides static bond-strength tests, theoretically clinically more relevant is in fact

to test adhesive interfaces dynamically, as in the clinical situation tooth-

composite bonds are seldom subjected to acute tensile/shear stresses. It is,

however, exposed to cyclic sub-critical loadings produced during chewing (De

Munck et al., 2005). Although fatigue tests are more labour intensive and time-

consuming than static bond-strength tests, a steadily growing, but still only low

number of fatigue tests have been tried out throughout recent years with regard

to their potential to predict clinical effectiveness. In the literature, six different

fatigue tests have been reported on, as there are, chronologically: (i) a macro-

push-out fatigue test (Frankenberger et al., 1999); (ii) a macro-shear fatigue test

(Erickson et al., 2009); (iii) a micro-rotary fatigue test (Van Meerbeek et al.,

2003); (iv) a micro-shear fatigue test (Braem, 2007); (v) a micro-4-point-bend

fatigue test (Staninec et al., 2008); and (vi) a micro-tensile fatigue test (Poitevin

et al., 2010). Despite the alleged need for more fatigue testing of adhesives and

even though several typical fatigue phenomena can be observed, little new

information on bonding effectiveness is provided than that revealed by the

easier and faster static bond-strength tests (Van Meerbeek et al., 2010). For

example, micro-rotary as well as micro-tensile fatigue testing revealed a similar

superior bonding effectiveness of the 3-step ‘gold-standard’ etch-and-rinse

adhesive OptiBond FL (Kerr, West Collins Orange, CA) over the 2-step ‘gold-

standard’ self-etch adhesive Clearfil SE Bond (Kuraray, Tokyo, Japan), that in

58

turn bonds significantly better than the 1-step adhesive G-Bond (GC, Tokyo,

Japan). In addition, these fatigue tests have largely been applied to dentine with

bonding to enamel being much more difficult to assess in fatigue (Van

Meerbeek et al., 2010).

The longevity of the bond upon ageing of the specimens is another aspect of

the performance of dental adhesives that requires particular attention. Several

studies highlighted very good instantaneous and short-term bonding

effectiveness either to enamel or dentine (Inoue et al., 2001), but durability and

stability of the resin-dentine bonded interfaces created by current adhesive

systems still remain unconvincing (De Munck et al., 2005). This shifted the

focus of researchers’ investigations to the evaluation of ageing mechanisms.

Accordingly, besides determining ‘immediate’ bond strength values, measuring

the ‘aged’ bond strength was decisive in order to estimate the clinical

effectiveness of this type of material (Breschi et al., 2008).

In vivo studies are ideally suited to assess both the performance and the

longevity of restorative materials (Hebling et al., 2005, Carrilho et al., 2007b),

but their feasibility is complicated or even precluded by the associated

bureaucratic requirements, they also require much more time to collect

significant information and a higher cost is involved in the procedure (Reinke et

al., 2012). Laboratory studies, on the other hand, offer the advantages of lower

costs, shorter duration, greater standardisation due to the possibility of isolation

of variables and have been widely used to predict the performance and

longevity of adhesive materials (De Munck et al., 2005, Van Noort, 1994,

Amaral et al., 2007).

59

Most of the knowledge we have about the longevity of dentine bonds are based

on in vitro studies, in which some kind of ‘ageing’ factor is added to the

investigation design (De Munck et al., 2005). This could range from examining

the effects of long-term storage in water, or some more aggressive solutions

(Lee et al., 1994, Yamauti et al., 2003, De Munck et al., 2007, Toledano et al.,

2006) along with the use of pH (Peris et al., 2007, Passalini et al., 2010),

thermal (Price et al., 2003, Nikaido et al., 2002, Bedran-de-Castro et al., 2004,

Lodovici et al., 2009), and mechanical loading cycling (Bedran-de-Castro et al.,

2004, Lodovici et al., 2009, Li et al., 2002, Osorio et al., 2005) as well as their

combinations (Grande et al., 2005, Bedran-de-Castro et al., 2004, Lodovici et

al., 2009) in order to recreate some of the challenges that these restorations are

prone to under clinical service for prolonged periods of time.

The immersion of micro-specimens in water is a well-validated method to

assess resin-dentine bond strength durability (De Munck et al., 2006). It usually

requires 6 months to detect drops on the μTBS values (De Munck et al., 2005),

but this period of time may be even shorter when daily water exchange is

performed (Skovron et al., 2010).

Doing so, it was reported that all classes of adhesives exhibited mechanical and

morphological evidence of degradation that resembled in vivo ageing (Shono et

al., 1999). Other water-storage studies confirmed that immediate resin-dentine

bond strength values do not always correlate with long term bond stability since

deterioration throughout the dentine bonded interface occurs at a fast pace

(Carrilho et al., 2005b, Garcia-Godoy et al., 2010, Hashimoto et al., 2010a).

The introduction of pH, thermal, and mechanical loading cycling are attempts to

simulate clinically relevant conditions; however, they still lack standardisation in

60

the number of cycles, temperature, dwell time, immersion time, load and load

frequency and this may hinder comparison of study results and lead to

contradictory findings (Amaral et al., 2007, Reinke et al., 2012).

Recently, an in situ model has been used for the evaluation of ageing

mechanisms involved in the degradation of resin-dentine bonded interfaces

created with two simplified etch-and-rinse adhesives [Adper Single Bond 2

(3MESPE, St. Paul, MN, USA) and Optibond Solo Plus (Kerr, Danburry, CT,

USA)] under more realistic conditions (Reinke et al., 2012).

Compared to the immediate results, where no restorations were included in the

intra-oral appliances used by volunteers and no ageing method was performed,

rapid deterioration in resin-dentine bond strength were observed after the 14-

day simulated cariogenic challenge accountable for a more intense and rapid

degradation rate of the collagen.

However, the findings of the present investigation could not be compared to

other durability studies since this was the first one that employed an in situ

model to investigate the degradation of resin-dentine bonds that occurs with

etch-and-rinse adhesives.

1.5.1 Assessment of sealing ability

The seal of a restorative material against the tooth structure, and the quality and

durability of the seal, are major considerations for the longevity of adhesive

composite restorations. Since the longevity of an adhesive composite

restoration is mainly affected by the leakage of oral fluids along the interface

between the restorative material and the tooth substrate (De Almeida et al.,

2003), it is very important to evaluate the capacity of a bonding system to

61

maintain the seal of the tooth-restoration interface. In case of bonding failure, it

would be desirable that the hybrid layer and resin tags remained in the dentine

surface rather than being pulled away, thereby maintaining a surface seal which

will continue to protect the pulp (Griffiths and Watson, 1995).

1.5.1.1 Micro-leakage and micro-permeability

The study of resistance to the diffusion of a substance into a fluid-filled gap or a

defect between filling material and tooth structure, has been of great concern in

restorative dentistry. The ability of a bonding system to maintain the tooth-

restoration interface seal can be evaluated using high-resolution leakage

studies on the μm scale (micro-leakage) or, alternatively, by micro-permeability

studies (Griffiths et al., 1998).

Micro-leakage is a typical example of structural integrity test, often assessed in

vitro on cross-sections using a tracer or dye that is able to infiltrate the

composite-tooth interface, such as silver nitrate. Silver nitrate has been

employed to assess the porosity of the hybrid layer because it is very soluble (a

50% aqueous solution is usually applied), the silver ion is very small, and, as

soon as it has diffused into a region and has been reduced to metallic silver, it

remains at that site and cannot diffuse away or fade as is frequent with water-

soluble dyes.

Micro-permeability tests also make use of fluorescent dyes, that are ‘loaded’ to

the pulp chamber in order to investigate at higher resolution the sealing ability of

adhesives at the interface itself. Employing confocal laser scanning microscopy,

fluorescent dyes have been shown to penetrate through dentine towards the

62

interfacial region thus indicating the intimacy of the hybrid layer to the bonded

interface (Sauro et al., 2009a).

In this method, the fluorophore presence can show with clarity the existence of

possible routes for micro-permeability either around resin tags, through the

porous base of the hybrid layer, and along the interface between the hybrid

layer and the adhesive agent.

The term 'micro-permeability' was first used by Sidhu and Watson (Sidhu and

Watson, 1998, Sauro et al., 2009b) in the evaluation of the interfacial

characteristics of resin-modified glass ionomers. Water movement from the pulp

chamber towards the resin-bonded dentine interface through dentinal tubules

was identified by using a solution of rhodamine B deposited in the pulp

chamber. This study revealed much information regarding porosity of the

interfacial bonded-layer, particularly in samples seemingly free of interfacial

gaps.

The extent of permeability is contingent on the penetration of the adhesive

components into etched dentine and on the development of gaps or porosities

in the bonded interface resulting from polymerisation shrinkage of the primer,

adhesive, or resin components.

1.5.1.2 Nano-leakage

In principle, resin hybridisation of dentine should protect hermetically the

collagen fibrils against the subsequent exogenous and endogenous

denaturation challenges, in a manner that is analogous to the protective

function of the apatite phases in mineralised dentine. This would ensure the

63

absence of spaces between the collagen fibrils and resins, making the

hybridised dentine long-lasting and resistant to degradation.

However Griffiths et al., using a control adhesive system that retained a

modified smear-layer, observed that fluorescent dye could penetrate not only

the porosities within the bonded interface but also the smear-layer itself

(Griffiths et al., 1999).

The term 'nano-leakage' was introduced by Sano et al. (1995) to distinguish

these findings from micro-leakage phenomena and indicate possible pathways

for permeability through the hybrid zone (Sano et al., 1995a).

Nano-leakage occurs in the absence of visibly discernible, 10-20 μm wide gaps

between restorative materials and cavity walls, through nanometre-sized

spaces of approximately 0.02 μm at the bottom of the hybrid layer, where resin

monomers interface with decalcified dentine.

These sites of incomplete interfibrillar resin infiltration were detected for the first

time by Sano et al. in hybrid layers created with both etch-and-rinse and self-

etch adhesives when these interfaces were immersed in an acidic (pH 4.5)

silver nitrate solution (Sano et al., 1995b). The silver nitrate tracer was reduced

into silver granules that were deposited as reticular patterns (so-called 'water

trees') within the interfibrillar spaces of the hybrid layer, which were then

observed using the back-scattered mode of a SEM or in thin sections prepared

for TEM. These were considered sites of incomplete water removal and

subsequent suboptimally polymerised resins. If the resin had perfectly filled all

the empty spaces between the hybrid layer and the underlying demineralised

dentine, there should have not been room available for silver ion penetration.

According to the theory of 'water tree' (Tay et al., 2004a), these reticular

64

patterns expand as a result of water movement from the underlying dentine

substrate into the partially polymerised adhesive resin matrix, in which the

polymer chains are not sufficiently cross-linked to resist their displacement by

the bulk free water. Thus water sorption fosters the transformation from the

original isolated silver grains to water-filled channels in the adhesive resin

matrices.

Since nano-leakage occurs in the deepest part of the hybrid layer, and spreads

throughout this structure, it has been identified as a special type of intertubular

dentine permeability within the defective nanometre-wide zones around the

collagen fibrils that were not completely enveloped by the resin. Nano-leakage

has also been observed using confocal light microscopy when a fluorescent dye

such as rhodamine B was used as a tracer (Pioch et al., 2001). Uninfiltrated

interfibrillar spaces are very small to permit bacterial penetration, but are large

enough to serve as a pathway for water movement within the adhesive-dentine

interface, leaving the hybrid layer with a large amount of porosity as a

predictable site of enzymatic and hydrolytic degradation over time (Tay and

Pashley, 2003b).

When a 50% ammoniacal silver nitrate tracer solution was employed instead of

regular aqueous solutions, two different modes of nano-leakage expression

were also seen. The first mode was made up of reticular interfibrillar deposits,

already characterised using the traditional silver nitrate solution, while the

second mode consisted of isolated spotted silver grains (Tay et al., 2002a).

Nano-leakage patterns can change over time. Tay and Pashley (2003)

hypothesised that water sorption and, afterward, hydrolytic degradation may be

evidenced by the variations in uptake of ammoniacal silver nitrate within resin-

65

dentine interfaces (Tay and Pashley, 2003b). Some of these variations are the

initial reduction of the reticular silver tracer patterns within the hybrid layer, and

the increment in size and density of the isolated spotted silver grains within the

adhesive. In fact, these represent sites where the cationic diamine silver ion

complexes interact with the anionic functional groups of the hydrophilic resin

monomers, and probably where subsequent water sorption within the polymer

matrix occurs via hydrogen bonding.

Interconnecting silver-filled channels have been reported within the adhesive

layers of 2-step etch-and-rinse and 1-step self-etch adhesives (Tay and

Pashley, 2003b). Water trees were predominantly located along, and

perpendicular to, the surface of the hybrid layer, extending into the overlying

adhesive layers. In their mildest forms, they seemingly expanded from the

surface of the bonded dentine into the adhesive layer. In their most severe

forms, commonly seen with 1-step self-etch adhesives, more than 50% of the

adhesive layer was filled with heavy silver deposits.

1.5.2 Bond strength measurement

While adhesive-enamel/dentine interfacial characterisation with electron

microscopy (possibly supplemented by chemical interfacial analysis) certainly

discloses a deeper insight into the underlying mechanisms of adhesion, the

actual bonding effectiveness of today’s adhesive approaches should be

determined using a mechanical bond-strength test. The rationale behind bond

strength measurements is that the stronger the adhesion, the better the material

will endure any stress imposed by resin polymerisation and oral function. By

definition, the ideal bond-strength test should be simple and reasonably fast. In

66

general, advantages of ‘laboratory testing’ are, among others, the relative

simplicity of the test methodologies commonly used, the relatively quick

collection of data on a specific parameter/property, to be able to test

simultaneously many experimental groups within one study set-up, to be able to

directly compare the performance of a new and/or experimental

material/technique with that of the current ‘gold-standard’, and the possibility to

measure one specific parameter, while keeping all other variables constant. The

final objective of a laboratory test should obviously be to gather data in

prediction of the eventual clinical outcome. In order to measure the bonding

effectiveness of adhesives, diverse methodologies can today be used (Burke et

al., 2008). The bond strength can be measured statically using a macro- or

micro-test set-up, basically depending upon the size of the bond area.

1.5.2.1 Macro-bond strength test

The macro-bond strength, with a bond area larger than 1 mm2, can be

measured using either ‘tensile’ or ‘push-out’ protocols and in ‘shear’ manner.

The macro-tensile bond-strength approach is the less popular and can be used

for instance to measure the bond strength of cements to hard materials such as

ceramics and metal alloys (Abreu et al., 2009, Kern et al., 2009). A push-out

approach has also been employed, in particular to dynamically test the fatigue

resistance of adhesive-dentine bonds (Drummond et al., 1996, Zicari et al.,

2008). It has however never been adopted as a universal bond-strength test

method, most likely because of the more laborious specimen preparation

involved as well as the more time-consuming methodology. This method

appeared however very useful to test the retention of posts luted in root canals

67

(Goracci et al., 2004, Zicari et al., 2008). Of all the adhesive tests used in

dentistry, the shear bond strength test has been one of the most popular

bonding experiments ever devised (Burke et al., 2008); it was found to have

been used in 26% of scientific papers reporting on bond strength (Van

Meerbeek et al., 2010). Its popularity has much to do with the simplicity with

which this experiment can be conducted as no further specimen processing is

required after the bonding procedure. When this experimental design was first

introduced the quality of the dentine bonding agents was quite poor relative to

the bonding agents available today. In those initial experiments the bond

strength was so poor that failure would generally occur at the adhesive interface

such that the bond strength, reported as a shear stress based on the ratio of

load to bonding area, could be used as a reliable measure of the quality of the

adhesion achieved. In this way different adhesives could be compared by

calculating the nominal shear bond strength as long as the experimental design

used was consistent in terms of size, shape and load application. As the quality

of dentine bonding agents improved, fractures no longer occurred at the

adhesive interface but would tend to be cohesive in nature, more often

regarding the dentine, and without an increase in the shear bond strength. The

reason for this is that the stresses generated predispose any occurring crack to

deviate into the dentine when confronted with a strong adhesive bond. In shear

bond strength tests this deviation of the fracture mode from the interface into

the dentine tends to occur at a nominal shear stress of some 20 MPa. This has

led some investigators to mistakenly suppose that the bond strength exceeds

the cohesive strength of the dentine and that the maximum bond strength that

can be achieved to dentine is of the order of 20 MPa, which is not in agreement

68

with the evidence that the ultimate tensile strength of dentine may be as high as

100 MPa (Bouillaguet et al., 2001).

1.5.2.2 Micro-bond strength test

Bond strength is typically measured in tensile by the micro-tensile bond strength

(μTBS) test, according to the procedure developed in 1994 by Sano (Sano et

al., 1994). It has been demonstrated that smaller specimens are stronger and

fail at higher stress than larger bonds, as a consequence of the reduction and

removal of inherent flaws in the system (Pashley et al., 1995b).

Defects in bonded interphases, such as air bubbles, water blisters or regions of

resin-solvent phase separations, act in fact as stress concentrators during bond

testing. These flaws permit initiation of local stresses that exceed the cohesive

strength of one of the components of the bonded interphase (resin composite-

adhesive layer; adhesive layer-top of hybrid layer; infiltrated resin-collagen; or

non infiltrated dentine-mineralised dentine junctions) and result in cracks that

propagate rapidly to cause catastrophic failure.

With smaller bonded specimens, the stress distribution throughout the resin-

dentine interface is more uniform and it is more likely for the bonds to fail

adhesively rather than cohesively in dentine (Sano et al., 1994).

In specimens used for micro-tensile bond strength tests, the bond area tested is

much smaller compared to that of the ‘macro’ tests, being about 1 mm2 or less.

After the bonding procedure, some further specimen processing or the actual

preparation of the micro-specimens is required, rendering the test more

laborious and technique-sensitive. Nevertheless, a long list of advantages is

typically ascribed to μTBS when compared to macro-bond-strength testing, of

69

which the most important are the better economic use of teeth (multiple micro-

specimens obtained from a single tooth enable large study set-ups), the better

control of regional differences as well as substrate variables (e.g. peripheral

versus central dentine) and the better stress distribution at the true interface

(Sano et al., 1994). In fact, all measurements are taken in the central part of the

specimen, well away from the clamping site, such that a uniform, uniaxial stress

is generated and homogenously distributed in the cross-sectional bonded area

(Soares et al., 2008). The maximum tensile stress is then calculated simply

dividing the load by the cross-sectional area. In other words, the microtensile

bond strength is calculated as the tensile load at failure divided by the cross-

sectional area of the bonded interface.

As the micro-tensile bond strength test may sample heterogeneous regions of

dentine, it is recommended that the multiple values for any tooth be averaged to

provide a mean and standard deviation around that mean.

The potential ability of the micro-tensile bond strength to calculate the average

tensile stress at the adhesive interface is a significant difference with the shear

bond strength test. This means that it is then possible to assess the quality of

the adhesion by comparing the tensile load at failure for different bonding

agents and evaluate the mode of failure. Another notable difference between

the micro-tensile test and the shear test is that the cohesive failure in tooth

substrate or composite occurs less frequently (Van Meerbeek et al., 2010). In

addition, given that the failure in shear occurs at lower stresses than failure in

tension, the 20 MPa ceiling, observed for the shear bond strength test, ceases

to exist and tensile bond strengths in excess of 40 MPa can be achieved. Thus,

a micro-tensile protocol appears to be able to discriminate adhesives better on

70

their fundamental behavioural characteristics and bonding performances than a

traditional shear bond-strength approach. This is the most likely reason for up to

60% of current scientific papers reporting on bond strengths having used the

μTBS approach (Van Meerbeek et al., 2010).

Van Meerbeek et al. reviewed the literature with regard to the relation between

bond strength testing and clinical effectiveness of adhesives in terms of

retention rates of Class-V restorations. Significant differences were found in the

‘pooled’ mean bond strength, as the weighted bond strength means of individual

adhesives ranged from about 12 MPa (for Absolute, Dentsply-Sankin, Tokyo,

Japan) to 49 MPa (for OptiBond FL, Kerr, West Collins Orange, CA, USA) and

the weighted bond-strength means per adhesive class ranged from 31 MPa for

3-step etch-and-rinse adhesives, to 29 MPa for 2-step self-etch adhesives, 26

MPa for 2-step etch-and-rinse adhesives, and 20 MPa for 1-step self-etch

adhesives (Van Meerbeek et al., 2010).

The weighted mean and large confidence interval for the (2-step) glass-

ionomers was less reliable, since only a few products were able to be included.

Glass-ionomer adhesives perform as well as the two-step self-etch adhesives,

yet the bond strength of glass-ionomers is scarcely tested, which could be due

to the well-known fact that during the test they tend to fail cohesively within the

material itself, rather than de-bonding from the tooth surface, so that the actual

bond strength to tooth tissue can hardly be determined. The poorer mechanical

properties of glass-ionomers also explain the lower scores achieved in bond-

strength tests when compared to those of resin-based adhesives.

Hence, 3-step etch-and-rinse adhesives bond more strongly to dentine than all

other adhesives that use simplified application procedures, even if some 2-step

71

self-etch adhesives may come close to the bonding effectiveness of etch-and-

rinse adhesives.

Also in order to generate as many specimens as possible from a single tooth, a

micro-shear bond-strength test (μSBS) was introduced in 2002 (Shimada et al.,

2002). This test combines the ease of manipulation with the ability to test

several specimens per tooth. The very fine composite build up (cylinder) with a

typical diameter of 0.7 mm, in combination with a relatively thick adhesive layer,

may however result in considerable bending and variable and non-uniform

loading conditions. This non-uniform stress distribution is probably even more

pronounced as compared to macro-shear bond testing. Furthermore, it is

impossible to confine the adhesive to the area tested, as required by ISO

Technical Specification No. 11405 (ISO/TS, 2003). Basically due to these major

shortcomings, the μSBS test has not been adopted very often, since only 7% of

recent bond-strength studies have used this protocol (Van Meerbeek et al.,

2010). In a recent study comparing both micro-bond methodologies, it was

shown that the micro-shear values were about 1/3 of the micro-tensile values,

while no difference in failure analysis was observed (Yildirim et al., 2008).

1.6 Classification of contemporary bonding systems

A good classification of adhesives is indispensable for maintaining an overview

of the current field. It has been mentioned how the main bonding mechanism of

current bonding systems can be regarded as an exchange process involving

substitution of inorganic tooth material by resin monomers which, with in situ

polymerisation, become micro-mechanically interlocked in the micro-porosities

created. Diffusion is the principal way to obtain such micro-mechanical

72

retention. Recently, more evidence has corroborated the potentially important

role of additional chemical interactions at the biomaterial-tooth interface,

especially with regard to bond stability (Yoshida et al., 2004b). Accordingly,

modern bonding strategies can be divided into (1) an etch-and-rinse [or total-

etch (Kanca, 1992)], (2) a self-etch (or etch&dry), and (3) nowadays also a self-

adhesive approach (Sarr et al., 2010). The strength of this classification lies in

its simplicity and its scientific basis. Each category is characterised by a specific

bonding mechanism, a specific and distinct application protocol and by a

specific interfacial ultrastructure as best imaged using TEM.

1.6.1 Etch-and-rinse

The multi-step etch-and-rinse approach requires a phosphoric acid conditioning

step that at enamel creates deep etch-pits in the HAp-rich substrate, and at

dentine demineralises up to a depth of 3-5 μm to expose a HAp-deprived

collagen mesh (Peumans et al., 2005). The next step includes either two

separate primer and adhesive resin steps, according to a 3-step procedure, or a

single priming step consisting in the application/curing of a combined

primer/adhesive resin, in accordance with a simplified 2-step procedure. In

other words, the latter approach combines the priming and the bonding steps

into one; these adhesives are frequently referred to as 'one-bottle adhesives'

and misleadingly suggest a single application step. The final purpose is the

micro-mechanical interlocking upon diffusion and in situ polymerisation of

monomers into the enamel etch-pits, the opened dentinal tubules and the

mainly organic substance remaining at acid-etched dentine. Ideally, resin tags

should bond to the tubule walls strongly enough to exceed the cohesive

73

strength of their resin components. This not only effectively seals the restoration

margins in the long term, but also safeguards the more vulnerable bond to

dentine against degradation (Peumans et al., 2005). On the contrary, etching

dentine is a rather aggressive procedure as it dissolves and removes (through

rinsing) the natural protection of collagen in preparation for creating a resin-

collagen complex that is susceptible of degradation. True chemical adhesion

between collagen and the methacrylate monomers is unlikely, because of the

inert nature of collagen fibrils and the low affinity of the monomers for HAp-

depleted collagen (Van Meerbeek et al., 1998). Consequently, the rather poor

adaptation of resin to the collagen fibrils leaves the nanometre-sized gaps

accountable for water sorption and degenerative processes (De Munck et al.,

2003). This is regarded as the major shortcoming of today’s etch-and-rinse

approach. Nevertheless, traditional 3-step etch-and-rinse adhesives are still

today considered as ‘gold-standard’ among adhesives, in spite of the rather

elaborate and lengthy working procedure. In fact, after ageing procedures in

durability studies, the bonding integrity of these adhesives is better maintained

(De Munck et al., 2003). For this reason, they are usually employed as control

in order to compare the performance of new-generation adhesives in many of

today’s bond-strength studies. Likewise, the clinical durability of 3-step etch-

and-rinse adhesives confirms their generally superior laboratory results (Van

Meerbeek et al., 2010). It has been extensively reported that 2-step etch-and-

rinse adhesives performed clinically less favourably than conventional 3-step

etch-and-rinse adhesives (Peumans et al., 2005). Laboratory studies have

corroborated these results, ascribing their poorer performance to their higher

hydrophilicity and reduced hybridisation potential. In fact, primed dentine is

74

covered with a layer of non-solvated hydrophobic adhesives in the final phase

of 3-step etch-and-rinse approach and the combination behaves as if it were

hydrophobic (Brackett et al., 2005, King et al., 2005). This measure reduces the

incorporation of water within the interface when the adhesive is applied to the

wet dentinal substrate, and at the same time minimises the susceptibility of the

bonded interfaces to water sorption in the long run. Conversely 2-step etch-and-

rinse formulations are somewhat hydrophilic (Tay and Pashley, 2003a). That is,

they contain too much HEMA and other hydrophilic monomers that create

hydrophilic copolymers. The hydrophilic polymers absorb too much water which

causes swelling of resins that are plasticised by water uptake, thereby reducing

their mechanical properties. Besides, 2-step etch-and-rinse formulations are

associated with greater technique sensitivity than their three steps counterparts,

which is understandable as a single solution combines the two separate

functions of primer and bonding resin.

The primer solvent within etch-and-rinse adhesives is a major factor affecting

the handling and performance properties of these materials. Water-based

adhesives are believed to be the most forgiving regarding application errors, but

the water content in the resultant interface jeopardises the durability. Acetone-

based adhesives, on the other hand, have water-free formulations, but require

the challenging ‘wet-bonding‘ technique (Tay et al., 1996).

Ethanol-based adhesives are thought to be an acceptable compromise with

regard to user-handling and performance (Van Meerbeek et al., 2003). It is

noteworthy that, irrespective of the number of application steps, acetone-based

etch-and-rinse adhesives have generally performed less satisfactorily than their

water/ethanol-based alternatives (Van Meerbeek et al., 2010). The use of

75

acetone-based primers presents monomer diffusion problems. One benefit of

acetone is that adhesive monomers are very soluble in this solvent. However,

during bonding procedures, the acetone present in the first layer of monomers

that is placed on water-saturated dentine mixes with the water (acetone and

water being very miscible), this blocks monomer penetration and makes it come

out of solution before it has a chance to diffuse far into the interfibrillar spaces

(Nakabayashi and Pashley, 1998).

With each subsequent application of primer, the acetone may re-dissolve the

monomer and permit it to diffuse further into the demineralised dentine. This

high technique-sensitivity of acetone-based adhesives must be the reason for

their compromised long term clinical data. One of the more recent

developments is the use in a 2-step etch-and-rinse adhesive of tert-butanol

solvent. This tertiary alcohol has similar vapor pressure to ethanol, but also a

better ability for chemical reaction with monomers. Tert-butanol-based XP Bond

(Dentsply DeTrey GmbH, Konstanz, Germany) has shown good micro-tensile

bond strength data, as well as performed well when tested following a

conventional shear bond strength (Manhart and Trumm, 2010).

1.6.2 Self-etch

Self-etching primers contain acidic monomers that only dissolve the smear-

layer, but do not remove the dissolved calcium phosphates, as there is no rinse

phase but only air-drying. The self-etch approach can be further subdivided into

a ‘strong’ (pH<1), an ‘intermediately strong’ (pH≈1.5), a ‘mild’ (pH≈2), and an

‘ultra-mild’ (pH≥2.5) self-etch approach depending on the etching

aggressiveness or demineralisation intensity (Van Meerbeek et al., 2011).

76

Consequently, the morphological features of the hybrid layer produced by self-

etch adhesives depend a great deal on the aggressiveness of the functional

monomers (Van Meerbeek et al., 2003). TEM images of 'strong' self-etch

adhesives applied on dentine strongly resemble the morphological aspect of an

etch-and-rinse adhesive with a thick hybrid layer, which is completely devoid of

HAp crystals, and with resin tags. The more aggressive the self-etching, the

deeper the hybridisation, the more calcium phosphates are dissolved (exposing

collagen) and embedded within the interfacial transition zone (Koshiro et al.,

2004). Such resin-encapsulated calcium phosphates within the exposed

collagen mesh are, however, not very hydrolytically stable and rather soluble,

thereby seriously jeopardising the bond longevity. This fact may account for the

significantly worse laboratory and clinical bonding efficiency, especially to

dentine, offered by 'strong' self-etch adhesives in comparison with the well-

documented and consistently good-performing 3-step etch-and-rinse (Brackett

et al., 2002). ‘Intermediately strong’ self-etch adhesives exhibit morphological

features that lie between the ‘strong’ and ‘mild’ self-etch adhesives. The latter

demineralise dentine only very superficially, giving rise to a rather shallow

hybrid layer with submicron dimensions, while still leaving substantial HAp-

crystals to protect the collagen fibrils (Van Meerbeek et al., 2010). Hence, the

resultant hybrid layer consists of a partially HAp-deprived collagen mesh

infiltrated by resin. The less intense the self-etching, the more bur-smear

occurs. This interferes with the eventual bonding performance (Ermis et al.,

2008, Cardoso et al., 2008);; nonetheless ‘mild’ self-etch adhesives seem to

cope relatively well with the smear-layer, even when produced by various

surface preparation methods such as diamond burs with different grit sizes, air-

77

abrasion and sono-abrasion (Van Meerbeek et al., 2011). The resin must

penetrate through it and engage intact dentine, but how deep through the sound

tissue is unknown. TEM revealed that ‘ultra-mild’ self-etch adhesives, such as

Clearfil S3 Bond (Kuraray, Tokyo, Japan), decalcified the dentine surface only

very superficially, for up to a few hundreds of nanometres, with few collagen

fibrils exposed at the interface (Fukuoka et al., 2011). This particular interfacial

feature has therefore been termed nano-interaction zone. The relationship

between hybrid layer of various thickness and bond strength has been

investigated, and neither the thickness of the hybrid layer, nor the length of the

resin tags seemed to play an important role regarding the bond strength (Inoue

et al., 2001). Similar bond strengths were obtained regardless of hybrid layer

depth (Finger et al., 1994, Yoshiyama et al., 1996) and this may be due to the

fact that resin retention, as measured by conventional bonding tests, is related

to the cohesive strength of adhesive resin that engages the very top of the

hybrid layer. Deeper penetration of resin may not increase the cross-sectional

area of resin engagement of collagen fibrils, although it may improve the

durability of the demineralised collagen. Be that as it may, in spite of the small

hybrid layer and the absence of resin tags (little micro-mechanical retention),

'mild' self-etch adhesives such as Clearfil SE Bond (Kuraray, Tokyo, Japan) can

reach satisfactory results in terms of bond strength to dentine (Inoue et al.,

2001). In two clinical studies, this system was reported to have high retention

rates in non-carious class V cavities after 2 and 5 years (Türkün, 2003,

Peumans et al., 2007). Together with the finding that the thickness of the hybrid

layer and the presence of resin tags do not overly affect the bonding

performance (Inoue et al., 2001), additional chemical interaction between

78

polycarboxyl- and phosphate-based monomers with residual HAp has been

proposed as a plausible explanation for the good performance of some smear-

layer incorporating adhesives (Yoshida et al., 2004a). The carboxylic and

phosphate groups that render these monomers hydrophilic and that function as

proton donors, have been proven to bond ionically with calcium in HAp (Yoshida

et al., 2000). Some functional monomers, like the 10-MDP (10-

methacryloyloxydecyl dihydrogen phosphate), have been shown to interact with

this residual HAp through primary ionic binding (Yoshida et al., 2004a). The

chemical bonding promoted by 10-MDP turned up to be not only more effective,

but also more stable in an aqueous environment than that produced by other

functional monomers like 4-MET (4-methacryloyloxyethyl trimellitic acid) and

phenyl-P (2-methacryloyloxyethyl phenyl phosphoric acid), in this order, as

revealed by AAS (atomic absorption spectroscopy) and XPS (X-ray

photoelectron spectroscopy) (Yoshida et al., 2004a). The hydrolytic stability of

the monomer itself is also important, especially with regard to bond durability.

Whereas micro-mechanical retention is thought to provide resistance to 'acute'

de-bonding stresses, the relevance of additional chemical bonding is suggested

to lie in durability and survival of adhesion (Van Meerbeek et al., 2003). This

two-fold micro-mechanical/chemical bonding mechanism closely resembles that

of glass-ionomers (Coutinho et al., 2006) and gives an explanation for the

actual bonding effectiveness of ‘mild’ self-etch adhesives.

With regard to the actual bonding effectiveness, it is now very clear that the in

vitro and in vivo performance of an adhesive greatly depends on its specific

ingredient composition. Since the ability to bond chemically is monomer-

specific, the interaction with tooth substrate is dependent to a great extent on

79

the kind of acidic functional monomer and eventually the overall composition of

the adhesive. For this reason, not all the self-etching adhesives available today

are equally effective, and there is a certain variation in bonding performance

between these agents. This variation is appreciably larger than that of etch-and-

rinse adhesives, all relying on the use of phosphoric acid and the subsequent

infiltration of monomers. There are two major classes of self-etching adhesives:

self-etching primer adhesives where the primer-treated dentine is covered with

a separate, solvent-free, more hydrophobic adhesive; and self-etching

adhesives that are not covered with more hydrophobic adhesive resins. 2-step

self-etch adhesives are those requiring the application of a different, more

hydrophobic adhesive resin to cover the dried hydrophilic primer. Likewise

conventional 3-step etch-and-rinse adhesives. In these bonding systems

hydrophilic and ionic resin monomers are only contained in the primers. Making

hydrophobic coatings upon hydrophilic resins better seals the interface,

improves the bond durability and the blend results hydrophobic like it happens

for 3-step etch-and-rinse adhesives (Brackett et al., 2005, King et al., 2005). As

a result, these adhesives tend to approach the bonding laboratory effectiveness

and the clinical performances of 3-step etch-and-rinse adhesives in terms of low

annual failure rates (Peumans et al., 2005).

Similar to the combined primer-adhesive resin solution of 2-step etch-and-rinse

adhesives, the most rapid 1-step self-etch formulations (all-in-one) are complex

mixtures of both hydrophilic and hydrophobic components. When applied to

dentine, the resulting homogenised resin films behave as hydrophilic resins and

consistently achieved lower bond strengths, compared with the multi-step self-

etch and etch-and-rinse versions (Inoue et al., 2001). Due to polymerisation

80

shrinkage of resin-based composites, a high configuration factor (C-factor) in

deep class I cavities leads to a certain amount of stress when the material is

bonded (Nikolaenko et al., 2004). Many 1-step self-etch adhesives might not be

able to withstand the resultant high polymerisation shrinkage stress, giving rise

to early failures and postoperative sensitivity (De Munck et al., 2005). The lower

bonding efficiency has been attributed to the fact that hydrophilic phases or

domains extended across the entire adhesive resin layer and allowed water

movement through it (Sauro et al., 2007). Additionally, quantitative fluid

permeability was consistent with the number of water droplets per unit area.

These hydrophilic domains may be visualised as fine, discrete silver grains (i.e.

the spotted mode of nano-leakage) in the adhesive after immersion in

ammoniacal silver nitrate (Tay et al., 2002a). Therefore, accumulating evidence

indicates that the use of solvated mixtures of hydrophilic and hydrophobic

monomers creates polymerised resin coatings permeable to water from the

outside environment as well as from the host dentine. In addition, these agents

are quite sensitive to application mistakes. In particular, the air-drying step

subsequent to their application is critical to minimise the amount of solvent and

water in the adhesive layer as much as possible. It is known that thick adhesive

layers of simplified adhesives decrease bond strengths massively (Zheng et al.,

2001). On the other hand, this air-drying operation should also be performed in

such a way that an adequate amount of monomers is kept at the surface to

provide satisfactory mechanical properties to the adhesive layer. In narrow and

complex cavities, the right balance between too much and not enough air-drying

is difficult to achieve; while the adhesive may still pool in the cavity corners, the

81

amount of monomers on the cavity walls may already be too low (Van Landuyt

et al., 2005).

In light of all the drawbacks attributed to simplified adhesives, conventional 3-

step etch-and-rinse adhesives and ‘mild’ 2-step self-etch adhesives are today

the benchmarks for high-quality adhesion to both enamel and dentine in routine

clinical practice. In spite of the improved ease-of-use and faster application, a

simplified application procedure so far seems to entail a reduced bonding

effectiveness, and the benefits of these adhesives should therefore be trade off

against their major shortcomings.

In particular, when bonding only to enamel, the etch-and-rinse approach is

definitely preferred, indicating that simple micromechanical interaction appears

sufficient to achieve a durable bond to this tissue. Altogether, when bonding to

both enamel and dentine, selective etching of enamel followed by the

application of the 2-step self-etch adhesive to both enamel and dentine

currently appears the best choice.

This is a consequence of the fact that whenever bonding solely to dentine is

required, a mild 2-step self-etch approach is superior, as the additional ionic

binding with residual HAp enhances bond durability (Van Meerbeek et al.,

2010).

In this case, the better durability may also be attributed to partial dissolution of

the apatite minerals which appear to exert a protective effect on collagen

degradation. The latter was corroborated by the minimal involvement of

endogenous MMP-2 and MMP-9 in interfaces bonded by the mild 2-step self-

etch adhesives (De Munck et al., 2010).

82

Nevertheless, mild 2-step self-etch adhesives are not totally immune to bond

degradation since they are hydrophilic and enzymatic hydrolysis of ester bonds

may still occur over time. Failure of water to be completely removed is probably

another reason why bond degradation may occur in this class of adhesives, as

shown by the presence of silver nano-leakage (Liu et al., 2011c).

1.6.3 Self-adhesive

The third adhesive approach is based on the technology of glass-ionomers and

their auto-adhesive capacity. These restorative materials have a specific

composition, containing polyacrylic acid (PAA), alkenoic copolymers, glass-filler

particles and water. Diverse formulas of glass-ionomers are on the market,

varying in their uses.

Conventional glass-ionomer cements (GICs) are two-component systems,

consisting of an ion-leachable flouroaluminosilicate glass powder and an

aqueous solution of polycarboxylic acid. GICs are set by means of an acid-base

neutralisation reaction, with the product of a hydrogel salt acting as a binding

matrix. On mixing of the two-components, hydrogen ions liberated from the

polymeric acid attack the glass, causing the release of metal ions such Al3+,

Ca2+ etc. In turn, these cations cross-link the acid to form an insoluble matrix

which subsequently bonds to the residual silicate structure formed on the glass

surface.

Polymeric materials are being increasingly introduced and used in dentistry as

cements, filling materials, dentine substitutes and treatment of early carious

lesions. Various efforts have therefore been directed also to the combination of

methacrylate technology and conventional glass-ionomer chemistry. When resin

83

components are added to GICs, these materials are referred to as hybrid-GICs,

resin-ionomers or resin-modified glass-ionomer cements (RMGICs) (Tyas and

Burrow, 2004).

The composition of RMGICs is variable but typically consists of vinyl-modified

polyalkenoic acid, water soluble methacrylates such as HEMA, an ion-leachable

glass and water (Aranha et al., 2006, Bertacchini et al., 1999, Nakabayashi and

Takarada, 1992).

Other ‘self-adhesive’ materials are the so-called self-adhesive luting composites

that have been introduced some years ago to adhesively lute indirect ceramic

restorations (Hikita et al., 2007, Radovic et al., 2008, Monticelli et al., 2008).

They are often mistakenly termed as ‘self-etching’, while they interact only very

superficially with dentine without clear signs of demineralisation (Van Meerbeek

et al., 2010).

GICs are the only true ‘self-adhesive’ materials as they can bond dentine

micromechanically, through infiltration of the collagen network which is exposed

by a short PAA pre-treatment, in combination with chemical bonding obtained

by ionic interaction of carboxyl groups from the acid with calcium ions of

remaining HAp crystals (Sidhu and Watson, 1998).

In detail, the weak acidic molecule removes the smear-layer and exposes

collagen fibrils up to a depth of about 0.5-1 µm. The abundant functional

carboxylic groups of PAA ‘grab’ HAp simultaneously at different and remote

sites, before the other glass-ionomer components diffuse and establish a micro-

mechanical bond following the principle of hybridisation. The ability of GICs to

provide a shallow but uniform hybrid layer, along with their additional capability

84

to chemically bond to the dentinal substrate, are considered convenient in terms

of resistance to long-term hydrolytic degradation.

Owing to the mild and partial demineralisation, HAp crystals can be micro-

morphologically distinguished around the collagen fibrils within the hybrid layer.

Typically, a 'gel phase' closely attached to the hybrid layer can be observed with

some glass-ionomers. This amorphous phase on top of the interface has been

reported to represent a salt of calcium polycarboxylate (Yoshida et al., 2001).

The basic difference with the resin based self-etch approach is that glass-

ionomers are self-etching through the use of a relatively high-molecular-weight

(from 8000 to 15,000) polycarboxyl-base polymer. This limits their infiltration

capacity and is the reason why only hybrid layers of little depth are created.

Moreover, because of this high molecular weight, they cannot penetrate

phosphoric-acid-decalcified dentine. In this case the etchant would demineralise

dentine to a depth greater than that penetrated by the adhesive, leaving the

collagen network unshielded by mineral or polymer and thus exposed to oral

fluids. Consequently, such aggressive conditioners should not be used in

conjunction with glass-ionomers (Van Meerbeek et al., 2003).

The chemical bonding and good adhesion to dental tissues is undeniably a

positive aspect of glass ionomers, in addition to the fact that the glass itself may

be used as a fluoride reservoir. Other advantages offered by conventional GICs

are biocompatibility and thermal expansion coefficients matched to tooth

structure. Also, the surface of these cements resists mild acid attack and

staining by certain agents, such as those that can occur in the mouth (Wilson,

1991).

85

Nonetheless, conventional GICs are far from ideal as restorative materials.

Glass ionomers lack the aesthetic benefits of resin - while they can attain a

match to tooth colour, the characteristic translucency of natural tooth structure

has been difficult to achieve with these materials. They offer low wear

resistance when placed on chewing surfaces, low tensile strength and fracture

toughness since they are brittle and tend to fracture relatively easily. In addition,

they are susceptible to attack by moisture in the initial stages of setting period

and this leads to crazing, lack of resistance to abrasion and susceptibility to

fracture under high shear stresses. Into the bargain, conventional GICs have

short working time, are porous and difficult to finish to a smooth surface (Attin et

al., 1996, Bell and Barkmeier, 1994, Croll et al., 1993, de Gee et al., 1998,

Kerby et al., 1997, Leevailoj et al., 1998, Mitchell et al., 1999).

In order to address the latter of these issues, metals have been added to the

cement to reinforce their structure (McLean, 1990); however, this practice is

diminishing in use (Tyas and Burrow, 2004).

The introduction of resin-modified glass-ionomer cements (RMGIC) has

resolved some of the problems inherent with GICs. Like conventional glass-

ionomer cements, RMGICs have a setting reaction including an acid-base

reaction between the ion-leachable glass and the polyalkenoic acid, but also a

photoactivated polymerisation reaction involving unsaturated side-chains on the

modified polyacid takes place. The two networks of polyacid and ionically cross-

linked polyalkenoate chains provides the structural integrity of the cement and

can be cross-linked through pendant methacrylate groups on the polyalkenoate

molecules (Attin et al., 1996, Allen et al., 1999).

86

Compared to conventional GICs, advantages of these materials include a

shortened setting time, early strength development, decreased early moisture

sensitivity and command set. However, RMGICs require many improvements

before they can be considered to be superior restorative material, as they have

been found to leach cytotoxic compounds (Geurtsen, 2000, Geurtsen, 1998).

Several in vitro studies demonstrated that most of the commercial resin-

modified glass-ionomer cements present more intense cytotoxic effects than

conventional glass-ionomer cements (Aranha et al., 2006). The high cytotoxicity

of resin-modified glass-ionomer cements is probably caused by leachable resin

components, such as HEMA, which has frequently been added to their chemical

composition. Leached residual monomer can easily diffuse through the dentinal

tubules due to its hydrophilic property and low molecular weight, and reach

dental pulp cells (Bouillaguet et al., 1996, Gerzina and Hume, 1996, Hamid et

al., 1998, Kan et al., 1997, Souza et al., 2006).

All light-cured systems suffer from the limited depth penetration of visible light.

Hence, layering techniques are required, despite the fact of being time-

consuming.

Another significant disadvantage of resin ionomer is the hydrophilic nature of

poly- hydroxyethyl methacrylate, which results in increased water absorption

and subsequent plasticity and hygroscopic expansion (Pashley et al., 1998, Yap

and Lee, 1997).

87

Chapter 2: Strategies for preventing resin-

dentine bond degradation

88

2.1 Introduction

The ultimate goal in the design and development of dental adhesives is to

render a stronger and more durable adhesion to hard dental tissues - despite

the severe conditions in the oral environment. Regarding the different

mechanisms of degradation, corresponding strategies to preserve the intact

hybrid layers have been proposed and practiced in vitro and in vivo. They are

as follows: (i) increasing the degree of conversion and esterase resistance of

hydrophilic adhesive (Marchesi et al., 2010); (ii) the use of broad-spectrum

inhibitors of collagenolytic enzymes, including novel inhibitor functional groups

grafted to methacrylate resins monomers to produce anti-MMP adhesives

(Breschi et al., 2010a); (iii) the use of cross-linking agents for improving the

resistance of uncross-linked or mildly cross-linked collagen matrices to

degradation by MMP and cathepsins (Castellan et al., 2010b); (iv) ethanol wet-

bonding to completely replace water from the extrafibrillar and intrafibrillar

collagen compartments, increase resin uptake and produce better sealing of the

collagen matrix, using hydrophobic monomers that absorb much less water over

time and lead to more durable bonds because of improved resistance to

hydrolytic attack (Sauro et al., 2010); and (v) progressive water replacement

from the resin-sparse regions of the hybrid layer with hierarchical deposition of

intrafibrillar and extrafibrillar apatite crystallites to exclude exogenous

collagenolytic enzymes and fossilise endogenous collagenolytic enzymes (Liu

et al., 2011a).

89

2.1.1 Improvement of degree of conversion and esterase resistance

Resin degradation is directly related to the absorption of water, given that

hydrolytic attack on ester linkages by increased acidity/basicity of resin

components or by salivary esterases can both take place in presence of water.

Human saliva incorporates sufficient amounts of pseudocholinesterase and

cholesterol esterase, which operate synergistically to degrade dimethacrylates

(Finer et al., 2004). Beside, the use of hydrophobic photoinitiators such as

camphorquinone for hydrophilic adhesives in the presence of water does not

lead to an optimal polymerisation of these adhesives (Ye et al., 2009a).

These challenges provided the rationale for the development of water-

compatible and esterase-resistant dentine adhesives (Spencer et al., 2010).

It has been suggested that the degree of conversion of hydrophilic adhesive

components could be enhanced using hydrophilic photoinitiators and

compatible accelerators, as well as experimental bulky/branched esterase-

resistant hydrophilic urethane-modified resin monomers (Hayakawa et al., 2005,

Ye et al., 2009a, Park et al., 2010, Park et al., 2008).

Other groups have also employed water-soluble photoinitiators (Cadenaro et

al., 2010, Ikemura et al., 2009) to improve the polymerisation of hydrophilic

adhesives within the confines of water-rich dentine substrates. Concomitant

cross-linking of polar functional groups on methacrylate side-chains may

augment the hydrophobic character of the hybrid layer after preliminary

infiltration of the hydrophilic resin monomers into the partially or completely

demineralised collagen matrix. Increasing the degree of conversion of mono-

and dimethacrylate resins, with the associated reduction in the number of

unreacted pendant functional groups, may also diminish the susceptibility to

90

esterase hydrolysis of the polymerised adhesive layer and resins that infiltrated

the hybrid layer.

The application of these new photoinitiators, accelerators, and cross-linking

resin monomers is likely to intensify the dynamic mechanical properties of the

resin-dentine interface immediately after polymerisation, as well as decreasing

the heterogeneity and nanophase separation of the polymer matrices in the

presence of water (Park et al., 2010).

These strategies, however, will not result in a quantum increase in the ability of

adhesive resin monomers to infiltrate the demineralised collagen matrix entirely,

particularly deep inside the aqueous compartments of the collagen fibrils (Liu et

al., 2011c).

2.1.2 Inhibition of enzyme-catalysed hydrolytic cleavage of collagen

Several authors have attempted to determine the benefits of employing

synthetic MMP inhibitors during bonding procedures (Liu et al., 2011c) in

relation to the dentine collagenolytic and gelatinolytic processes, responsible for

the degradation of collagen fibrils within incompletely resin-infiltrated hybrid

layers (Zhang and Kern, 2009) and the loss of quasi-static mechanical

properties of the collagen matrix (Tezvergil-Mutluay et al., 2010a).

The quest for developing specific MMP inhibitors, capable of potently and

selectively inhibiting or blocking the uncontrolled activity of individual MMPs, is

a nearly three-decade endeavour and only few selective and effective drugs

with the desired properties have emerged (Li and Wu, 2010). A number of

rationally designed MMP inhibitors have shown some promise in the treatment

of pathological conditions in which MMPs are suspected to be involved, such as

91

cancer, arthritis, and other diseases associated with tissue remodelling. Most of

these inhibitors, however, have performed poorly in clinical trials. The reasons

for this failure include shortcomings in the chemistry of these compounds,

namely, broad MMP sub-type selectivity, toxicity and metabolic lability. With

respect to the selection of appropriate non-specific inhibitors for dentine matrix-

bound MMPs, the structural homology of the catalytic domains among members

of the MMP family represented a favourable characteristic because the

application of such inhibitors to acid-etched dentine was analogous to a topical

application method and involved nanogram quantities, making systemic toxicity

a less critical issue, albeit still important.

Chlorhexidine (CHX), a widely used antimicrobial agent, has been successfully

employed to suppress the activities of MMP-2, MMP-8, and MMP-9 (Gendron et

al., 1999). Several investigators have used this cationic bis-biguanide as a non-

specific MMP-inhibitor, demonstrating CHX-related improvement in terms of

bond strength preservation and hybrid layer stability, revealed as a lower

interfacial nano-leakage expression compared to control specimens after

different periods of in vivo ageing. In split-mouth design experiments (Carrilho et

al., 2007b, Hebling et al., 2005), teeth of a quadrant were treated with 2% CHX

between acid-etching and bonding procedures; control teeth on the other side

had no CHX treatment. After 6 to 14 months, control hybrid layers revealed

extensive loss of cross-banded collagen with the formation of voids in the hybrid

layer, while the CHX-treated teeth had normal hybrid layers without any signs of

degradation. Fourteen-month results of microtensile bond strength revealed that

the mean (±standard deviation) bond strength had decreased from 29.3±9.0 to

19.0±5.2 MPas in the control group (Carrilho et al., 2007b).

92

In contrast, the experimental CHX-treated teeth showed a mean 1-day bond

strength of 32.7±7.6 MPa, while the 14-month value was 32.2±7.2 MPa. Thus,

there was virtually no loss of bond strength in the CHX-treated specimens. In a

fluorescein-labelled functional assay, collagenolytic activity of mineralised

dentine powder was suppressed for the 73% by protease inhibitors, and almost

100% by 0.2% concentrations of CHX (Moszner et al., 2005). Another in vitro

study confirmed the protective role of 0.2% CHX after 1 year of storage of

bonded specimens in artificial saliva, suggesting that lower concentration of

CHX can be equally effective compared to 2% concentrations (Breschi et al.,

2009).

One of the advantages of using CHX is represented by its substantivity since it

is able to bind to mineralised dentine for at least 12 weeks (Mohammadi and

Abbott, 2009). Additionally, it has been shown that demineralised dentine can

bind more CHX than mineralised dentine and that it may remain attached to

demineralised dentine after bonding (Kim et al., 2010d). The same study proved

that CHX is not de-bound by HEMA and this may be the reason for the long-

term effectiveness of CHX as a MMP inhibitor in resin dentine bonds. Relatively

large amounts of CHX remained bound to partially and completely

demineralised dentine incubated in phosphate-buffered saline for at least 8

weeks, and no de-binding occurred after the first half-hour of incubation

(Carrilho et al., 2010). Since the binding mechanism is electrostatic in nature

and is reversible (Blackburn et al., 2007), CHX molecules could be eventually

displaced by competing cations derived from dentinal fluid or saliva and leach

out of the denuded collagen matrix. There is a prevailing notion that its binding

to demineralised dentine simply postpones instead of permanently arresting

93

bond degradation (Mai et al., 2009). This may account for the discovery that

bonds made to CHX pre-treated acid-etched dentine with commercial adhesives

and water-wet bonding techniques were conserved after 9 months but not after

18 months, with severe hybrid layer degradation at the 18th month observation

(Sadek et al., 2010a).

Such a concern provided the rationale for chemically grafting CHX to resin

monomers to produce CHX-methacrylates (Luthra and Sandhu, 2005) and

including these resin monomers with anti-MMP potentials into dentine

adhesives.

The introduction of quaternary ammonium methacrylate resin monomers with

assorted anti-MMP properties (Liu et al., 2011c) has recently attracted the

attention of investigators. A study employing one of the best-known

commercially available quaternary ammonium methacrylates, 12-

methacryloyloxydodecylpyridinium bromide (MDPB), confirmed that the

experimental antibacterial adhesive systems employing MDPB-containing

primer or/and bonding-resin could produce an effective bond under in vivo

conditions (Imazato et al., 2007). This could have accounted, in hindsight, for

the in vitro and in vivo observations that resin-dentine bonds degraded after one

year when Clearfil SE Bond (Kuraray Medical Inc., Tokyo, Japan) was used as

the self-etching primer, while bonds created in the same study with the MDPB-

containing self-etching primer Clearfil Protect Bond (Kuraray) were well

preserved after one year, with more pronounced water treeing observed in the

former adhesive under both ageing conditions (Donmez et al., 2005).

These resin monomers were included into dentine adhesives to avoid the

displacement of electrostatically bound CHX and other potential non-resin-

94

conjugated MMP inhibitors from denuded collagen matrices of the hybrid layers

during ageing. Copolymerisation of CHX-methacrylates or quaternary

ammonium methacrylates with other dimethacrylate resin monomers made

CHX unable to de-bind and leach out from incompletely resin-infiltrated hybrid

layers. With respect to self-etch adhesives, chlorhexidine has been included

directly into primers (De Munck et al., 2009) (Zhou et al., 2010).

However, there were critical restrictions when CHX was directly incorporated

into polymerisable resin monomer formulations. Although incorporation of 1-2%

CHX directly into resin blends did not have any significant effect on their degree

of conversion, such an approach adversely influenced the mechanical

properties of the polymerised resins. It was demonstrated that the addition of

even 1% CHX to a variety of resin blends with different hydrophilicity reduced

the modulus of elasticity (i.e., stiffness) of the polymerised resins by 27-48%

(Cadenaro et al., 2009a). This decreasing effect continued over time as CHX

release from the resins was related to water-induced swelling (Hiraishi et al.,

2008). In addition, release of CHX from a polymer matrix was pH-dependent,

with more being released at lower pH values (Anusavice et al., 2006). More

important still, CHX proved to be beneficial only for etch-and-rinse adhesives,

as it could not bind to the collagen matrix in the presence of an acidic

environment (Curtis and Watson, 2008).

Another non-specific MMP inhibitor, GM 6001 (galardin), is often used as a

reference inhibitor in generic MMP assay kits. Breschi et al. aimed to determine

the effect of a synthetic MMPs inhibitor (galardin) used as an experimental

primer on acid-etched dentine prior to the application of an etch-and-rinse

adhesive (Breschi et al., 2010a). The inhibitory effect of galardin on dentinal

95

MMPs was confirmed by zymographic analysis, as complete inhibition of both

MMP-2 and -9 was observed. Interfacial nano-leakage expression after ageing

revealed reduced silver deposits in galardin-treated specimens compared to

controls. Furthermore, the use of galardin had no effect on immediate bond

strength, while it significantly decreased bond degradation after 1 year. Others

have shown that polyvinylphosphonic acid (PVPA) (Tezvergil-Mutluay et al.,

2010b) and benzalkonium chloride (BAC) (Tezvergil-Mutluay et al., 2011) also

possess generic anti-MMP activities and could be used as an alternative to

CHX to prevent collagen degradation within hybrid layers.

Chemically modified tetracyclines (CMTs) (i.e., tetracyclines that lack antibiotic

activities but retain their anti-MMP activities) are effective non-specific MMP

inhibitors and have been used as MMP inhibitors in experimental caries

(Sulkala et al., 2001). This study suggested that MMPs could have an important

role in dentine caries pathogenesis, and that MMP inhibitors could be useful in

the prevention of caries progression. However, CMTs have not been used to

prevent the degradation of hybrid layers, since they may stain teeth with a

purple hue after photo-oxidation of the tetracycline.

Much work has also been done on designing cathepsin K inhibitors (Teno and

Masuya, 2010). Selective inhibitors of cathepsin K could be in fact promising

therapeutic agents for the treatment of diseases characterised by excessive

bone loss, such as osteoporosis. Representative inhibitors have demonstrated

antiresorptive activity both in vitro and in vivo and therefore are promising leads

for therapeutic agents (Thompson et al., 1997). Expansion of these inhibitor

concepts can be envisioned for the many other cysteine proteases implicated

for therapeutic intervention. However, little is known about whether these drugs

96

that are targeted at inhibiting specific cathepsins are useful for direct application

to acid-etched dentine or incorporation into self-etch adhesives for preventing

the degradation of type I dentine collagen scaffolds. In contrast with this, CHX

has been reported to be effective against cysteine proteinases produced by

Porphyromonas gingivalis (Houle et al., 2003) and human recombinant cysteine

cathepsins (Scaffa et al., 2012). Thus, it is important to determine if CHX could

be used as a non-specific inhibitor for both MMPs and cysteine cathepsins

associated with intact and carious dentine.

2.1.3 Use of collagen cross-linking agents

As the major component of the dentine organic matrix, fibrillar type I collagen

plays a number of structural roles, such as to provide the tissue with

viscoelasticity, forming a rigid, strong space-filling biomaterial (Cheng et al.,

1996).

Type I collagen is a heterotrimeric molecule composed of two α1 chains and

one α2 chain that is comprised of 3 domains: the NH2-terminal non-triple helical

(N-telopeptide), the central triple helical, and the COOH-terminal non-triple

helical (C-telopeptide) domains (Bedran-Russo et al., 2009).

Dentine collagen fibrils are stabilised by lysyloxidase-mediated covalent inter-

and intramolecular cross-links (Yamauchi and Shiiba, 2002) which increase the

tissue resistance to thermal denaturing and enzymatic degradation (Kuboki and

Mechanic, 1982). In addition, intrinsic collagen cross-links provide the tensile

properties of collagen molecules (Yamauchi, 2000).

On this account several in vitro studies suggested that introducing

supplementary cross-links to acid-demineralised dentine collagen could help to

97

reduce the susceptibility of additionally cross-linked dentine collagen to

enzymatic degradation by collagenases, improve its short-term mechanical

properties and increase the stability of the resin-dentine interface.

The effects of selective collagen cross-linkers, such as glutaraldehyde, genipin,

carbodiimide and proanthrocyanidin, applied during adhesive procedures has

been investigated in several studies over the last few years (Al-Ammar et al.,

2009, Macedo et al., 2009, Bedran-Russo et al., 2009, Bedran-Russo et al.,

2010, Castellan et al., 2010a, Castellan et al., 2010b).

Glutaraldehyde, a synthetic cross-linking agent, is widely used as fixative agent

(Nimni et al., 1998) and has been reported to improve mechanical properties of

various collagen-based tissues (Bedran-Russo et al., 2008, Bedran-Russo et

al., 2007, Charulatha and Rajaram, 2003, Ritter et al., 2001, Yannas, 1992,

Silver, 1994). However, despite its ability to induce cross-links in collagen,

glutaraldehyde is also known for its cytotoxicity (Sung et al., 1999).

Genipin, a naturally occurring cross-linking agent, not only has shown to

improve the mechanical properties of various protein-based biomaterials

(Bedran-Russo et al., 2007, Tsai et al., 2002), but also presented low toxicity

when compared with glutaraldehyde (Frujikawa et al., 1987).

Increased resistance to collagenase challenge and mechanical properties of

collagen-based materials have been reported following treatment with

carbodiimide, a cyanamide isomer, able to assemble amino acids into peptides.

It presents very low cytotoxicity when compared to glutaraldehyde as the urea

derivative, released when the cross-link is generated, is easily rinsed from the

collagen, leaving no residual chemicals (Khor, 1997).

98

Proanthocyanidin, widely present in fruits, vegetables, nuts, seeds and flowers

is a potent antioxidant cross-linking agent with vast biological activities,

including inhibition of MMP-2 and MMP-9 (Matchett et al., 2005). Recently, the

use of a grape seed extract, mainly composed of proanthocyanidin, has been

shown to improve the mechanical properties of demineralised dentine (Bedran-

Russo et al., 2008, Bedran-Russo et al., 2007, Frujikawa et al., 1987).

Since dentine collagen is already highly cross-linked, it is questionable if the

increase in resin-dentine bond longevity can be explained only by augmentation

in cross-linking density. Current literature suggests that this MMP resistance

may be attributed to silencing of MMPs and probably other exogenous collagen

degradation enzymes via conformational changes in the enzyme 3-D structure

(Busenlehner and Armstrong, 2005). Theoretically, this may be achieved via

irreversible changes induced within the catalytic domain or allosteric inhibition of

other modular domains that co-participate in collagen degradation (Sela-

Passwell et al., 2010).

Undeniably, the use of cross-linking agents increases the resistance of uncross-

linked or mildly cross-linked collagen matrices to degradation by bacterial

collagenases (Avila and Navia, 2010, Ma et al., 2010).

Even though there is no evidence that the catalytic domain of collagenolytic

MMPs can be cross-linked to inactivate their functions, oxidative cross-linking of

adjacent tryptophan and glycine residues in the catalytic domain of MMP-7 by

hypochlorous acid, a potent oxidant produced by the myeloperoxidase system

of phagocytes, resulted in inactivation of the catabolic activity of this enzyme

(Fu et al., 2004). These observations indicated that specific structural motifs are

99

important for controlling protein modification by oxidants and suggested that

pericellular oxidant production by phagocytes might limit MMP activity.

The use of cross-linking agents may also play a part in MMP silencing via

allosteric control of non-catalytic domains. Lauer-Fields et al. indicated that

hemopexin-like domains collaborate with catalytic domains in collagen

catabolism by properly aligning the triple helix and coupling conformational

states to facilitate hydrolysis. Therefore, the catalytic domains in collagenolytic

MMPs can cleave non-collagen substrates, but the hemopexin-like domain of

these enzymes is necessary for them to first unwind and then cleave the three

triple-helical fibrillar elements of the collagen molecule in succession (Lauer-

Fields et al., 2009).

The MMP family has developed at least two distinct mechanisms for collagen

unwinding and cleavage. These distinctive mechanisms underly a drastically

different mode of interaction with triple helical fibrillar collagen I, according to

which the MMP domain is involved in binding (Gioia et al., 2007). One

mechanism is characterised by binding (likely through the hemopexin-like

domain) and cleavage of alpha-1 and/or alpha-2 chains without distinguishing

between them and keeping the gross conformation of the triple helix (at least

during the first cleavage step). The other instead involves preferentially binding

of the alpha-1 chains (likely through the fibronectin-like domain, grossly altering

the whole triple helical arrangement of the collagen molecule and cleaving

preferentially the alpha-2 chain.

Regardless of which of the two collagen-binding mechanisms is involved, cross-

linking of either the hemopexin-like or fibronectin-like domains may contribute to

100

inactivation of the associated MMPs and reduction in their collagenolytic

efficacy.

This hypothesis appears to be supported by the results of a study that analysed

the tissue properties of pericardium from young calves and pigs after Cross-

linking with glutaraldehyde or carbodiimide. Cross-linking with glutaraldehyde

completely abolished gelatinase activities, while the use of carbodiimide was

less effective; but, interestingly, a relative reduction of MMP-9 versus MMP-2

was detected (Calero et al., 2002).

Cathepsin K is also allosterically regulated by modifiers, such as sulphated

glycosaminoglycans (GAGs), that bind outside of its active catalytic site

(Novinec et al., 2010).

It has been shown that at physiological plasma pH the enzyme fluctuates

between multiple conformations which are differently susceptible to

macromolecular inhibitors and can be manipulated by varying the ionic strength

of the medium.

Thus, GAGs may act as natural allosteric modifiers of cathepsin K, exploiting

the conformational flexibility of the enzyme to regulate its activity and stability

against autoproteolysis. It is possible that the use of cross-linking agents may

alter the GAG-cathepsin allosteric interaction and “trap” the enzyme in a certain

conformation that inactivates its collagenolytic activity.

Cross-linking may also affect MMP activities that are usually modified by non-

collagenous proteins (Malla et al., 2008).

MMPs, like other proteinases, can undergo autolytic degradation once activated

in vivo. In dentine, MMP activities and resistance to degradation may be

regulated by serum glycoprotein fetuin-A, a member of the cystatin superfamily

101

(Ray et al., 2003), SIBLINGs Bone Sialoprotein and Dentine Matrix Protein-1

(Fedarko et al., 2004).

Taken together, these data suggest that cross-linking of these non-collagenous

proteins may indirectly silence MMPs via inactivation of the functional domains

of these glycoproteins.

The mechanical properties of dentine are a fundamental aspect of restorative

procedures, since dentine constitutes the greatest volume of tooth structure.

Similarly to the utilisation of non-specific inhibitors, the major limitation in the

use of cross-linking agents is that a water-rich, resin-sparse collagen matrix with

poor mechanical properties is retained within the hybrid layer. Demineralised

dentine collagen has a modulus of elasticity of less than 8 MPa. Even if these

demineralised collagen fibrils can be stiffened 50X with cross-linking agents, the

resulting modulus of elasticity (ca. 0.4 GPa) is still far inferior to that of resin-

infiltrated dentine (ca. 3-5 GPa) (Ito et al., 2005, Chiaraputt et al., 2008) and

mineralised dentine (ca. 20 GPa) (Kinney et al., 2003).

Besides, chemical cross-linking of collagen does not alter the intrinsic collagen

molecular stiffness (Liao et al., 2005).

Thus, these very flaccid collagen fibrils are susceptible to creep and subsequent

fatigue rupture after prolonged function (Fung et al., 2009). This accentuated

the need for alternative strategies that enable the dynamic mechanical

properties of denuded collagen matrices to be improved or regained as a

mechanism to prevent the degradation of resin-dentine bonds (Liu et al.,

2011c).

102

2.1.4 Ethanol-wet bonding technique

Two main reasons have been advocated to interpret the relatively poor

infiltration of etch-and-rinse adhesives into water wet-dentine: (1) substantial

discrepancies between molar concentrations of water (55.6 moles/L) and that of

typical dentine adhesives comonomer blends (2.3-3.9 moles/L), making it nearly

impossible for comonomers to expel and replace all of the water from collagen

fibrils (Pashley et al., 2007); (2) presence of hydrophobic dimethacrylates in

dental adhesives with low solubility in water, that can give rise to macro, micro

or nanophase separation when applied on acid-etched water wet-dentine

(Eliades et al., 2001).

After rinsing the acid-etched dentine, the water is replaced with absolute ethyl

alcohol. Being completely miscible with water, and since the volume of ethanol

applied is far in excess of the amount of water in the collagen matrix, ethanol

replaces all the water in the interfibrillar spaces and at the top of the dentinal

tubules.

As a result, ethanol-solvated hydrophobic resin blend consisting of BisGMA and

TEGDMA should penetrate deep into ethanol-saturated collagen without

undergoing phase changes (Sadek et al., 2008), and any residual layer of

ethanol would allow the infiltrating monomers to dissolve and create a more

intimate and resistant association between collagen and resin (Tay et al., 2007).

Whereas etch-and-rinse adhesives applied to water-saturated dentine invariably

resulted in a diffusion gradient of resin infiltration within the collagen matrix,

studies with two-photon laser confocal microscopy (Sauro et al., 2009b) and

micro-Raman spectral analysis (Shin et al., 2009) indicated that a relatively

103

homogeneous distribution of hydrophobic resins within the hybrid layer could be

achieved with ethanol-wet bonding.

Replacement of water by ethanol increased resin uptake of hydrophobic Bis-

GMA/TEGDMA mixtures and produced better sealing of the collagen matrix by

bringing the Hoy solubility parameter for δt of the ethanol-saturated matrix very

close to that of ethanol-solvated bis-GMA/TEGDMA mixtures. The δt value of

water-saturated mixtures is in fact too far away from the δt of Bis-

GMA/TEGDMA, indicating that the solvated comonomers are not miscible with

water-wet bonding (Pashley et al., 2007).

Two variants of the ethanol-wet bonding technique have been described. In the

simplified version, 100% ethanol was applied to water-saturated acid-etched

dentine for 1 min preceding the application of ethanol-solvated hydrophobic

resin comonomer blends (Nishitani et al., 2006, Sauro et al., 2010). The

rationale was to provide a method of application of hydrophobic resin

comonomers to acid-etched dentine within a clinically relevant time frame.

However, this approach proved to be extremely susceptible to operator-specific

handling and did not completely reduce dentine permeability without the use of

adjunctive tubular occlusion agents (Sadek et al., 2007, Cadenaro et al., 2009b,

Sauro et al., 2009c), even after three absolute ethanol applications (Sadek et

al., 2010b).

In the progressive ethanol replacement adaptation of the technique, water was

gradually removed from the collagen matrix via a series of ascending ethanol

concentrations (Sadek et al., 2007, Sadek et al., 2010a). Nevertheless, this

technique version was time-consuming and impractical for clinical application.

104

In both cases, when ethanol replacement is not meticulously performed to

prevent water-saturated collagen from exposure to air, the surface tension

present along the air-collagen interface can easily result in collapse of the

collagen matrix and prevent optimal infiltration of the adhesive monomers

(Osorio et al., 2010).

Also, the technique is sensitive to moisture contamination (Sadek et al., 2007,

Sadek et al., 2008). Since hydrophobic monomers are immiscible with water,

contamination of these resin monomers with as little as 5% of water on the

ethanol-saturated dentine substrate resulted in a 25% reduction in tensile

strength of the experimental hydrophobic adhesive to dentine (Sadek et al.,

2008). As a result, whether experimental ethanol-wet bonding can become a

clinically applicable technique for bonding to deep, vital dentine remains

uncertain. Moreover, ethanol-wet bonding is not suitable for self-etch adhesives,

with being water a prerequisite for the ionisation of their acidic resin monomer

components (Hiraishi et al., 2005).

Ethanol-wet bonding is now considered a philosophy rather than a proper

bonding technique, due to its clinical impracticality. However, it represented a

major contribution to adhesive technology, since the reasoning behind it

revealed the critical barrier to overcome in dentine bonding with contemporary

etch-and-rinse and self-etch adhesives.

Ethanol replacement of water-saturated dentine provided an opportunity for

higher resin uptake and made it possible to obtain hybrid layers characterised

by wider interfibrillar spaces alongside collagen fibrils with reduced fibrillar

diameter (Tay et al., 2007, Hosaka et al., 2009).

105

Given that these intrafibrillar spaces were penetrated by resin during infiltration,

no space was left empty after polymerisation (Liu et al., 2011c). As a matter of

fact, when ethanol-wet bonding was meticulously performed, neither nano-

leakage (Sadek et al., 2008) nor intrafibrillar remineralisation (Kim et al., 2010a)

could be detected.

Such a phenomenon has never been observed in hybrid layers created with

hydrophilic etch-and-rinse adhesives bonded to water-wet dentine, or with

water-containing self-etch adhesives (Liu et al., 2011c). Generally, milder

versions of self-etch adhesives have the ability to infiltrate the interfibrillar

spaces of a partially demineralised collagen matrix via the mechanism of

simultaneous etching and resin infiltration. Nevertheless, intrafibrillar spaces

created by both self-etch and etch-and-rinse adhesives are amenable to

remineralisation by apatite crystallites (Kim et al., 2010b). Such an observation

provided indirect evidence that both classes of commercially available

adhesives are incapable of completely replacing water from the intrafibrillar

collagen compartment, and that resin monomers can entirely fill this

compartment if it is saturated with ethanol, but not if it contains water.

2.1.5 Restoring the mineral phase of the collagen matrix

Sadek et al. proved that bonds made to ethanol-saturated dentine with an

experimental hydrophobic adhesive did not degrade over an 18-month ageing

period with preservation of hybrid layer integrity (Sadek et al., 2010a). This was

an important observation, since it addressed the issue that MMPs are not

capable of collagenolysis in the absence of water as a functional medium.

106

Even if a decreasing gradient of resin monomer diffusion within acid-etched

dentine was not present, with a resultant phase of demineralised collagen

matrix at the base of the hybrid layer (Pashley et al., 2011), and hydrophilic

adhesives could infiltrate to the full extent immobilising collagenolytic enzymes,

resin hydrolysis via esterases and water sorption would subsequently result in

reactivation of those immobilised enzymes.

Encapsulating MMPs and cathepsins with hydrophobic resins represented

another innovative way of immobilising the catalytic and allosteric domains of

these enzymes to inactivate their functional activities, and highlighted the

concept of water replacement from the collagen intrafibrillar compartments as

the ultimate goal in extending the longevity of resin-dentine bonds (Liu et al.,

2011c).

Such a method is conceptually similar to the technique of molecular imprinting

to produce abiotic polymers with enzymatic functions (Takeuchi and Hishiya,

2008), with the exception that the enzyme substrate is not removed to expose

their functional sites.

During collagen mineralisation, bulk water and loosely bound water are

progressively removed from the internal compartments of the collagen fibrils

and replaced by apatite crystallites (Chesnick et al., 2008).

Collagen fibrils that are stabilised by intrafibrillar and interfibrillar apatite

crystallites in mineralised tissues do not degrade over time. Hard tissue fossils

from the late Cretaceous era (65 million years ago) were well preserved at the

microstructural and molecular levels and responded to collagenase digestion

(Avci et al., 2005). Analysis of these data suggested that MMP-bound collagen

107

must have been functionally immobilised by apatite crystallites for the integrity

of those collagen fibrils to be preserved.

Apart from increases in mechanical properties (Kinney et al., 2003, Balooch et

al., 2008), a major role played by the hierarchical deposition of apatite in

mineralised collagen is the exclusion of molecules larger than water (ca. 18 Da)

from the mineral-protein biocomposite (Lees and Page, 1992).

Conversely, when apatites are replaced by water in clinical dentine bonding, the

life span of these resin-dentine bonds has seldom been shown to last for more

than 10 years (Van Dijken et al., 2007). For these reasons, replacement of

water within incompletely resin-infiltrated hybrid layers, accompanied by

inactivation or silencing of collagenolytic enzymes via remineralisation, emerged

as a viable strategy to overcome the critical barriers currently restraining the

longevity of resin-dentine interfaces (Liu et al., 2011c).

The physical exclusion of exogenous collagenolytic enzymes (activated

cathepsin K, 27 kDa; activated MMP-2, 67 kDa; bacterial collagenase, 68-130

kDa; activated MMP-9, 85 kDa) by apatite represents the postulate of the

“enzyme exclusion” mechanism that preserved archeological collagen from

degradation (Nielsen-Marsh et al., 2000). Studies on collagenase hydrolysis of

dentine have corroborated the protective role played by the mineral phase on

collagen degradation (Klont and ten Cate, 1991).

Similar to a host of growth factors and signaling molecules, endogenous MMPs

and cathepsins become “fossilised” (Smith, 2003) but retain inside the

mineralised dentine matrix their biologic characteristics.

These are restored upon removal of the mineral phase, provided that the

demineralisation agent is not strong enough to denature these molecules. As

108

collagen mineralises, free and loosely bound water is progressively replaced by

apatite. This physiologic dehydration mechanism (Chesnick et al., 2008)

guarantees that the internal environment of the mineralised fibril continues to be

relatively dry to protect the integrity of the entrapped bioactive molecules.

2.1.5.1 Guided tissue remineralisation

It has been recently suggested that the evolutional mechanism of molecular

immobilisation of the functional activity of collagenolytic enzymes could be

recapitulated in man-made resin-dentine bonds with a guided tissue

remineralisation strategy. To this end, Tay and coworkers (2011) have

proposed that acid-etched dentine should be treated with a biomimetic

remineralising primer just before bonding. This primer would contain

nanoparticles of amorphous calcium phosphate and other reagents that would

remineralise any water-rich, resin-poor regions in hybrid layers (Kim et al.,

2010c, Kim et al., 2010f, Liu et al., 2011b, Liu et al., 2011c, Ito et al., 2012).

Biomimetic mineralisation (bioremineralisation) is a strategy that utilises

nanotechnology principles to mimic what occurs in biomineralisation (Tay and

Pashley, 2008). This method replaces water from resin-sparse regions of the

hybrid layer with apatite crystallites that are small enough to occupy the

extrafibrillar and intrafibrillar compartments of the collagen matrix (Tay and

Pashley, 2009). By restoring the enzyme exclusion and fossilisation properties

of mineralised dentine, it should be possible to preserve the longevity of resin-

dentine bonds (Kim et al., 2010a).

Enormous progress has been made over the last few decades in understanding

the biomineralisation processes in mammalian tissues such as bone, enamel

109

and dentine. Some knowledge has been acquired about the role of proteins in

achieving a morphologically controlled deposition of mineral as opposed to

precipitation of unstructured agglomerates of crystals.

Biologically mineralised tissues have remarkable hierarchical structures that

have evolved over time in order to achieve great functions in a large variety of

organisms. Mineralised crystals are typically formed in an organic matrix with

precise regulation of synthetic mechanisms through proteins. These proteins

are in dynamic equilibrium with their environment, thus resulting in fluctuations

and tissue remodelling. In consequence, several conditions are required to

restore lost mineral constituents in demineralised dentine.

The internal consistency of the collagen structure is the primary requisite.

Organic phases play in fact a key role in templating the structure of mineralised

tissues; therefore, their matrices should be sound as a scaffold for the mineral

crystals to grow (Kusboki et al., 1977, Gonzalez-Cabezas, 2010).

Second, there should be residual mineral crystals to serve as growth centres,

or, at the very least, there should be newly formed nucleation sites in case of

complete demineralisation (Xu et al., 2010).

Last but not least, mineral sources containing calcium and phosphorous should

be supplied to the lesion (Peters, 2010).

In the biomineralising scheme, there are no apatite seed crystallites present in

the organic scaffold. In other words, this approach cannot rely on the presence

of remnant dentine matrix proteins within a demineralised collagen matrix and is

compelled to mineralise reconstituted collagen that is devoid of mineralisation-

promoting proteins (Kim et al., 2010g).

110

As a result, biomineralisation has to proceed through an unconventional

sequence of chemical reactions that includes homogeneous nucleation. The

latter is not a thermodynamically favourable process and necessitates

alternative kinetically driven protein/polymer-modulated pathways in order to

lower the activation energy barrier for crystal nucleation (Wang and Nancollas,

2009).

For most vertebrates, collagen fibrils alone cannot initiate tissue mineralisation.

These controls, imposed by matrix and soluble proteins leading to the

sequential steps of phase transformation, entail sequestration of the amorphous

mineral phase by poly(anionic) extracellular matrix proteins (ECM). Non-

collagenous ECM derived from the secretory calcium-binding phosphoprotein

family are necessary for regulating bone and dentine mineralisation. They are

also needed for controlling the dimension, order and hierarchy of carbonated

apatite apposition within mineralised hard tissues.

Collagen is an active scaffold in the formation of oriented hydroxyapatite

platelets, with domains of charged amino acids at the border of the gap and

overlap zones acting as nucleation sites. Modelling of collagen fibrils shows that

these nanosized, positively charged regions are used for mineral infiltration as

well as charge-charge attraction. This leads to the deposition of a dense

network of pre-nucleation clusters bound by polyanions within any nanosized

region, and their subsequent transformation into amorphous calcium phosphate

and, finally, oriented crystalline hydroxyapatite inside the fibrils (Barrère et al.,

2006).

Over the last two decades, there has been intensive research into isolating

ECM and examining their role in biomineralisation (He and George, 2004).

111

However, the complexity of biological proteins limits our understanding of the

functional groups. Meanwhile, use of native or recombinant matrix proteins as

nucleating agents for in situ biomineralisation is not yet economically viable in

routine restorative dentistry therapies.

Thus, scientists have resorted to the use of poly(anionic) acid molecules to

mimic the functional domains of naturally occurring mineralisation-promoting

proteins (Liu et al., 2011c). Additional acidic matrix phosphoproteins are

employed as templates to promote crystal nucleation and growth within the

collagen structure (George and Veis, 2008). In nanotechnology terminology,

such pathways are examples of a bottom-up synthesis (Wong et al., 2009),

wherein nanoscale materials are created by a particle-based self-assembly

process (Gower, 2008).

Bioremineralisation of resin-dentine bonds adopted the guided bottom-up

assembly by using two poly(anionic) analogs of acidic matrix proteins to

separately mimic the sequestration and templating functional motifs that are

present in naturally occurring matrix protein molecules (Liu et al., 2011c).

Blocks from different silicate-based materials (as a source of calcium and

hydroxyl ions) were immersed in a biomimetic analog of matrix proteins made of

simulated body fluid containing a polycarboxylic acid such as polyacrylic acid

(Girija et al., 2004, Tay and Pashley, 2008, Kim et al., 2010e) and a

phosphorus-based analog of matrix phosphoproteins such as

polyvinylphosphonic acid (George and Veis, 2008, Gu et al., 2011).

The role of polyacrylic acid is to mimic aspartic acid as well as serine-rich

mineralisation-promoting C-terminal domain (ASARM) of cleaved DMP-1

112

(Gericke et al., 2004) and stabilise amorphous calcium phosphate into fluidic

nanoprecursor particles (Gower, 2008, Nudelman et al., 2010).

This makes possible the inflitration of the microfibrillar spaces by amorphous

calcium phosphate and the formation of initially amorphous, intertwining mineral

strands within the collagen fibrils (Kim et al., 2010g).

Instead, polyvinylphosphonic acid is used as a collagen-binding template to

induce the nucleation and growth of apatite from the initial amorphous mineral

phase (Gajjeraman et al., 2008). In addition, this molecule has been shown to

possess MMP-inhibiting properties that prevented collagen degradation during

remineralisation (Tezvergil-Mutluay et al., 2010b).

The two biomimetic analogs (sequestration and templating) must be utilised in

concert to reproduce the dimension and hierarchy of the apatite crystallites that

are found in natural mineralised dentine. In fact, when only polyaspartic acid

was used as the sequestration analog, infiltration of amorphous calcium

phosphate nanoprecursors into reconstituted collagen fibrils produced

intrafibrillar mineralisation that apparently lacked the hierarchical order of

apatite arrangement in natural mineralised collagen (Deshpande and Beniash,

2008).

Conversely, large extrafibrillar mineral spheres were deposited around the

collagen matrix when only a templating analog was employed (Li and Chang,

2008).

Hybrid layers created by both etch-and-rinse and self-etch adhesives have

shown to be remineralisable with the bottom-up biomimetic approach. For etch-

and-rinse adhesives, Mai and co-workers detected, after one month, apatite

crystallites in both extrafibrillar and intrafibrillar spaces of the denuded collagen

113

matrices, with hybrid layers remineralised to 80-90% of their thickness after 2-4

months (Mai et al., 2010). Apatite deposition was almost exclusively identified

from the intrafibrillar spaces in case of self-etch bonding systems, as

interfibrillar spaces were filled with adhesive resin (Kim et al., 2010b, Kim et al.,

2010c).

Unfortunately, guided bottom-up mineral assembly using biomimetic analogs of

dentine matrix proteins is a slow process, since it involves at least two

kinetically driven pathways and usually takes 3-4 months to complete. In

experimental studies, long times were also required because remineralisation

was performed via a lateral diffusion mechanism by the immersion of specimen

slabs in the medium containing dissolved biomimetic analogs without direct

dentine-material contact (Liu et al., 2011c).

Thus, while the mineral constituents is being restored, the region of the exposed

collagen fibrils is still susceptible to hydrolytic degradation by MMPs.

Furthermore, the water-filled denuded collagen fibrils remain flaccid and exhibit

weak mechanical properties (Bertassoni et al., 2009).

The modulus of elasticity of resin-infiltrated dentine beams increased

hundredfold after biomimetic remineralisation as a result of intrafibrillar

remineralisation of the collagen matrices (Gu et al., 2010). A limitation of that

study, however, was that the resin-infiltrated dentine beams (macro-hybrid

layers) were evaluated en masse by three-point bending. Since remineralisation

does not occur in locations of the collagen matrices that are occupied by resins,

three-point bending is insensitive for evaluating the localised increases in

modulus of elasticity of the remineralised collagen.

114

Although the reincorporation of mineral into the demineralised dentine matrix

does not represent a full recovery of its functionality, it still plays a very

important role, since the remineralised remnant crystallites in the subsurface of

the tissue may be much more resistant to subsequent acid attacks

(Featherstone, 1996).

An experimental calcium-silicate-based hydrophilic composite has been

developed for the sustained release of calcium and hydroxyl ions (Kim et al.,

2010g). The inclusion of reactive calcium-silicate mineral powder as filler in

hydrophylic resin may be an innovative method for the biomimetic

remineralisation of apatite-depleted dentine surfaces and to prevent the

demineralisation of hypomineralised/carious dentine, with potentially great

advantage in clinical applications. However, while this method proved to be

useful for remineralising resin-dentine interfaces with thin adhesive layers,

remineralisation was hampered in presence of thick adhesive layers (Dickens

and Flaim, 2008).

Thus, alternative strategies are being developed to deliver calcium, hydroxyl,

and phosphate ions to the base of the hybrid layers created by etch-and-rinse

adhesives, where remineralisation is most needed.

2.1.5.2 Top-down remineralisation via epitaxial growth

Remineralisation of apatite-depleted, partially demineralised dentine is not new.

Reports on remineralisation of carious dentine appeared in the dental literature

more than half a century ago. Currently, dental literature abounds with reports

on the use of bioactive glass particles (Vollenweider et al., 2007), calcium-

phosphate-based composites (Peters et al., 2010) and fluoride-releasing glass-

115

ionomer cements (Ngo, 2010) for remineralising partially demineralised carious

dentine. Fluoride, which is not a functional motif for biomineralisation, has been

shown to enhance conventional remineralisation over existing seed crystallites

(Liu et al., 2011c).

In nanotechnology terminology, remineralisation techniques currently used in

dentistry represent a top-down approach (Wong et al., 2009). This approach

creates materials using scaled down versions of a bulk material that

incorporates nanoscale details of the original material. Partial demineralisation

of mineralised collagen matrix by acids derived from bacteria or dentine bonding

procedures creates the seed crystallites necessary for this top-down

remineralisation approach. The orientation of those remineralised crystalline

lattices is pre-determined by the lattice of the original seed crystallites.

However, remineralisation does not occur in locations where nuclei for

crystallisation are completely absent, as demonstrated with the use of a

strontium-based glass-ionomer cement (Kim et al., 2010f).

Non-denatured intact collagen is remineralisable as long as seed crystallites are

present as nidi for heterogeneous nucleation of calcium phosphate phases

(Koutsoukos and Nancollas, 1981).

In fact, partially demineralised dentine is considered to have a better capacity to

restore its original mineralised state because it contains remnant mineral

crystals and noncollagenous phosphoproteins that can act as nucleation sites

for remineralisation (Clarkson et al., 1991, Bertassoni et al., 2011).

Detailed electron-microscopic analysis of crystallites in various zones of caries

lesions has also confirmed that remineralisation occurs by growth of existing

crystals (Featherstone, 1996).

116

Moreover, the mineral precipitated may work as a constant site for further

nucleation of mineral ions present in the oral cavity, facilitating continuous

remineralisation over time.

Unlike the biomimetic bottom-up remineralisation approach that takes months of

protracted chemical reactions to complete, top-down mineral fabrication of

apatite crystals proceeds rapidly via epitaxial growth over existing seed

crystallites (Liu et al., 2011c).

Furthermore, traditional remineralisation by epitaxial growth is a

thermodynamically favourable process that overcomes the energy barrier of

homogeneous nucleation (Jiang and Liu, 2004).

Despite their complex hierarchical structures, the basic building blocks of teeth

and bones during both initial and later formation stages are nanosized mineral

particles.

Initial contacts between the organic matrix and mineral nuclei is a key factor for

the control of crystallisation pathways and the organisation of hierarchical

structures of teeth and bones at different length scales (Wang and Nancollas,

2008a, Wang and Nancollas, 2008b).

The earliest heterogeneous nucleation events in the presence of the organic

template and subsequent growth involve various possible precursor phases

(amorphous or crystalline) to the final mineral phase.

Classic crystallisation theory assumes that crystals nucleate and grow from

elementary building blocks (ions, molecules) in a supersaturated solution,

although phase transformations may also occur in the later stages. The

association of solution species to form “metastable intermediate precursors”

117

(Onuma and Ito, 1998), or “growth units”, that subsequently dissolve as the

precipitation reactions proceed, is a crucial initial step (Eanes et al., 1965).

The nature of the primary mineral phase leading up to mature bone and dentine

mineral apatite remains controversial.

Both brushite and octacalcium phosphate have been implicated as possible

precursors to the formation of biogenic hydroxyapatite with trace of carbonate

and fluoride. Moreover, in vivo calcifications have also suggested the

involvement of an initial amorphous calcium phosphate phase (ACP) followed

by transformation to the final HAp product (Mahamid et al., 2008).

According to Posner, mineral apatite derives from calcium phosphate clusters

[Ca9(PO4)6] packing randomly with interfacial water to form ACP precursors

(Posner, 1985). This theory is supported by the presence of several calcium

phosphate growth inhibitors such as magnesium that stabilise the amorphous

state (Barrére et al., 1999, Root, 1990).

Although extensive investigations of calcium phosphate crystallisation have

been performed, many have studied the final structures and morphologies and

have not emphasised the need to consider the molecular contacts between

mineral and matrix that drive nucleation or the thermodynamic and kinetic

controls imposed by matrix and soluble proteins during the nucleation stage.

Current results and concepts of crystal nucleation and growth at the molecular

level, and the role of site-specific interactions in crystallisation, provide possible

mechanisms of calcium phosphate crystallisation that are related to the

mineralisation of teeth and bones (Wang and Nancollas, 2008a, Wang and

Nancollas, 2008b).

118

However, the detailed physical and chemical processes by which nucleation

control is established and the thermodynamic and kinetic parameters that define

those processes remain largely unknown.

These precursor nanoparticle phases grow further in size via ion-by-ion

attachment (Niederberger and Cölfen, 2006) and aggregation, with local loss of

solvent, undergoing amorphous-crystalline transformations or phase

transformations en route to a thermodynamically stable macro-crystal.

According to Ostwald’s rule (Ostwald, 1897), normally the first occurring phase

in polymorphism is the least stable and closest in free energy to the mother-

phase, followed by other phases in order of increasing stability.

The formation, dissolution, and transformation of calcium phosphates depend

on the nature of the calcium phosphate body (particle size, crystallographic

features, density) and the nature of the solution (composition, pH, temperature).

Most calcium phosphates are sparingly soluble in water, and some are very

insoluble, but all dissolve in acids. Their solubility, defined as the amount of

dissolved solute contained in a saturated solution when particles of solute are

continually passing into solution (dissolving) while other particles are returning

to the solid solute phase (growth) at exactly the same rate (Wu and Nancollas,

1998), decreases with the increase in temperature and in pH (de Groot, 1983).

Each calcium phosphate phase possesses its own thermodynamical solubility.

For example, at pH=7 and 37ºC, HAp is the most stable phase (Barrère et al.,

2006).

However, these thermodynamical considerations are under equilibrium

conditions, and therefore they do not take into account kinetics that dictate the

formation of one or the other phase under dynamic conditions.

119

In vivo, the interactions between calcium phosphates and their “biological

surroundings” are highly complex due to the non-equilibrium conditions and due

to the undefined amount of compounds playing a role in these interactions. The

second important factor in the stability of the calcium phosphates is the

characteristics of the solution in which these salts are formed or placed, namely

the solution supersaturation in free calcium and phosphates ions (Tang et al.,

2001).

At a given pH and temperature, a free calcium and phosphate ion containing

solution can be categorised in three different states: (i) the stable

(undersaturated) condition, when crystallisation is impossible; (ii) the metastable

condition, when spontaneous crystallisation of calcium phosphate salt is

improbable, although the concentrations are higher than the ones

corresponding to the salt solubility. If a crystal seed were placed in such a

metastable solution, growth would occur on the seed; (iii) the unstable

condition, when spontaneous crystallisation of calcium phosphate is probable,

but not inevitable (Barrère et al., 2006).

Extracellular fluids that are supersaturated for calcium and phosphate may

induce the nucleation and growth of new calcium phosphate crystals (Barrère et

al., 2006).

Almost all mineralised tissues are highly hierarchical at many different length

scales. At the lowest level they often consist of crystals formed of thin plates of

irregular shapes. Their sizes range in length from 20 Å for the smallest particles,

to 1100 Å for the largest particles (Kim et al., 1995).

120

This size is not arbitrary; rather, it seems to give biominerals such as bone and

tooth remarkable physical characteristics (Wang and Nancollas, 2008a, Wang

and Nancollas, 2008b).

These mineral crystals expose a very large surface area to the extracellular

fluids, which is critically important for the rapid exchange of ions with these

fluids. Minerals start to nucleate into the holes and pores present in the collagen

fibrils (Glimcher, 1987). This heterogeneous nucleation is catalysed by the

presence of phosphated esters groups (Glimcher et al., 1984) and carboxylate

groups (Rhee et al., 2000) present in the collagen fibrils. Subsequently, the

growth, or mineralisation, takes place along the collagen fibrils, eventually

interconnecting all of the collagen fibrils.

There is uncertainty over the specific morphology of the mineral crystallites. The

shape of each inorganic crystal is often related to the intrinsic unit cell structure,

correspondingly the same material can exhibit diverse crystal morphologies due

to different surface energies of the faces and the external growth environment.

Kinney et al. (Kinney et al., 2001) used small-angle X-ray scattering to indirectly

assess the micromorphology of the apatite crystallites in dentine and suggested

that minerals have a rod-like shape near the pulp while they are more plate-like,

with approximately 5 nm thickness, nearer the dentine-enamel junction.

Similarly, Nalla et al. (Nalla et al., 2005) using transmission electron microscopy

confirmed early observations from Boyde (Boyde, 1974) and suggested the

presence of needle-like crystallites in the intertubular dentine region.

Lowenstam and Weiner (Lowenstam and Weiner, 1980), also using TEM,

evaluated the ultrastructure of the crystallites in bone (which has a similar

model of mineralisation) after removal of its organic structures and found that

121

the average length and width of the crystallites was 50 and 25 nm instead, with

an approximate thickness of 2-3 nm, resembling plate-like structures.

The mineral phase is classified as intrafibrillar and extrafibrillar, according to its

location with respect to the collagen fibrils. The former is confined within or

immediately adjacent to the gap zones of collagen fibrils, extending amid

tropocollagen molecules, while the latter lies in the interstitial spaces between

the fibrils. Structurally intact collagen fibrils that retain their banded

characteristics and intermolecular crosslinks are considered to be

physiologically remineralisable (Landis et al., 1993, Landis, 1996).

The concept of remineralisation through the growth of existing apatite crystals is

based on the incorporation of ions (calcium, hydroxyl, and phosphate)

spontaneously from the oral fluid or, alternatively, from external sources with

tailored treatments that deliver the same ions into the surrounding fluids and

elicit a positive response at the remnant crystallites within the subsurface

(Featherstone, 1990, ten Cate and Featherstone, 1991).

The newly formed minerals, in their turn, may act as sites for further nucleation

promoting a continuous remineralisation over time when in presence of

environmental mineral ions. The capability of a material to induce the formation

of apatite on demineralised dentine (remineralisation ability) is strictly related to

the biointeractivity and bioactivity, i.e. the ability to evoke a positive response

from the biological environment.

Bioactive materials have been proven to promote bone formation and to form a

stable bond to bone (Hench, 1997) and prediction of a material's bioactivity can

be made both in vivo and in vitro (Kokubo and Takadama, 2006).

122

The spontaneous formation of HAp on a material surface upon storage in a

simulated body fluid (SBF) is taken as evidence of bioactivity (Piskounova et al.,

2009).

The control of metallic ion release from dentine bonding agents must be

considered as an attractive approach to enhance the biological capability of

adhesives for dental tissue engineering.

A bioactive dentine bonding agent which forms HAp on the surface would have

several benefits including remineralisation of adjacent tooth substance, possible

closure of gaps between material and tooth and potentially better bond strength

over time (less degradation of bond). In general, today’s dental restorative

materials are not considered to be bioactive and no dentine bonding systems

with proven capability to induce dentine remineralisation are available on the

market (Gandolfi et al., 2011b). Currently, only a few attempts to develop

bioactive dental materials have been published based on either calcium

aluminate (ceramics) or via addition of calcium phosphates to resin based

materials (Engqvist et al., 2004, Skrtic et al., 1996).

2.1.5.3 Key objectives in the design of bioactive dentine bonding systems

There is a necessity for restorative materials that bond strongly and durably to

dentine; at the same time, there is the demand for these materials to be

aesthetically pleasing. Resin-based restoratives meet the latter qualification, but

there is much room for refinement in the case of the former. It is possible that

the incorporation of additional materials into the resin-bonding procedure can

produce this enhancement - specifically, incorporating bioactive agents.

Because of the ability of such materials to promote the formation of apatite in

123

aqueous environments that contain calcium and phosphate (e.g. saliva), their

presence at the bonded interface could ameliorate the quality of the resin-

dentine bond through mineral deposition throughout the course of nano-leakage

events.

Restorative compounds capable of filling voids by crystal deposition can be

categorised as self-healing materials and may protect resin-tooth bonds

because they heal nanometre- and micrometre-sized voids before and after

ageing. The ability of restorative materials to fill in voids, making them self-

repairing, has great potential to enable restorations to have longer lifetimes due

to this self-healing ability. In view of the clinical demand on engineered dental

tissue, new biologically active adhesive/primer formulations have to exhibit:

I. light-curable characteristics with controlled water sorption and solubility

behaviour of the cross-linked network in water and oral fluids.

II. A hydrophilic nature to interact with oral fluids and absorb water for the

necessary ion movement.

III. Ability to release calcium, hydroxyl, and phosphate ions in a sustained

manner (bioavailability of remineralising ions).

IV. Potential to completely replace water from the resin-sparse regions of the

hybrid layer with redeposition of thermodynamically stable, apatitic tooth

mineral (bioactivity).

V. Alkalinising activity within the resin-dentine interface to prevent MMPs

activity. Hydroxyl ions released during the hydration reaction may also

create unfavourable conditions for bacterial survival and proliferation.

Antibacterial properties are primarily required at the dentine-restoration

interfacial region. The presence of residual bacteria within dentine further

124

increases the risk of reinfection and secondary caries, in particular when

using dental composites lacking any antimicrobial activity.

These said features thus form the requisites for a successful, advanced ion-

releasing dentine bonding system.

2.2 Development of ion-releasing adhesives comprising bioactive fillers

Many modifications and improvements have been made to resin bonding

products in the years since their inception. Several “generations” of bonding

agents have been developed in the last few decades, but the advancements

have mainly increased ease of use and reduced technique sensitivity. To date,

however, little progress has been made with regard to reducing the propensity

of the adhesive towards nano-leakage, though the achievement of improved

bond strengths has significantly reduced gross gaps from occurring (Pashley et

al., 2011).

One possible modification that may significantly reduce the occurrence and

extent of nano-leakage within the hybrid layers is represented by the

incorporation of biologically active agents into the bonding process; materials

which may create a chemical bond in addition to the micromechanical one, or

potentially precipitate mineral into any open space that remains after

polymerisation of the adhesive.

The development of a bioactive ion-releasing dentine bonding system with

therapeutic ability to remineralise mineral-depleted sites within the bonded-

dentine interface is currently one of the main targets of the dental biomaterial

research (Tay and Pashley, 2008, Liu et al., 2011b, Peters et al., 2010,

Bresciani et al., 2010, Moshaverinia et al., 2011).

125

In the remineralising process the bioavailability of mineral ions from restorative

materials is the basic requirement to enhance the apatite formation and the

mineralisation of the dentinal tissue in the presence of mineral ions (calcium,

phosphate, fluoride) from the oral fluid.

Gaining full understanding of the remineralisation process may be difficult, since

both physical and chemical processes are relevant. The site and the amount of

mineral deposition are probably determined by the physical condition (mineral

distribution profile and transport mechanism) and by the chemical process

(deposition). Notwithstanding this, there is little doubt that significant

remineralisation of dentine lesions can occur under both in vitro and in vivo

conditions.

In vitro (Dickens and Flaim, 2008, Dickens et al., 2003, Ryou et al., 2011) and in

vivo (Peters et al., 2010, Bresciani et al., 2010) studies suggested in fact that

the amount of minerals and the mechanical properties (i.e. hardness and

modules of elasticity) of dental hard tissues may be favourably increased.

In particular, Bertassoni et al. (Bertassoni et al., 2009) proved that a continuous

delivery of calcium (Ca2+) and phosphates (PO4-3) ions may induce

remineralisation and mechanical recovery of mineral-deficient dentine due to a

fine association of the minerals with the organic matrix.

Agents that enhance and/or promote remineralisation of dentine lesions are part

of a new era of dentistry aimed at controlling the

demineralisation/remineralisation cycle, depending upon the microenvironment

around the tooth (Rao and Malhotra, 2011).

Today a variety of agents that aid in remineralisation of dental hard tissues are

available commercially. Remineralising agents can be incorporated into different

126

products and restorative materials could become vehicles for these

remineralising agents as it occurs with pit-and-fissure sealants, dentifrices,

chewing gums, and rinses.

Peters et al. (Peters et al., 2010) demonstrated by means of electron probe

elemental microanalysis (EPMA) techniques an increased mineral content in

caries affected dentine treated with resin-based materials containing calcium

phosphate cements.

Furthermore, a number of studies (Moshaverinia et al., 2011, Ryou et al., 2011,

ten Cate and van Duinen, 1995) have recently demonstrated that mineral

precipitation in residual caries affected dentine can also be encouraged by GICs

or RMGICs.

Conventional GICs offer adhesion to tooth structure and therefore they are not

as prone to leakage as are resin-bonded restorations, when used in appropriate

situations (Masih et al., 2011). They have been used as liner/base materials,

gaining popularity because of beneficial properties such as biocompatibility and

ability to release fluoride, available for the formation of a less soluble

fluorapatite (Wiegand et al., 2007).

Despite the great mass of information on the positive effects of fluoride on

enamel, no data have demonstrated the effectiveness of fluoride ions to induce

new mineralisation of demineralised dentine and no nucleation of new apatite

crystallites within an apatite-free dentine has been identified in the

demineralised dentine immersed in a calcium- and -phosphate containing

remineralisation media in presence of a glass ionomer cement (Kim et al.,

2010g).

127

RMGICs do possess positive qualities, but in this case there is much room for

improvement where leaching of resin components, cytotoxicity and durability

are concerned (Pashley et al., 1998, Yap and Lee, 1997).

Hydroxyapatite itself and some other Ca-containing materials exhibit excellent

biocompatibility manifested in minimal tissue toxicity and foreign-body reaction,

osteoinductivity, and osteogenicity (Hench, 1999, Kokubo, 1990, Kokubo, 1991,

Yuan et al., 2000). The reason for this characteristic may be their ability to

release calcium and phosphate ions, which are critical factors in bone

metabolism.

It has been reported that HAp particles may biomineralise on the surface of

chitosan/gelatin network films through hydrogen bonding between COOH, OH-,

and NH2 groups of the film and OH- group of the HAp crystals (Li et al., 2009).

Therefore, inducing the same interactions between HAp nanoparticles and

collagen fibres of demineralised dentine could be an especially promising

prospect for mineral replenishment of incompletely resin-infiltrated hybrid layers.

Although the remineralisation process would strongly depend on the

concentration of the HAp particles adjacent to the collagen fibres and pH of the

media, Sadat-Shojai et al. suggested that hydroxyapatite nanorods may be

regarded as alternative fillers for dentine bonding systems (Sadat-Shojai et al.,

2010). They found that the use of HAp nanorods in low percentages provided

adhesives with improved properties. The increase in bond strength might be

explained by the fact that the nanofillers can reinforce the adhesive layer at the

resin-dentine interface (Atai et al., 2009).

However, even if HAp nanoparticles are currently used as coatings in

orthopedic and dental implants (Domingo et al., 2003, Ong and Chan, 2000),

128

their poor mechanical properties along with the high water uptake of

hydroxyapatite remain a matter of concern and have limited its application for

load-bearing materials (Labella et al., 1994, Santos et al., 2001, Domingo et al.,

2003).

2.2.1 Calcium/sodium phosphate-phyllosilicates fillers

Recently, surface-active glasses have been introduced in many fields of

dentistry (Hench, 2006, Margonar et al., 2012, Sauro et al., 2012b).

Bioactive glasses have numerous novel features, most important of which are

their ability to act as biomimetic mineralisers, matching the body’s own

mineralising traits, while also affecting cell signals in a way that benefits the

restoration of tissue structure and function (Hench and Paschall, 1974).

Bioactive glasses, as opposed to most technical glasses, are characterised by

the materials’ reactivity in aqueous environments that contain calcium and

phosphate. This bioactivity is derived from their reactions with tissue fluids,

resulting in the formation of a carbonated hydroxyapatite layer on the

glass/tissue interface, which makes it possible to bond bone and soft tissue

without toxicological consequences (Hench and Paschall, 1974).

The best-studied and -characterised bioactive glass is the commercially

available Bioglass® (formula 45S5), a high biocompatible calcium/sodium

phosphate-phyllosilicate originally developed as osteoconductive material but

chemically similar to natural tooth mineral (Hench and Andersson, 1993). It is

one of the foremost bioactive compounds with an excellent ability to promote

calcium phosphate (Ca/P) precipitation and subsequent crystallisation into HAp,

when immersed in simulated body fluid, a protein-free solution with ion

129

concentrations similar to those of human blood plasma, or saliva (Andersson

and Kangasniemi, 1991).

The components in calcium/sodium phosphate-phyllosilicates are basically

oxides of calcium, sodium, phosphorus, and silicon at certain weight ratios. The

chemical composition is significant. Silica maintains the glass structure, while

sodium, soluble in an aqueous environment, aids in maintenance of

physiological ionic balance and pH (Hench et al., 1971). Calcium and

phosphate are required, as they are the principal constituents of apatite mineral.

The proportions of all these constituents have been varied with differing results

(Hench et al., 1971).

Bioglass® 45S5 is endowed with unique compositional characteristics, namely

45 wt% SiO2, 24.5 wt% Na2O and CaO, and 6 wt% P2O5. Bioactive glasses

have traditionally kept the P2O5 fraction constant while varying the SiO2 content.

In fact, the network breakdown of silica by OH- was found to be time dependant

upon the concentration of SiO2. It is now understood that keeping the silica

content below 60 wt% and maintaining a high CaO/P2O5 ratio guarantees a

highly reactive surface.

So far, most studies on bioactive glasses have been focused on orthopedic

research, due to their ability to form a bone-like apatite layer on their surfaces in

the body environment (Gatti et al., 1994, Heikkila et al., 1993, Turunen et al.,

1994, Heikkila et al., 1995). The material is reported to have a long record of

safety and efficacy as as bone grafting material and has been cleared by the

FDA for use in orthopedic surgery (Bauer and Smith, 2002).

130

More than three decades of study have revealed a distinct, though as yet not

completely understood, series of chemical reactions that takes place on the

material's surface when it is brought into contact with body fluids.

These reactions do not depend on the presence of tissue and, within minutes of

exposure to the calcium-phosphate-rich environment, begin with leaching and

exchange of cations followed by loss of soluble silica. Both steps are

characterised by formation of silanols (SiOH). Polycondensation of SiOH to

form a hydrated silica gel precedes the formation of an amorphous calcium

phosphate layer. The reaction is concluded by the crystallisation of a

hydroxycarbonate apatite (HCAp) layer (Hench and Andersson, 1993,

Andersson et al., 1988, Kokubo, 1992).

More in detail, leaching is characterised by release of alkali or alkaline earth

elements (Na+ or K+), usually through cation exchange with H+ or H3O+ ions,

and de-alkalinisation of the glass surface layer. Ion exchange is quite rapid

because these cations are not part of the glass network, they only impart minor

changes to the network by forming non-bridging oxygen bonds. This ion

exchange process leads to an increase in interfacial pH, to values > 7.4.

Network dissolution occurs concurrently by breaking of -S-O-Si-O-Si- bonds of

the glass structure through the action of hydroxyl ions (base catalysed

hydrolysis of -S-O-Si-O-Si- bonds). Breakdown of the network is localised and

releases silica into the solution in the form of silicic acid (SiOH4). Last stages

involve the development of silica rich and amorphous calcium phosphate layers

respectively. Hydrated silica formed on the glass surface by these reactions

undergoes rearrangement by polycondensation of neighboring SiOH,

culminating in a silica-rich gel layer.

131

During the precipitation reaction, calcium and phosphate ions liberated from the

glass along with those of the solution combine to create a calcia-phosphate-rich

(CaP) layer on the silica gel. The calcium phosphate phase that stratifies in the

gel surface is initially amorphous (a-CaP). Within 3-6 h in vitro, the a-CaP will

crystallise into a HCAp structure by incorporating carbonate anions from the

solution within the a-CaP phase. Chemically and structurally, this apatite is

nearly identical to bone and tooth mineral, thus allowing the body to attach

directly to it. These surface reactions from contact with a calcium-phosphate-

rich fluid to the 100-150 μm formation of a HCAp layer takes in total 12 to 24 h

(Hench and Andersson, 1993).

Although Bioglass® 45S5 has also been successfully used for dentine

remineralisation when directly applied on the dentinal tissue (Curtis et al.,

2010), the development of mineralising dentine bonding systems containing

bioactive glass ceramic micro-fillers remains an important target to accomplish.

When calcium/sodium phosphate-phyllosilicates are utilised for pulp-capping

procedures and to reduce tooth sensitivity, the particle size can be as large as

300 μm (Oonishi et al., 1997). Such large particles would be inadequate for

incorporation into the resin bonding process, being the dentine tubule diameters

on the order of 1-2.5 μm (Marshall et al., 1997).

The spaces in the three-dimensional mesh of the collagen network exposed

through demineralisation are smaller still. The kinetics of dissolution of small (~5

μm) particles of Bioglass® 45S5 have been studied (Sepulveda et al., 2002),

and the material retains its ability to precipitate apatite even at that small size

(Vollenweider et al., 2007).

132

Including bioactive glasses in dentine bonding procedures, or in case of

methacrylate resins containing bioactive filler phases for potential application as

adhesive materials, it is expected that precipitated apatite would bond to

specific amino acids within exposed dentine collagen as occurs in bone.

Evidence has emerged suggesting that bonds to bone are mechanically strong

and believed to be chemical in nature; force is required to separate the tissue

from the glass (Wilson et al., 1981). The exact character of the bonds is not

known, though it has been suggested that the forming mineral bonds to specific

amino acids within the collagen, that is the main structural component of bone

(Hench et al., 1971, Hench and Paschall, 1973).

Bioactive glasses have already been shown to react favourably with dentine,

creating a mechanical bond (Efflandt et al., 2002), forming HCAp similar to tooth

mineral (Aoba et al., 1992, Yoshiyama et al., 1996) and displaying antibacterial

properties that would be beneficial to the prevention of secondary caries

(Zehnder et al., 2004).

Filling the adhesive with glass is also expected to minimise polymerisation

shrinkage (Say et al., 2006), which is likely to improve marginal integrity and

diminish leakage phenomena. To date, the glasses used in these applications

have been conventional silica glass or fluoride-releasing, but the use of

calcium/sodium phosphate-phyllosilicates as fillers in dentine bonding systems

is yet to be attempted.

Previously, two types of bioactive glass have been shown to reduce leakage

without detrimental effects on the bond strength (Zeiger, 2008). However, the

adopted method of application of the glass -vacuum deposition- is not possible

in the clinic.

133

2.2.2 Filler phase consisting of calcium silicate cements

Ion-releasing adhesives comprising bioactive fillers are meant to be in intimate

contact with biological fluid, soft and hard human tissues, hence, they should

principally present high biocompatibility as well as potential osteogenic

induction.

Based on the same principle, experimental remineralising resin-based calcium

silicate cements (ion-leaching composites) have been proposed as dental

materials with tailored remineralising properties, to be used as restorative base-

liner materials in sandwich restorations (Gandolfi et al., 2011b). Interestingly,

Gandolfi et al. demonstrated that the presence of experimental composites in

contact with demineralised dentine surfaces induced a significant

remineralisation of the phosphorous-depleted demineralised dentine surface

(Gandolfi et al., 2011b).

These findings contributed to and endorse the development strategy for an

advanced dentine bonding system containing biointeractive calcium-silicate

cements, and thus able to improve the longevity of hybrid layers through self-

sealing caused by the formation of apatite in the presence of nano-leakage.

Moreover, hydroxyl ions released during their hydration reaction may create

unfavourable conditions for bacterial survival and proliferation at the dentine-

restoration interfacial region.

Calcium silicate cements, such as the Portland cement, have been introduced

in dentistry as materials for different endodontic clinical applications due to their

biocompatibility and bioactive properties (Parirokh and Torabinejad, 2010). In

particular, mineral trioxide aggregate (MTA), a mechanical mixture of Portland

cement (75%), bismuth oxide (20%), gypsum (5%) with trace amounts of silica,

134

calcium oxide, magnesium oxide, potassium sulfate and sodium sulfate, was

introduced as a root-end filling material (Torabinejad et al., 1995), subsequently

expanding to many other applications of root repair and bone healing. These

applications included direct pulp capping, repair of root and furcation

perforations, and apexification (Torabinejad and Chivian, 1999, Schwartz et al.,

1999, Torabinejad et al., 1993).

Collectively, calcium-silicate cements are hydrophylic materials able to tolerate

moisture (hydraulic materials) and to polymerise and harden (setting) also in the

presence of biological fluids. They are in fact porous media, partially or

completely saturated with a pore solution. Due to this porosity, material-

environment mass exchanges occur, in particular ion diffusion through the

porosity.

The apatite-forming ability (i.e. bioactivity) of these cements have been

adequately documented. Their biological behaviour is mainly related to calcium

release, to the presence of silicon ions on their surface and to the formation of

bonelike apatite.

Portland cements are mixtures of dicalcium silicate, tricalcium silicate, tricalcium

aluminate, and tetracalcium aluminoferrite (Sarkar et al., 2005). They are ion-

leaching materials able to release calcium and hydroxyl ions (alkalinising

activity) into the surrounding fluids, stimulating the formation of new apatite-

containing tissues (Gandolfi et al., 2010a, Gandolfi et al., 2010b, Gandolfi et al.,

2010c, Taddei et al., 2011).

However, since calcium-silicate cements lack phosphate, they are able to

induce apatite formation only in presence of phosphate-containing fluids (e.g.

blood, plasma, saliva, dentinal fluid) (Qi et al., 2012). In a similar environment,

135

they dissolve, releasing all of their major cationic components. Of all the ions

released, Ca2+ is the most dominant. Because it is sparingly soluble in biologic

fluids, it leads to the precipitation of HAp (Sarkar et al., 2005).

The hydration process of Portland cements represents one of the most

important factors for the setting and maturation of the cements (Taylor, 1997).

Notably, while calcium-silicate particles react with water, a solid-liquid interface

forms on the mineral particles and ion dissolution occurs almost immediately.

Ca2+ ions are rapidly released (calcium hydroxide formation) and migrate into

the solution. A high-pH solution containing Ca2+, OH- and silicate ions is

established. Silicates are attacked by OH- ions (hydrolysis of SiO44- groups in

alkaline environment) and a CSH phase (calcium-silicates hydrates) forms on

mineral particles (Parirokh and Torabinejad, 2010).

CSH is a porous, fine-grained and highly disorganised hydrated silicate gel

layer containing Si-OH silanol groups and negative surface charges. Actually, at

alkaline pH, the deprotonation of silanol groups should predominate (Sanchez

and Zhang, 2008) with the consequent formation of SiO- negative groups. The

attraction between CSH particles has been reported as a consequence of the

very high negative charge density of the CSH particles and the presence of

Ca2+ ions (Plassard et al., 2005). The SiO- negative groups induce

heterogeneous nucleation of apatite by bonding calcium ions from the mineral

particles on the silica-rich CSH surface.

While the calcium silicate hydrate gel sets over time to form a solid network, the

release of calcium hydroxide increases the alkalinity of the surrounding medium

(Pellenq et al., 2009).

136

It was demonstrated how a material covered with an apatitic layer in proximity to

calcified tissues forms a chemical bond with the latter (Hench, 1999, Kokubo,

1990, Kokubo, 1992, Yuan et al., 2000)

Calcium silicates appeared to bond chemically to dentine when placed against

it, possibly via a diffusion-controlled reaction between their apatitic surface and

dentine (Sarkar et al., 2005). Sarkar et al. proposed that after the placement of

MTA in root canals and its gradual dissolution, with the release all of its major

cationic components, HAp crystals nucleate and grow, filling the microscopic

space between MTA and the dentinal wall. Initially, this seal is mechanical. With

time, they speculated that a diffusion-controlled reaction between the apatite

layer and dentine could lead to their chemical bonding. The result is the creation

of a seal at the MTA-dentine interface (Sarkar et al., 2005).

The first step in the development of alternative Portland-based filling/adhesive

materials requires the analysis of the physical characteristics of basic Portland

cements used as a principal constituent. In fact, other components may be

added for their beneficial effects either on manufacturing processes or on

biologic responses and mode of action.

The chemical composition and the hydration process may therefore play an

important role in the development of a new generation of biomedical materials

which would influence physical, chemical and structural characteristics (Gombia

et al., 2010).

Although different Portland cements (clinkers) are classified under the same

UNI EN 197/1 specification (UNI EN, 2007), they may nevertheless present

different physical characteristics and this may influence their basic properties

and behaviour. For instance, variable amounts of calcium sulfate are usually

137

added to regulate the setting time of Portland cements. Indeed, the presence of

calcium sulfate increases the setting time up to a minimum of 45 minutes

(Gombia et al., 2010) by reacting with tricalcium aluminate to form hexacalcium

aluminate trisulfate, a reactive salt known as ettringite (Gombia et al., 2010).

This is normally formed during the setting reaction of Portland cement. It sets in

contact with water and undergoes a hydration process in which Ca2+ and OH-

ions are released from tricalcium silicate into the surrounding environment. At

supersaturation levels, it forms calcium hydroxide (portlandite) precipitate and

amorphous calcium silicate hydrate gel (Camilleri, 2008).

This reaction is initially extremely fast with a subsequent decrease in velocity

due to the formation of an excessive layer of calcium sulfate dihydrate (gypsum)

which covers and protects the calcium aluminates. On the contrary, the

absence of gypsum would leave the tricalcium aluminates to react faster with

water forming hydrated calcium aluminates with a cubic structure (Taylor,

1997).

2.2.3 Dye-assisted confocal microscopy imaging of remineralised hard

tissues

As with many imaging techniques in optical microscopy, the main function of a

confocal imaging system is to improve image contrast: to delineate structures

that would otherwise be difficult to see. Confocal microscopy offers several

advantages over conventional optical microscopy to evaluate the

ultramorphology of resin-dentine interfaces, including controllable depth of field,

elimination of image degrading out-of-focus information, and the ability to collect

serial optical sections from thick specimens. The key to the confocal approach

138

is the use of spacial filtering to eliminate out-of-focus light or flare that occurs in

thick fluorescently labelled specimens. In a conventional widefield microscope,

the entire specimen is flooded evenly in light from a light source. All parts of the

specimen in the optical path are excited at the same time and the resulting

fluorescence is detected by an image capture device or photographic film

including a large unfocused background part. In other words, secondary

fluorescence emitted by the specimen that appears away from the region of

interest often interferes with the resolution of those features that are in focus.

This is especially problematic for specimens having a thickness greater than

circa 2 μm. In contrast, a confocal microscope eliminates out-of-focus signal as

only light produced by fluorescence very close to the focal plane can be

detected, and the image's optical resolution, particularly in the sample depth

direction, results much better than that of wide-field microscopes (Watson,

1997). The word 'confocal' derives from the use of an aperture in the optically

conjugate focal plane in front of the detector, in both the illuminating and

imaging pathways of the microscope. The area surrounding the aperture rejects

stray light returning from areas that are not in the focal plane of the lens.

Furthermore, this microscope presents the optical sectioning characteristic,

which enables the detection of subsurfaces components. The terminology refers

to the noninvasive method by which the instrument collects images, using

focused light rather than physical means to section the specimen (Watson,

1997). Consequently, high-resolution confocal microscopic images may derive

from either the surface of a sample or beneath the surface. These images are

thin (>0.35 μm) optical slices up to 200 μm in depth. With microscopes running

under "normal" conditions, the optical section thickness can be >1 μm and the

139

effective penetration into dentine a maximum of 100 μm. At this distance the

sharpness and contrast of the image may be poor, the best images being

derived from structures just below the surface (<20 μm). Imaging samples from

their surfaces without the need for thin section preparation has also significant

advantages when it comes to biomaterials/tooth interfaces as relatively large

intact tooth samples can be placed on the microscope stage. All that is required

is to section the sample once and observe directly the subsurface structures,

taking advantage of the optical sectioning feature.

All microscopic techniques involve the interaction of the sample under

examination with the interrogating system being used to probe it. The simplest

technique for using confocal microscopy in dental materials research is to

highlight the distribution of components within the bonded interface with

fluorescent labels. Fluorescent dyes, also called fluorophores or/and

fluorochromes, are useful as tracers to identify the path or the current location

of a compound in clinical as well as laboratory investigations (Watson, 1997).

The principle of fluorescence involves the absorption by the dye molecule of

energy in the form of light at one wavelength range (excitation), which causes

electrons to move into higher energy shells away from the nucleus. When these

electrons return to more stable, lower-energy shells the energy is dissipated as

both heat and light. Because the emitted photon has less energy than the

excitation photon, the wavelength of the emission is longer by Stokes' Law, and

the difference between the wavelengths of the excitation and emission maxima

is known as the Stokes' Shift (Rost, 1995).

Most of the dyes used in dental research are water-soluble and easily

detectable, even in a low concentration. They are made of very small particles

140

(0.60-0.80 nm in diameter) that may erratically permeate throughout dentine

and hybrid layer.

Rhodamine B, a water-soluble molecule with a molar mass of 479 g mol-1, is

one of the most frequently used fluorochromes. This compound is excited using

green light (540 nm) and emits red in colour (590 nm). Rhodamine B is effective

in very low concentration, fairly labile, moves freely across the bonded interface,

and is easily detected microscopically with appropriate filters. The compound is

also stable under various pH conditions.

In the early 1970s, Rahn & Perren introduced xylenol orange (Rahn and Perren,

1971), calcein blue (Rahn and Perren, 1970) and alizarin complexone (Rahn

and Perren, 1972). These chemicals fluoresce at different wavelengths and

colours, and all possess adjacent donor sites for calcium chelation. This means

that the label can indicate any area of mineralisation in the body, including a site

of bone formation or dentine deposition (Rahn and Perren, 1971).

Tetracyclines are also strong chelating agents, able to sequester a metallic ion

such as calcium and firmly bind it into a ring (Lee et al., 2003).

The use of fluorochromes in remineralisation studies is a widely accepted

technique that dates back to the 1950s. Several pioneers, such as Harold Frost

(Frost, 1963), have thoroughly investigated the potential of labelling with

fluorescent calcium chelators for the study of bone formation and bone

remodelling dynamics. Since the development of bone tissue engineering, a

renewed interest in the benefits of fluorochrome use was perceived.

Fluorochrome use in animal models makes it possible to determine the onset

time and location of osteogenesis, which are the fundamental parameters in

bone tissue engineering studies (Van Gaalen et al., 2010).

141

Since most of the fluorescent labels present one or another disadvantage, the

possibility of selection from a larger number increases the probability of finding

the one best suited for a special purpose. Important requirements are that (i) the

label should be slow to bleach and stable in an aqueous environment,

remaining deposited at sites of new mineralisation; (ii) the fluorescence should

be clearly outlined and produce a strong signal; (iii) the fluorescent dye should

be non-destructive to the substrate or material to which it comes into contact,

namely, there should be little interference with the experiment from local effects

on calcification; and (iv) the fluorescence should be resistant to chemicals and

to the light used for its excitation. Easy photographic recording and reasonable

cost of the dye are further considerations for selection.

2.2.4 Aims of the study

The aims of this study were:

x To evaluate the chemical-physical properties, bioactivity and adhesive

effectiveness of novel light-curable methacrylate-based dentine bonding

agents either incorporating calcium/sodium phosphosilicate (Bioglass®

45S5) or three distinct hydrated blends of experimental calcium-silicate

cements.

x To implement a prospective adhesive procedure involving the preliminary

application of a bioactive surface reactive glass-ceramic material

(Bioglass® 45S5).

The following investigations were planned:

142

I. A measurement of the changes in dimension and weight

resulting from water sorption in the experimental light-cured

resin-based materials.

II. A thermoanalytical study (differential scanning calorimetry) to

assess the relationship between the degree of hydration and

thermal properties.

III. An attenuated total reflection Fourier transform infrared

spectroscopy study of the bioactive micro-filled materials as a

function of soaking time in phosphate-containing solutions.

IV. A microtensile bond strength test and scanning electron

microscopic study of resin-dentine specimens created with the

experimental protocol and adhesives bonded to flat mid-

coronal dentine after immersion in phosphate buffered saline or

simulated body fluid solutions for 24 h or 6 months.

V. A morphological and nano-leakage confocal laser scanning

microscopic observation of the experimental resin-dentine

interfaces after ageing in phosphate buffered saline or

simulated body fluid solutions for 24 h or 6 months.

VI. A micro-indentation hardness analysis (micro-hardness testing)

of the experimental resin-dentine interfaces (hybrid layer and

its surroundings) after ageing in phosphate buffered saline or

simulated body fluid solutions for 24 h or 6 months.

143

Section II - Experimental projects

144

Chapter 3: Chemical-physical properties and

apatite-forming ability of experimental dental

resin cements containing bioactive fillers

145

3.1 Introduction

The clinical success of composite restorations is compromised by the

incomplete infiltration of currently used resinous materials into the

demineralised dentine that leaves unprotected collagen fibrils below and within

the resultant hybrid layer (Spencer et al., 2010). In an attempt to improve the

durability of resin-dentine interfaces, it was speculated that dentine bonding

agents (DBAs) containing Bioglass® 45S5 (BAG) as ion-releasing micro-filler

could be considered a promising approach for a possible therapeutic/protective

effect associated with the precipitation of mineral compounds (i.e apatite) within

the collagen matrix (Profeta et al., 2012). Likewise, light-curable resins

containing calcium-silicate Portland cements have been proposed to deliver

calcium and hydroxyl ions (alkalinising activity) into the surrounding fluids and

elicit a positive response at the interface from the biological environment (Tay

and Pashley, 2009).

It is common knowledge that release of ions from fillers in dentine adhesives

and restoratives depends on the rate of water sorption and the segmental

mobility of the polymer chains within the copolymerised, highly cross-linked

resin matrix (Ikemura et al., 2003). Their ion-leaching potential thus relies on the

hydrophilic nature of the resin phase to absorb water for the necessary ion

movement (Cattani-Lorente et al., 1999).

An important property in polymeric materials science is the glass transition

temperature (Tg) of the cured matrix which indicates the degree of cross-linking,

physical state and final mechanical properties of synthetic organic materials

used as plastics and resins. It has been suggested that water absorbed into the

adhesive polymers would result in a reduction of the Tg and a weakening of the

146

polymer network (Ito et al., 2005, Tay et al., 2002a, Dhanpal et al., 2009). The

Tg reached after the incorporation of fillers into the adhesive would also depend

on the amount of filler particles contained in its composition (Moraes and

Grandini, 2011).

However, because of the novelty of these adhesives, the influence of the

amount of water sorption on their thermal properties is still unknown. As such,

the relationship between Tg and degree of hydration needs to be determined.

Recent studies selected vibrational spectroscopy to investigate and identify a

large range of components associated with dental materials (Wang et al., 2012,

Kim et al., 2010e). Attenuated Total Reflection Fourier Transform Infrared

Spectroscopy (ATR-FTIR) is often used to evaluate the in vitro apatite-forming

ability of bioactive materials as a function of soaking time in phosphate-

containing solutions (Dulbecco’s Phosphate Buffered Saline, DPBS) (Gandolfi

et al., 2011a).

Thus, the objective of this study was to investigate the performance of four new

light-curable methacrylate-based DBAs either incorporating BAG or three

different experimental calcium-silicate cements, with particular emphasis on the

water sorption and solubility behaviour of the cross-linked networks under a

simulated wet oral environment. For each material, the Tg was characterised by

using differential scanning calorimetry (DSC) directly after curing and following

60 days of storage in deionised water. The in vitro apatite-forming ability was

assessed by ATR-FTIR technique after soaking in DPBS for 60 days.

The null hypotheses to be tested in this study were: (i) there is no difference

between DBAs with respect to water sorption, solubility and water uptake. (ii)

There is no effect of water uptake and micro-filler content on the thermal

147

properties (Tg) of the tested adhesives. (iii) Milled DBAs possess bioactivity with

the ability to form hydroxycarbonate apatite (HCA) after immersion in DPBS.

3.2 Materials and methods

3.2.1 Experimental micro-fillers and resin blends formulation

A comonomer blend was formulated from commercially available monomers by

using 28.75 wt% of hydrophilic 2-hydroxyethyl methacrylate (HEMA: Aldrich

Chemical, Gillingham, UK) and 40 wt% of cross-linking dimethacrylate 2, 2-

bis[4(2-hydroxy-3-methacryloyloxy-propyloxy)-phenyl] propane (Bis-GMA:

Esstech Essington, PA, USA). 30 wt% of 2,5-

dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid (PMDM:

Esstech Essington) was included to obtain a dental bonding system with

chemical affinity to Calcium (Ca2+) (Table 3.1). To make the resins light-curable,

0.25 wt% camphorquinone (CQ: Aldrich Chemical) and 1.0 wt% 2-ethyl-

dimethyl-4-aminobenzoate (ETDA: Aldrich Chemical) were also added.

A calcium sodium phosphosilicate (BAG: Bioglass® 45S5, SYLC, OSspray Ltd,

London, UK) and three modified calcium-silicate cements were used as calcium

and hydroxyl ions releasing micro-fillers (20-30 μm-sized particles). The first

calcium-silicate filler (HOPC) was created by mixing 82.5 wt% of a type I

ordinary Portland (OPC: Italcementi Group, Cesena, Italy), mainly constituted

by tri-calcium silicate (Alite: 3CaO x SiO2), di-calcium silicate (Belite: 2CaO x

SiO2), tri-calcium aluminate (3CaO x Al2O3) and gypsum (CaSO4 x 2H2O), with

7.5 wt% of phyllosilicate consisting of sodium-calcium-aluminum-magnesium

silicate hydroxide hydrate [(Na,Ca)(Al,Mg)6(Si4010)3(OH)6-nH2O; Acros

148

Organics, Fair Lawn, NJ, USA] in deionised water (Ratio 2:1). The last two

fillers were formulated as following: i) HPCMM: 90 wt% OPC with the addition of

7.5 wt% phyllosilicate and 2.5 wt% hydrotalcite containing aluminum-

magnesium-carbonate hydroxide hydrate [(Mg6Al2(CO3)(OH)16·4(H2O); Sigma-

Aldrich] in deionised water (Ratio 2:1); ii) HPCTO: 80 wt% OPC plus 7.5 wt%

phyllosilicate, 2.5 wt% hydrotalcite and 10 wt% titanium oxide (TiO2: Sigma-

Aldrich, Gillingham, UK)] in deionised water (Ratio 2:1). The three experimental

Portland-base silicates were allowed to set in incubator at 37°C for 24 h, ground

in an agate jar and finally sieved to obtain < 30 μm-sized micro-particles. Hybrid

photopolymerisable adhesive agents were prepared by mixing with a spatula for

30 s on a glass plate each individual filler (40 wt%) and the neat resin (60 wt%)

(Hashimoto et al., 2010b) in order to form a homogeneous paste. A generic

label for every experimental material has been proposed to refer to the main

components (Res-Contr, Res-BAG, Res-HOPC, Res-HPCMM and Res-

HPCTO). This made the accounting easier but it is non-presumptive about

molecular formula and structure.

149

Table 3.1 - Chemical structures of the constituent monomers and composition (wt%) of the experimental adhesives used in this study. Abbreviations. Bis-GMA: 2, 2-bis[4(2-hydroxy-3-methacryloyloxy-propyloxy)-phenyl] propane; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid; CQ: camphoroquinone; EDAB: 2-ethyl-dimethyl-4-aminobenzoate; BAG: Bioglass® 45S5; HOPC: set Portland cement and smectite; HPCMM: Portland cement, Smectite and Hydrotalcite; HPCTO: set Portland cement, Smectite, Hydrotalcite and Titanium Oxide.

150

3.2.2 Specimen preparation

Each co-monomer mixture was directly dispensed drop-by-drop into a Teflon

split ring mould (7 mm in diameter and 1 mm thick) in an ordinary laboratory

environment at ~23°C taking care to make the adhesive bubble-free. The Teflon

mold was then sandwiched between two glass slides covered with Mylar sheets

to exclude atmospheric oxygen and the complete assembly was clamped. The

resin was light-cured for 20 s on each side using a quartz-tungsten-halogen

light-curing unit (Optilux VLC, Demetron Research Co., CT, USA). The light-

curing unit had an exit-window diameter of 5 mm and was operated at

600mWcm-2. The output intensity of the curing unit was verified before testing

with a halogen radiometer (Optilux Radiometer Model 100P/N - 10503;

Demetron Research Co.). Standardisation of the distance between the light

source and the specimen was provided by the thickness of the glass slide (1

mm), and the glass slides also provided a smooth surface for the testing

process. Selection of curing time was determined in a pilot experiment by

measuring a baseline micro-hardness of the surface of the resin disks

(unpublished data). With the adopted total curing time (40 s), resins exhibited a

mean Knoop micro-hardness of 20 ± 2 KHN that was sufficient to allow

specimens to be removed from the Teflon mold without undergoing permanent

deformation. Any specimen with visible voids was discarded. The excess

material around the disks was removed using a scalpel blade and the margins

were rounded and finished using 1000-grit silicon carbide grinding paper. The

ultimate specimens (n = 10 per group) were flat and had very smooth surfaces.

The thickness and diameter of the specimens were measured at four points

using a digital caliper (Mitutoyo Corporation, Tokyo, Japan), rounded to the

151

nearest 0.01 mm, and these measurements were used to calculate the volume

(V) of each disk (in mm3). The resin disks were then stored in a desiccator with

silica gel at 37°C to ensure dryness for measurement of the initial mass. Each

disk was weighed in an electronic analytical balance (Model AD6, PerkinElmer,

Shelton, CT, USA) with a reproducibility of ± 0.1 mg until a constant mass (M0)

was obtained (i.e. variation lower than 0.2 mg in 24 h), which was the baseline

weight for the absorption-desorption cycle.

3.2.3 Water sorption and solubility evaluation

The protocol for this study was determined according to the ISO Standard

Specification No. 4049 (ISO Standard, 2000) except for the dimension of the

specimen disks and the period of water storage that had been extended up to

60 days.

The specimens were individually immersed in Paraffin-sealed glass vials

containing 20 ml of deionised water (pH 7.2) at 37 ± 0.5°C. At noted intervals

(3h, 5h, 10h, 24h, 2d, 3d, 6d, 9d, 12d, 15d, 18d, 21d), each specimen was

taken out of the glass vial using tweezers, blotted on a Whatman's filter paper to

remove excess fluid, weighed and restored in fresh deionised water. The uptake

of water was recorded until there was no significant change in weight, i.e.

maximum wet mass of the surface-blotted water-equilibrated specimens (Ms)

was attained.

The specimens then underwent desorption in a desiccator, as previously

described, and weighed daily until a dried constant mass (MD) was obtained.

The values for water sorption (WS) and solubility (SL), (μg/mm3), were

calculated for each specimen using the following equations:

152

WS = (Ms − M0) / V

SL = (M0 − MD) / V

(Ito et al., 2005)

Because the mass variation of resin disks is the net result of both an increase in

mass owing to water penetration and the decrease in mass owing to elution of

low-molecular-weight monomers, and of oligomers, the net water uptake was

calculated as the sum of maximum water sorption and solubility (Unemori et al.,

2003).

3.2.4 Differential scanning calorimetry (DSC)

The Tg of the experimental bonding systems were characterised by the use of a

Perkin-Elmer Jade DSC system (Perkin-Elmer Corp., Waltham, MA, USA). The

system was calibrated with zinc/indium.

Additional specimens (n = 4 per group) were produced under the same

photoactivation conditions used in the water sorption kinetics experiments and

divided into two groups. The Tg values of the first groups were measured

immediately after curing, and the Tg values of the second groups were

measured after 60 days. The second groups of specimens were stored in light-

proof boxes after the curing procedure to prevent further exposure to light and

in distilled water at 37 ± 0.5°C.

Two crimpled aluminium pans with perforations (diameter = 4mm, 1.2 thick)

were placed in the sample holder of the DSC furnace. Adhesive films (10-13

mg) were obtained from each specimen and these were sealed in one of two

aluminum pans while the other pan was left empty as reference. Each

experimental adhesive was tested five times at a rate of 10°C min-1 under

153

nitrogen gas (20 ml min-1) over a two step heating/cooling cycle temperature

range of -20 to 200°C, to eliminate any residual water. The analysis of DSC

curves was carried out for the second heating run data in order to exclude the

effect of any possible impurity, which might influence the thermal properties of

the materials during the first heating cycle. The glass transition temperature was

determined by using the inflection midpoint of the initial S-shaped transition

slope, and determined from the onset with the aid of Perkin-Elmer Spectrum

computer software. For each thermograph the Tg was calculated three times

and the average value was used as the result and for the purpose of

comparison.

3.2.5 Statistics

Kruskal-Wallis analysis of variance (ANOVA) was used to evaluate whether

there were any differences between groups for maximum water uptake,

solubility, net water uptake and the percentage change in Tg values. The

percentage change in Tg values was used to compensate for some differences

in initial Tg values. Where a statistically significant p value was observed (p <

0.05), post-ANOVA pair-wise comparisons were conducted using Mann-

Whitney-U tests with p < 0.01 regarded as statistically significant to compensate

for multiple comparisons.

3.2.6 ATR-FTIR spectroscopy

Four disks (n = 4) were made per group according to the procedure previously

described. In vitro bioactivity was determined by immersing the materials in

sealed cylindrical polystyrene holder (3 cm high and 4 cm in diameter)

154

containing 5 mL of DPBS fluid. DPBS is a physiological-like buffered (pH 7.4)

Ca- and Mg-free solution with the following composition (mM): K+ (4.18), Na+

(152.9), Cl- (139.5), PO43- (9.56, sum of H2PO4

- 1.5mM and HPO42- 8.06 mM).

The solution was replaced every 72 h. Infrared spectra were obtained

immediately after curing and after 60 days of DPBS ageing at 37°C (Gandolfi et

al., 2010a), using a Spectrum One FTIR spectrophotometer (Perkin-Elmer

Corp., Norwalk, Conn., USA) equipped with a diamond crystal attenuated total

reflection (ATR) accessory. Prior to the spectrophotometric analysis, each

sample was rinsed with distilled water for 30 s to stop the exchange reactions

and then completely air dried. ATR/FTIR spectra were acquired from Bioglass®

45S5 powder as well as the three hydrated blends of calcium-silicate cements

alone used for preparing the filled adhesive. The ATR area had a 2 mm

diameter. The IR radiation penetration was about 2 μm. Spectra were collected

in the range of 650-4000 cm-1 at 4 cm-1 resolution for a total of 88 scans for

each spectrum, and processed by smoothing, baseline correction, and

normalisation with Spectrum One Software Version 5.0.1 (Perkin-Elmer Corp.,

Norwalk, Conn., USA). To avoid complications deriving from potential lack of

homogeneity of the samples, five spectra were recorded on each specimen.

The reported IR spectra were the average of the spectra recorded on five

different points.

3.3 Results

3.3.1 Water sorption and solubility evaluation

All the specimens remained intact after the absorption and desorption cycles.

155

No visible signs of discoloration, crazes or cracks were observed in the resin

disks.

Table 3.2 shows a summary of mean (± standard deviation) data for mass gain

(maximum water uptake) and mass loss (solubility). Since these processes

occur simultaneously (Tay and Pashley, 2003b), they were added together to

provide an estimate of the net water uptake.

Kruskal-Wallis analysis of variance indicated that maximum and net water

uptake showed potential differences between groups (p < 0.001). Post-ANOVA

contrasts indicated that all groups were different from each other with the

exception that the Res-HPCTO group was not different from the groups Res-

HOPC and Res-HPCMM.

Overall, the neat resin (Res-Contr) exhibited values of maximum water uptake,

solubility and net water uptake that were significantly lower when compared to

its corresponding filled versions. With respect to water sorption, the Res-Contr

group absorbed the least amount of water (Res-Contr < Res-HPCMM = Res-

HPCTO = Res-HOPC < Res-BAG), while Res-BAG showed the highest values

and both differed statistically from all the other DBAs (p < 0.05). Res-HPCTO

presented intermediate values and did not differ statistically from Res-HPCMM

and Res-HOPC groups (p > 0.05).

Following the same trend, the lowest value of net water uptake (89.5 μg/mm3)

was observed for the comonomer blend with no filler (Res-Contr < Res-HPCMM

= Res-HPCTO = Res-HOPC < Res-BAG), which was significantly different from

the values obtained with the other DBAs (p < 0.05). No differences on net water

uptake were detected when the value obtained for Res-HPCTO (212.5 μg/mm3)

was compared with the results of Res-HPCMM (185.5 μg/mm3) and Res-HOPC

156

groups (218.9 μg/mm3) (p > 0.05), respectively.

The loss of dry mass following water sorption was defined as solubilisation of

resin. No statistically significant differences between all the DBSs were

observed for this parameter (p > 0.05).

Table 3.2 - Summary of maximum water uptake, solubility and net water uptake data. Values are mean (± standard deviation) in relative terms (1 μg/mm3 = 0.001 mg/mm3 x 100 = 0.1 mg/100 mm3 = 0.1%) and in absolute terms (μg/mm3) to provide comparisons to literature values which include both expressions. For each parameter investigated, same superscript lower case letters (analysis in columns) indicate no statistically significant differences (p > 0.05). Negative solubility values indicate that the dried constant mass obtained after final desiccation (MD) was higher than the initial adhesive polymer mass before water immersion (M0), suggesting that the absorbed water may have not been completely eliminated.

157

3.3.2 Differential scanning calorimetry (DSC) DSC thermograms displayed two heating/cooling cycles for each specimen

during the heating programme. The upper part of the DSC curve represented

the heating cycles whereas, the lower part represented the cooling cycles. The

thermal heat-capacity changes of the Tg obtained for the control and the four

experimental DBAs during the second heating run are reported in Table 3.3.

This shows a summary of the mean data for initial Tg, Tg after ageing and the

percentage change in Tg.

For the resin control, the initial glass transition temperature was 115.3°C

whereas initial Tg values of Res-BAG, Res-HOPC, Res-HPCMM and Res-

HPCTO were 119.7°C, 117.2°C, 114.5°C and 114.4°C, respectively. After the

ageing period, the Tg for the Res-Contr group increased to 121.3°C. The Tg

also increased at the end of the period of water storage for all the other

experimental DBAs, ranging between 131.2°C and 147.8°C.

Kruskal-Wallis analysis of variance indicated that the percentage change in Tg

showed potential differences between groups (p < 0.001). Post-ANOVA

contrasts indicated that all groups were different from each other except that the

lowest percentage change in Tg presented by Res-Contr group was not

statistically different from the percentage change of the Res-HOPC group and

that the percentage change in Tg for Res-BAG group was not statistically

different from the Res-HOPC and Res-HPCMM groups, respectively. Lastly, the

Res-HPCTO group exhibited the highest percentage change in Tg.

158

Table 3.3 - Means and standard deviations for Tg initially, after the ageing period and percentage change as determined by DSC analysis. Numbers in parentheses are standard deviations. Same letter indicates no differences in columns. * Five measurements

159

3.3.3 ATR-FTIR spectroscopy

The results of ATR-FTIR analyses are shown in Figure 3.1. Band assignments

have been given according to the literature (Rueggeberg et al., 1990, Wang et

al., 2011).

In the unmilled comonomer blend, the group assignments of the IR absorption

bands confirmed prominent bands at 1716, 1637, 1454, 1162-1078 cm-1

attributed to the C=O, C=C, CH2, and C-O-C vibrations (HEMA), respectively.

Absorption bands at 1260, 1385, 1520-1610, 3000 and 3440 cm-1 were

attributed to C-O-C6H4, CH2, C6H4, C-H and OH vibrations (Bis-GMA).

The IR spectra of the hydrated powders presented similarities for the calcium-

silicate cements. Calcium carbonate bands at about 1457-1460 cm-1 (HOPC,

HPCTO and HPCMM) and 1420 cm-1 (HOPC) were present. The band at 1110

cm-1 can be attributed to the SO42- group. The 870-850 cm-1 bands are

attributable to belite.

The HPCTO and HPCMM powders were characterised by a higher relative

intensity of the bands at about 1460 cm-1; in the group constituted by type I

ordinary Portland cement, the band at 1460 cm-1 presented lower intensity than

in the other cements (i.e. lower carbonate content).

IR spectra of Bioglass® 45S5 powder were also obtained to characterise its

chemical structure. Bioglass® 45S5 powder presented vibrational bands at 731,

920, 1030 and 1457 cm-1 corresponding to CO32-, Si-O, Si-O-Si and CO3

2-

stretches, respectively (Kim et al., 1989).

Figure 3.1 also shows the FTIR-ATR spectra recorded on the outer surfaces of

freshly prepared materials immediately after curing and of the samples soaked

160

in DPBS for 60 days, revealing the presence of their respective constitutive

elements.

With the exception of samples made of pure comonomer blend with 0 wt%

micro-filler content (Res-Contr), used as references to evaluate the structural

evolution and the bioactivities of the prepared biomaterials, FTIR analyses

proved the presence of carbonated apatite on all the experimental adhesives

after the ageing period. It was observed the beginning of apatite deposition

(1030-1465 cm-1) for Res-HPCTO and Res-HPCMM groups, with the latter

exhibiting a more crystalline apatite phase at 1465 cm-1.

A more crystalline apatite phase was also observed on the adhesives

incorporating type I ordinary Portland cement (Res-HOPC), as primarily

revealed by the higher resolution of the phosphate bending bands at 960, 1025

and 1465 cm-1.

Following the ageing period, the band present in the Bioglass® 45S5 powder at

731 and 920 cm-1 disappeared on the Res-BAG adhesive. A small carbonate

(HCA) band was present at 870 cm-1 (Lusvardi et al., 2009); this carbonate

band is usually taken as an indication for carbonate being incorporated into

apatite, resulting in HCA, rather than stoichiometric HAp (Lu and Leng, 2005). It

is difficult to distinguish whether this band is split, in which case it would be B-

type substitution (i.e. carbonate replacing a phosphate group). However, broad

CO32- bands are present in the region starting from 1410-1440 cm-1 indicating

B-type substitution (Brauer et al., 2010).

Spectra of Res-BAG sample after the soaking period also showed a shoulder at

1080-1090 cm-1 due to the P-O stretch which is observed in B-type substituted

HCA (LeGeros et al., 1969). In the same samples, the presence of

161

orthophosphate at 60 days was confirmed by a clearly pronounced

orthophosphate band at 1027 cm-1, superimposed on the Si-O-Si stretch band

at 1030 cm-1 which increased but did not markedly sharpen.

The absorption band at 1716 cm-1, attributed to C=O stretching, can clearly be

seen in the IR spectra of all the milled adhesives. This indicates that the C=O of

the amide groups along the chains of HEMA was actively involved in the

formation of the polymeric networks.

162

163

Figure 3.1 - ATR-FTIR spectra of the unmilled comonomer blend, of Bioglass® 45S5, HOPC, HPCTO and HPCMM powders and of the hybrid experimental adhesives immediately after curing and following 60 days in DPBS.

164

3.4 Discussion

The water absorption characteristics of any polymeric material, whether filled or

not, is of great importance for dental applications. Water is known to swell the

resin polymer network and hydrolytically cleave methacrylate ester bonds (Salz

et al., 2005). This enables water percolation through the hybrid layer and may

favour the rapid and catastrophic degradation of resin-dentine bonds (Pashley

et al., 2004).

However, some water absorption may not always be entirely disadvantageous

(Wei et al., 2011) and this study explored the inclusion of four different ion-

releasing fillers in an experimental dental adhesive as a way to benefit from the

resin's hydrophilic behaviour and open up the potential to create favourably

biointeractive dental adhesives. Hence a precise evaluation of the water

absorption characteristics of the polymeric biomaterials was clearly important.

By using the data obtained with the neat resin (Res-Contr) as a parameter for

the relationship between water sorption and type of micro-filler, maximum water

uptake and net water uptake were found to increase by adding Bioglass® 45S5

or calcium-silicate cements to the polymer network, leading the anticipated first

null hypothesis to be partially rejected and corroborating the results of previous

studies (Yiu et al., 2004).

After 60 days of storage in deionised water, all the experimental adhesives

provided statistically significant increments relative to their initial glass transition

temperature that compelled us to reject also the second null hypothesis.

Conversely, the third null hypothesis was accepted, since ATR-FTIR analyses

proved the presence of carbonated apatite on all the experimental adhesives

after prolonged storage in DPBS.

165

In ISO 4049 (ISO standard, 2000), resin solubility is calculated as a loss of dry

mass in specimens that have been immersed in water over time. Unreacted

monomers or oligomers can leach out of the polymer during water sorption and

subsequent polymer expansion. Generally, increases in water sorption are

associated with increases in solubility. This is in contrast with the results of the

present study in which no significant differences were observed for the latter

between all the tested materials (Table 3.2). A possible explanation for this

controversy might be that the dried constant mass obtained after final

desiccation was higher than the initial adhesive polymer mass before water

immersion, suggesting that the absorbed water may have not been completely

eliminated and thus giving either poor (Res-BAG, Res-HOPC and Res-HPCTO)

or negative solubility values (Res-Contr, Res-HPCMM). However, this does not

mean that no solubility occurred but rather that water sorption was greater than

solubility.

Since calcium and phosphate ion release is slow at later times (Abou Neel et

al., 2010), this could in part provide a further reason of the slight reduction in

material mass and volume after 60 days.

Water diffuses into polymers at different rates depending on the polarity of their

molecular structure, the degree of cross-linking and the presence of residual

monomers and/or other water-attracting species, e.g. glass surfaces (Fabre et

al., 2007). According to these molecular and microstructural factors, the

mechanism of water diffusion can be summarised in two main theories: (i) free

volume theory, according to which water diffuses through nanometre spaces

within the polymer and (ii) interaction theory, according to which water binds to

specific ionic groups of the polymer chain. In this case, water diffusion occurs

166

according to water-affinity of these groups (Malacarne et al., 2006).

Water absorbed into resin polymers has been demonstrated to decrease the

glass transition temperature of dental adhesives. However, the damage is

reversible at the outset with removal of the sorbed water (Dhanpal et al., 2009).

After light-curing, the presence of Bis-GMA monomers in the experimental resin

created a polymeric network able to stabilise the outer surface of the resin

disks. Nonetheless, water-filled regions and/or hydrophilic polymer domains

were also present in the cured resin (Tay et al., 2002b), and these channels

were thought to be responsible for water movement within the adhesives. Due

to the hydrophilic pendant groups of HEMA, once immersed in aqueous

medium the designed resin matrix was permeable enough to absorb water that

was reused for hydrating the ionic products (PO43- and Ca2+) released from the

water-soluble micro-fillers. This may explain why the addition of the bioactive

particles to the control resin showed no reduction of the Tg for the experimental

adhesives.

As already pointed out in a previous work (Min et al., 1993), Tg variation has

also been attributed to various molecular parameters, such as molecular weight

and stiffness of the cross-linked chains. Along with a sustained and effective

assimilation of water, the reactive fillers introduced phosphate and silicate

groups that had a hindering effect on chain mobility; furthermore, secondary

interactions were generated by the possible formation of oxygen bridges

between close proximity phosphate and silicate groups within the copolymeric

network, requiring a greater amount of energy to free the chains and thus

raising the thermal heat-capacity of the material (Kemal et al., 2011).

Intraoral temperatures that exceed the Tg may result in softening of the material

167

and, consequently, in failure of the clinical procedure. It is equally important to

observe that all the Tg values obtained are well above the range of oral

environment temperature.

The ability to release biologically active ions (biointeractivity) is a prerequisite

for a material to be bioactive and trigger the formation of apatite. Hashimoto et

al. demonstrated that crystal formation was a common behavior of adhesives

prepared with 40 mass% bioglass and 60 mass% of light-cured resin containing

Bis-GMA and HEMA during long-term water storage (Hashimoto et al., 2010b).

The occurrence of similar chemical-physical events was recently reported in a

light-curable MTA-based material containing an amphiphilic resin immersed in

phosphate-containing solutions (Gandolfi et al., 2011a).

In this study, ATR-FTIR spectroscopy technique yielded informations on the

chemical composition change which occurred on the surface of all the resin

disks during the ageing period. As showed in Figure 3.1, IR spectra of the

methacrylate-based filled resin disks soaked in DPBS for 60 days confirmed the

presence of carbonate ions in different chemical phases, mainly as apatite

deposits at 960, 1025 and 1465 cm-1 (Res-HOPC); 1030 and 1465 cm-1 (Res-

HPCTO and Res-HPCMM groups); 870, 1027, 1080 and 1440 cm-1 (Res-BAG)

It appears from the preceding discussion that the similarity in mode of biologic

action of the micro-fillers employed in this study stems from one common

characteristic they all possess: their propensity to release Ca2+ and ability to

form hydroxycarbonate apatite (HCA) on the polymerised methacrylate.

It is important to consider that Bioglass® 45S5 and the silicate phases of

Portland cements when hydrated underwent a series of physicochemical

reactions with subsequent precipitation of a polycondensated silica-rich layer

168

“Si-gel” (Res-BAG) (Andersson and Kangasniemi, 1991) or resulting in the

formation of a nanoporous matrix/gel of calcium silicate hydrates “C-S-H

phases” together with a soluble fraction of calcium hydroxide Ca(OH)2 and

portlandite (Camilleri, 2008) (Res-HOPC, Res-HPCMM and Res-HPCTO).

Further incorporation of various mineral ions from the surrounding fluid helped

the amorphous Si-gel and C-S-H phases to finally evolve into apatite crystals.

This process is governed mainly by a tailored surface charge; it has been

previously demonstrated that negatively charged polar groups must be present

for a catalytic effect on apatite nucleation (Welch et al., 2010). Presumably,

crystal formation also depends on the ion product being greater than the

solubility product constants of the solid phase formed on the material surface

(Filgueiras et al., 1993). Indeed, the Si-gel and C-S-H phases provided the

negatively charged sites for the migration of Ca2+ ions, which in turn led to an

over-saturated solution that exceeded the solubility product constants for a

number of mineral forms, inducing crystal growth (Kokubo et al., 2003, Kim et

al., 2005).

There is morphological evidence that the extent of permeability through the

hybrid layer (nano-leakage) increases after 1 year of water storage (Tay et al.,

2003, Hashimoto et al., 2002), in consequence of the adsorption of water,

hydrolysis of the ester bonds and component release processes in methacrylate

materials (Spencer et al., 2010).

As the crystallisation process can bind water and encourage reprecipitation of

less soluble species in material regions from which components have been

released, it can also be anticipated that this might limit nano-leakage and

reduction in bond strength during long-term function.

169

3.5 Conclusion

Within the limitations of this study, it can be concluded that experimental

methacrylate-based DBAs either incorporating BAG or calcium-silicate cements

are not inert materials in a simulated oral environment, and the precipitation of

apatite deposits in the surrounding fluid or contiguous dental tissues may occur

in the intra-oral conditions.

A bioactive dental material which forms HCA on the surface would have several

benefits including closure of gaps forming at the resin-dentine interface and

potentially better bond strength over time (less degradation of bond).

Future experiments will explore the functional consequences of the water

sorption values observed in this study on microtensile bond strength when the

same experimental adhesives are applied to acid-etched water-saturated

dentine. Microscopy ultra-morphological analysis needs to be also performed.

170

Chapter 4: Bioactive effects of a calcium/sodium

phosphosilicate on the resin-dentine interface: a

microtensile bond strength, scanning electron

microscopy, and confocal microscopy study

171

4.1 Introduction

Bioactive materials are often used in operative dentistry due to their ability to

interact actively with dental hard tissues, inducing calcium-phosphates (Ca/P)

deposition in the presence of body fluids or saliva (Kim et al., 2010e, Ryou et

al., 2011, Hench and Andersson, 1993, Sauro et al., 2011a).

Whereas remineralisation of enamel lesions can be achieved predictably (Shen

et al., 2001, Lagerweij and ten Cate, 2002), there is little information on whether

it is possible to remineralise specific mineral-deficient areas within the resin-

dentine interface (i.e. hybrid layers) (Ryou et al., 2011). Some glass ionomer

based, fluoride-releasing adhesive resins may induce crystal growth within gaps

in the bonded interface after long-term storage in water (Hashimoto et al.,

2008). Furthermore, bioactive, ion-releasing materials, such as calcium-

phosphate (Ca/P) cements, have the potential to encourage dentine

remineralisation by mineral precipitations (Ngo et al., 2006, Dickens et al., 2003,

Dickens and Flaim, 2008, Arends et al., 1997).

Peters et al. (Peters et al., 2010) showed the presence of a higher mineral

content [determined by electron probe elemental micro-analysis (EPMA)

techniques] and an increase in micro-hardness along the interface of resin-

bonded caries-affected dentine, following the application of materials containing

Ca/P cements. Bioactive calcium/sodium (Ca/Na) phosphosilicates, such as

Bioglass® 45S5 (BAG), are able to induce deposition of hydroxycarbonate

apatite (Sauro et al., 2011a, Vollenweider et al., 2007, Efflandt et al., 2002,

Hench and Paschall, 1973). Although bioactive glasses have previously been

used for dentine remineralisation by direct application onto demineralised

dentinal tissue when dispersed in water solutions (Sauro et al., 2011a, Efflandt

172

et al., 2002), there is little information about the potential therapeutic effects of

BAG on the resin-dentine interface when used during etch-and-rinse bonding

procedures.

Therefore, this study was devised to assess the bioactive effects of BAG during

etch-and-rinse dentine-bonding procedures on the resin-dentine interface. This

aim was accomplished by evaluating the microtensile bond strength (μTBS) of

specimens after 24 h and 6 months of storage in PBS. Fractographic analysis

was also performed through scanning electron microscopy (SEM).

The ultramorphology and nano-leakage analysis of the resin-bonded dentine

was executed using confocal laser scanning microscopy (CLSM).

The null hypotheses to be tested in this study were: (i) the use of BAG

employed during bonding procedures has no effect on the bond strength, and

(ii) the presence of BAG does not reduce nano-leakage within the

demineralised ‘poorly-infiltrated’ areas within the resin-dentine interface.

4.2 Materials and methods

4.2.1 Specimen preparation

Caries-free human third molars, extracted for surgical reasons from 20 to 40 yr.

old patients under a protocol approved by the institutional review board of Guy's

Hospital (South East London ethical approval 10/H0804/056), were used in this

study. The treatment plan of any of the involved patients, who had given

informed consent for use of their extracted teeth for research purposes, was not

altered by this study.

173

The study was conducted in accordance with the ethical guidelines of the

Research Ethics Committee (REC) for medical investigations.

The teeth were stored in deionised water (pH 7.1) at 4°C and used within 1

month after extraction. The coronal dentine specimens were prepared by

sectioning the roots 1 mm beneath the cemento-enamel junction (CEJ) with a

hard tissue microtome (Accutom-50; Struers, Copenhagen, Denmark) using a

slow-speed, water-cooled diamond wafering saw (330-CA RS-70300; Struers).

A 180-grit silicon carbide (SiC) abrasive paper mounted on a water-cooled

rotating polishing machine (Buehler Meta-Serv 3000 Grinder-Polisher; Buehler,

Düsseldorf, Germany) was used (30 s) to remove the diamond saw smear layer

and to replace it with a standard and more clinically relevant smear layer

(Oliveira et al., 2003).

4.2.2 Experimental bonding procedures and formulation of resin

adhesives

A resin co-monomer blend was formulated by using a hydrophobic, cross-

linking dimethacrylate monomer - bisphenyl-A-glycidyl methacrylate (Bis-GMA;

Esstech, Essington, PA, USA) - and a hydrophilic monomer - 2-hydroxyethyl

methacrylate (HEMA; Sigma-Aldrich, Gillingham, UK). In order to obtain a

dental bonding system with chemical affinity to calcium (present in dentine and

BAG), an acidic functional monomer - 2,5-dimethacryloyloxyethyloxycarbonyl-

1,4-benzenedicarboxylic acid (PMDM; Esstech Essington) - was also included

within the composition of the resin blend. Subsequently, the resin blend was

made light-curable by a binary photoinitiator system based on

camphoroquinone (CQ; Sigma-Aldrich) and 1,2-ethyl-dimethyl-4-aminobenzoate

174

(EDAB; Sigma-Aldrich). This resin co-monomer blend was used to formulate the

experimental primer and the bonds used in this study (Table 4.1).

A BAG (Sylc; OSspray, London, UK) with particle size < 10 μm was employed

in the etch-and-rinse bonding procedures using two different experimental

approaches: (i) BAG-AD (30 wt% BAG included within the composition of a

resin adhesive as a bioactive microfiller), and (ii) BAG-PR (BAG applied directly

onto H3PO4-etched/wetted dentine before bonding procedures). The neat

adhesive, with no BAG, served as the control (RES-Ctr) (Fig. 4.1).

In detail, a water wet-bonding dentine substrate was achieved by water-rinsing,

for 15 s, the dentine surfaces acid-etched with 37% phosphoric acid solution

(H3PO4) (Sigma-Aldrich) and gently blowing off (for 2 s) excess water to leave a

wet reflective-surface.

The control bonding procedure (RES-Ctr) was accomplished by applying two

consecutive coats of an ethanol-solvated resin primer [50 wt% absolute ethanol

(Sigma-Aldrich) and 50 wt% of neat co-monomer resin blend] and a layer of the

neat co-monomer resin blend (Table 4.1) within a period of 20 s. Light-curing

was immediately performed for 30 s (> 600 mW/cm-2, Optilux VLC; Demetron,

Danbury, CT, USA).

The first experimental bonding procedure (BAG-AD) was performed by applying

the same ethanol-solvated resin primer onto H3PO4-etched dentine, as

previously described, followed by a layer of bonding resin containing BAG

(Table 4.1; Fig. 4.1). The bonding and the light-curing procedures were

executed as previously described for the RES-Ctr group.

The second experimental bonding procedure (BAG-PR) was performed as

follows. The 37% H3PO4 solution (Sigma-Aldrich) was applied onto the dentine

175

surface for 15 s. Then, 0.05 g of BAG powder was placed onto the H3PO4-

etched wet dentine surface, spread immediately for 10 s using a cotton pellet,

and finally rinsed with copious amounts of deionised water for 15 s (Fig. 4.1).

The primer/bond application and the light-curing procedures were performed as

previously described for the RES-Ctr group.

A final composite build-up (5 mm) was constructed on each specimen using a

light-cured resin composite (Filtek Z250; 3M-ESPE, St Paul, MN, USA) in five

incremental layers (of 1 mm thickness). Each layer of composite was

individually light cured for 20 s. The resin-bonded dentine specimens were

stored in PBS for 24 h or 6 months at 37°C. The PBS was composed of (in g/l)

CaCl2 (0.103), MgCl2.6H2O (0.019), KH2PO4 (0.544), KCl (17), and HEPES

(acid) buffer (4.77), and the pH was 7.4.

176

Figure 4.1 - Schematic illustrating the experimental study design. Human third molars were used to prepare standardised dentine surfaces. The three different bonding approaches were performed using specific components and application procedures. Abbreviations. Bis-GMA: bisphenyl-A-glycidyl methacrylate; HEMA: 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid.

177

Table 4.1 - Composition of the experimental bonding procedures/adhesive systems used in this study. Abbreviations. Bis-GMA: bisphenyl A glycidyl methacrylate; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid. At the end of the formulation of the resins, 0.25 wt% camphoroquinone (CQ) and 1.0 wt% 2-ethyl-dimethyl-4-aminobenzoate (EDAB) were added to the resin mixture.

178

4.2.3 μTBS and SEM fractography and failure analysis

Twenty dentine-bonded specimens from each group were sectioned using a

slow-speed water-cooled diamond wafering blade (Struers) mounted on a hard-

tissue microtome (Isomet 11/1180; Buehler) in both x and y directions across

the adhesive interface to obtain matchsticks with cross-sectional areas of 0.9

mm2. By excluding peripheral beams showing the presence of residual enamel,

only the remaining matchsticks (n = 10-15) were selected to create three groups

with the same total number of resin-dentine specimens in each group (n = 280).

The exact width of each matchstick was checked using a calliper (Mitutoyo

CD15; Mitutoyo, Kawasaki, Japan) and half of them (n = 140) were tested after

24 h of storage in PBS and the remaining half (n = 140) were tested after 6

months of storage in PBS at 37°C. The μTBS test was performed using a

microtensile jig in a LAL300 linear actuator (SMAC Europe; Horsham, UK) with

a LAC-1 high-speed controller single axis with a built-in amplifier and at the

following settings: stroke length = 50 mm, peak force = 250 N, displacement

resolution = 0.5 mm, and crosshead speed = 1 mm min-1. Bond-strength data

were calculated and expressed in MPa, the μTBS values of sticks from the

same restored teeth were averaged, and the mean bond strength was used as

one statistical unit for the statistical analysis. The μTBS (mean-MPa) data for

each group were analysed using a repeated-measures ANOVA and Tukey’s

post-hoc test for pairwise comparisons (α = 0.05).

The mode of failure was classified as percentage of adhesive, mixed, or

cohesive. The failed bonds were examined at x30 magnification using a

stereoscopic microscope (Leica M205A; Leica Microsystems, Wetzlar,

Germany).

179

Five representative debonded specimens for each group that failed in mixed or

adhesive modes were selected for ultramorphology analysis of the fractured

surface (SEM Fractography). They were dried overnight and mounted on

aluminium stubs with carbon cement, then sputter-coated with gold (SCD 004

Sputter Coater; Bal-Tec, Vaduz, Liechtenstein) and examined using SEM (S-

3500; Hitachi, Wokingham, UK) with an accelerating voltage of 15 kV and a

working distance of 25 mm at increasing magnifications.

4.2.4 Confocal microscopy ultramorphology and nano-leakage evaluation

A further three dentine specimens from each group were bonded, as previously

described, with the primer/bond resins doped with 0.1 wt% rhodamine-B (Rh-B:

Sigma-Aldrich, St Louis, MO, USA) and employed for the confocal microscopy

analysis (Sauro et al., 2012a, Sauro et al., 2012b). The specimens were serially

sectioned across the adhesive interface to obtain resin-dentine slabs (of 1 mm

thickness). The resin-dentine slabs (n = 10 per group) were then divided into

two subgroups based on the period of storage in PBS (24 h or 6 months) (Fig.

4.2). Subsequent to the storage period, the specimens were coated with two

layers of fast-setting nail varnish applied 1 mm from the resin-dentine interfaces

and immersed in 1 wt% aqueous fluorescein (Sigma-Aldrich) solution for 24 h.

The specimens were then treated in an ultrasonic water bath for 2 min and

polished using SiC abrasive papers of ascending grit (#1200 to #4000)

(Versocit; Struers) on a water-cooled rotating polishing machine (Buehler Meta-

Serv 3000 Grinder-Polisher; Buehler). A final treatment in an ultrasonic water

bath (5 min) completed the specimen preparation for the confocal microscopy

evaluation (Fig. 4.2).

180

The microscopy examination was performed using a confocal laser scanning

microscope (Leica SP2 CLSM; Leica, Heidelberg, Germany) equipped with a 63

X /1.4 NA oil-immersion lens and using 488-nm argon/helium (fluorescein

excitation) or 568-nm krypton (rhodamine excitation) laser illumination. The

reflection imaging was performed using the argon/helium laser. Confocal laser

scanning microscopy reflection and fluorescence images were obtained with a

1-μm z-step to section optically the specimens to a depth up to 20 μm below the

surface (Sauro et al., 2012a).

The z-axis scans of the interface surface were arbitrarily pseudo-coloured by

two selected operators and compiled into single projections using the Leica

image-processing software (Leica). The configuration of the system was

standardised and used at the same settings for the entire investigation. Each

resin-dentine interface was completely investigated and then five optical images

were randomly captured. Micrographs representing the most common features

of nano-leakage observed along the bonded interfaces were captured and

recorded (Sauro et al., 2012b).

181

Figure 4.2 - Schematic illustrating the composite-tooth matchsticks (1 mm) prepared using a water-cooled diamond saw, stored in PBS for 24 h or 6 months, and then subjected to microtensile bond strength (μTBS) testing and scanning electron microscopy failure analysis. This schematic also illustrates how composite-tooth slabs were prepared, stored in PBS for 24 h or 6 months, and evaluated by confocal laser scanning microscopy.

182

4.3 Results

4.3.1 μTBS and SEM fractography and failure analysis

The BAG-bonding technique versus storage time was statistically significant

only for the BAG-AD group (P = 0.001); no significant reduction of the μTBS

values was observed after 6 months of storage in PBS (P > 0.05). On the other

hand, significant μTBS reductions were observed in both the BAG-PR and

RES-Ctr groups (P < 0.05) after prolonged storage in PBS (6 months). The

μTBS results (expressed as Mean and SD) are presented in Table 4.2.

High μTBS values were achieved in all groups after 24 h of storage in PBS, with

failures occurring mainly in cohesive mode in all groups; in contrast, important

changes in the μTBS were observed after 6 months of storage in PBS. For

instance, the μTBS of specimens in the RES-Ctr group (no BAG) showed a

significant (P < 0.05) decrease after 6 months of storage in PBS and failed

mostly in adhesive mode (66%). The specimens stored for 24 h in PBS that

fractured in mixed mode were characterised by the presence of exposed

dentinal tubules with spare extruded resin tags (Fig. 4.3A2). Conversely, the

surface of the specimens that failed in adhesive mode after 6 months of storage

in PBS presented several ‘funnelled’ dentinal tubules with no exposed collagen

fibrils (Fig. 4.3A3). The resin-dentine specimens of the BAG-AD group

maintained a high μTBS (P > 0.05) after 6 months of storage in PBS (23.89 ±

7.74 MPa). In these specimens the failure was prevalent in cohesive (43%) and

mixed (40%) modes (Fig. 4.3B1) and the SEM analysis of the fractured surface

revealed a dentine surface predominantly covered by residual resin and mineral

crystals embedded within a resin/collagen network (Fig. 4.3B3).

183

The specimens of the BAG-PR group, where the BAG powder was applied onto

acid-etched/wetted dentine before application of the adhesive system, showed

a significant decrease in μTBS values (P < 0.05) after prolonged storage in PBS

(Table 4.2). These specimens failed mainly in adhesive mode (56%) after 6

months of storage in PBS, and the SEM fractographic analysis showed that the

fracture during μTBS testing occurred along the intertubular dentine, leaving an

intact peritubular dentine and a consistent precipitation of mineral inside the

dentinal tubules (Fig. 4.3C3).

184

Table 4.2 - Means and standard deviations (SD) of the microtensile bond strength values (MPa) obtained for the different experimental groups and percentage distribution of failure modes after microtensile bond strength testing; total number of beams (tested stick/pre-load failure). For each horizontal row: values with identical numbers indicate no significant difference. For each vertical column: values with identical letters indicate no significant difference using Student-Newman-Keuls test (P > 0.05).

185

Figure 4.3 - Scanning electron microscopy images of failure modes of the resin-bonded specimens created using the three different bonding approaches tested. (A) Micrograph of the failure mode (cohesive) of the resin control bonded to etched dentine (37% H3PO4) after 24 h of storage in PBS (A1). At higher magnification (A2) it was possible to observe the presence of some exposed dentinal tubules (t), but most remained obliterated by resin tags (rt). No exposed collagen fibrils were visible on the dentine surface, and a well resin-hybridised hybrid layer was present (pointer). At 6 months (A3), the resin-dentine interfaces created with the control resin (RES-Ctr; containing no bioactive filler) showed only a few resin tags inside the dentinal tubules and no collagen fibrils were visible on a dentine surface characterised by funnelled dentinal tubules (pointer). (B) Micrograph of the failure mode (mixed) of the calcium/sodium phosphosilicate-containing adhesive (BAG-AD) bonded to dentine, after 6 months of storage in PBS (B1). At higher magnification (B2) no exposed dentinal tubules or exposed collagen fibrils were observed; the dentine surface was well resin-hybridised (pointer). After 6 months of storage in PBS (B3), the debonded resin-dentine interface showed the presence of resin tags remaining inside the dentinal tubules (rt) and mineral crystals embedded within a preserved collagen network (pointer). (C) Micrograph of the failure mode (adhesive) of BAG applied directly onto H3PO4-etched/wetted dentine before bonding (BAG-PR) after 24 h of storage in PBS (C1). At higher magnification (C2) it was possible to observe the presence of some exposed dentinal tubules (t), while most remained obliterated by resin tags (rt) containing few BAG particles. No exposed collagen fibrils were present on the dentine surface (pointer). At 6 months testing (C3), the resin-dentine interface created with the BAG-PR showed a dentine surface characterised by the presence of remineralised dentinal tubules (t) obliterated by mineral crystals (cr). It is interesting to note how the fracture occurred along the intertubular dentine leaving an intact peritubular dentine around the mineral-obliterated dentinal tubule (pointer).

186

4.3.2 Confocal microscopy ultramorphology and nano-leakage evaluation

The CLSM investigation showed that all the bonding procedures used in this

study were able to create a resin diffusion within the demineralised dentine

(hybrid layer 7-9 μm) and several resin tags into the dentinal tubules (Fig. 4.4A).

Nevertheless, the resin-dentine interfaces of the specimens created in the three

groups showed evident fluorescein penetration (nano-leakage) within the hybrid

layer and along the dentinal tubules after 24 h of storage in PBS (Fig. 4.4B,C).

The experimental bonding approach used to bond the specimens of the BAG-

PR group created resin-dentine interfaces characterised by the presence of

mineral deposits inside the dentinal tubules and within the hybrid layer (Fig.

4.4D).

The prolonged storage in PBS induced important changes in terms of

ultramorphology and nano-leakage.

For instance, the resin-dentine interface of the RES-Ctr group specimens was

affected by severe nano-leakage within the hybrid layer and the presence of a

continuous gap between dentine and composite (Fig. 4.5A).

Conversely, the specimens of the BAG-AD group showed the presence of a

strong reflective mineral material and partial dye penetration within the hybrid

layer (Fig. 4.5B). The resin-dentine interface of specimens in the BAG-PR group

was affected by partial dye penetration within a crystallised hybrid layer.

However, gaps were also observed between the hybrid and adhesive layers

(Fig. 4.5C), probably caused by the sample preparation procedure before the

CLSM analysis.

187

Figure 4.4 - Confocal laser scanning microscopy (CLSM) images showing the interfacial characterisation and nanoleakage, after 24 h of storage in PBS, of the resin-dentine interfaces created using the three different bonding approaches tested. (A) Confocal laser scanning microscopy three-dimensional (3D) single-projection (fluorescence mode) image exemplifying the interfacial characteristics of the resin-dentine interface created using the control adhesive system (RES-Ctr) applied onto H3PO4-etched dentine. It is possible to observe a clear hybrid layer (hl) (approximate thickness 9 μm) (pointer) located underneath a thick adhesive layer (a) and long resin tags (rt). (B) This CLSM 3D single-projection (fluorescence/reflection mode) image of the resin-dentine interface created using the bioactive calcium/sodium phosphosilicate-containing adhesive (BAG-AD) shows an intense nanoleakage signal from the hybrid layer (pointer) located underneath a thick adhesive layer (a) characterised by the presence of BAG microfiller (fl). The presence of long resin tags (rt) is also evident. (C) The resin-dentine interface created using the bonding procedure where the BAG is applied directly onto H3PO4-etched/wetted dentine (BAG-PR) shows evident dye penetration within the hybrid layer (pointer). Short resin tags (rt) are visible underneath a thick adhesive layer. The reason why only short resin tags could be created during this type of bonding procedure is shown in (D) where it is possible to observe a strong reflective signal from the demineralised dentine layer (pointer) and inside the dentinal tubules (t), indicating the presence of mineral particles.

188

Figure 4.5 - Confocal laser scanning microscopy (CLSM) images showing the interfacial characterisation and nanoleakage, after 6 months of storage in PBS, of the resin-dentine interfaces. (A) Confocal laser scanning microscopy three-dimensional single-projection (fluorescence/reflection mode) image of the resin-dentine interface created using the control adhesive system (RES-Ctr) applied onto H3PO4-etched dentine. It is possible to note the presence of evident dye diffusion (nanoleakage) within the hybrid layer and inside the dentinal tubules (pointer). A gap (g) is present between the dentine and the composite (c). (B) The resin-dentine interface created using the bonding approach where the bioactive calcium/sodium phosphosilicate-containing adhesive (BAG-AD) is applied onto H3PO4-etched dentine shows partial dye diffusion within a hybrid layer characterised by a strong reflective signal (pointer). (C) The resin-dentine interface created using the bonding procedure where the BAG is applied directly onto H3PO4-etched/wet dentine (BAG-PR) shows a crystallised reflective layer (pointer) characterised by low dye penetration (nanoleakage). A pronounced gap (g) can be seen between the adhesive layer (a) and the composite (c). It is also possible to observe the remaining reflective mineral materials on the fractured edge of the adhesive layer (arrows).

189

4.4 Discussion

Hybrid layers created using etch-and-rinse adhesives include water-rich, resin-

sparse regions that account for 2-3% of their entire volume, which increase

subsequent to prolonged ageing in fluids (Reis et al., 2007). The water-rich,

resin-sparse regions represent essentially the nanoporosities within the

demineralised collagen fibrils, created during adhesive application as a result of

incomplete replacement of water by resin infiltration (Pashley et al., 2011). This

undisplaced water may act as a functional medium for the hydrolysis of

suboptimally polymerised resin matrices by esterases and denaturation of

collagen via the activation of host-derived matrix metalloproteinases (MMPs),

jeopardising the durability of the resin-dentine interfaces (Pashley et al., 2011,

Pashley et al., 2004, Breschi et al., 2010b).

Several methods have been advocated to increase the longevity of these resin-

dentine interfaces, including the inhibition of the MMPs within the hybrid layer

(Pashley et al., 2004, Breschi et al., 2010b) and enhancement of the resin

infiltration within the demineralised collagen fibril using more hydrophobic resin

monomers and ethanol wet-bonding (Pashley et al., 2011).

Based on the results obtained in this study, the first null hypothesis must be

partially rejected because the use of BAG produced bioactive/protective effects

on the bond strength only when used as resin microfiller within the adhesive

composition. The second null hypothesis must be totally rejected as both the

experimental bonding approaches based on the use of BAG were able to

reduce the nano-leakage within the demineralised ‘poorly infiltrated’ areas

within the resin-dentine interface.

190

In detail, the control bonding procedure (RES-Ctr) and the two experimental

bonding approaches used (BAG-AD and BAG-PR) to bond the acid-etched

dentine produced comparably high μTBS values after 24 h of storage in PBS

(Table 4.2). Conversely, a significant decrease in μTBS (P > 0.05) occurred in

all groups after storage in PBS for 6 months, except for the specimens bonded

using the resin adhesive containing BAG microfiller (BAG-AD).

The SEM analysis of the fractured specimens of the RES-Ctr group showed,

after 24 h of storage in PBS, a dentine surface characterised by a hybrid layer

that was highly hybridised with resin and no presence of demineralised collagen

fibrils exposed (Fig. 4.3A2). Conversely, these resin-dentine specimens stored

for 6 months in PBS had a fractured surface characterised by ‘funnelled’

dentinal tubules, indicating degradation of the demineralised peritubular dentine

(Fig. 4.3A3). In contrast, the bonded-dentine specimens of the BAG-AD group

immersed in PBS for 6 months had a fractured (adhesive mode) dentine

surface, with mineral crystals embedded within a preserved collagen network

and no evidence of ‘funnelled’ dentinal tubules (Fig. 4.3B3).

The SEM ultramorphology analysis of the fractured specimens (adhesive mode)

of the RES-PR group stored for 24 h in PBS demonstrated the presence of

dentinal tubules obliterated by resin tags and no exposed collagen fibrils (Fig.

4.3C2). Interestingly, when this type of dentine-bonded specimen was

immersed in PBS for 6 months it was possible to detect a fractured dentine

surface characterised by dentinal tubules obliterated by mineral crystals and a

distinctive fracture along the intertubular dentine, which left an intact peritubular

dentine (Fig. 4.3C3).

191

Possible explanations for such longevity attained in dentine-bonded specimens

created using BAG-AD after prolonged storage in PBS may be as follows:

I. The presence of BAG within the resin-dentine interface may have

induced the release of a silicic acid, such as Si(OH)4, and a subsequent

polycondensation reaction between the silanols compounds and the

demineralised collagen via electrostatic, ionic, and/or hydrogen bonding

(Vollenweider et al., 2007, Välimäki et al., 2005, Zhong et al., 1994),

which interfered with the ability of MMPs - although BAG is not a direct

MMP inhibitor - to execute their collagenolytic and gelatinolytic activities.

A study by Osorio et al. (Osorio et al., 2011) showed that it is possible to

reduce the collagen-degradation process by using specific chemical

compounds, such as zinc oxide, which interfere with the zinc-binding and

calcium-binding catalytic domains of MMPs.

II. The precipitation of an amorphous calcium phosphate (ACP) on the

polycondensate SiO2-rich template of nucleation (Hench and Andersson,

1993, Shen et al., 2001, Vollenweider et al., 2007, Sauro et al., 2012a)

induced by the dissolution and immediate reaction between Ca2+ and

PO43- species from BAG may have also favoured the formation of a high-

molecular-weight complex (Ca/P-MMPs), which restricted the activities of

MMP-2 and MMP-9 within the hybrid layer (Kremer et al., 1998).

However, the ability of specific bioactive glass, such as Bioglass® 45S5,

to modulate and/or reduce the presence of collagens I, II, and III,

osteocalcin, osteonectin, and osteopontin, has also been demonstrated

in bone-regeneration studies (Välimäki et al., 2005).

192

III. The release of Na+ and Ca2+ ions from BAG, and the incorporation of

H3O+ protons into the glass particles, may have created an optimal

alkaline environment (Shen et al., 2001, Sauro et al., 2012a) within the

resin-dentine interface that interfered with the activity of MMPs, which

are very acidic-pH dependent (Pashley et al., 2004, Breschi et al.,

2010b).

IV. The bioactive remineralisation induced by BAG may have decreased the

distribution of the water-rich, resin-sparse regions within the hybrid layer

(Ryou et al., 2011, Sauro et al., 2012a) via silanols polycondensation

and subsequent ACP/AH remineralisation, which probably interfered with

the water-dependent hygroscopic and hydrolytic degradation of the

polymer network (Ferracane, 2006).

The confocal microscopy evaluation performed after 6 months of storage in

PBS indicated that both the experimental bonding approaches used in this

study (BAG-AD and BAG-PR) created a resin-dentine interface affected only by

partial dye penetration (nano-leakage) within a hybrid layer characterised by the

deposition of a strong reflective mineral (Fig. 4.5B,C).

Whereas it is reasonable to believe that the hybrid layer of the specimens

created using the BAG-AD approach remineralised as a result of the

bioactive/biomimetic activity of Bioglass® 45S5 after prolonged (6 months)

storage in PBS (Bakry et al., 2011a, Sauro et al., 2012a, Zhong et al., 1994), a

completely different bioactive phenomenon may have occurred within the resin-

dentine interface created by directly applying the BAG on the demineralised

193

H3PO4-wetted dentine, as a significant decrease of μTBS was attained after

prolonged storage in PBS (Table 4.2).

In this case, a possible explanation for the reduced confocal nano-leakage may

be due to the chemical nature of mineral precipitation that occurred within the

resin-dentine interface created as a result of the experimental bonding

procedures (BAG-PR). Our hypothesis is that the chemical reaction between

BAG and H3PO4 solution (Fig. 4.4D) may have induced the precipitation of

dicalcium-phosphate salts (i.e. brushite and monetite). Bakry et al. (Bakry et al.,

2011a, Bakry et al., 2011b) showed that the acid-base chemical reaction

between BAG and H3PO4 may induce the formation of brushite via combination

of the phosphate (released from the BAG and H3PO4) and calcium ions

(released from BAG and etched dentine). The precipitation reaction of the

brushite may be responsible for the creation of an acidic environment (Mandel

and TAS, 2010), which may have evoked the activation of MMPs (Pashley et

al., 2004, Breschi et al., 2010b); this situation is also aggravated by the fact that

BAG no longer has the ability to create a localised ‘protective’ alkaline pH within

the resin-dentine interface.

Moreover, it is also possible that the BAG/ H3PO4 reaction may have altered the

chemical and/or physical characteristics of BAG, in particular those responsible

for the polycondensation of silanols and ACP/HA precipitation (Vollenweider et

al., 2007, Välimäki et al., 2005, Zhong et al., 1994), which may be fundamental

in altering the activity of MMPs (Kremer et al., 1998), as previously described.

However, even supposing that the reaction between hydroxyl ions and Si(OH)4

formed the silanols compounds and induced the polycondensation reaction,

194

they may have been washed out by application of the air-water jet before

application of the primer and bond (Bunker et al., 1988).

Furthermore, a slightly acidic environment may have remained in loco within the

resin-dentine interface during the prolonged storage in PBS as a result of the

release of H+ from the acidic monomer (2,5-dimethacryloyloxyethyloxycarbonyl-

1,4-benzenedicarboxylic acid) contained within the resin adhesives (Wang and

Spencer, 2005, Sauro et al., 2011b, Bayle et al., 2007), causing a long-

standing, MMP-mediated degradation of collagen in both the RES-Ctr and BAG-

PR groups. In addition, a durable acidic environment may have induced

supplementary precipitation of dicalcium or octocalcium phosphates

(Jayaraman and Subramanian, 2002, Mandel and TAS, 2010) during buffered

condition (replacement of PBS) within the microporosities generated by the

degradation of the dentine collagen fibrils (Fig. 4.5C). Indeed, as a result of this

probable additional precipitation of mineral over time, the interface created

using the BAG-PR bonding technique may have achieved mechanical

characteristics similar to those created using glass ionomer cements (GICs)

applied onto polyacrylic acid-treated dentine (Sauro et al., 2012a, Liu et al.,

2009, Hewlett et al., 1991). This is probably why bond strength reduction and

gap formation were observed in the BAG-PR specimens. The GIC-bonded

interfaces can reach a tensile or shear bond strength of approximately 5 MPa

and frequently prefail during specimen preparation (Hewlett et al., 1991, Berry

and Powers, 1994).

Yip et al. (Yip et al., 2001) affirmed that the results obtained from tensile testing

of GICs bonded to dentine do not represent the actual strength of such stiff

bonded interfaces and that only an accurate ultramorphology analysis using

195

electron microscopy may reveal the proper bonding ability of such restorative

materials.

However, it is also important to consider that the hydrophilic characteristics

conferred by specific resin monomers, such as HEMA and PMDM, within the

tested adhesives (Fig. 4.1) may have compromised the mechanical properties

(i.e. modulus of elasticity) of the hybrid layers (Giannini et al., 2004, Bedran-

Russo et al., 2007) as a result of polymer hydrolysis and swelling tensions

generated within the polymer chains. In contrast, the BAG microfiller contained

within the adhesive used in the BAG-AD group may have absorbed and used

the water not required by the hydrophilic/acid monomers for the bioactive

processes of conversion into apatite (Sauro et al., 2012a), thus preventing the

polymer network from considerable hygroscopic/hydrolytic degradation

(Ferracane, 2006).

4.5 Conclusion

In conclusion, this study provided preliminary evidence for the use of bioactive

Ca/Na phosphosilicate, such as Bioglass® 45S5, in dentine-bonding

procedures in order to enhance the durability of the resin-dentine interfaces.

However, further in vitro (i.e. transmission electron microscopy as well as

computer-controlled indentation techniques) and long-term clinical studies are

required to confirm the protective/therapeutic effects of BAG on the resin-

dentine interface. Confocal Raman analysis will be also necessary to confirm

the chemical nature of the mineral precipitates observed within the bonded-

dentine interfaces created with the two experimental BAG-bonding procedures.

196

Chapter 5: Experimental etch-and-rinse

adhesives doped with calcium silicate-based

micro-fillers to generate therapeutic bioactivity

within resin-dentine interfaces

197

5.1 Introduction

The durability of resin-dentine interfaces represents one of the main concerns in

the contemporary adhesive dentistry as they are affected by severe degradation

processes associated with water contact. Bond degradation occurs via water

sorption (Ito et al., 2005), hydrolysis of monomer methacrylates ester bonds

caused by salivary esterases (Moszner et al., 2005), and hydrolysis of collagen

fibrils which may be enhanced by activation of endogenous dentine matrix

metalloproteinases (MMPs) (Pashley et al., 2004). Regarding the different

mechanisms of degradation, strategies to preserve the intact hybrid layers such

as ethanol-wet bonding (Sadek et al., 2007, Tay et al., 2007) and the use of

MMP inhibitors (Carrilho et al., 2007a) have been proposed. Nevertheless,

current attempts to extend the longevity of resin-dentine bonds via incorporation

of more hydrolytically stable resin monomers (Moszner et al., 2006) and/or the

use of matrix metalloproteinase inhibitors (Carrilho et al., 2010) fail to address

two fundamental issues in the degradation of resin-dentine bonds: 1)

replacement of the mineral phase within the demineralised dentine collagen; 2)

protection of the collagen from biodegradation (Bakry et al., 2011b).

The use of bioactive materials which promptly interact with dental hard tissues

through therapeutic/protective effects may provide a feasible means to extend

the longevity of resin-dentine bonds (Profeta et al., 2012). Furthermore,

experimental resin-based calcium-phosphate cements have been advocated as

potential therapeutic restorative base-liner materials due to their ability to induce

remineralisation of caries-affected dentine-bonded interfaces (Peters et al.,

2010). Nevertheless, alternative strategies are being developed in order to

enhance calcium, hydroxyl, and phosphate ions delivery within and beneath the

198

hybrid layers. Calcium-silicate cements are Portland-derived cements able to

release calcium and hydroxyl ions (alkalinising activity) (Huan et al., 2008,

Coleman et al., 2009), so creating favourable conditions for the remineralisation

of dental hard tissues (i.e. dentine and enamel). These materials possess a

bioactive behavior since they are able to develop apatite on their surface in a

short induction period (Taddei et al., 2011) eliciting a positive response at the

interface from the biological environment (Parirokh and Torabinejad, 2010).

However, the use of the Portland cement-based materials in operative dentistry

is still debated due to clinical limitations related to their long setting time (Taddei

et al., 2011, Mandel and TAS, 2010), high dissolution rate and poor mechanical

properties (Bunker et al., 1988). In contrast, the incorporation of resin specific

monomers such as 2-hydroxyethyl methacrylate (HEMA), triethyleneglycol

dimethacrylates (TEGDMA) and urethane dimethacrylates (UDMA) in silicate-

based materials has been proposed to improve the mechanical properties, bond

strength to dental tissues and reduce the setting time (Taddei et al., 2011,

Wang and Spencer, 2005).

Since there is little information concerning the use of such “hybrid” resin-base

light-curable adhesive materials, this study was purposed to assess the

therapeutic/bioactive effects of three newly developed experimental bonding

agents containing modified Portland cement-based micro-fillers on resin-dentine

interfaces. This aim was accomplished by evaluating the microtensile bond

strength (µTBS) after simulated body fluid solution (SBS) storage (24 h or 6

months). Scanning electron microscopy (SEM) fractography on the de-bonded

specimens and confocal microscopy (CLSM) analysis of the ultramorphology

and nano-leakage of the resin-dentine interface were executed. The null

199

hypotheses to be tested were that the inclusion of tested micro-fillers within the

composition of the experimental bonding agent induces: (i) no effect on the

bond strength durability; (ii) no mineral precipitation and nano-leakage reduction

within the demineralised ‘poorly resin-infiltrated’ areas within the resin-dentine

interface.

5.2 Materials and methods

5.2.1 Preparation of the experimental bioactive resin-base bonding agents

A type I ordinary Portland cement (82.5 wt%), (OPC: Italcementi Group,

Cesena, Italy) mainly consisting of tri-calcium silicate (Alite: 3CaO x SiO2), di-

calcium silicate (Belite: 2CaO x SiO2), tri-calcium aluminate (3CaO x Al2O3) and

gypsum (CaSO4 x 2H2O) was mixed with 7.5 wt% of phyllosilicate consisting of

sodium-calcium-aluminum-magnesium silicate hydroxide hydrate

[(Na,Ca)(Al,Mg)6(Si4010)3(OH)6-nH2O; Acros Organics, Fair Lawn, NJ, USA] in

deionised water (Ratio 2:1) to create the first experimental filler (HOPC). The

second experimental filler (HPCMM) was created by mixing 90 wt% of type I

OPC, 7.5 wt% phyllosilicate and 2.5 wt% of hydrotalcite consisting of aluminum-

magnesium-carbonate hydroxide hydrate [(Mg6Al2 (CO3)(OH)16·4(H2O); Sigma-

Aldrich]. The third filler (HPCTO) used in this study was created by mixing OPC

(80 wt%), phyllosilicate (7.5 wt%), hydrotalcite (2.5 wt%) and 10 wt% titanium

oxide (TiO2: Sigma-Aldrich, Gillingham, UK). The three modified Portland-base

silicates were mixed with deionised water (Ratio 2:1) and allowed to set in

incubator at 37°C for 24 h. Subsequently, they were ground in an agate jar and

sieved to obtain < 30 μm-sized micro-filler particles.

200

A resin co-monomer blend was prepared as a typical three-step, etch-and-rinse

bonding agent including a neat resin blend as bond and a 50 wt% ethanol-

solvated resin mixture as primer (Res-Ctr - no filler).

The neat resin blend was formulated by using 40 wt% of a hydrophobic cross-

linking dimethacrylate 2,2-bis[4(2-hydroxy-3-methacryloyloxy-propyloxy)-

phenyl]-propane (Bis-GMA; Esstech, Essington, PA, USA) and 28.75 wt% of

hydrophilic 2-hydroxyethyl methacrylate (HEMA; Aldrich Chemical, Gillingham,

UK). An acidic functional monomer 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-

benzenedicarboxylic acid (PMDM: Esstech Essington) was also added (30 wt%)

to the blend solution to obtain a dental bonding system with chemical affinity to

Calcium (Ca2+) present in dentine and in the micro-fillers (Fig. 5.1). The neat

resin was made light-curable by adding 0.25 wt% of camphoroquinone (CQ;

Aldrich), 0.5 wt% of 2-ethyl-dimethyl-4-aminobenzoate (EDAB; Aldrich) and

0.5% diphenyliodonium hexafluorophosphate.

The resin co-monomer blend was used as control filler-free or mixed with each

micro-filler in order to formulate three experimental resin-based agents (GB

patent application No 1118138.5 - filed on 20th October 2011): i) Res-HOPC:

60 wt% of neat resin and 40 wt% of HOPC; ii) Res-HPCMM: 60 wt% of neat

resin and 40 wt% of HPCMM; iii) Res-HPCTO: 60 wt% of neat resin and 40 wt%

of HPCTO filler (Table 5.1). The hybrid calcium silicate-based adhesives

systems were prepared by mixing the neat resin and the fillers for 30 s on a

glass plate to form a homogeneous paste prior the bonding procedures.

201

Figure 5.1 - Chemical structures of the methacrylate monomers used in the tested resin blends. Abbreviations. Bis-GMA: Bisphenol A diglycidyl methacrylate; HEMA: 2-hydroxyethyl methacrylate; TEGDMA: triethylene-glycol dimethacrylate; PDMD: Bis(2-Methacryloyloxyethyl) Pyromellitate.

202

Table 5.1 - Chemical composition (wt%) and application mode of the experimental adhesive system used in this study. Abbreviations. Bis-GMA: bisphenyl A glycidyl methacrylate; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid; HOPC: set Portland cement and smectite; HPCMM: Portland cement, Smectite and Hydrotalcite; HPCTO: set Portland cement, Smectite, Hydrotalcite and Titanium Oxide.

*Three discs for each experimental resin-base material (6 mm in diameter and 1 mm thick) were light-cured for 30 s, immersed in 25 ml of H2O (pH 6.7) at 37°C and maintained for 30 days; the pH/alkalinising activity was evaluated using a professional pH electrode (Mettler-Toledo, Leicester, UK) at room temperature (~ 24° C)

203

5.2.2 Specimen preparation and bonding procedures

Caries-free human molars (age 20-40 yr.), extracted for periodontal reasons

under a protocol approved by the institutional review board of Guy's Hospital

(South East London ethical approval 10/H0804/056), were used in this study.

The treatment plan of any of the involved patients, who had given informed

consent that their extracted teeth could be used for research purposes, was not

altered by this investigation. This study was conducted in accordance with the

ethical guidelines of the Research Ethics Committee (REC) for medical

investigations.

The teeth were stored in deionised water (pH 7.1) at 4°C and used within 1

month after extraction. A flat mid-coronal dentine surface was exposed using a

hard tissue microtome (Accutom-50; Struers, Copenhagem, Denmark)

equipped with a slow-speed, water-cooled diamond wafering saw (330-CA RS-

70300; Struers). The roots were sectioned 1 mm beneath the cemento-enamel

junction (CEJ) using the slow-speed diamond saw. A 180-grit silicon carbide

(SiC) abrasive paper mounted on a water-cooled rotating polishing machine

(Buehler Meta-Serv 3000 Grinder-Polisher, Düsseldorf, Germany) was used (30

s) to remove the diamond saw smear layer and to replace it with a standard and

more clinically related smear layer (Koibuchi et al., 2001). The specimens were

divided into four groups (n = 5/group) based on the tested materials (Table 5.1).

The specimens were etched using a 37% phosphoric acid solution (H3PO4;

Aldrich Chemical) for 15 s followed by a copious water rinse. The etched-

dentine surfaces were gently air-dried for 2 s to remove the excess of water and

leave a wet reflective-surface. The control and experimental adhesives (Res-

Ctr; Res-HOPC; Res-HPCMM; Res-HPCTO) were applied within a period of 20

204

s. The specimens were immediately light-cured for 30 s using a quartz-

tungsten-halogen (QTH) lamp (600mW/cm-2, Optilux VLC; Demetron, CT,

USA). Five 1-mm-thick incremental build-up were performed using a resin

composite (Filtek Z250; 3M-ESPE, St Paul, MN, US) light-activated for 20 s

each step with a final curing of 60 s (Figure 5.2). The specimens were finally

stored in SBS solutions (Oxoid, Basingstoke, Hampshire, UK) for 24 h and 6

months at 37°C.

Figure 5.2 - Schematic illustrating the resin-dentine match-sticks prepared using a water-cooled diamond saw, stored in SBS for 24 h or 6 months, and then subjected to microtensile bond strength (µTBS) testing and scanning electron microscopy fractography. This schematic also illustrates how composite-tooth slabs were prepared, stored in SBS for 24 h or 6 months, immersed in fluorescein (nanoleakage) or Xylenol Orange (Calcium-binding dye) and finally evaluated by confocal laser scanning microscopy (CLSM).

205

5.2.3 μTBS and SEM observations of the failed bonds

The specimen of each group were sectioned perpendicular to the adhesive

interface with a slow speed water-cooled diamond wafering blade (Accutom-50;

Struers) mounted on a hard tissue microtome (Isomet 11/1180; Buehler).

Subsequently, match-sticks with cross-sectional adhesive area of 0.9 mm2 were

created (Fig. 5.2). As each tooth yielded 16 beams, there were 80 match-sticks

in total per bonding material. For each group, half of the match-sticks (n = 40)

were tested after 24 h and the remaining half (n = 40) after 6 months of SBS

storage at 37°C. Each resin-dentine match-stick was attached to a testing

apparatus with a cyanoacrylate adhesive (Zapit, Dental Ventures, CA, USA). A

tensile load was applied with a customised micro-tensile jig in a LAL300 linear

actuator (SMAC Europe, Horsham, West Sussex, UK) with LAC-1 high speed

controller single axis with built-in amplifier, that has a stroke length of 50 mm,

peak force of 250 N, displacement resolution of 0.5 mm and crosshead speed

of 1 mm-1 (Sauro et al., 2012b). The load (N) at failure and the cross-sectional

area of each failed beam (measured with a digital micrometer: Mitutoyo CD15;

Mitutoyo, Kawasaki, Japan) permitted calculation of the μTBS that was

expressed in MPa. The μTBS (mean-MPa) data for each group were subjected

to a repeated measures ANOVA and Tukey's post-hoc test for pair-wise

comparisons (α = 0.05). Fisher's least significant difference (LSD) test was used

to isolate and compare the significant differences (P < 0.05) between the

groups. Premature failures were included in the statistical analysis as zero

values.

Modes of failure were classified as percentage of adhesive (A), mixed (M), or

cohesive (C) when the failed bonds were examined at x 30 magnification with a

206

stereoscopic microscope (Leica M205A; Leica Microsystems, Wetzlar,

Germany). For each group, five representative de-bonded specimens, depicting

the most frequent failure modes, were chosen for SEM ultra-morphology

analysis of the fractured surfaces. They were dried overnight and mounted on

aluminum stubs with carbon cement. They were sputter-coated with gold (SCD

004 Sputter Coater; Bal-Tec, Vaduz, Liechtenstein) and examined using a

scanning electron microscope (SEM) (S-3500; Hitachi, Wokingham, UK) with an

accelerating voltage of 15 kV and a working distance of 25 mm at increasing

magnifications from x 60 to x 5000.

5.2.4 Dye-assisted CLSM evaluation

Three further dentine-bonded specimens were prepared as previously

described for each group with the primer/bond resins doped with 0.1 wt%

Rhodamine-B (Rh-B: Sigma-Aldrich, Munich, Germany) and then serially

sectioned across the adhesive interface to obtain resin-dentine slabs (n = 12

per group) with a thickness of approx. 1 mm (Fig. 5.2). The resin-dentine slabs

were then allocated to two subgroups (n = 6/group) based on the period of

storage in SBS (24 h or 6 months). Following each ageing period, the

specimens were coated with two layers of fast-setting nail varnish applied 1 mm

away from the resin-dentine interfaces. Three specimens from each subgroup

were immersed in 1 wt% aqueous fluorescein (Sigma-Aldrich) and the other

three specimens in 0.5 wt% Xylenol Orange solution (XO: Sigma-Aldrich) for 24

h at 37°C. The latter is a calcium-chelator fluorophore commonly used in

mammals bone remineralisation studies (Rahn and Perren, 1971), due to its

ability to form complexes with divalent Ca2+ ions. The specimens were then

207

treated in an ultrasonic water bath for 2 min and polished using ascending

(#1200 to #4000) grit SiC abrasive papers (Versocit; Struers) on a water-cooled

polishing device (Buehler Meta-Serv 3000 Grinder-Polisher; Buehler). A final

ultrasonic treatment (5 min) concluded the specimen preparation for the

confocal microscopy analysis which was immediately performed using a

confocal laser scanning microscope (Leica SP2 CLSM; Leica, Heidelberg,

Germany) equipped with a 63x / 1.4 NA oil immersion lens. The xylenol orange

and the fluorescein were excited at 488-nm using an argon/helium laser. The

ultramorphology evaluation (resin-diffusion) was executed using a 568-nm

krypton (rhodamine excitation) laser. CLSM images were obtained with a 1 μm

z-step to optically section the specimens to a depth up to 20 μm below the

surface (Sauro et al., 2012a). The z-axis scans of the interface surface were

arbitrarily pseudo-colored by the same operator for better exposure and

compiled into single projections using the Leica image-processing software

(Leica). The configuration of the system was standardised and used at the

same settings for the entire investigation. Each resin-dentine interface was

completely investigated and then five optical images were randomly captured.

Micrographs representing the most common features of nano-leakage observed

along the bonded interfaces were captured and recorded (Profeta et al., 2012).

5.3 Results

5.3.1 μTBS and SEM observations of the failed bonds

The interaction between bonding system versus SBS storage was statistically

significant only for the Res-HOPC and Res-HPCMM groups (P = 0.001); no

208

significant reduction of the µTBS values was observed after 6 months of SBS

ageing (P > 0.05). Conversely, significant drops in µTBS values were observed

in both Res-HPCTO and Res-Ctr groups (P < 0.05) after prolonged storage in

SBS (6 months). The µTBS results (expressed as Mean and SD) and modes of

failures obtained for each group are summarised in Table 5.2.

All the tested materials showed high μTBS values after 24 h of SBS storage

with failures occurring mainly in the cohesive mode. However, only the resin-

dentine specimens of the Res-HOPC and Res-HPCMM groups maintained high

μTBS values (P > 0.05) after 6 months of storage in SBS (31.3 ± 11.5 and 24 ±

12.7 MPa, respectively) and they debonded prevalently in cohesive mode (57%

and 54%, respectively). The SEM analysis of the fractured surfaces at 24h of

SBS storage revealed the absence of both exposed dentine tubules and

collagen fibrils indicating a good hybridisation of dentine (Figures 5.3A and

5.3B, respectively). After 6 months of SBS storage, the dentine surfaces were

characterised by embedded mineral crystals and remnant resin presenting filler

lacunas (Res-HOPC: Figures 5.3A1 and 5.3A2; Res-HPCMM: Figures 5.3B1

and 5.3B2). In contrast, a significant drop (P < 0.05) in μTBS was observed

after 6 months of storage in SBS, with the specimens created using the Res-Ctr

group (filler-free) and with those created with the Res-HTCPO. These latter

specimens showed, after 24 h, a well-hybridised de-bonded surface embedding

micro-fillers (Fig. 5.3C). On the contrary, the specimens de-bonded after 6

months of SBS ageing showed a de-bonded surface with a few dentinal tubules

but with no sign of clear degradation and well-hybridised peritubular dentine

with resin tags (Figures 5.3C1 and 5.3C2). The Res-Ctr specimens tested after

24 h of SBS storage and analysed with SEM presented few exposed dentinal

209

tubules but mostly were obliterated by resin tags or covered by resin remnants

(Fig. 5.3D). Conversely, the surface of the specimens de-bonded after 6 months

of SBS exhibited no collagen fibrils on the dentine surface with rare resin tags

and degraded funneled dentinal tubules (Figures 5.3D1 and 5.3D2).

Table 5.2 - Mean and standard deviation (SD) of the μTBS (MPa) to dentine.

Values are mean ± SD in MPa. In each row, same numbers indicate no differences (p > 0.05) after 24 h and 6 m of SBS storage. In columns, same capital letter indicates no statistically significant differences between each group (p > 0.05). Premature failures were included in the statistical analysis as zero values and are indicated in parentheses (for instance 5/35 means that there were 5 premature failures and 35 testable beams). The modes of failure are expressed in percentage in the brackets [adhesive/mix/cohesive].

210

Figure 5.3 - SEM failure analysis of debonded specimens. (A): SEM micrograph (1000x magnification) of an adhesively fractured stick bonded with Res-HOPC after 24 h of SBS storage. Observe the dentine entirely covered adhesive resin (ar) with some fillers’ lacunas (pointer) and rarely found opened dentinal tubules (dt). (A1) In another specimen, note the presence of resin adhesive (ar) onto dentine without unprotected collagen fibrils. Some fillers were detached during SEM preparation (pointer) and initial mineral precipitation may be observed (white asterisk). (A2)After 6 months, the debonded dentine surface at higher magnification (2.500X) kept covered with adhesive resin (ar).Mineral crystals (asterisk) embedded within a preserved collagen network were vastly encountered albeit some fillers were detached (pointer). (B): SEM micrograph of a specimen bonded with Res-HPCMM after 24h. Note the presence of adhesive resin (ar) mostly covering the dentine and dentinal tubules (dt);; some fillers’ lacunas are also observed (pointer). (B1)At another region of the same specimen, more lacunas are observed due to filler detaching, but also initial mineral crystallisation was depicted (asterisk). (B2) After 6 months storage, a failure mode observed under higher magnification (2.500X) similar to that found in B1 is disclosed showing the very slow bonding degradation. More mineral precipitation was observed (asterisk) and the fillers’ lacunas (pointer) are wider due to expansion of the fillers when exposed to water. (C): Micrograph of a debonded stick from group Res-HPCTO after 24h showing dentine completely covered with rare filler detachment (pointer). (C1) In other region, note the presence of few exposed dentinal tubules as well as intact resin tags (rt). (C2) After 6 months, the fractured dentine surface bonded with Res-HPCTO showed a de-bonding at hybrid layer and/or at its bottom. A lot of resin tags (rt) were observed well-hybridised with peritubular dentine (black pointer) suggesting potential remineralisation albeit some funneled dentinal tubules were encountered without resin tags (black asterisk). (D): SEM micrograph from a specimen bonded with Res-Ctr and presenting a perfectly hybridised and resin covered dentine. (D1) Other fractured stick from the same group showed an adhesive failure at the bottom of hybrid layer with remnants of resin adhesive (ar) and some resin tags (rt). (D2) The control adhesive after 6 months showed many signs of degradation since most collagen fibrils were degraded, the funneled dentinal tubules (asterisk) were often found along with poorly hybridised resin tags (pointer). Symbols. White finger: filler lacuna; white asterisk: mineral precipitation; black asterisk: funneled degraded peritubular dentine; black finger: hybridisation between tags and peritubular.

211

5.3.2 Dye-assisted CLSM evaluation

CLSM imaging of the bonded-dentine interfaces subsequent to 24 h of SBS

storage revealed relevant ultramorphology and nano-leakage information for all

groups. It was observed that all tested materials were able to diffuse within the

demineralised dentine, creating an hybrid layer 7-10 μm thick, with a multitude

of resin tags penetrating the dentinal tubules (Fig. 5.4). Nevertheless, all these

interfaces were affected by conspicuous fluorescein penetration (nano-leakage)

within porous hybrid layers and through dentinal tubules (Fig. 5.4). Furthermore,

the resin-dentine interface created using the experimental bonding agents

containing the micro-fillers and the acidic functional monomer PMDM showed

presence of calcium-chelator dye (XO: Xylenol orange) also within the hybrid

layer and inside the dentinal tubules (Fig. 5.5). On the contrary, the acid-etched

dentine bonded using the resin control (Res-Ctr, filler-free) showed no presence

of XO along the interface.

Significant ultramorphological changes were observed subsequent to prolonged

SBS storage. For instance, the CLSM analysis revealed no gap and limited

fluorescein penetration (nano-leakage) within the resin-dentine interfaces

created using the Res-HOPC and Res-HPCMM (Figures 5.6A and 5.6B). In

addition, XO produced a clearly outlined fluorescence due to a consistent Ca-

minerals deposition within the resin-dentine interface and inside the dentinal

tubules (Figures 5.5A and 5.5B). The resin-dentine interfaces created using

Res-HPCTO showed less nano-leakage within the hybrid layer; resin

degradation of the adhesive layer was also observed (Fig. 5.6C). Intense nano-

leakage and constant gaps affected the resin-dentine interfaces created using

the Res-Ctr (Fig. 5.6D). When the same interfaces were investigated employing

212

xylenol orange, only the walls of the dentinal tubules were stained by the

fluorescent calcium-chelator dye (Fig. 5.5E).

213

Figure 5.4 - Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterisation and nanoleakage, after 24 h of storage in SBS. Images (1) indicate the projection of fluorescein dye whereas the images (2) disclose the projection of rhodamine B dye. The images (3) are depicting the projections of both dyes. (A1, A2, A3): CLSM images showing the interfacial characteristics of the bonded-dentine interface created using Res-HOPC. It is possible to observe a clear hybrid layer (hl) with long resin tags (rt) penetrating the dentinal tubules (dt) underneath an adhesive layer (ad) characterised by evident mineral fillers (FL). Intense fluorescein uptake was observed within the entire resin-dentine interface as well as the adhesive layer. (B1, B2, B3): Micrographs showing the interfacial characteristics of the bonded-dentine interface created using Res-HPCMM. Similarly to images from Res-HOPC, these images presented high dye uptake throughout the entire resin-dentine interface as well as the adhesive layer. (C1, C2, C3): The resin-dentine interface created using the Res-HPCTO bonding system was characterised by a clear hybrid layer (hl) located underneath the adhesive layer (ad) containing the experimental filler (FL). Long resin tags (rt) penetrating the dentinal tubules (dt) were observed as well as evident nanoleakage and dye uptake along the entire interface and adhesive layer. (D1, D2, D3): CLSM images showing the bonded-dentine control interface (RES-Ctr) characterised by a thick and fluorescent hybrid layer (hl) (approximately 8 μm thickness) located underneath an adhesive layer (ad) devoid of fillers.

214

Figure 5.5 - CLSM single-projection images disclosing the fluorescent calcium-chelators dye xylenol orange. All images were obtained from specimens immersed in simulated body-fluid solution for 24 h or 6 months. (A): CLSM image of the resin-dentine interface created with Res-HOPC after 24 h of SBS storage. Mineral deposition can be visualised within the adhesive layer (ad), the hybrid layer (hl) along the walls of dentinal tubules (dt) and the filler inside the resin tags (rt). (B): CLMS image of the resin-dentine interface created with Res-HPCMM and immersed in SBS for 6 months where it is possible to observe a clear fluorescence signal due to a consistent presence of Ca-deposits within the adhesive layer (ad), hybrid layer, walls of the dentinal tubules (dt) and resin tags (rt). (C): Image of the resin-dentine interface created with Res-HPCTO and immersed in SBS for 24h. Xylenol Orange was able to stain the Ca-minerals within adhesive layer, hybrid layer and dentinal tubule (dt). Note the intense calcium deposition at bottom of hybrid layer. (D): Image of the resin-dentine interface created with Res-HPCTO and immersed in SBS for 6 months showing also in this case Ca-mineral presence at the bottom and within the hybrid layer, dt and rt. (E): Image of the resin-dentine interface created with Res-Ctr (no filler) in which one may note absence of calcium deposition both within the hybrid (hl) and adhesive layer (ad). Only the walls of the dentinal tubule tubules (dt) were stained by the fluorescent dye.

215

Figure 5.6 - Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterisation and nanoleakage after 6 months of SBS storage. (A): Image showing the resin-dentine interface created using Res-HOPC characterised by reduced nanoleakage within the hybrid layer (hl). Note the absence of fluorescein uptake within the adhesive layer (ad). (B): Image showing the interfacial features of the bonded-dentine interface created using the Res-HPCMM. Note the low overall nanoleakage with very little fluorescein uptake in hybrid (hl) and adhesive layers (ad). (C): Image of the resin-dentine interface created using Res-HPCTO. Despite the mineral deposition and reduced nanoleakage within adhesive layer (ad), the weak bond strength created gaps (gap) in which the fluorescein was deposited. The gaps may be induced by the cutting procedures as the resin degradation was replaced by mineral precipitation creating an interface with low elasticity (high stiffness properties). The nanoleakage was only observed within the hybrid layer (hl). (D): Micrograph of the resin-dentine interface created using the control adhesive system (RES-Ctr). Note the presence of intense dye uptake (nanoleakage) within the hybrid layer (hl) and at the bottom of adhesive layer (ad). In this case the presence of gaps frequently observed between hybrid and adhesive was very likely due to hybrid layer degradation (reduced thickness). Other symbols. dt: dentinal tubules; rt: resin tags; c: composite.

216

5.4 Discussion

The resin-dentine interfaces created using contemporary “simplified” etch-and-

rinse bonding agents are affected by bond strength reduction subsequent to

prolonged water ageing (Breschi et al., 2008). This phenomenon occurs due to

the inability of such materials to completely replace loosely bound and bulk-free

water from the apatite-depleted dentine collagen matrix during bonding

procedures which cause hygroscopic swelling effects and hydrolytic

degradation of polymer networks and favours dentinal collagenolytic

metalloproteinases (MMPs)-mediated activity upon water ageing (Breschi et al.,

2008, Kim et al., 2010d, Pashley et al., 2011, Breschi et al., 2010b, Liu et al.,

2011a). Nevertheless, the presence of water may be essential to facilitate

apatite nucleation within the gap zones of collagen fibrils and fossilisation of the

host-derived, collagen-bound enzymes (MMPs) (Kim et al., 2010d, Pashley et

al., 2011, Tay and Pashley, 2009). Recent investigations have demonstrated

that it is possible to reduce the nano-leakage and micropermeability within the

resin-dentine interface and maintain the bond strength (Profeta et al., 2012) of

bioactive resin-base materials applied to H3PO4 acid-etched dentine

subsequent to simulated body fluid storage for 3-6 months (Wang and Spencer,

2005, Sauro et al., 2012a). Ryou et al. (Ryou et al., 2011) demonstrated that

using a biomimetic remineralisation approach it is feasible to remineralise the

dentine collagen within the resin-dentine interface via slow release of calcium

ions from set white Portland cement and subsequent interaction of these ions

with phosphate species from SBS or dentine substrate. Portland cements

designed for dental applications, also called hydraulic silicate cements or MTA,

mainly constituted by Alite (3CaO x SiO2), Belite (2CaO x SiO2) and tri-calcium

217

aluminate (3CaO x Al2O3), exhibit outstanding biological properties and high

bioactivity when immersed in SBF (Wang and Spencer, 2005, Ryou et al., 2011,

Darvell and Wu, 2011).

In the present study, modified Portland cement-based micro-fillers (< 20 μm)

were included within the composition of a representative three-step/etch-and-

rinse bonding agent in order to create a material with therapeutic remineralising

effects on the mineral-deficient areas along the resin-dentine interface. Based

on the results obtained in this study, the first null hypothesis that the inclusion of

tested micro-fillers within the composition of the experimental bonding agent

has no effect on the bond strength durability must be rejected as only the use of

Res-HOPC and Res-HPCMM bonding agents preserved the bond strength

durability. The second null hypothesis that no mineral precipitation and nano-

leakage reduction will be observed within the demineralised ‘poorly resin-

infiltrated’ areas within the resin-dentine interface must be also rejected.

In detail, the three experimental bonding agents containing experimental micro-

fillers (Res-HOPC, Res-HPCMM and Res-HPCTO) and the control co-monomer

blend (RES-Ctr) used to bond the acid-etched dentine produced comparably

high μTBS values (P > 0.05) following 24 h of storage in SBS (Table 5.2).

Conversely, after 6 months of storage in SBS a significant decrease in μTBS (P

< 0.05) was observed for the RES-Ctr and Res-HPCTO groups, while the

specimens bonded using Res-HOPC or Res-HPCMM maintained consistent

long-term bond strength values (P > 0.05) compared to the control group (24 h

SBS storage). The specimens of the Res-HOPC and Res-HPCMM groups de-

bonded after 6 months of SBS storage showed, during SEM fractography

examination, residual resin presence and newly formed mineral-bodies (Figures

218

5.3A1 and 5.3A2, 5.3B1 and 5.3B2). Several morphological differences were

observed in the specimens of the Res-HPCTO group which presented a de-

bonded surface characterised by very few partially exposed dentinal tubules

and an important precipitation of mineral crystals after 6 months of SBS storage

(Figures 5.3C1 and 5.3C2). The SEM analysis revealed that the de-bonded

dentine surface of the specimens in RES-Ctr group was well resin-hybridised

and characterised by no exposed collagen fibrils after 24 h of SBS storage (Fig.

5.3D). In contrast, the prolonged SBS storage (6 months) induced radical

changes; the dentine surface presented funneled dentinal tubules as a sign of

degradation of the “poorly resin-infiltrated” demineralised peritubular dentine

(Figures 5.3D1 and 5.3D2). These results were also supported by the CLSM

analysis performed to evaluate the nano-leakage and the presence of calcium-

compounds within the resin-dentine interface subsequent to SBS storage (24 h

or 6 months). Indeed, further evidence of the therapeutic bioactivity of the

experimental bonding agents containing the modified Portland cement-based

micro-fillers were attained; reduced fluorescent-dye uptake (nano-leakage) was

observed along the entire resin-dentine interface after 6 months of storage in

SBS (Figures 5.5A, 5.5B and 5.5C). These latter observations along with the

strong xylenol orange signal from the hybrid layer and the dentinal tubules (Fig.

5.5), clearly indicated the remineralisation of those areas which were previously

detected within the resin-dentine interface as mineral-deficient/poor-resin

infiltrated zones.

It is hypothesised that the therapeutic remineralising effects observed within the

mineral-depleted resin-dentine interface were essentially due to the bioactivity

of the experimental micro-fillers. Indeed, the reaction mechanism of the

219

Portland cement-based micro-fillers involved the reaction of the polymerised

calcium-silicate hydrate gel with water to release calcium hydroxide and to the

consequent increase of the alkalinity of the surrounding environment

(Lawrence, 1998); this increase of pH was confirmed in this study (Table 5.2).

This localised increase in pH within the resin-dentine interface may have

interfered with the activity of MMPs (Pashley et al., 2004, Breschi et al., 2008).

Furthermore, the interaction between the phosphate ions present in the ageing

solution (SBF) or in the dentine substrate, and the calcium released from the

experimental Portland-based micro-fillers may have enhanced the formation of

new apatite deposits upon existing mineral constituents within the dentine

matrix (biocatalysation) (Gandolfi et al., 2010c). However, it is well known that

the increase in environmental pH and the presence of free OH− may facilitate

apatite nucleation and reduce the solubility of intermediate Ca/P species formed

during apatite formation (Sauro et al., 2011b). The most appropriate pH to

support the formation of stoichiometric hydroxyapatite (HA) in vitro (Bayle et al.,

2007) and in vivo (Jayaraman and Subramanian, 2002) falls in a range between

8 and 9. At higher pH it is common to obtain a Ca-deficient HA (lower solubility

then stoichiometric hydroxyapatite) characterised by higher concentrations of

PO43- and lower Ca2+ ions (Liu et al., 2009). Furthermore, the presence of

carboxylic species (R-COO-) within the acidic functional monomer (PMDM)

used in this study may have acted as sequestering agent for Ca/P cluster

favouring the precipitation of nano-apatite within the polymer network and

dentine collagen (Tay and Pashley, 2009, Ryou et al., 2011). Moreover, the R-

COO- species of PMDM may have interacted with the remnant calcium present

along the front of demineralisation at the bottom of the hybrid layer acting as a

220

sort of biomimetic template primer which promoted precipitation of Ca-

compounds (Wang and Spencer, 2005, Sauro et al., 2012a). During all these

processes for Ca/P nucleation, apatite precipitation may have reduced the

distribution of water-rich regions within the resin-dentine interface (Wang and

Spencer, 2005, Qi et al., 2012), interfering with the hydrolytic and hygroscopic

mechanisms involved in the degradation of dental polymers (Ferracane, 2006).

In addition to the formation of apatite crystals, the nanostructure of the calcium

silicate hydrate may also have contributed to seal the dentinal tubules due the

small-scale volume of the forming gels, along with a slight expansion of the

calcium silicate-based materials once immersed in SBS (Skinner et al., 2010).

In particular, the phyllosilicates (i.e. smectite) and hydrotalcite, which were

contained in the micro-fillers used in this study, have the ability to expand

considerably following water sorption into the interlayer molecular spaces

(Bhattacharyya and Gupta, 2008). The amount of expansion is due largely to

the type of exchangeable cation contained in the micro-filler; the uptake kinetics

of cation exchange is fast and the presence of Na+, as the predominant

exchangeable cation, can result in material swelling. In this condition, the

exceeding water is removed, thereby preventing hygroscopic effects and

hydrolytic degradation of the polymer chains (Malachová et al., 2009). Also, it is

reasonable to expect that the metallic ions intercalated on phyllosilicate were

easily released by ion-exchange with cations present in the surrounding

solutions and acted as effective antibacterial substances in the long term

(Ferracane, 2006). In contrast, the bond strength reduction observed in the

resin-dentine interfaces created using the Res-HPCTO bonding agent after

prolonged storage in SBS (Table 5.2) may be due to the high hydrophilicity of

221

the TiO2. Micro-fine titanium oxide (TiO2) have been used as an inorganic

additive of resin composites to match the opaque properties of teeth (Yu et al.,

2009) and as nano-particles to increase the micro-hardness and flexural

strength of dental composites (Bhattacharyya and Gupta, 2008). However, TiO2

has been advocated as a super-hydrophilic component, in particular under

ultraviolet (UV) light irradiation (Xia et al., 2008, Chadwick et al., 1994, Bonding

et al., 1987). Therefore, a possible explanation for the μTBS reduction may be

attributed to this high hydrophilicity which may have permitted excessive water

adsorption which induced severe hydrolytic resin and collagen degradation as

well as the extraction of water-soluble un-reacted monomers or oligomers from

the resin-matrix (Karuppuchamy and Jeong, 2005). Moreover, the replacement

of the degraded resin by mineral crystallisation within the Res-HPCTO bonded-

dentine interface (Figures 5.4 and 5.5) over prolonged SBS storage may have

conferred mechanical characteristics related to bond strength comparable to

those created by conventional glass-ionomer cements (GICs) applied onto

polyacrylic acid-etched dentine and submitted to tensile tests (Spencer et al.,

2010, Hewlett et al., 1991). Indeed, several studies indicated that the bond

strength of GICs when tested using tensile or shear methods was approximately

5 MPa; these values results do not reflect the true adhesive strength to dentine

(Berry and Powers, 1994). These factors may have been also responsible for

the formation of gaps within the resin-dentine interface created by the Res-

HPCTO during the cutting/sample preparation (Profeta et al., 2012).

222

5.5 Conclusion

In conclusion, as the results of this study demonstrated that the resin-dentine

bond may be maintained over time by inducing a therapeutic remineralisation of

the bonding interface, specific experimental resin bonding systems containing

bioactive micro-fillers, such as Res-HOPC, Res-HPCMM or Res-HPCTO may

offer the possibility to improve the durability of the resin-dentine interfaces. The

characteristic of promoting bioactivity should also open up the potential to

create therapeutic restorative materials able to reduce the incidence of

secondary caries. Indeed, it is important to consider that restorative materials

containing bioactive fillers may be effective in killing a wide selection of aerobic

bacteria due to the increase of the local pH and concentration of alkaline ions

(Yip et al., 2001, Hewlett et al., 1991). The antibacterial properties are

potentially of great importance as the infiltration of microorganisms may cause

secondary caries which jeopardise the longevity of resin-dentine interface

leading to the replacement of dental restorations (Giannini et al., 2004, Sauro et

al., 2006). Further studies are ongoing in order to evaluate the species-specific

antibacterial effects and biocompatibility of the materials tested in this study.

223

Chapter 6: In vitro micro-hardness of resin-

dentine interfaces created by etch-and-rinse

adhesives comprising bioactive fillers

224

6.1 Introduction

Current concepts of resin/dentine adhesion imply that chemicals are applied

before bonding to alter the structure of dentine and favour resin infiltration

(Nakabayashi and Pashley, 1998). Subsequently, resin hybridisation should

restore the biological and mechanical properties of the partially demineralised

dentine to approximate those of the original undemineralised dentine.

Unfortunately, despite significant improvements in adhesive systems, the

resin/dentine bonded interface formed by a mixture of collagen organic matrix,

residual hydroxyapatite crystallites and resin monomers still remains the

weakest area of adhesive restorations (Breschi et al., 2008).

The discrepancy between the etching depth and the adhesive system

penetrating capacity make the collagen-rich zone underlying the hybrid layer

susceptible to nano-infiltration (Sano et al., 1995a). This may lead to the

formation of pathways in which oral fluid and endogenous proteolytic enzymes

concur to degrade each component of the resin-dentine bonds (Hashimoto et

al., 2003).

Recently, progressive removal of water by apatite deposition emerged as a

viable strategy to address the fundamental issue of replacing mineral that is

iatrogenically depleted during the acid etching phase (Kim et al., 2010e, Ryou et

al., 2011). This should lead to a more durable form of tissue engineered dentine

that is capable of preserving its organic components and maintain adhesive

strength after ageing.

Remineralisation of incompletely resin-infiltrated collagen matrices can be

promoted by the use of bioactive, ion-releasing materials, e.g., glass-ionomer

cements (Endo et al., 2010) or resin-based calcium-phosphate (CaP) cements

225

(Ngo et al., 2006).

Bresciani et al. (Bresciani et al., 2010) revealed significantly increased Knoop

hardness along the interface of resin-bonded dentine, confirming the capacity of

materials containing CaP cements to promote remineralisation of caries-

affected residual dentine.

Silicate compounds, including calcium/sodium phosphosilicates, such as

Bioglass® 45S5 (BAG), and certain calcium-silicate cements (often referred to

as Cal-Sil) are known to release ions in aqueous solution and induce deposition

of carbonated hydroxyapatite (HCA). Whereas BAG has been successfully used

for dentine hypersensitivity (Greenspan, 2010, Salian et al., 2010) and Cal-Sil

cements are already employed in dentistry for different endodontic clinical

applications (Taddei et al., 2011), the development of resin-based restorative

materials containing bioactive micro-fillers with remineralising effects on the

mineral-depleted areas within the bonded-dentine interface remains an

important target to accomplish.

The present study further investigated the performance of four experimental

adhesives either incorporating BAG (Chapter 3) or three distinct hydrated

blends of Cal-Sil cements (Chapter 4), for their potential benefits with respect to

the improvement of micro-hardness in resin-dentine interfaces created with

etch-and-rinse bonding techniques. This aim was accomplished by comparing

micro-hardness values of the area that was considered to represent the resin-

dentine interface (the hybrid layer and its surroundings) after 24 h and 6 months

of storage in phosphate buffered solution (PBS). Our null hypothesis was that

treatment with adhesives containing BAG or Cal-Sil cements has a preserving

effect on the micro-hardness of resin infiltrated dentine surfaces.

226

6.2 Materials and methods

6.2.1 Teeth collection and preparation

Intact fresh human third molars extracted for surgical reasons were collected for

this study with the informed consent of the donors (20 to 40 yr. old), following a

protocol approved by the ethical guidelines of the Research Ethics Committee

(REC) for medical investigations. The teeth were stored in deionised water (pH

7.1) at 4°C, and they were used within 1 month following extraction.

Each tooth was prepared by exposing a flat mid-coronal dentine surface and

removing the roots 1 mm beneath the cemento-enamel junction (CEJ) at an

angle of 90° to their longitudinal axis using a slow-speed, water-cooled diamond

wafering saw (330-CA RS-70300; Struers) mounted on a hard tissue microtome

(Accutom-50; Struers, Copenhagem, Denmark). A 180-grit silicon carbide (SiC)

abrasive paper installed on a water-cooled rotating polishing machine (Buehler

Meta-Serv 3000 Grinder-Polisher; Buehler, Düsseldorf, Germany) was used (30

s) to remove the diamond saw smear layer and to replace it with a standard and

more clinically relevant smear layer (Oliveira et al., 2003).

6.2.2 Formulation of the comonomer resin adhesive blend

The resin co-monomer blend used in this study as dentine bonding agent

represents the formulation of a typical three-step, etch-and-rinse adhesive. It

was prepared from commercially available monomers - 2, 2-bis[4(2-hydroxy-3-

methacryloyloxy-propyloxy)-phenyl] propane (Bis-GMA; Esstech, Essington,

PA, USA) and 2-hydroxyethyl methacrylate (HEMA; Aldrich Chemical,

Gillingham, UK) - and included a 50 wt% ethanol-solvated resin mixture used as

227

primer [Bis-GMA, HEMA, PMDM, 50% absolute ethanol (Sigma-Aldrich)]. To

obtain a dental bonding system with chemical affinity to Calcium (Ca2+)

available in dentine and in each micro-filler, we included the acidic functional

monomer 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid

(PMDM: Esstech Essington). A binary photoinitiator system based on

camphoroquinone (CQ; Aldrich) and 2-ethyl-dimethyl-4-aminobenzoate (EDAB;

Aldrich) made the neat resin light-curable (Table 6.1).

228

Table 6.1 - Chemical composition (wt%) of the experimental adhesive systems used in this study. Abbreviations. Bis-GMA: bisphenyl A glycidyl methacrylate; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid; CQ: camphoroquinone; EDAB: 2-ethyl-dimethyl-4-aminobenzoate; BAG: Bioglass® 45S5; HOPC: set Portland cement and smectite; HPCMM: Portland cement, Smectite and Hydrotalcite; HPCTO: set Portland cement, Smectite, Hydrotalcite and Titanium Oxide.

229

6.2.3 Bioactive fillers and experimental bonding systems

Bioglass® 45S5 (Sylc; OSspray, London, UK) with an average particle size of

20 μm was included within the composition of the neat resin as bioactive micro-

filler (60 wt% resin blend/40 wt% BAG) to create the first experimental dentine

bonding system which was called Res-BAG (Table 6.1). Subsequently, three

calcium-silicate micro-fillers were designed and prepared. A type I ordinary

Portland cement (82.5 wt%), (identified as OPC: Italcementi Group, Cesena,

Italy) constituted by tri-calcium silicate (Alite: 3CaO x SiO2), di-calcium silicate

(Belite: 2CaO x SiO2), tri-calcium aluminate (3CaO x Al2O3) and gypsum

(CaSO4 x 2H2O) was mixed with 7.5 wt% of phyllosilicate consisting of sodium-

calcium-aluminum-magnesium silicate hydroxide hydrate

[(Na,Ca)(Al,Mg)6(Si4010)3(OH)6-nH2O; Acros Organics, Fair Lawn, NJ, USA] in

deionised water (Ratio 2:1) to create the first Portland-base experimental filler

(HOPC). The second experimental filler (HPCMM) was created by mixing 90

wt% of type I OPC, 7.5 wt% phyllosilicate and 2.5 wt% of hydrotalcite consisting

of aluminum-magnesium-carbonate hydroxide hydrate [(Mg6Al2

(CO3)(OH)16·4(H2O); Sigma-Aldrich]. The third calcium silicate-based filler

(HPCTO) used in this study was created by mixing OPC (80 wt%), phyllosilicate

(7.5 wt%), hydrotalcite (2.5 wt%) and 10 wt% titanium oxide (TiO2: Sigma-

Aldrich, Gillingham, UK). Each cement was combined with deionised water

(Ratio 2:1), allowed to set in an incubator at 37°C for 24 h and finally grinded in

an agate ball mill as well as being sieved to obtain 20-30 μm-sized hydrated

silicate fillers (HOPC, HPCMM, HPCTO).

Ultimately, three Cal-Sil cements-based experimental adhesive systems were

also prepared (GB patent application No 1118138.5 - filed on 20th October

230

2011): Res-HOPC (60 wt% resin blend/40 wt% HOPC); Res-HPCMM (60 wt%

resin blend/40 wt% HPCMM); Res-HPCTO (60 wt% resin blend/40 wt%

HPCTO) (Table 6.1).

Each hybrid etch-and-rinse adhesive was prepared by mixing the neat resin and

the fillers for 30 s on a glass plate to form a homogeneous paste prior to the

bonding procedures. Overall, four experimental bonding systems were created

in this study, while the application of the neat resin adhesive with no filler served

as a control group (Res-Contr).

6.2.4 Bonding procedures

The specimens were etched using 37% phosphoric acid solution (H3PO4;

Aldrich Chemical) for 15 s followed by copious water-rinsing. The etched-

dentine surfaces were gently air-dried for 2 s to remove the excess of water and

leave a wet reflective substrate. Each time the bonding procedure was

accomplished by applying two consecutive coats of the ethanol-solvated primer

and a layer of the control or experimental bonding resin (Res-Contr, Res-BAG,

Res-HOPC, Res-HPCMM, Res-HPCTO) within a period of 20 s. Light-curing

was immediately performed for 30 s using a quartz-tungsten-halogen (QTH)

lamp (600mWcm-2, Optilux VLC; Demetron, CT, USA). Following adhesive

treatment, five 1-mm-thick increments of resin composite (Filtek Z250; 3M-

ESPE, St Paul, MN, US) were built up and individually light-activated for 20 s.

The resin bonded specimens were stored in PBS solutions (Oxoid, Basingstoke,

Hampshire, UK) for 24 h and 6 months at 37°C.

231

6.2.5 Knoop micro-hardness (KHN) analysis

Three dentine-bonded specimens were created for each group and every single

sample was subsequently sectioned across the adhesive interface using a

sectioning machine (Accutom-50; Struers) with a water-cooled diamond

wafering saw (330-CA RS-70300; Struers) to obtain two 4 mm-thick resin-

dentine slabs (n = 6 per group).

Following 24 h of storage in PBS, each slab was treated in an ultrasonic water

bath for 2 min and polished using ascending #500, #1200, #2400 and #4000 grit

SiC abrasive papers (Versocit, Struers A/S, Copenhagen, Denmark) on a water-

cooled rotating polishing machine (Buehler Meta-Serv 3000 Grinder-Polisher;

Buehler). Between each polishing step the slabs were cleansed in an ultrasonic

bath containing deionized water for 5 min.

The Knoop micro-hardness evaluation (Duramin-5, Struers A/S, DK-2750

Ballerup, Denmark) was performed using a 25 g load and 15 s dwell time to

produce indentations with long diagonals of a suitable size for accurate

measurements in relation to the thickness of the hybrid layer, while minimising

surface damage.

To minimise errors caused by tilting and to avoid the introduction of stresses

during the micro-hardness testing, each section was mounted and supported on

a glass slide using green-stick compound and a paralleling device. The metal

chuck containing the dentine slices was clamped onto the stage of the testing

machine and the surfaces were oriented perpendicular to the diamond indenter

axis.

Each polished surface received 15 indentations performed immediately after

polishing to provide a more uniform surface for reading and to improve the

232

precision of the indentations. These were arranged in three widely separated

straight lines starting from hybrid layer and placed perpendicular to the resin-

dentine interface. As a result, any interference of the deformation areas caused

by neighboring marks was avoided. Five measurements were executed along

each line, every 30 µm up to 115 µm in depth for sufficient hardness data to be

subjected to statistical analysis (Bresciani et al., 2010, Reinke et al., 2012)

(Figure 6.1). The dentine surface was covered with a wet tissue paper for 1 min

after each indentation to avoid dehydration of the surface (Xu et al., 1998). The

length of the long diagonal of each indentation was determined immediately to

avoid possible shrinkage caused by mechanical recovery of the tooth surfaces

with a resolution of 0.1 μm (Duramin-5 software; Struers). The main criteria for

accepting an indentation were clarity of outline and visibility of its apices.

The values obtained were converted into Knoop Hardness Numbers according

to the following formula:

KHN = 14,229 P/d2

where P = applied load in g, and d = length of the longest diagonal in μm.

After 6 months of PBS storage, hardness measurements were executed again

from the same sections not far away from the first indentations. In total, 900

data points were obtained, 450 after 24 h and 450 after the prolonged ageing

period, respectively.

All micro-hardness values of the slabs obtained from the same tooth were

averaged, and just one value per tooth was used in the statistical analysis.

Mean (±SD) KHN numbers were treated with two-way analysis of variance

(ANOVA) to determine differences between materials and the effect of ageing.

Subsequent one-way ANOVA was performed to assess differences between

233

materials for the two different storage times separately. Post-ANOVA contrasts

were performed using a Bonferroni test for multiple comparisons.

Figure 6.1 - Optical images obtained during the micro-hardness test along the resin-dentine interface. A): Picture illustrating how the five measurements (indentations) were taken along each line every 30 µm up to 115 µm in depth. B): At high magnification it is possible to observe that the first indentation was performed exactly on a hybrid layer (arrow) located between the adhesive resins (a) and the dentine surface (d).

234

6.3 Results

6.3.1 Knoop micro-hardness (KHN) analysis

The indenter load produced micro-indentations with long diagonals that enabled

accurate Knoop micro-hardness measurement. There was no major surface or

subsurface damage evident on all the examined specimens, and microscopic

inspection revealed no evidence of cracks radiating from the apices of the

indentations. The biomechanical properties (surface micro-hardness) were

influenced by dentine treatment, position along the resin-dentine interface and

storage time. Two-way analysis of variance indicated that there was a

statistically significant interaction between materials and the effect of ageing.

Statistical comparisons of mean KHN values (±SD) at different depths obtained

after 24 h and 6 months of PBS storage are shown in Table 6.2.

Analysis of the present data showed a statistical reduction of KHN values after

prolonged PBS storage (p < 0.001) within the hybrid resin-dentine zones

created with the Res-Contr adhesive containing no filler (24 h: 24.4 ± 0.5 KHN;

6 m: 15.9 ± 2.9 KHN) and the experimental bonding system Res-HPCTO (24 h:

21.3 ± 3 KHN; 6 m: 14.1 ± 2.5 KHN). On the contrary, no significant KHN

decrease of the average superficial micro-hardness (p > 0.001) was observed

after 6 m in all the other indentations taken away from the composite resin layer

for both groups (Table 6.2, Group 6 m).

The experimental bonding agents of the group Res-BAG, Res-HOPC and Res-

HPCMM maintained high KHN values with no statistical difference after the

ageing period in all the tested locations (p > 0.001). The mean surface micro-

hardness values of the resin-dentine regions were 17.5 ± 0.8 at 24 h and 17.6 ±

0.3 after 6 months for Res-BAG; 18.8 ± 3.6 at 24 h and 24.2 ± 1.7 after 6

235

months for Res-HOPC; 17.5 ± 3 at 24 h and 13.1 ± 2.4 after 6 months for Res-

HPCMM, respectively. Likewise, all the other KHN values corresponding to 25,

55, 85 and 115 µm in depth did not show a statistical decrease in the mean

hardness of the surface after 6 m of PBS storage (Table 6.2, Group 6 m).

236

Table 6.2 - The results of the micro-hardness measurements for each bonding system after 24 hours and 6 months of PBS storage. In each row, same capital letter indicate no statistical difference between mean KHN values (SD). Asterisk symbol (*) indicates a statistically significant reduction (p < 0.001) of mean KHN values (SD) subsequent to 6 m of PBS storage.

I : Mean KHN values (SD) at the resin-dentine bonded interface. II : Mean KHN values (SD) at the points 25 μm distant from the composite resin layer in the direction of the dentine. III : Mean KHN values (SD) at the points 55 μm distant from the composite resin layer in the direction of the dentine. IV : Mean KHN values (SD) at the points 85 μm distant from the composite resin layer in the direction of the dentine. V : Mean KHN values (SD) at the points 115 μm distant from the composite resin layer in the direction of the dentine.

237

6.4 Discussion

There are enduring challenges in adhesive dentistry due to the incomplete

infiltration of wet dentine with resin monomers that yield resin-dentine bonds

prone to degradation (Breschi et al., 2008). Major concerns have been

expressed regarding interfacial ageing caused by absorption of water,

hydrolysis of the resin and disruption of the collagen network (Spencer et al.,

2010).

At the same time, in vitro studies have suggested possible strategies to reduce

the flaws inherent in the dentine bonding systems, each one having its own

merits and limitations (Liu et al., 2011c).

The development of bioactive ion-releasing restorative materials with a

therapeutic ability to fill micro- and nano-sized voids by crystal deposition is

currently one of the main targets of dental biomaterial research (Tay and

Pashley, 2008, Liu et al., 2011b, Peters et al., 2010, Bresciani et al., 2010).

In our previous work, the bonding durability and morphological changes induced

on resin-infiltrated dentine were evaluated using four novel etch-and-rinse

adhesives, either incorporating BAG (Chapter 4) or three distinct hydrated

blends of calcium-silicate cements (Chapter 5), to gain perspective on their

potential clinical use.

In view of the mechanism of its formation, the resin-dentine interface represents

a continuous structure from the hybrid layer (in which decalcified dentine

impregnated by resin and that not impregnated by resin are considered to be

mixed) to the healthy dentine (Nakazawa et al., 1999). Due to the complexity of

these regions, we have resorted to the use of a indentation technique for

appraising regional differences in micro-hardness associated with mineral

238

deposition over time (Van Meerbeek et al., 1993).

The results of this study indicated that incorporation of the aforementioned

bioactive fillers had an effect on the superficial hardness profile of water-rich,

resin-sparse regions within the bonded interface after prolonged storage in

PBS. However, it appeared that only treatment with Res-BAG, Res-HOPC and

Res-HPCMM did not result in a statistically significant change in KHN with

ageing. On the other hand, the reduction of resistance to local deformation in

the hybrid resin-dentine zones created with the Res-Contr and Res-HPCTO

bonding systems (p < 0.001) requires the null hypothesis of this study to be

partially rejected.

Various methods have been suggested for evaluating the extent of

remineralisation in dental tissues, mostly based on the determination of

changes in mineral content. Although several of these experimental

methodologies are able to analyse mineral levels in detail [transverse micro-

radiography (TMR), micro-computer tomography (CT), X-ray micro-tomography

(XMT), etc.], comparable, objective relative dentine hardness measurements

are often undertaken in vitro using micro-hardness indenters, resulting in

reproducible data sets with minimal damage to the sample (Banerjee et al.,

1999, Ogawa et al., 1983).

Micro-hardness measurements can be correlated with mechanical properties

such as modulus of elasticity, fracture resistance (Perinka et al., 1992), and

yield strength (Currey and Brear, 1990, Mahoney et al., 2000). A positive

correlation also exists between Knoop micro-hardness and bond strength and it

was proposed that adhesion mechanisms for both enamel and dentine are

controlled, to a major extent, by the mineral content of the tooth (Panighi and

239

G’Sell, 1993, Yoshiyama et al., 2002).

In addition, Bresciani et al. (Bresciani et al., 2010) hypothesised that it may be

possible to evaluate the therapeutic ability of bioactive calcium phosphate-

based materials in repairing the demineralised intertubular and intrafibrillar

dentine collagen by re-establishment of the superficial micro-hardness.

In our study, cross-sectional Knoop micro-indentation measurements provided

an average hardness of each surface and gave valuable additional information

regarding the behaviour of dentine/restoration interfaces because any variation

observed reflected a quantitative difference in mineral content (Hu and

Featherstone, 2005, Owens and Miller, 2000, Iijima et al., 2012).

However, many factors may influence the hardness results such as the dentine

depth and the relative quantities of the tubular, peritubular or intertubular areas

which vary considerably with location (Fuentes et al., 2003, Hosoya et al., 2000,

Pashley et al., 1985). Consequently, Knoop test measurements were obtained

from the same specimens and close to previously assessed locations after 6 m

of PBS storage. This was done in order to minimise the effect of the structural

variations within the same tooth, and to establish a reasonable baseline for

evaluation.

In bonding systems using phosphoric acid, the area 10 μm away from the

interface in the direction of healthy dentine might be within the decalcified layer

not impregnated by resin, resulting in lower hardness (Nakazawa et al., 1999).

Overall mean KHN values calculated in sound, mineralised dentine adjacent to

restorations (25, 55, 85 and 115 μm from the composite resin layer) were in

agreement with those reported by other authors (Craig and Peyton, 1958,

Fusayama et al., 1966, Meredith et al., 1996) .The similarity of reported values

240

is certainly due to the reproducible micro-indentation technique employed

(Fuentes et al., 2003).

For Knoop hardness, upon unloading, elastic recovery occurs mainly along the

shortest diagonal, but the longest diagonal remains relatively unaffected

(Shannon and Keuper, 1976, Marshall et al., 1982). Therefore, the hardness

measurements obtained by this method are virtually insensitive to the elastic

recovery of the material.

Another chief characteristic of the Knoop hardness test is its sensitivity to

surface effects and textures (Lysaght and DeBellis, 1969, Knoop et al., 1939).

Accordingly, wider impressions of the tool mark were found on all the resin-

dentine interfaces, these regions being more elastic and softer than the healthy

dentine, and lower KHN values were obtained. Even these observations

correlate with previous findings indicating that, in the absence of intrafibrillar

mineralisation, the hardness and modulus of elasticity are inferior to those of

mineralised dentine (Balooch et al., 2008). In fact, the modulus of elasticity of

wet demineralised dentinal matrix is only about 5 MPa (Bedran-Russo et al.,

2008) which is more than 1000 times lower than that of mineralised dentine.

This different histological configuration is held accountable for hardness

reduction in the resin-dentine interface and therefore for the little resistance

offered to the testing indenter.

Interestingly, specimens created with the experimental adhesives Res-BAG,

Res-HOPC and Res-HPCMM did not restore micro-hardness to the level of

sound dentine in these zones but maintained the same KHN values and no

statistical difference reduction was found following 6 m of PBS storage. The

only statistically significant change occurred in the resin-dentine interfaces

241

bonded with either Res-Contr or Res-HPCTO that were subjected to a reduction

of KHN values with ageing. These results are in agreement with our previous

findings (Chapter 4-5) where apatite precipitation and concomitant reduction of

water-rich regions within the resin-dentine interface appeared to restrict the

collagenolytic and hydrolytic mechanisms responsible for loss of mechanical

stability.

Remineralisation, defined as restoration of lost mineral content (Bresciani et al.,

2010), should redevelop the mechanical properties to approximate those of

original undemineralised dentine. Even so, the stiffness of resin-dentine

interfaces cannot be compared with natural mineralised dentine, as the

adhesive resins used to infiltrate the collagen matrices are much more

viscoplastic than bioapatites. It is equally suggested that the lack of mechanical

reconstitution may be attributed to an heterogeneous arrangement of the newly

deposited mineral within the demineralised organic network. In fact,

mineralisation patterns that differ from the usual organisation and orientation of

minerals might lead to different mechanical properties of the resultant

substrates (Bresciani et al., 2010, Bertassoni et al., 2011).

Strengths of this study include the fact that every effort was made to

standardise the experimental process. A thoughtfully planned pilot study was

carried out to develop an effective experimental design.

Nonetheless, the relationship between hardness and mineral content remains

complex, not yet fully understood and necessitates further detailed investigation

(Bresciani et al., 2010). Additional long-term studies and examination of the

resin-dentine interface at narrower intervals (i.e. atomic force microscopy nano-

indentation) will give us better insight into remineralisation dynamics, rate of

242

mineral uptake and hardness modifications induced by novel dentine bonding

systems containing bioactive fillers.

6.5 Conclusions

Within the limitations of this in vitro study the following conclusions were made:

I. In terms of micro-hardness, the hybrid layer of the resin-dentine

interfaces created in the Res-BAG, Res-HOPC and Res-HPCMM groups

resisted water degradation showing no statistical change in KHN values

after 6 months of PBS ageing. Conversely, the use of Res-Contr and

Res-HPCTO bonding systems showed a statistically significant micro-

hardness drop in the hybrid layer after 6 months of PBS storage.

II. Remineralisation of imperfect mineral-depleted sites where resin

monomers border on decalcified dentine may be a potential means for

preserving the micro-mechanical properties of resin-dentine bonds. This

particular approach may be suitable for the contemporary concept of

Conservative Dentistry where minimally invasive treatments are followed

by therapeutic restorations which stabilise the carious lesion and/or

create an optimal environment for the protection of the remaining dental

hard tissues.

243

Chapter 7: General discussion and conclusion

244

7.1 Summary

Over the past three decades, bonding of resin-based composite restorations

has been revolutionised by continuing advances in dental adhesive technology.

Improvements have been made in the areas of aesthetic appeal, ease of use

and reduction of technique sensitivity. However, the process was, and still is,

not without its faults, and chief among those is the reduced durability of resin-

dentine bonds compared with resin-enamel bonds, owing to the fact that

dentine bonding relies on organic components (Marshall et al., 1997). The

infiltration of hydrophilic resin monomers into demineralised collagen matrix, to

produce a hybrid layer that couples adhesives/resin composites to the

underlying mineralised dentine, provides ample opportunity for nano-leakage to

occur beneath the restoration: oral fluids penetrate any poorly infiltrated area

and serve as a functional medium for esterases and collagenolytic enzymes

that act synergistically to increase the biodegradation of both polymer matrices

and exposed collagen (Spencer et al., 2010). While each experimental strategy

that attempted to overcome these problems has its own benefits and reciprocal

limitations (Liu et al., 2011c), progressive water replacement by apatite

generated during dentine remineralisation may be a suitable strategy for

extending the service life of resin-based dentine bonding procedures and its

actualisation has been a source of conjecture until now (Liu et al., 2011a). In

this case, nano-leakage might only be a temporary phenomenon that could be

solved by new hard-tissue formation (Tay and Pashley, 2009). Water may

provide the aqueous environment for the infiltration of amorphous calcium

phosphate precursors into the gap zones of collagen fibrils to initiate nucleation

and growth of newly formed mineral crystals. As remineralisation proceeds,

245

water then becomes less readily available for the functioning of collagen

degrading matrix metalloproteinases (Sadek et al., 2010a). Moreover, these

host-derived, collagen-bound proteases (Mazzoni et al., 2007) may be fossilised

by the intrafibrillar apatite crystallites that are deposited within and on the

collagen fibrils (Liu et al., 2011a). Previous research has concentrated on

nanotechnology principles and the use of biomimetic analogs of matrix

phosphoproteins to mimic what occurs in biomineralisation, particularly in areas

devoid of seed crystallites (Zhang et al., 2012).

The work presented in this thesis set out to answer questions about whether the

presence of designed reactive silicate mineral powders at the bonded interface

could have a significant impact on the processes occurring in their vicinity, and

the formation of hydroxycarbonate apatite deposits on hypomineralised

adjacent dentine surfaces might take place in presence of phosphate buffered

saline or simulated body fluid solutions. This hypothesis was based on the

established fact that the experimental methacrylate-based adhesives employed

in this study, either when incorporating calcium/sodium phosphate-

phyllosilicates or calcium silicate cements, demonstrated to possess bioactive

characteristics in an aqueous environment that contained calcium and

phosphate ions (Section II - Experimental projects, Chapter 3).

At this stage, the first objective was to establish if the bioactive glass powder of

the well-characterised 45S5 formulation and of an average particle size of < 10

μm could be included in the resin-to-dentine bonding process. When 0.05 g of

Bioglass® 45S5 were applied onto H3PO4-etched wet dentine surfaces before

246

the bonding procedures (Chapter 4), a commercially available adhesive was

able to form an apparently normal hybrid layer. Notably, after 6 months of

storage in phosphate buffered saline, the occurrence and extent of nanometre-

sized voids within the hybrid layers was reduced due to the chemical nature of

the mineral precipitation (dicalcium-phosphate salts) within the resin-dentine

interface. Conversely, severe water uptake was observed within the resin-

bonded dentine interfaces created with the control resin both after 24 h and six

months of storage. However, a concomitant decrease in microtensile bond

strength over time was also observed. On this account, the crystallised bonded-

dentine interfaces may have attained mechanical characteristics comparable to

those created by conventional glass-ionomer cements applied onto polyacrylic

acid-etched dentine when submitted to tensile tests (Yip et al., 2001), not

reflecting their true adhesive strength. Following this promising start, it was

decided to include 30 wt% of Bioglass® 45S5 within the composition of a resin

adhesive as bioactive micro-filler. This approach produced bioactive/protective

effects not only in terms of nano-leakage reduction but also preserving high

adhesive strength and joint integrity. Analysis of the failure modes displayed a

pronounced tendency for these samples to fail cohesively, while the negative

controls tended towards adhesive failure. Another important trend was

confirmed in this study: the ultramorphology analysis of the fractured specimens

demonstrated the formation of mineral crystals embedded within a well-

preserved collagen network over the course of the ageing period.

Recently, hydraulic calcium silicate cements have been investigated with

respect to their potential use for the biomimetic remineralisation of apatite-

247

depleted dentine surfaces and to prevent the demineralisation of

hypomineralised/carious dentine (Gandolfi et al., 2011b). Thus, it remained to

include even these materials in an adhesive system, and to see if the

subsequent bonds to dentine would be of long duration or not. In chapter 5,

small amounts of three derivatives and proprietary formulations based on the

composition of an ordinary Portland cement were added to a resin representing

a typical 3-step etch-and-rinse adhesive, and teeth were bonded for the

purpose of carrying out bond strength tests, ultramorphology and nano-leakage

studies. The most favourable results were obtained with the addition of set type-

I Portland cement modified using either sodium-calcium-aluminum-magnesium

silicate hydroxide or aluminum-magnesium-carbonate hydroxide hydrates. The

use of resin bonding agents containing these two tailored micro-fillers promoted

a therapeutic mineral deposition mechanism within the hybrid layer: nano-

leakage did indeed appear to be reduced after soaking in simulated body fluid

for 6 months, and consistent bond strength values were maintained when

compared to negative controls. The reduced fluorescent-dye uptake

observations, along with the strong calcium chelating fluorophore (xylenol

orange) signal from the hybrid layer and the dentinal tubules, clearly indicated

the remineralisation of those areas which were previously detected as mineral-

deficient/porous zones. A further supporting result was found in the scanning

electron microscopy: the ultramorphological analysis performed on fractured

match-sticks demonstrated the presence of mineral crystals within the resin-

dentine matrix subsequent to the storage period. Despite no specific biomimetic

analogue agents being used in this study, there may have been a sort of

biomimetic activity evoked by the main products of the two silicate-based fillers

248

hydration. This process caused an alkaline condition which may have interfered

with the activity of metalloproteinases within the demineralised collagen matrix

and may have played a supplementary role in the remineralisation kinetics

enhancing the growth of apatite crystals (Somasundaran et al., 1985). Addition

of a Portland cement-based micro-filler containing titanium oxide to the same

adhesive was less successful. Again, it was shown that an apparently normal

hybrid layer was able to form, and the strength of the bonds formed was

comparably high following 24 h of storage in simulated body fluid. However,

long-term evaluation of the micro-mechanical strength was less definitive than

in the other two etch-and-rinse adhesives doped with calcium silicates. Although

nano-leakage was substantially reduced, specimens of this experimental

adhesive group showed diminution of bond strength durability due to the high

hydrophilicity of titanium oxide.

The mechanical properties of resin-dentine bonds are a fundamental aspect of

restorative procedures. The study in chapter 6 found that the same

experimental bonding agents, which appeared to restrict the collagenolytic and

hydrolytic mechanisms responsible for reduction of microtensile bond strength

to dentine, were also able to maintain unchanged micro-hardness values within

the resin-dentine interfaces submitted to prolonged storage (6 months) in

phosphate buffered solution. These results suggested that a positive correlation

exists between Knoop micro-hardness and bond strength and that adhesion

mechanisms for both enamel and dentine may be controlled, to some extent, by

the mineral content of the tooth.

249

7.2 Research contributions

The current ethos in minimally invasive operative treatment requires the

execution of therapeutic restorations that may combat the carious process and

remineralise the dental hard tissues (Peters and McLean, 2001). The

bioavailability and crystallographic inclusion of foreign ions is the basic

requirement for apatite formation (biocatalysation) in presence of aqueous

environments that contain calcium and phosphate (e.g. saliva). The concept of

a therapeutic restoration may be satisfied only if innovative bioactive materials

with the ability to release specific ions within the bonding interface are

employed, so evoking a positive response from the biological environment and

inducing protection and/or remineralisation of the mineral-depleted dental

tissues (Qi et al., 2012).

This series of experiments provided preliminary evidence that the durability of

the resin-dentine interfaces may be enhanced by modified bonding agents in a

clinically relevant manner. The inclusion of highly reactive silicate compounds,

such as commercially available bioactive glass and calcium-silicate Portland-

derived cements, within the composition of a representative etch-and-rinse

bonding system conferred on the hybrid adhesives attractive basic properties,

such as:

I. light-curing ability with controlled solubility in water and oral fluids as

detailed in Chapter 3.

II. A tendency to tolerate humidity during placement and to interact with oral

fluids and wet tooth structures (hydrophylic nature).

250

III. High bioavailability and propensity to release various remineralising

species; when activated with water, each micro-filler releases the

predominant ions of its composition enhancing mineral ion delivery within

and beneath the hybrid layers (mineral enrichment effect).

IV. A potential to displace water from the resin-sparse regions of the hybrid

layer with redeposition of apatitic tooth minerals (bioactivity).

V. Alkalinising activity to buffer the environmental acids as well as to

interfere with the activity of matrix metalloproteinases, and antibacterial

properties.

VI. Providing nano-leakage reduction and lower rates of enzymatic

degradation, two of the major causes of restoration failure, while

producing no negative effects on bond strength over time (Chapter 4-5).

VII. A capacity to preserve the mechanical properties and stability of resin-

dentine bonded interfaces (Chapter 6).

The initial evaluation of bioactive silicate compounds as a possible addition to

the resin-dentine bonding process has shown their value; a new generation of

biologically active dental materials able to induce physicochemical reactions

yielding apatite formation in demineralised dentine has been obtained as

promising adhesives to be tested in future studies and clinical trials. These

results support the possibility of prospective applications in clinical practice for

direct resin-composite restorations, dentine hypersensitivity treatment and as

cavity liners. This material may also have a potential use in apical root, root

canal and root perforation treatments.

251

7.3 Recommendations for future research

While the first steps toward the improvement of dentine bonding technology with

regard to nano-leakage have been made, much work remains to be done.

Continuing studies should be performed in order to optimise the appropriate

solvent for the adhesive along with the amount of micro-filler to include, its

particle size and composition. Particle size, or fineness, affects hydration rate:

the smaller the particle size, the greater the surface-area-to-volume ratio, and

thus, the more area available for water-cement interaction per unit volume. The

addition of specific chemical compounds, such as zinc oxide, might increase the

therapeutic/protective effects against the breakdown of collagen matrices

mediated by matrix metalloproteinases within the aged resin-bonded dentine

(Osorio et al., 2011). A further protective effect might as well be due to the

release of Zn2+ ions exerting an antibacterial action within the bonded-dentine

interface (Swetha et al., 2012). Reactions and bioactivity mechanisms

of bioactive glasses depend on the glass composition. It was recently shown

that fluoride-containing bioactive glasses are able to form fluorapatite, which is

much less vulnerable to acid attack than is carbonated apatite (Lynch et al.,

2012). In addition to the traditional bioactive glass components (silica, sodium,

calcium, and phosphate), magnesium can also be added for its synergistic

effects on the crystallinity and solubility of apatites: slowdown in the deposition

of calcium phosphate, which is believed to lead to better-controlled (and better

quality) mineralisation (Diba et al., 2012).

The relationship between hardness and mineral content necessitates further

detailed investigation. Examination of the resin-dentine interface at narrower

252

intervals will give us better insight into remineralisation dynamics, rate of

mineral uptake and hardness modifications induced by novel dentine bonding

systems containing bioactive fillers. Probing biomaterial properties on a

nanoscale requires comparatively low indentation loads and an ability to

examine materials in a hydrated state. It has been advocated that atomic force

microscopy nano-indentation performed in hydrated dentine may be a suitable

method for the determination of the visco-elasticity of the demineralised dentine

and its effective remineralisation (Bertassoni et al., 2009).

Similarly, confocal micro-Raman spectroscopy can be considered as a valuable

method for future in vitro investigations of the dentine/adhesive interface. This

technique would offer distinct advantages, including minimal sample preparation

and both qualitative and quantitative analysis at ∼ 1 μm spatial resolution. This

would enable spatially resolved chemical analysis of the interface areas

between the modified adhesives and the underlying dentine, via direct

examination of specimens without compromising their integrity. The non-

destructive nature of this analysis would also allow investigation of the same

specimen using complementary techniques. In combination with StreamLine™

imaging techniques, changes in the mineral composition of the hybrid layer

associated with ageing of the specimens can be investigated.

Additionally, it would be important to learn the durability of these bonds over a

greater time span; bond strengths and nano-leakage should be examined after

storage of the samples for more than six months, or even years. It would also

be useful to investigate interface-degradation patterns under more realistic

253

conditions using an in situ model. This method, considered as an intermediate

stage between in vitro and in vivo studies, would permit to ageing the resin-

bonded interfaces in a relatively short period of time mimicking most of the

challenging conditions adhesive restorations are submitted to in the oral

environment (Reinke et al., 2012). Finally, clinical studies should be conducted

to support laboratory data, in order to point out if modified adhesive systems

may really provide restoration longevity.

254

Bibliography

255

ABOU NEEL, E. A., PALMER, G., KNOWLES, J. C., SALIH, V. & YOUNG, A. M.

2010. Chemical, modulus and cell attachment studies of reactive calcium

phosphate filler-containing fast photo-curing, surface-degrading, polymeric bone

adhesives. Acta Biomater, 6, 2695-703.

ABREU, A., LOZA, M. A., ELIAS, A., MUKHOPADHYAY, S., LOONEY, S. &

RUEGGEBERG, F. A. 2009. Tensile bond strength of an adhesive resin cement

to different alloys having various surface treatments. J Prosthet Dent, 101, 107-

18.

AGEE, K. A., BECKER, T. D., JOYCE, A. P., RUEGGEBERG, F. A., BORKE, J. L.,

WALLER, J. L., TAY, F. R. & PASHLEY, D. H. 2006. Net Expansion of dried

demineralized dentin matrix produced by monomer/alcohol infiltration and

solvent evaporation. J Biomed Mater Res A, 79, 349-58.

AIDA, M., ODAKI, M., FUJITA, K., KITAGAWA, T., TESHIMA, I. & SUZUKI, K.

2009. Degradation-stage effect of self-etching primer on dentin bond durability.

J Dent Res, 88, 443-8.

AIMES, R. T. & QUIGLEY, J. P. 1995. Matrix metalloproteinase-2 is an interstitial

collagenase. Inhibitor-free enzyme catalyzes the cleavage of collagen fibrils and

soluble native type I collagen generating the specific 3/4- and 1/4-length

fragments. J Biol Chem, 270, 5872-6.

AKINMADE, A. O. & NICHOLSON, J. W. 1993. Glass-ionomer cements as

adhesives. Part I. Fundamental aspects and their clinical relevance. J Mater Sci

Mater Med, 4, 95-101.

AL-AMMAR, A., DRUMMOND, J. L. & BEDRAN-RUSSO, A. K. 2009. The use of

collagen cross-linking agents to enhance dentin bond strength. J Biomed Mater

Res B Appl Biomater, 91, 419-24.

256

ALLEN, E. P., BAYNE, S. C., BECKER, I. M., DONOVAN, T. E., HUME, W. R. &

KOIS, J. C. 1999. Annual review of selected dental literature: report of the

Committee on Scientific Investigation of the American Academy of Restorative

Dentistry. J Prosthet Dent, 82, 27-66.

AMARAL, F. L., COLUCCI, V., PALMA-DIBB, R. G. & CORONA, S. A. 2007.

Assessment of in vitro methods used to promote adhesive interface degradation:

a critical review. J Esthet Restor Dent, 19, 340-53.

ANDERSSON, O. & KANGASNIEMI, I. 1991. Calcium phosphate formation at the

surface of bioactive glass in vitro. J Biomed Mater Res A, 25, 1019-30.

ANDERSSON, O., KARLSSON, K. H., KANGASNIEMI, K. & YLI-URPO, A. 1988.

Model for physical properties and bioacticity of phosphate opal glasses.

Glastech Ber, 61, 300-5.

ANUSAVICE, K. J., ZHANG, N. Z. & SHEN, C. 2006. Controlled release of

chlorhexidine from UDMA-TEGDMA resin. J Dent Res, 85, 950-4.

AOBA, T., MORENO, E. C. & SHIMODA, S. 1992. Competitive Adsorption of

Magnesium and Calcium-Ions onto Synthetic and Biological Apatites. Calcif

Tissue Int, 51, 143-50.

ARANHA, A. M., GIRO, E. M., SOUZA, P. P., HEBLING, J. & DE SOUZA COSTA,

C. A. 2006. Effect of curing regime on the cytotoxicity of resin-modified glass-

ionomer lining cements applied to an odontoblast-cell line. Dent Mater, 22, 864-

9.

ARENDS, J., RUBEN, J. L. & INABA, D. 1997. Major Topics in Quantitative

Microradiography of Enamel and Dentin: R Parameter, Mineral Distribution

Visualization, and Hyper-Remineralization. Adv Dent Res, 11, 403-14.

257

ASMUSSEN, E., HANSEN, E. K. & PEUTZFELDT, A. 1991. Influence of the

solubility parameter of intermediary resin on the effectiveness of the gluma

bonding system. J Dent Res, 70, 1290-3.

ASMUSSEN, E. & UNO, S. 1993. Solubility parameters, fractional polarities, and bond

strengths of some intermediary resins used in dentin bonding. J Dent Res, 72,

558-65.

ATAI, M., L., S., NODEHI, A., MIRABEDINI, S. M., KASRAEI, S. & AKBARI, K.

2009. PMMA-grafted nanoclay as novel filler for dental adhesives. Dent Mater,

25, 339-47.

ATTAL, J. P., ASMUSSEN, E. & DEGRANGE, M. 1994. Effects of surface treatment

on the free energy of dentin. Dent Mater, 10, 259-64.

ATTIN, T., VATASCHKI, M. & HELLWIG, E. 1996. Properties of resin-modified

glass-ionomer restorative materials and two poly-acid-modified resin composite

materials. Quintessence Int, 27, 203-9.

AVCI, R., SCHWEITZER, M. H., BOYD, R. D., WITTMEYER, J. L., TERÁN ARCE,

F. & CALVO, J. O. 2005. Preservation of bone collagen from the late

Cretaceous period studied by immunological techniques and atomic force

microscopy. Langmuir, 21, 3584-90.

AVILA, M. Y. & NAVIA, J. L. 2010. Effect of genipin collagen crosslinking on

porcine corneas. J Cataract Refract Surg, 36, 659-64.

BAIER, R. E. 1992. Principles of adhesion. Oper Dent Suppl, 5, 1-9.

BAKRY, A. S., TAKAHASHI, H., OTSUKI, M., SADR, A., YAMASHITA, K. &

TAGAMI, J. 2011a. CO2 laser improves 45S5 bioglass interaction with dentin.

J Dent Res, 90, 246-50.

258

BAKRY, A. S., TAMURA, Y., OTSUKI, M., KASUGAI, S., OHYA, K. & TAGAMI,

J. 2011b. Cytotoxicity of 45S5 bioglass paste used for dentine hypersensitivity

treatment. J Dent, 39, 599-603.

BALOOCH, M., HABELITZ, S., KINNEY, J. H., MARSHALL, S. J. & MARSHALL,

G. W. 2008. Mechanical properties of mineralized collagen fibrils as influenced

by demineralization. J Struct Biol, 162, 404-10.

BANERJEE, A., SHERRIFF, M., KIDD, E. A. & WATSON, T. F. 1999. A confocal

microscopic study relating the autofluorescence of carious dentine to its

microhardness. Br Dent J, 187, 206-10.

BARBOSA, S. V., SAFAVI, K. E. & SPÅNGBERG, S. W. 1994. Influence of sodium

hypochlorite on the permeability and structure of cervical human dentine. Int

Endod J, 27, 309-12.

BARKMEIER, W. W. & COOLEY, R. L. 1992. Laboratory evaluation of adhesive

systems. Oper Dent, 5, 50-61.

BARRÉRE, F., LAYROLLE, P., VAN BLITTERSWIJK, C. A. & DE GROOT, K.

1999. Biomimetic calcium phosphate coatings on Ti6AI4V: a crystal growth

study of octacalcium phosphate and inhibition by Mg2+ and HCO3-. Bone, 25,

107S-11S.

BARRÈRE, F., VAN BLITTERSWIJK, C. A. & DE GROOT, K. 2006. Bone

regeneration: molecular and cellular interactions with calcium phosphate

ceramics. Int J Nanomedicine, 1, 317-32.

BAUER, T. W. & SMITH, S. T. 2002. Bioactive materials in orthopaedic surgery:

overview and regulatory considerations. Clin Orthop Relat Res, 395, 11-22.

259

BAYLE, M. A., GRÉGOIRE, G. & SHARROCK, P. 2007. The role of

acrylophosphonic acid monomers in the formation of hybrid layers based on

self-etch adhesives. J Dent, 35, 302-8.

BEDRAN-DE-CASTRO, A. K., PEREIRA, P. N., PIMENTA, L. A. & THOMPSON, J.

Y. 2004. Effect of thermal and mechanical load cycling on microtensile bond

strength of a total-etch adhesive system. Oper Dent, 29, 150-6.

BEDRAN-RUSSO, A. K., PASHLEY, D. H., AGEE, K., DRUMMOND, J. L. &

MIESCKE, K. J. 2008. Changes in stiffness of demineralized dentin following

application of collagen crosslinkers. J Biomed Mater Res B Appl Biomater, 86,

330-4.

BEDRAN-RUSSO, A. K., PEREIRA, P. N., DUARTE, W. R., DRUMMOND, J. L. &

YAMAUCHI, M. 2007. Application of crosslinkers to dentin collagen enhances

the ultimate tensile strength. J Biomed Mater Res B Appl Biomater, 80, 268-72.

BEDRAN-RUSSO, A. K., VIDAL, C. M., DOS SANTOS, P. H. & CASTELLAN, C.

S. 2010. Long-term effect of carbodiimide on dentin matrix and resin-dentin

bonds. J Biomed Mater Res B Appl Biomater, 94, 250-5.

BEDRAN-RUSSO, A. K., YOO, K. J., EMA, K. C. & PASHLEY, D. H. 2009.

Mechanical properties of tannic-acid-treated dentin matrix. J Dent Res, 88, 807-

11.

BELL, R. B. & BARKMEIER, W. W. 1994. Glass-ionomer restoratives and liners:

shear bond strength to dentin. J Esthet Dent, 6, 129-34.

BERRY, E. A. & POWERS, J. M. 1994. Bond strength of glass ionomers to coronal

and radicular dentin. Oper Dent, 19, 122-6.

260

BERTACCHINI, S. M., ABATE, P. F., BLANK, A., BAGLIETO, M. F. & MACCHI,

R. L. 1999. Solubility and fluoride release in ionomers and compomers.

Quintessence Int, 30, 193-7.

BERTASSONI, L. E., HABELITZ, S., KINNEY, J. H., MARSHALL, S. J. &

MARSHALL, G. W. 2009. Biomechanical perspective on the remineralization

of dentin. Caries Res, 43, 70-7.

BERTASSONI, L. E., HABELITZ, S., MARSHALL, S. J. & MARSHALL, G. W.

2011. Mechanical recovery of dentin following remineralization in vitro--an

indentation study. J Biomech, 44, 176-81.

BHATTACHARYYA, K. G. & GUPTA, S. S. 2008. Adsorption of a few heavy metals

on natural and modified kaolinite and montmorillonite: a review. Adv Colloid

Interface Sci, 140, 114-31.

BLACKBURN, R. S., HARVEY, A., KETTLE, L. L., MANIAN, A. P., PAYNE, J. D.

& RUSSELL, S. J. 2007. Sorption of chlorhexidine on cellulose: mechanism of

binding and molecular recognition. J Phys Chem B, 111, 8775-84.

BONDING, N., GRAEM, N., RYGAARD, J. & DABELSTEEN, E. 1987. Effects of

irradiation with dental light curing units on Langerhans cells in human stratified

epithelium in heterotransplanted skin. Scan J Dent Res, 95, 463-6.

BOSKEY, A. L. 2007. Mineralization of Bones and Teeth. Elements Magazine, 3, 385-

91.

BOUILLAGUET, S., CIUCCHI, B., JACOBY, T., WATAHA, J. C. & PASHLEY, D.

2001. Bonding characteristics to dentin walls of class II cavities, in vitro. Dent

Mater, 17, 316-21.

261

BOUILLAGUET, S., WATAHA, J. C., HANKS, C. T., CIUCCHI, B. & HOLZ, J.

1996. In vitro cytotoxicity and dentin permeability of HEMA. J Endod, 22, 244-

8.

BOUSHELL, L. W., KAKU, M., MOCHIDA, Y., BAGNELL, R. & YAMAUCHI, M.

2008. Immunohistochemical localization of matrixmetalloproteinase-2 in human

coronal dentin. Arch Oral Biol, 53, 109-16.

BOWEN, R. L. 1963. Properties of a silica-reinforced polymer for dental restorations. J

Am Dent Assoc, 66, 57-64.

BOWEN, R. L., BENNETT, P. S., GROH, R. J., FARAHANI, M. & EICHMILLER, F.

C. 1996. New surface-active comonomer for adhesive bonding. J Dent Res, 75,

606-10.

BOYDE, A. 1974. Transmission electron microscopy of ion beam thinned dentine. Cell

Tissue Res, 152, 543-50.

BRACKETT, W. W., COVEY, D. A. & ST-GERMAIN, H. A. 2002. One-year clinical

performance of a self-etching adhesive in class V resin composites cured by two

methods. Oper Dent, 27, 218-22.

BRACKETT, W. W., ITO, S., TAY, F. R., HAISCH, L. D. & PASHLEY, D. H. 2005.

Microtensile dentin bond strength of self-etching resins: Effect of a hydrophobic

layer. Oper Dent, 30, 734-9.

BRAEM, M. 2007. Microshear fatigue testing of tooth/adhesive interfaces. J Adhes

Dent, 9, 249-53.

BRANNSTRÖM, M. & ASTRÖM, A. 1972. The hydrodynamics of the dentine; its

possible relationship to dentinal pain. Int Dent J, 22, 219-27.

BRÄNNSTRÖM, M., JOHNSON, G. & NORDENVALL, K. J. 1981. Resin

impregnation of sensitive dentin surfaces. Odontoiatr Prat, 16, 5-18.

262

BRAUER, D. S., KARPUKHINA, N., O’DONNELL, M. D., LAW, R. V. & HILL, R.

G. 2010. Fluoride-containing bioactive glasses: effect of glass design and

structure on degradation, pH and apatite formation in simulated body fluid. Acta

Biomater, 6, 3275-82.

BRAZEL, C. S. & PEPPAS, N. A. 1999. Dimensionless analysis of swelling of

hydrophilic glassy polymers with subsequent drug release from relaxing

structures. Biomaterials, 20, 721-32.

BRESCHI, L., CAMMELLI, F., VISINTINI, E., MAZZONI, A., VITA, F.,

CARRILHO, M., CADENARO, M., FOULGER, S., MAZZOTI, G., TAY, F.,

DI LENARDA, R. & PASHLEY, D. 2009. Influence of chlorhexidine

concentration on the durability of etch-and-rinse dentin bonds: a 12-month in

vitro study. J Adhes Dent, 11, 191-8.

BRESCHI, L., MARTIN, P., MAZZONI, A., NATO, F., CARRILHO, M.,

TJÄDERHANE, L., VISINTINI, E., CADENARO, M., TAY, F. R., DE

STEFANO DORIGO, E. & PASHLEY, D. H. 2010a. Use of a specific MMP-

inhibitor (galardin) for preservation of hybrid layer. Dent Mater, 26, 571-8.

BRESCHI, L., MAZZONI, A., NATO, F., CARRILHO, M., VISINTINI, E. T., L.,

RUGGERI, A. J. R., TAY, F. R., S., D. & PASHLEY, D. H. 2010b.

Chlorhexidine stabilizes the adhesive interface: a 2-year in vitro study. Dent

Mater, 26, 320-5.

BRESCHI, L., MAZZONI, A., RUGGERI, A., CADENARO, M., DI LENARDA, R. &

DORIGO, E. 2008. Dental adhesion review: aging and stability of the bonded

interface. Dent Mater, 24, 90-101.

263

BRESCIANI, E., WAGNER, W. C., NAVARRO, M. F., DICKENS, S. H. & PETERS,

M. C. 2010. In vivo dentin microhardness beneath calcium-phosphate cement. J

Dent Res, 89, 836-41.

BUNKER, B. C., TALLANT, D. R. & HEADLEY, T. J. 1988. The structure of leached

sodium borosilicate glass. Phys Chem Class, 29, 106-20.

BUONOCORE, M. G. 1955. A simple method of increasing the adhesion of acrylic

filling materials to enamel surfaces. J Dent Res, 34, 849-53.

BUONOCORE, M. G. & QUIGLEY, M. 1958. Bonding of synthetic resin material to

human dentin: preliminary histological study of the bond area. J Am Dent Assoc,

57, 807-11.

BURKE, F. J. T., HUSSAIN, A., NOLAN, L. & FLEMING, G. J. P. 2008. Methods

used in dentine bonding tests: an analysis of 102 investigations on bond

strength. Eur J Prosthodont Rest Dent, 16, 158-65.

BUSENLEHNER, L. S. & ARMSTRONG, R. N. 2005. Insights into enzyme structure

and dynamics elucidated by amide H/D exchange mass spectrometry. Arch

Biochem Biophys, 433, 34-46.

CADENARO, M., ANTONIOLLI, F., CODAN, B., AGEE, K., TAY, F. R., DORIGO,

E., PASHLEY, D. H. & BRESCHI, L. 2010. Influence of different initiators on

the degree of conversion of experimental adhesive blends in relation to their

hydrophilicity and solvent content. Dent Mater, 26, 288-94.

CADENARO, M., BRESCHI, L., RUEGGEBERG, F. A., AGEE, K., DI LENARDA,

R., CARRILHO, M., TAY, F. R. & PASHLEY, D. H. 2009a. Effect of adhesive

hydrophilicity and curing time on the permeability of resins bonded to water vs.

ethanol-saturated acid-etched dentin. Dent Mater, 25, 39-47.

264

CADENARO, M., PASHLEY, D. H., MARCHESI, G., CARRILHO, M.,

ANTONIOLLI, F., MAZZONI, A., TAY, F. R., DI LENARDA, R. &

BRESCHI, L. 2009b. Influence of chlorhexidine on the degree of conversion

and E-modulus of experimental adhesive blends. Dent Mater, 25, 1269-74.

CALERO, P., JORGE-HERRERO, E., TURNAY, J., OLMO, N., LÓPEZ DE

SILANES, I., LIZARBE, M. A., MAESTRO, M. M., ARENAZ, B. &

CASTILLO-OLIVARES, J. L. 2002. Gelatinases in soft tissue biomaterials.

Analysis of different crosslinking agents. Biomaterials. Biomaterials, 23, 3473-

8.

CAMILLERI, J. 2008. Characterization of hydration products of mineral trioxide

aggregate. Int Endod J, 41, 408-17.

CARDOSO, M. V., COUTINHO, E., ERMIS, R. B., POITEVIN, A., VAN

LANDUYT, K. & DE MUNCK, J. 2008. Influence of dentin cavity surface

finishing on micro-tensile bond strength of adhesives. Dent Mater, 24, 492-501.

CARRILHO, M. R., CARVALHO, R. M., DE GOES, M. F., DI HIPÓLITO, V.,

GERALDELI, S., TAY, F. R., PASHLEY, D. H. & TJÄDERHANE, L. 2007a.

Chlorhexidine preserves dentin bond in vitro. J Dent Res, 86, 90-4.

CARRILHO, M. R., CARVALHO, R. M., SOUSA, E. N., NICOLAU, J., BRESCHI,

L., MAZZONI, A., TJÄDERHANE, L., TAY, F. R., AGEE, K. & PASHLEY,

D. H. 2010. Substantivity of chlorhexidine to human dentin. Dent Mater, 26,

779-85.

CARRILHO, M. R., CARVALHO, R. M., TAY, F. R., YIU, C. & PASHLEY, D. H.

2005a. Durability of resin-dentin bonds related to water and oil storage. Am J

Dent, 18, 315-9.

265

CARRILHO, M. R., GERALDELI, S., TAY, F., DE GOES, M. F., CARVALHO, R.

M., TJÄDERHANE, L., REIS, A. F., HEBLING, J., MAZZONI, A., BRESCHI,

L. & PASHLEY, D. H. 2007b. In vivo preservation of the hybrid layer by

chlorhexidine. J Dent Res, 86, 529-33.

CARRILHO, M. R. O., F.R., T., PASHLEY, D. H., TJADERHANE, L. &

CARVALHO, R. M. 2005b. Mechanical degradation of resin dentin bond

components. Dent Mater, 21, 232-41.

CARVALHO, R. M., CHERSONI, S., FRANKENBERGER, R., PASHLEY, D. H.,

PRATI, C. & TAY, F. R. 2005. A challenge to the conventional wisdom that

simultaneous etching and resin infiltration always occurs in self-etch adhesives.

Biomaterials, 26, 1035-42.

CARVALHO, R. M., SANO, H., HOMER, J. A., BREWER, P. D. & PASHLEY, D. H.

1996. Comparison of interfacial morphology and strength of bonds made to

superficial and deep dentin. Am J Dent, 8, 297-302.

CASTELLAN, C. S., PEREIRA, P. N., GRANDE, R. H. & BEDRAN-RUSSO, A. K.

2010a. Mechanical characterization of proanthocyanidin-dentin matrix

interaction. Dent Mater, 26, 968-73.

CASTELLAN, C. S., PEREIRA, P. N., VIANA, G., CHEN, S. N., PAULI, G. F. &

BEDRAN-RUSSO, A. K. 2010b. Solubility study of phytochemical cross-

linking agents on dentin stiffness. J Dent, 38, 431-6.

CATTANI-LORENTE, M. A., DUPUIS, V., MOYA, F., PAYAN, J. & MEYER, J. M.

1999. Comparative study of the physical properties of a polyacid-modified

composite resin and a resin-modified glass ionomer cement. Dent Mater, 15, 21-

32.

266

CHADWICK, R. G., TRAYNOR, N., MOSELEY, H. & GIBBS, N. 1994. Blue light

curing units – a dermatological hazard. Br Dent J, 176, 17-31.

CHARULATHA, V. & RAJARAM, A. 2003. Influence of different crosslinking

treatments on the physical properties of collagen membranes. Biomaterials, 24,

759-67.

CHAUSSAIN-MILLER, C., FIORETTI, F., GOLDBERG, M. & MENASHI, S. 2006.

The role of matrix metalloproteinases (MMPs) in human caries. J Dent Res, 85,

22-32.

CHENG, H., CATERSON, B., NEAME, P. J., LESTER, G. E. & YAMAUCHI, M.

1996. Differential distribution of lumican and fibromodulin in tooth cementum.

Connect Tissue Res, 34, 87-96.

CHERSONI, S., SUPPA, P., GRANDINI, S., GORACCI, C., MONTICELLI, F., YIU,

C., HUANG, C., PRATI, C., BRESCHI, L., FERRARI, M., PASHLEY, D. H.

& TAY, F. R. 2004. In vivo and in vitro permeability of one-step self-etch

adhesives. J Dent Res, 83, 459-64.

CHESNICK, I. E., MASON, J. T., GIUSEPPETTI, A. A., EIDELMAN, N. &

POTTER, K. 2008. Magnetic resonance microscopy of collagen mineralization.

Biophys J, 95, 2017-26.

CHIARAPUTT, S., MAI, S., HUFFMAN, B. P., KAPUR, R., AGEE, K. A., YIU, C.

K., CHAN, D. C., HARNIRATTISAI, C., AROLA, D. D., RUEGGEBERG, F.

A., PASHLEY, D. H. & TAY, F. R. 2008. Changes in resin-infiltrated dentin

stiffness after water storage. J Dent Res, 87, 655-60.

CHIGIRA, H., YUKITANI, W. & HASEGAWA, T. 1994. Self-etching dentin primers

containing Phenyl-P. J Dent Res, 73, 1088-95.

267

CLARKSON, B. H., FEAGIN, F. F., MCCURDY, S. P., SHEETZ, J. H. & SPEIRS, R.

1991. Effects of phosphoprotein moieties on the remineralization of human root

caries. Caries Res, 25, 166-73.

COLEMAN, N. J., AWOSANYA, K. & NICHOLSON, J. W. 2009. Aspects of the in

vitro bioactivity of hydraulic calcium (alumino)silicate cement. J Biomed Mater

Res A, 90, 166-74.

COUTINHO, E., VAN LANDUYT, K., DE MUNCK, J., POITEVIN, A., YOSHIDA,

Y. & INOUE, S. 2006. Development of a self-etch adhesive for resin-modified

glass ionomers. J Dent Res, 85, 349-53.

CRAIG, R. G. & PEYTON, F. A. 1958. The micro-hardness of enamel and dentin. J

Dent Res, 37, 661-8.

CROLL, T. P., KILLIAN, C. M. & HELPIN, M. L. 1993. A restorative dentistry

renaissance for children: light-hardened glass ionomer/resin cement. ASDC J

Dent Child, 60, 89-94.

CURREY, J. D. & BREAR, K. 1990. Young’s modulus and yield strength in

mammalian mineralized tissues. J Mater Sci Mater Med, 1, 14-20.

CURTIS, A. R., WEST, N. X. & SU, B. 2010. Synthesis of nanobioglass and formation

of apatite rods to occlude exposed dentine tubules and eliminate

hypersensitivity. Acta Biomater, 6, 3740-6.

CURTIS, R. V. & WATSON, T. F. 2008. Dental biomaterials: imaging, testing and

modeling. Abington Hall, England: Woodhead Publishing Ltd.

CUSSLER, E. L. 1976. “Multicomponent Diffusion”. Elsevier Scientific Pub.,

Amsterdam, Netherlands.

DARVELL, B. W. & WU, R. C. 2011. "MTA"-an hydraulic silicate cement: review

update and setting reaction. Dent Mater, 27, 407-22.

268

DE ALMEIDA, J. B., PLATT, J. A., OSHIDA, Y., MOORE, B. K., COCHRAN, M. A.

& ECKERT, G. J. 2003. Three different methods to evaluate microleakage of

packable composites in Class II restorations. Oper Dent, 28, 453-60.

DE GEE, A. J., LELOUP, G., WERNER, A., VREVEN, J. & DAVIDSON, C. L. 1998.

Structural integrity of resin-modified glass ionomers as affected by the delay or

omission of light activation. J Dent Res, 77, 1658-63.

DE GROOT, K. 1983. Ceramics of calcium phosphates: preparation and properties.

Bioceramics of calcium phosphate (de Groot K. ed.) CRC Press Inc, 100-11.

DE MUNCK, J., ERNIS, R. B., KOSHIRO, K., INOUE, S., IKEDA, T., SANO, H.,

VAN LANDUYT, K. L. & VAN MEERBEEK, B. 2007. NaOCl degradation of

a HEMA-free all-in-one adhesive bonded to enamel and dentin following two

air-blowing techniques. J Dent, 35, 74-83.

DE MUNCK, J., MINE, A., VAN DEN STEEN, P. E., VAN LANDUYT, K. L.,

POITEVIN, A. & OPDENAKKER, G. 2010. Enzymatic degradation of

adhesive-dentin interfaces produced by mild self-etch adhesives. Eur J Oral Sci,

118, 494-501.

DE MUNCK, J., SHIRAI, K., YOSHIDA, Y., INOUE, S., VAN LANDUYT, K.,

LAMBRECHTS, P., SUZUKI, K., SHINTANI, H. & VAN MEERBEEK, B.

2006. Effect of water storage on the bonding effectiveness of 6 adhesives to

Class I cavity dentin. Oper Dent, 31, 456-65.

DE MUNCK, J., VAN DEN STEEN, P. E., MINE, A., VAN LANDUYT, K. L.,

POITEVIN, A., OPDENAKKER, G. & VAN MEERBEEK, B. 2009. Inhibition

of enzymatic degradation of adhesive-dentin interfaces. J Dent Res, 88, 1101-6.

269

DE MUNCK, J., VAN LANDUYT, K., PEUMANS, M., POITEVIN, A.,

LAMBRECHTS, P. & BRAEM, M. 2005. A critical review of the durability of

adhesion to tooth tissue: methods and results. J Dent Res, 84, 118-32.

DE MUNCK, J., VAN MEERBEEK, B., YOSHIDA, Y., INOUE, S., VARGAS, M.,

SUZUKI, K., LAMBRECHTS, P. & VANHERLE, G. 2003. Four-year water

degradation of total-etch adhesives bonded to dentin. J Dent Res, 82, 136-40.

DESHPANDE, A. S. & BENIASH, E. 2008. Bio-inspired Synthesis of Mineralized

Collagen Fibrils. Cryst Growth Des, 8, 3084-90.

DHANPAL, P., YIU, C. K., KING, N. M., TAY, F. R. & HIRAISHI, N. 2009. Effect of

temperature on water sorption and solubility of dental adhesive resins. J Dent,

37, 122-32.

DIBA, M., TAPIA, F., BOCCACCINI, A. R. & STROBEL, L. A. 2012. Magnesium-

containing bioactive glasses for biomedical applications. IJAGS, 2, 221-53.

DICKENS, S. H. & FLAIM, G. M. 2008. Effect of a bonding agent on in vitro

biochemical activities of remineralizing resin-based calcium phosphate cements.

Dent Mater, 24, 1273-80.

DICKENS, S. H., FLAIM, G. M. & TAKAGI, S. 2003. Mechanical properties and

biochemical activity of remineralizing resin-based Ca-PO4 cements. Dent

Mater, 19, 558-66.

DICKINSON, D. P. 2002. Cysteine peptidases of mammals: their biological roles and

potential effects in the oral cavity and other tissues in health and disease. Crit

Rev Oral Biol Med, 13, 238-75.

DOMINGO, C., ARCIS, R. W., OSORIO, E., OSORIO, R., FANOVICH, M. A. &

RODRIGUEZ-CLEMENTE, R. 2003. Hydrolytic stability of experimental

hydroxyapatite-filled dental composite materials. Dent Mater, 19, 478-86.

270

DONMEZ, N., BELLI, S., PASHLEY, D. H. & TAY, F. R. 2005. Ultrastructural

correlates of in vivo/in vitro bond degradation in self-etch adhesives. J Dent Res,

84, 355-9.

DRUMMOND, J. L., SAKAGUCHI, R. L., RACEAN, D. C., WOZNY, J. &

STEINBERG, A. D. 1996. Testing mode and surface treatment effects on dentin

bonding. J Biomed Mater Res, 32, 533-41.

EANES, E. D., GILLESSEN, I. H. & POSNER, A. S. 1965. Intermediate states in the

precipitation of hydroxyapatite. Nature, 208, 365-7.

EDDLESTON, C., HINDLE, A., RUEGGEBERG, F. A., AGEE, K., CARVALHO, R.,

TAY, F. R. & PASHLEY, D. H. 2003. Dimensional changes in acid-

demineralized matrices following the use of HEMA-water vs. HEMA-alcohol

primers. J Biomed Mater Res A, 67, 267-82.

EFFLANDT, S. E., MAGNE, P., DOUGLAS, W. H. & FRANCIS, L. F. 2002.

Interaction between bioactive glasses and human dentin. J Mater Sci Mater Med,

13, 557-65.

EICK, J. D., COBB, C. M., CHAPELL, R. P., SPENCER, P. & ROBINSON, S. J.

1991. The dentinal surface: its influence on dentinal adhesion. Part I.

Quintessence Int, 22, 967-77.

EICK, J. D., GWINNETT, A. J., PASHLEY, D. H. & ROBINSON, S. J. 1997. Current

Concepts On Adhesion To Dentin. Crit Rev Oral Biol Med, 8, 306-35.

EICK, J. D., MILLER, R. G., ROBINSON, S. I., BOWLES, C. O., GUTSHALL, P. L.

& CHAPPELOW, C. C. 1996. Quantitative analysis of the dentin adhesive

interface by Auger spectroscopy. J Dent Res, 75, 1027-33.

271

ELIADES, G., VOUGIOUKLAKIS, G. & PALAGHIAS, G. 2001. Heterogeneous

distribution of single-bottle adhesive monomers in the resin-dentin interdiffusion

zone. Dent Mater, 17, 277-83.

ENDO, K., HASHIMOTO, M., HARAGUCHI, K. & OHNO, H. 2010. Crystal growth

by restorative filling materials. Eur J Oral Sci, 118, 489-93.

ENGQVIST, H., SCHULTZ-WALZ, J. E., LOOF, J., BOTTON, G. A., MAYER, D. &

PHANEUF, M. W. 2004. Chemical and biological integration of a mouldable

bioactive ceramic material capable of forming apatite in vivo in teeth.

Biomaterials, 25, 2781-7.

ERICKSON, R. L. 1989. Mechanism and clinical implication of bond formation for two

dentin bonding agents. Am J Dent, 2, 117-23.

ERICKSON, R. L., BARKMEIER, W. W. & KIMMES, N. S. 2009. Fatigue of enamel

bonds with self-etch adhesives. Dent Mater, 25, 716-20.

ERMIS, R. B., DE MUNCK, J., CARDOSO, M. V., COUTINHO, E., VAN

LANDUYT, K. L. & POITEVIN, A. 2008. Bond strength of self-etch adhesives

to dentin prepared with three different diamond burs. Dent Mater, 24, 978-85.

FABRE, H. S., FABRE, S., CEFALY, D. F., DE OLIVEIRA CARRILHO, M. R.,

GARCIA, F. C. & WANG, L. 2007. Water sorption and solubility of dentin

bonding agents light-cured with different light sources. J Dent, 35, 253-8.

FEATHERSTONE, J. D. 1990. An updated understanding of the mechanism of dental

decay and its prevention. Nutr Q, 14, 5-11.

FEATHERSTONE, J. D. 1996. Modeling the caries-inhibitory effects of dental

materials. Dent Mater, 12, 194-7.

272

FEDARKO, N. S., JAIN, A., KARADAG, A. & FISHER, L. W. 2004. Three small

integrin binding ligand N-linked glycoproteins (SIBLINGs) bind and activate

specific matrix metalloproteinases. FASEB J, 18, 734-6.

FENG, L., LI, S., LI, Y., LI, H., ZHANG, L., ZHAI, J., SONG, Y., LIU, B., JIANG, L.

& ZHU, D. 2002. Super-Hydrophobic Surfaces: From Natural to Artificial. Adv

Mater, 14, 1857-60.

FERRACANE, J. L. 1994. Review Elution of leachable components from composites. J

Oral Rehabil, 21, 441-52.

FERRACANE, J. L. 2006. Hygroscopic and hydrolytic effects in dental polymer

networks. Dent Mater, 22, 211-22.

FILGUEIRAS, M. R., LATORRE, G. P. & HENCH, L. L. 1993. Solution effects on the

surface reactions of a bioactive glass. J Biomed Mater Res, 27, 445-53.

FINER, Y., JAFFER, F. & SANTERRE, J. P. 2004. Mutual influence of cholesterol

esterase and pseudocholinesterase on the biodegradation of dental composites.

Biomaterials, 25, 1787-93.

FINGER, W. J., INOUE, M. & ASMUSSEN, E. 1994. Effect of wettability of adhesive

resins on bonding to dentin. Am J Dent, 7, 35-8.

FRANKENBERGER, R., KRÄMER, N. & PETSCHELT, A. 1999. Fatigue behaviour

of different dentin adhesives. Clin Oral Investig, 3, 11-7.

FRANKENBERGER, R., PASHLEY, D. H., REICH, S. M., LOHBAUER, U.,

PETSCHELT, A. & TAY, F. R. 2005. Characterisation of resin-dentine

interfaces by compressive cyclic loading. Biomaterials, 26, 2043-52.

FROST, H. M. 1963. Measurement of human bone formation by means of tetracycline

labelling. Can J Biochem Physiol, 41, 31.

273

FRUJIKAWA, S., YOKOTA, T., KONGA, K. & KUMADA, J. 1987. The continuous

hydrolysis of geniposide to genipin using immobilized betaglucosidase on

calcium alginate gel. J Biotechnol Lett, 9, 697-702.

FU, X., J.L., K., BERGT, C., KASSIM, S. Y., HUQ, N. P., D'AVIGNON, A., PARKS,

W. C., MECHAM, R. P. & HEINECKE, J. W. 2004. Oxidative cross-linking of

tryptophan to glycine restrains matrix metalloproteinase activity: specific

structural motifs control protein oxidation. J Biol Chem, 279, 6209-12

FUENTES, V., TOLEDANO, M., OSORIO, R. & CARVALHO, R. M. 2003.

Microhardness of superficial and deep sound human dentin. J Biomed Mater Res

A, 66A, 850-3.

FUKUOKA, A., KOSHIRO, K., INOUE, S., YOSHIDA, Y., TANAKA, T., IKEDA,

T., SUZUKI, K., SANO, H. & MEERBEEK, B. V. 2011. Hydrolytic stability of

one-step self-etching adhesives bonded to dentin. J Adhes Dent, 13, 243-8.

FUNG, D. T., WANG, V. M., LAUDIER, D. M., SHINE, J. H., BASTA-PLJAKIC, J.,

JEPSEN, K. J., SCHAFFLER, M. B. & FLATOW, E. L. 2009. Subrupture

tendon fatigue damage. J Orthop Res, 27, 264-73.

FUSAYAMA, T. 1980. Concepts in Operative Dentistry. Tokio: Quintessence.

FUSAYAMA, T., OKUSE, K. & HOSODA, H. 1966. Relationship between hardness,

discoloration, and microbial invasion in carious dentin. J Dent Res, 15, 1033-46.

GAJJERAMAN, S., HE, G., NARAYANAN, K. & GEORGE, A. 2008. Biological

assemblies provide novel templates for the synthesis of hierarchical structures

and facilitate cell adhesion. Adv Funct Mater, 18, 3972-80.

GANDOLFI, M. G., TADDEI, P., SIBONI, F., MODENA, E., CIAPETTI, G. &

PRATI, C. 2011a. Development of the foremost light-curable calcium-silicate

274

MTA cement as root-end in oral surgery. Chemical-physical properties,

bioactivity and biological behavior. Dent Mater, 27, e134-57.

GANDOLFI, M. G., TADDEI, P., SIBONI, F., MODENA, E., DE STEFANO, E. D. &

PRATI, C. 2011b. Biomimetic remineralization of human dentin using

promising innovative calcium-silicate hybrid "smart" materials. Dent Mater, 27,

1055-69.

GANDOLFI, M. G., TADDEI, P., TINTI, A., DORIGO DE STEFANO, E., ROSSI, P.

L. & PRATI, C. 2010a. Kinetics of apatite formation on a calcium-silicate

cement for root-end filling during ageing in physiological-like phosphate

solutions. Clin Oral Investig, 14, 659-68.

GANDOLFI, M. G., TADDEI, P., TINTI, A. & PRATI, C. 2010b. Apatite-forming

ability of ProRoot MTA. Int Endod J, 43, 917-29.

GANDOLFI, M. G., VAN LANDUYT, K., TADDEI, P., MODENA, E., VAN

MEERBEEK, B. & PRATI, C. 2010c. Environmental scanning electron

microscopy connected with energy dispersive x-ray analysis and Raman

techniques to study ProRoot mineral trioxide aggregate and calcium silicate

cements in wet conditions and in real time. J Endod, 36, 851-7.

GARBEROGLIO, R. & BRÄNNSTRÖM, M. 1976. Scanning electron microscopic

investigation of human dentinal tubules. Arch Oral Biol, 21, 355-62.

GARCIA-GODOY, F., KRAMER, N., FEILZER, A. J. & FRANKENBERGER, R.

2010. Longterm degradation of enamel and dentin bonds: 6-year results in vitro

vs. in vivo. Dent Mater, 26, 1113-8.

GARCÍA-GODOY, F., TAY, F. R., PASHLEY, D. H., FEILZER, A., TJÄDERHANE,

L. & PASHLEY, E. L. 2007. Degradation of resin-bonded human dentin after 3

years of storage. Am J Dent, 20, 109-13.

275

GARNERO, P., BOREL, O., BYRJALSEN, I., FERRERAS, M., DRAKE, F. H. &

MCQUENEY, M. S. 1998. The collagenolytic activity of cathepsin K is unique

among mammalian proteinases. J Biol Chem, 273, 32347-52.

GATTI, A., VALDRE, G. & ANDERSSON, O. 1994. Analysis of the in vivo reactions

of a bioactive glass in soft and hard tissue. Biomaterials, 15, 208-12.

GENDRON, R., GRENIER, D., SORSA, T. & MAYRAND, D. 1999. Inhibition of the

activities of matrix metalloproteinases 2, 8, and 9 by chlorhexidine. Clin Diagn

Lab Immunol, 6, 437-9.

GEORGE, A. & VEIS, A. 2008. Phosphorylated proteins and control over apatite

nucleation, crystal growth, and inhibition. Chem Rev, 108, 4670-93.

GERICKE, A., QIN, C., SUN, Y., REDFERN, R., REDFERN, D., FUJIMOTO, Y.,

TALEB, H., BUTLER, W. T. & BOSKEY, A. L. 2004. Different forms of

DMP1 play distinct roles in mineralization. J Dent Res, 89, 355-9.

GERZINA, T. M. & HUME, W. R. 1996. Diffusion of monomers from bonding resin-

resin composite combinations through dentine in vitro. J Dent, 24, 125-8.

GEURTSEN, W. 1998. Substances released from dental resin composites and glass

ionomer cements. Eur J Oral Sci, 106, 687-95.

GEURTSEN, W. 2000. Biocompatibility of resin-modified filling materials. Crit Rev

Oral Biol Med, 11, 333-55.

GIANNINI, M., SOARES, C. J. & DE CARVALHO, R. M. 2004. Ultimate tensile

strength of tooth structures. Dent Mater, 20, 322-9.

GIOIA, M., MONACO, S., FASCIGLIONE, G. F., COLETTI, A., MODESTI, A.,

MARINI, S. & COLETTA, M. 2007. Characterization of the mechanisms by

which gelatinase A, neutrophil collagenase, and membrane-type

276

metalloproteinase MMP-14 recognize collagen I and enzymatically process the

two alpha-chains. J Mol Biol, 368, 1101-13.

GIRIJA, E. K., YOKOGAWA, Y. & NAGATA, F. 2004. Apatite formation on collagen

fibrils in the presence of polyacrylic acid. J Mater Sci Mater Med, 15, 593-9.

GLIMCHER, M. J. 1987. The nature of the mineral component of bone and the

mechanism of calcification. Instr Course Lect, 36, 49-69.

GLIMCHER, M. J., KOSSIVA, D. & BRICKLEY-PARSONS, D. 1984.

Phosphoproteins of chicken bone matrix. Proof of synthesis in bone tissue. J

Biol Chem, 259, 290-3.

GOMBIA, M., BORTOLOTTI, V., DE CARLO, B., MONGIORGI, R., ZANNA, S. &

FANTAZZINI, J. 2010. Nanopore structure buildup during endodontic cement

hydration studied by time-domain nuclear magnetic resonance of lower and

higher mobility 1H. J Phys Chem B, 114, 1767-74.

GONZALEZ-CABEZAS, C. 2010. The chemistry of caries: remineralization and

demineralization events with direct clinical relevance. Dent Clin North Am, 54,

469-78.

GORACCI, C., TAVARES, A. U., FABIANELLI, A., MONTICELLI, F.,

RAFFAELLI, O. & CARDOSO, P. C. 2004. The adhesion between fiber posts

and root canal walls: comparison between microtensile and push-out bond

strength measurements. Eur J Oral Sci, 112, 353-61.

GOWER, L. B. 2008. Biomimetic model systems for investigating the amorphous

precursor pathway and its role in biomineralization. Chem Rev, 108, 4551-627.

GRANDE, R. H., REIS, A., LOGUERCIO, A. D., SINGER JDA, M., SHELLARD, E.

& NETO, P. C. 2005. Adhesive systems used for sealing contaminated surfaces:

a microleakage evaluation. Braz Oral Res, 19, 17-22.

277

GREENSPAN, D. C. 2010. NovaMin and tooth sensitivity--an overview. J Clin Dent,

21, 61-5.

GRIFFITHS, B. M. & WATSON, T. F. 1995. Resin–dentin interface of scotchbond

multi-purpose dentin adhesive. Am J Dent, 8, 212-6.

GRIFFITHS, B. M., WATSON, T. F. & SHERRIFF, M. 1998. The influence of dentine

bonding systems and their handling characteristics on the morphology and

micropermeability of dentine adhesive systems. J Dent, 27, 63-71.

GRIFFITHS, B. M., WATSON, T. F. & SHERRIFF, M. 1999. Variable polymerisation

shrinkage and the interfacial micropermeability of a dentin bonding system. J

Adhes Dent, 2, 119-31.

GU, L., KIM, Y. K., LIU, Y., TAKAHASHI, K., ARUN, S. & WIMMER, C. E. 2011.

Immobilization of a phosphonated analog of matrix phosphoproteins within

cross-linked collagen as a templating mechanism for biomimetic mineralization.

Acta Biomater, 7, 268-77.

GU, L. S., HUFFMAN, B. P., AROLA, D. D., KIM, Y. K., MAI, S. & ELSALANTY,

M. E. 2010. Changes in stiffness of resin-infiltrated demineralized dentin after

remineralization by a bottom-up biomimetic approach. Acta Biomater, 6, 1453-

61

GYSI, A. 1900. An attempt to explain the sensitiveness of dentin. Brit J Dent Sci, 43,

865-8.

HAGGAR, O. 1951. Sevitron cavity seal. British patent 687 299.

HAMID, A., OKAMOTO, A., IWAKU, M. & HUME, W. R. 1998. Component release

from light-activated glass ionomer and compomer cements. J Oral Rehabil, 25,

94-9.

278

HANNAS, A. R., PEREIRA, J. C., GRANJEIRO, J. M. & TJÄDERHANE, L. 2007.

The role of matrix metalloproteinases in the oral environment. Acta Odontol

Scand, 65, 1-13.

HASHIMOTO, M., FUJITA, S., NAGANO, F., OHNO, H. & ENDO, K. 2010a. Ten-

year degradation of resin-dentin bonds. Eur J Oral Sci, 118, 404-10.

HASHIMOTO, M., IIJIMA, M., NAGANO, F., OHNO, H. & ENDO, K. 2010b. Effect

of monomer composition on crystal growth by resin containing bioglass. J

Biomed Mater Res B Appl Biomater, 94, 127-33.

HASHIMOTO, M., NAKAMURA, K., KAGA, M. & YAWAKA, Y. 2008. Crystal

growth by fluoridated adhesive resins. Dent Mater, 24, 457-63.

HASHIMOTO, M., OHNO, H., SANO, H., KAGA, M. & OGUCHI, H. 2003. In vitro

degradation of resin-dentin bonds analysed by microtensile bond test, scanning

and transmission electron microscopy. Biomaterials, 24, 3795-803.

HASHIMOTO, M., OHNO, H., SANO, H., TAY, F. R., KAGA, M., KUDOU, Y.,

OGUCHI, H., ARAKI, Y. & KUBOTA, M. 2002. Micromorphological changes

in resin-dentin bonds after 1 year of water storage. J Biomed Mater Res, 63, 306-

11.

HAYAKAWA, T., KIKUTAKE-SUGIYAMA, K. & NEMOTO, K. 2005. Efficacy of

water-soluble photoinitiator on the adhesion of composite resin to bovine teeth

in all-in-one bonding system. Dent Mater J, 24, 213-8.

HE, G. & GEORGE, A. 2004. Dentin matrix protein 1 immobilized on type I collagen

fibrils facilitates apatite deposition in vitro. J Biol Chem, 279, 11649-56.

HEBLING, J., PASHLEY, D. H., TJÄDERHANE, L. & TAY, F. R. 2005.

Chlorhexidine arrests subclinical degradation of dentin hybrid layers in vivo. J

Dent Res, 84, 741-6.

279

HEIKKILA, J. T., AHO, H. J., YLI-URPO, A., ANDERSSON, O. H., AHO, A. J. &

HAPPONEN, R. P. 1993. Bioactive glass is as good as hydroxylapatite in

reconstruction of experimental osteochondral defects. Acta Orthop Scand, 64,

672-8.

HEIKKILA, J. T., AHO, H. J., YLI-URPO, A., HAPPONEN, R. P. & AHO, A. J. 1995.

Bone formation in rabbit cancellous bone defects filled with bioactive glass

granules. Acta Orthop Scand, 66, 463-7.

HENCH, L.L. 1997. Biomaterials: a forecast for the future. European Materials

Research Society 1997 meeting, symposium G - Biomaterials - perspectives for

research and industry at the century change 1419-23.

HENCH, L.L. 1999. Bioactive glasses and glass-ceramics. Shackelford JF, Ed.

Bioceramics: applications of ceramic and glass materials in medicine. Enfield,

NH: Trans Tech Publications, 37-64.

HENCH, L.L. 2006. The story of Bioglass. J Mater Sci Mater Med, 17, 967-78.

HENCH, L.L. & ANDERSSON, O. 1993. Bioactive glasses. Advanced Series in

Ceramics - Vol. 1. An introduction to Bioceramics. World Scientific Singapore

Hench LL, Wilson J eds, 45-7.

HENCH, L.L. & PASCHALL, H. 1974. Histological responses at a biomaterial's

interface. J Biomed Mater Res, 8, 49-64.

HENCH, L.L. & PASCHALL, H. A. 1973. Direct chemical bond of bioactive glass-

ceramic materials to bone and muscle. J Biomed Mater Res, 7, 25-42.

HENCH, L.L., SPLINTER, R. J., ALLEN, W. C. & GREENLEE, T. K. 1971. Bonding

Mechanisms at the Interface of Ceramic Prosthetic Materials. J Biomed Mater

Res, 5, 117-41.

280

HEWLETT, E. R., CAPUTO, A. A. & WROBET, D. C. 1991. Glass ionomer bond

strength and treatment of dentin with polyacrylic acid. J Prosthet Dent, 66, 767-

72.

HIKITA, K., VAN MEERBEEK, B., DE MUNCK, J., IKEDA, T., VAN LANDUYT,

K. & MAIDA, T. 2007. Bonding effectiveness of adhesive luting agents to

enamel and dentin. Dent Mater, 23, 71-80.

HIRAISHI, N., NISHIYAMA, N., IKEMURA, K., YAU, J. Y., KING, N. M.,

TAGAMI, J., PASHLEY, D. H. & TAY, F. R. 2005. Water concentration in

self-etching primers affects their aggressiveness and bonding efficacy to dentin.

J Dent Res, 84, 653-8.

HIRAISHI, N., YIU, C. K., KING, N. M., TAY, F. R. & PASHLEY, D. H. 2008.

Chlorhexidine release and water sorption characteristics of chlorhexidine-

incorporated hydrophobic/hydrophilic resins. Dent Mater, 24, 1391-9.

HOSAKA, K., NISHITANI, Y., TAGAMI, J., YOSHIYAMA, M., BRACKETT, W.

W., AGEE, K. A., TAY, F. R. & PASHLEY, D. H. 2009. Durability of resin-

dentin bonds to water- vs. ethanol-saturated dentin. J Dent Res, 88, 146-51.

HOSOYA, Y., MARSHALL, S. J., WATANABE, L. G. & MARSHALL, G. W. 2000.

Microhardness of carious deciduous dentin. Oper Dent, 25, 81-9.

HOULE, M. A., GRENIER, D., PLAMONDON, P. & NAKAYAMA, K. 2003. The

collagenase activity of Porphyromonas gingivalis is due to Arg-gingipain. FEMS

Microbiol Lett, 221, 181-5.

HU, W. & FEATHERSTONE, J. D. 2005. Prevention of enamel demineralization: an

in-vitro study using light-cured filled sealant. Am J Orthod Dentofacial Orthop,

128, 592-600.

281

HUAN, Z., CHANG, J. & HUANG, X. H. 2008. Self-setting properties and in vitro

bioactivity of Ca2SiO4/CaSO4.1/2H2O composite bone cement. J Biomed

Mater Res B Appl Biomater, 87, 387-94.

HUME, W. R. & GERZINA, T. M. 1996. Review Bioavailability of components of

resin-based materials which are applied to teeth. Crit Rev Oral Biol Med, 7, 172-

9.

IIJIMA, M., HASHIMOTO, M., NAKAGAKI, S., MUGURUMA, T., KOHDA, N.,

ENDO, K. & MIZOGUCHI, I. 2012. Bracket bond strength and cariostatic

potential of an experimental resin adhesive system containing Portland cement.

Angle Orthod, 82, 900-6.

IKEMURA, K., ICHIZAWA, K., FUCHIGAMI, K., ITO, S. & ENDO, T. 2009. Design

of a new dental adhesive--effect of a water-soluble sodium acylphosphine oxide

with crown ether on adhesion to dental hard tissues. Dent Mater J, 28, 267-76.

IKEMURA, K., TAY, F. R., KOURO, Y., ENDO, T., YOSHIYAMA, M., MIYAI, K.

& PASHLEY, D. H. 2003. Optimizing filler content in an adhesive system

containing pre-reacted glass-ionomer fillers. Dent Mater, 19, 137-46.

IMAZATO, S., TAY, F. R., KANESHIRO, A. V., TAKAHASHI, Y. & EBISU, S.

2007. An in vivo evaluation of bonding ability of comprehensive antibacterial

adhesive system incorporating MDPB. Dent Mater, 23, 170-6.

INOUE, S., VARGAS, M. A., ABE, Y., YOSHIDA, Y., LAMBRECHTS, P. V., G. ,

SANO, H. & VAN MEERBEEK, B. 2001. Microtensile bond strength of eleven

contemporary adhesives to dentin. J Adhes Dent, 3, 237-45.

ISHIGURO, K., YAMASHITA, K., NAKAGAKI, H., IWATA, K. & HAYAKAWA,

T. 1994. Identification of tissue inhibitor of metalloproteinases-1 (TIMP-1) in

282

human teeth and its distribution in cementum and dentine. Arch Oral Biol, 39,

345-9.

ISO Standard 4049: 2000 (E). Dentistry-polymer-based filling, restorative and luting

materials. Geneva: International Organization for Standardization; 2000, 18-21.

ISO/TS 11405: 2003. Dental materials - testing of adhesion to tooth structure. Geneva:

International Organization for Standardization/Technical Specification; 2003.

ITO, S., HASHIMOTO, M., WADGAONKAR, B., SVIZERO, N., CARVALHO, R.

M., YIU, C., RUEGGEBERG, F. A., FOULGER, S., SAITO, T., NISHITANI,

Y., YOSHIYAMA, M., TAY, F. R. & PASHLEY, D. H. 2005. Effects of resin

hydrophilicity on water sorption and changes in modulus of elasticity.

Biomaterials, 26, 6449-59.

ITO, S., IIJIMA, M., MOTAI, F., MIZOGUCHI, I. & SAITO, T. 2012. Effects of

calcium salts of acidic monomers on mineral induction of phosphoprotein

immobilized to agarose beads. J Biomed Mater Res A, 100, 2760-5.

JACOBSEN, T. & SODERHOLM, K. J. 1995. Some effects of water on dentin

bonding. Dent Mater, 11, 132-6.

JAYARAMAN, M. & SUBRAMANIAN, M. V. 2002. Preparation and characterization

of two new composites: collagen-brushite and collagen octa-calcium phosphate.

Med Sci Monit, 8, 481-7.

JIANG, H. & LIU, X. Y. 2004. Principles of mimicking and engineering the self-

organized structure of hard tissues. J Biol Chem, 297, 41286-93.

KAN, K. C., MESSER, L. B. & MESSER, H. H. 1997. Variability in cytotoxicity and

fluoride release of resin-modified glass-ionomer cements. J Dent Res, 76, 1502-

7.

283

KANCA, J. R. 1992. Resin bonding to wet substrate. 1. Bonding to dentin.

Quintessence Int, 23, 39-41.

KARUPPUCHAMY, S. & JEONG, J. M. 2005. Super-hydrophilic amorphous titanium

dioxide thin film deposited by cathodic electrodeposition. Mater Chem Phys, 93,

251-4.

KEMAL, E., ADESANYA, K. O. & DEB, S. 2011. Phosphate based 2-hydroxyethyl

methacrylate hydrogels for biomedical applications. J Mater Chem, 21, 2237-45.

KERBY, R. E., KNOBLOCH, L. & THAKUR, A. 1997. Strength properties of visible-

light-cured resin-modified glass-ionomer cements. Oper Dent, 22, 79-83.

KERN, M., BARLOI, A. & YANG, B. 2009. Surface conditioning influences zirconia

ceramic bonding. J Dent Res, 88, 817-22.

KHOR, E. 1997. Methods for the treatment of collagenous tissues for bioprostheses.

Biomaterials, 18, 95-105.

KIM, C. Y., CLARK, A. E. & HENCH, L. L. 1989. Early stages of calcium-phosphate

layer formation in bioglasses. J Non-Cryst Solids, 113, 195-202.

KIM, H. M., HIMENO, T., KOKUBO, T. & NAKAMURA, T. 2005. Process and

kinetics of bonelike apatite formation on sintered hydroxyapatite in a simulated

body fluid. Biomaterials, 26, 4366-73.

KIM, H. M., REY, C. & GLIMCHER, M. J. 1995. Isolation of calcium-phosphate

crystals of bone by nonaqueous methods at low-temperature. J Bone Miner Res,

10, 1589-601.

KIM, J., AROLA, D. D., GU, L., KIM, Y. K., MAI, S., LIU, Y., PASHLEY, D. H. &

TAY, F. R. 2010a. Functional biomimetic analogs help remineralize apatite-

depleted demineralized resin-infiltrated dentin via a bottom-up approach. Acta

Biomater, 6, 2740-50.

284

KIM, J., UCHIYAMA, T., CARRILHO, M., AGEE, K. A., MAZZONI, A., BRESCHI,

L., CARVALHO, R. M., TJÄDERHANE, L., LOONEY, S., WIMMER, C.,

TEZVERGIL-MUTLUAY, A., TAY, F. R. & PASHLEY, D. H. 2010b.

Chlorhexidine binding to mineralized versus demineralized dentin powder. Dent

Mater, 26, 771-8.

KIM, J., VAUGHN, R. M., GU, L., ROCKMAN, R. A., AROLA, D. D., SCHAFER, T.

E., CHOI, K. K., PASHLEY, D. H. & TAY, F. R. 2010c. Imperfect hybrid

layers created by an aggressive one-step, self-etch adhesive in primary dentin

are amendable to biomimetic remineralization in vitro. J Biomed Mater Res A,

93, 1225-34.

KIM, J. R., GU, L., BRESCHI, L., TJÄDERHANE, L., CHOI, K. K., PASHLEY, D. H.

& TAY, F. R. 2010d. Implication of ethanol wet bonding in hybrid layer

remineralization. J Dent Res, 89, 575-80.

KIM, Y. K., GU, L. S., BRYAN, T. E., KIM, J. R., CHEN, L. & LIU, Y. 2010e.

Mineralization of reconstituted collagen using polyvinylphosphonic

acid/polyacrylic acid templating matrix protein analogues in the presence of

calcium, phosphate and hydroxyl ions. Biomaterials, 31, 6618-27.

KIM, Y. K., MAI, S., MAZZONI, A., LIU, Y., TEZVERGIL-MUTLUAY, A.,

TAKAHASHI, K., ZHANG, K., PASHLEY, D. H. & TAY, F. R. 2010f.

Biomimetic remineralization as a progressive dehydration mechanism of

collagen matrices—Implications in the aging of resin-dentin bonds. Acta

Biomater, 6, 3729-39.

KIM, Y. K., YIU, C. K. Y., KIM, J. R., GU, L., KIM, S. K. & WELLER, R. N. 2010g.

Failure of a glass ionomer to remineralize apatite-depleted dentin. J Dent Res,

89, 230-5.

285

KING, N. M., TAY, F. R., PASHLEY, D. H., HASHIMOTO, M., ITO, S.,

BRACKETT, W. W., GARCIA-GODOY, F. & SUNICO, M. 2005. Conversion

of one-step to two-step self-etch adhesives for improved efficacy and extended

application. Am J Dent, 18, 126-34.

KINNEY, J. H., HABELITZ, S., MARSHALL, S. J. & MARSHALL, G. W. 2003. The

importance of intrafibrillar mineralization of collagen on the mechanical

properties of dentin. J Dent Res, 82, 957-61.

KINNEY, J. H., POPLE, J. A., MARSHALL, G. W. & MARSHALL, S. J. 2001.

Collagen orientation and crystallite size in human dentin: a small angle X-ray

scattering study. Calcif Tissue Int, 69, 31-7.

KLONT, B. & TEN CATE, J. M. 1991. Susceptibility of the collagenous matrix from

bovine incisor roots to proteolysis after in vitro lesion formation. Caries Res, 25,

46-50.

KNOOP, F., PETERS, C. G. & EMERSOON, W. B. 1939. A sensitive pyramidal-

diamond tool for indentation measurements. J Res Nat Bur Stand, 23, 39.

KOIBUCHI, H., YASUDA, N. & NAKABAYASHI, N. 2001. Bonding to dentin with a

self-etching primer: the effect of smear layers. Dent Mater, 17, 122-6.

KOKUBO, N., KIM, H. M. & KAWASHIMA, M. 2003. Novel bioactive materials with

different mechanical properties. Biomaterials, 24, 2161-75.

KOKUBO, T. 1990. Surface chemistry of bioactive glass-ceramics. J Non-Cryst Solids,

120, 138-51.

KOKUBO, T. 1991. Bioactive glass ceramics: properties and applications.

Biomaterials, 12, 155-63.

286

KOKUBO, T. 1992. Bioactivity of glasses and glass ceramics. Ducheyne P, Kokubo T,

van Blitterswijk C, editors. Bone bonding biomaterials. Leiderdrop: Reed

Healthcare Communications, 31-46.

KOKUBO, T. & TAKADAMA, H. 2006. How useful is SBF in predicting in vivo bone

bioactivity? Biomaterials, 27, 2907-15.

KOSHIRO, K., INOUE, S., TANAKA, T., KOASE, K., FUJITA, M., HASHIMOTO,

M. & SANO, H. 2004. In vivo degradation of resin-dentin bonds produced by a

self-etch vs. a total-etch adhesive system. Eur J Oral Sci, 112, 368-75.

KOSTORYZ, E. L., DHARMALA, K., YE, Q., WANG, Y., HUBER, J. & PARK, J. G.

2009. Enzymatic biodegradation of HEMA/bisGMA adhesives formulated with

different water content. J Biomed Mater Res B Appl Biomater, 88, 394-401.

KOUTSOUKOS, P. G. & NANCOLLAS, G. H. 1981. Crystal growth of calcium

phosphates—epitaxial considerations. J Cryst Growth, 53, 10-9.

KRAMER, I. R. H. & MC LEAN, J. W. 1952. Alterations in the staining reactions of

dentine resulting from a constituent of a new self-polymerizing resin. Br Dent J,

92, 150-3.

KREMER, E. A., CHEN, Y., SUZUKI, K., NAGASE, H. & GORSKI, J. P. 1998.

Hydroxyapatite induces autolytic degradation and inactivation of matrix

metalloproteinase-1 and -3. J Bone Miner Res, 13, 1890-902.

KUBOKI, Y. & MECHANIC, G. L. 1982. Comparative molecular distribution of cross-

link in bone and dentin collagen. Structure-function relationships. Calcif Tissue

Int, 34, 306-8.

KUSBOKI, Y., OHGUSHI, K. & FUSAYAMA, T. 1977. Collagen biochemistry of the

two layers of carious dentin. J Dent Res, 56, 1233-7.

287

LABELLA, R., BRADEN, M. & DEB, S. 1994. Novel hydroxyapatite-based dental

composites. Biomaterials, 15, 1197-200.

LAGERWEIJ, M. D. & TEN CATE, J. M. 2002. Remineralisation of enamel lesions

with daily applications of a high-concentration fluoride gel and a fluoridated

toothpaste: an in situ study. Caries Res, 36, 270-4.

LANDIS, W. J. 1996. Mineral characterization in calcifying tissues: atomic, molecular

and macromolecular perspectives. Connect Tissue Res, 34, 239-46.

LANDIS, W. J., SONG, M. J., LEITH, A., MCEWEN, L. & MC-EWEN, B. F. 1993.

Mineral and organic matrix interaction in normally calcifying tendon visualized

in three dimensions by high-voltage electron microscopic tomography and

graphic image reconstruction. J Struct Biol, 110, 39-54.

LAUER-FIELDS, J. L., CHALMERS, M. J., BUSBY, S. A., MINOND, D., GRIFFIN,

P. R. & FIELDS, G. B. 2009. Identification of specific hemopexin-like domain

residues that facilitate matrix metalloproteinase collagenolytic activity. J Biol

Chem, 284, 24017-24.

LAWRENCE, C. D. 1998. Physicochemical and mechanical properties of Portland

cements. LEA’s Chemistry of Cement and Concrete, P. C.Hewlett,

Ed.,Woburn,Mass, USA, Butterworth & Heinemann, 4th edition, 343-420.

LEE, S. Y., GREENER, E. H., MUELLER, H. J. & CHIU, C. H. 1994. Effect of food

and oral simulating fluids on dentine bond and composite strength. J Dent, 22,

352-9.

LEE, T. C., MOHSIN, S., TAYLOR, D., PARKESH, R., GUNNLAUGSSON, T.,

O'BRIEN, F. J., GIEHL, M. & GOWIN, W. 2003. Detecting microdamage in

bone. J Anat, 203, 161-72.

288

LEES, S. & PAGE, E. A. 1992. A study of some properties of mineralized turkey leg

tendon. Connect Tissue Res, 28, 263-87.

LEEVAILOJ, C., PLATT, J. A., COCHRAN, M. A. & MOORE, B. K. 1998. In vitro

study of fracture incidence and compressive fracture load of all-ceramic crowns

cemented with resin-modified glass ionomer and other luting agents. J Prosthet

Dent, 80, 699-707.

LEGEROS, R. Z. 1991. Calcium phosphates in oral biology and medicine. Myers HM,

ed. Monographs in Oral Science. Basel: Karger.

LEGEROS, R. Z., TRAUTZ, O. R., KLEIN, E. & LEGEROS, J. P. 1969. 2 Types of

carbonate substitution in apatite structure. Experientia, 25, 5-7.

LI, H., BURROW, M. F. & TYAS, M. J. 2002. The effect of load cycling on the

nanoleakage of dentin bonding systems. Dent Mater, 18, 111-9.

LI, J., DOU, Y., YANG, J., YIN, Y., ZHANG, H. & YAO, F. 2009. Surface

characterization and biocompatibility of micro- and nano-

hydroxyapatite/chitosan–gelatin network films. Mater Sci Eng C, 29, 1207-15.

LI, X. & CHANG, J. 2008. Preparation of bone-like apatite-collagen nanocomposites

by a biomimetic process with phosphorylated collagen. J Biomed Mater Res A,

85, 293-300.

LI, X. & WU, J. F. 2010. Recent developments in patent anti-cancer agents targeting the

matrix metalloproteinases (MMPs). Recent Pat Anticancer Drug Discov, 5, 109-

41.

LIAO, J., YANG, L., GRASHOW, J. & SACKS, M. S. 2005. Molecular orientation of

collagen in intact planar connective tissues under biaxial stretch. Acta Biomater,

1, 45-54.

289

LIN, B. A., JAFFER, F., DUFF, M. D., TANG, Y. W. & SANTERRE, J. P. 2005.

Identifying enzyme activities within human saliva which are relevant to dental

resin composite biodegradation. Biomaterials, 26, 4259-64.

LIU, G., ZHAO, D., TOMSIA, A. P., MINOR, A. M., SONG, X. & SAIZ, E. 2009.

Three-dimensional biomimetic mineralization of dense hydrogel templates. J Am

Chem Soc, 29, 9937-9.

LIU, Y., LI, N., QI, Y., NIU, L. N., ELSHAFIY, S., MAO, J., BRESCHI, L.,

PASHLEY, D. H. & TAY, F. R. 2011a. The use of sodium trimetaphosphate as

a biomimetic analog of matrix phosphoproteins for remineralization of artificial

caries-like dentin. Dent Mater, 27, 465-77.

LIU, Y., MAI, S., LI, N., YIU, C. K. Y., MAO, J., PASHLEY, D. H. & TAY, F. R.

2011b. Differences between top-down and bottom-up approaches in

mineralizing thick, partially demineralized collagen scaffolds. Acta Biomater, 7,

1742-51.

LIU, Y., TJÄDERHANE, L., BRESCHI, L., MAZZONI, A., LI, N., MAO, J.,

PASHLEY, D. H. & TAY, F. R. 2011c. Limitations in bonding to dentin and

experimental strategies to prevent bond degradation. J Dent Res, 90, 953-68.

LODOVICI, E., REIS, A., GERALDELI, S., FERRACANE, J. L., BALLESTER, R. Y.

& RODRIGUES FILHO, L. E. 2009. Does adhesive thickness affect resin-

dentin bond strength after thermal/load cycling? Oper Dent, 34, 58-64.

LOWENSTAM, H. A. & WEINER, S. 1980. On biomineralization New York, Oxford

University Press.

LU, X. & LENG, Y. 2005. Theoretical analysis of calcium phosphate precipitation in

simulated body fluid. Biomaterials, 26, 1097-108.

290

LUSVARDI, G., MALAVASI, G., MENABUE, L., AINA, V. & MORTERRA, C.

2009. Fluoride-containing bioactive glasses: surface reactivity in simulated body

fluids solutions. Acta Biomater, 5, 3548-62.

LUTHRA, A. K. & SANDHU, S. S. 2005. Methods and clinical devices for the

inhibition or prevention of mammalian cell growth. United States Patent

Application: 6,929,818.

LYNCH, E., BRAUER, D. S., KARPUKHINA, N., GILLAM, D. G. & HILL, R. G.

2012. Multi-component bioactive glasses of varying fluoride content for treating

dentin hypersensitivity. Dent Mater, 28, 168-78.

LYSAGHT, V. E. & DEBELLIS, A. 1969. Microhardness testing. Wilkes B, editor.

Hardness testing handbook. New York: American Chain and Cable Co., 76-105.

MA, D. H., LAI, J. Y., CHENG, H. Y., TSAI, C. C. & YEH, L. K. 2010. Carbodiimide

cross-linked amniotic membranes for cultivation of limbal epithelial cells.

Biomaterials, 31, 6647-58.

MACEDO, G. V., YAMAUCHI, M. & BEDRAN-RUSSO, A. K. 2009. Effects of

chemical cross-linkers on caries-affected dentin bonding. J Dent Res, 88, 1096-

100.

MACIEL, K. T., CARVALHO, R. M., RINGLE, R. D., PRESTON, C. D., RUSSELL,

C. M. & PASHLEY, D. H. 1996. The effects of acetone, ethanol, HEMA, and

air on the stiffness of human decalcified dentin matrix. J Dent Res, 75, 1851-8.

MAHAMID, J., SHARIR, A., ADDADI, L. & WEINER, S. 2008. Amorphous calcium

phosphate is a major component of the forming fin bones of zebrafish:

Indications for an amorphous precursor phase. Proc Natl Acad Sci USA, 105,

12748-53.

291

MAHONEY, E., HOLT, A., SWAIN, M. & KILPATRICK, N. 2000. The hardness and

modulus of elasticity of primary molar teeth: An ultra-microindentation study. J

Dent Res, 28, 589-94.

MAI, S., GU, L. & LING, J. Q. 2009. Current methods for preventing degradation of

resin-dentin bonds. Hong Kong Dent J, 6, 83-92.

MAI, S., KIM, Y. K., KIM, J., YIU, C. K., LING, J., PASHLEY, D. H. & TAY, F. R.

2010. In vitro remineralization of severely compromised bonded dentin. J Dent

Res, 89, 405-10.

MALACARNE, J., CARVALHO, R. M., DE GOES, M. F., SVIZERO, N., PASHLEY,

D. H., TAY, F. R., YIU, C. K. & CARRILHO, M. R. 2006. Water

sorption/solubility of dental adhesive resins. Dent Mater, 22, 973-80.

MALACHOVÁ, K., PRAUS, P., PAVLÍČKOVÁ, Z. & TURICOVÁ, M. 2009.

Activity of antibacterial compounds immobilised on montmorillonite. Applied

Clay Science, 43, 364-8.

MALLA, N., S., S., WINBERG, J. O., HADLER-OLSEN, E. & UHLIN-HANSEN, L.

2008. Biological and pathobiological functions of gelatinase dimers and

complexes. Connect Tissue Res, 49, 180-4.

MANDEL, S. & TAS, A. C. 2010. Brushite (CaHPO4×2H2O) to Octacalcium

Phosphate (Ca8(HPO4)2(PO4)4×5H2O) Transformation in DMEM (Dulbecco’s

Modified Eagle Medium) Solutions at 36.5°C. Mater Sci Eng C, 30, 245-54.

MANHART, J. & TRUMM, C. 2010. Marginal adaptation of an etch-and-rinse

adhesive with a new type of solvent in class II cavities after artificial aging. Clin

Oral Investig, 14, 699-705.

MARCHESI, G., NAVARRA, C. O., CADENARO, M., CARRILHO, M. R., CODAN,

B., SERGO, V., DI LENARDA, R. & BRESCHI, L. 2010. The effect of ageing

292

on the elastic modulus and degree of conversion of two multistep adhesive

systems. Eur J Oral Sci, 3, 304-10.

MARGONAR, R., QUEIROZ, T. P., LUVIZUTO, E. R., MARCANTONIO, É., LIA,

R. C., HOLZHAUSEN, M. & MARCANTONIO-JÚNIOR, É. 2012. Bioactive

glass for alveolar ridge augmentation. J Craniofac Surg, 23, e220-2.

MARSHALL, D. B., NOMA, T. & EVANS, A. G. 1982. A simple method for

determining elastic-modulus to hardness ratios using Knoop indentation

measurements. Commun ACS, 65, 175-6.

MARSHALL, G. W. 1993. Dentin: Microstructure and characterization. Quintessence

Int, 24, 606-17.

MARSHALL, G. W., MARSHALL, S. J., KINNEY, J. H. & BALOOCH, M. 1997. The

dentin substrate: structure and properties related to bonding. J Dent Res, 25,

441-58.

MARSHALL, S. J., BAYNE, S. C., BAIER, R., TOMSIA, A. P. & MARSHALL, G.

W. 2010. A review of adhesion science. Dent Mater, 26, e11-6.

MASIH, S., THOMAS, A. M., KOSHY, G. & JOSHI, J. L. 2011. Comparative

evaluation of the microleakage of two modified glass ionomer cements on

primary molars. An in vivo study. J Indian Soc Pedod Prev Dent, 29, 135-9.

MATCHETT, M. D., MACKINNON, S. L., SWEENEY, M. I., GOTTSCHALL-PASS,

K. T. & HURTA, R. A. 2005. Blueberry flavonoids inhibit matrix

metalloproteinase activity in DU145 human prostate cancer cells. Biochem Cell

Biol, 83, 637-43.

MAZZONI, A., MANNELLO, F., TAY, F. R., TONTI, G. A., PAPA, S., MAZZOTTI,

G., DI LENARDA, R., PASHLEY, D. H. & BRESCHI, L. 2007. Zymographic

293

analysis and characterization of MMP-2 and -9 forms in human sound dentin. J

Dent Res, 86, 436-40.

MAZZONI, A., PASHLEY, D. H. & NISHITANI, Y. 2006. Reactivation of inactivated

endogenous proteolytic activities in phosphoric acid-etched dentine by etch-and-

rinse adhesives. Biomaterials, 27, 4470-6.

MCLEAN, J. W. 1990. Cermet cements. J Am Dent Assoc, 120, 43-7.

MEREDITH, N., SHERIFF, M., SETCHELL, D. J. & SWANSON, S. A. V. 1996.

Measurement of the microhardness and Young’s modulus of human enamel and

dentine using an indentation technique. Arch Oral Biol, 41, 539-45.

MIN, B. G., STACHURSKI, Z. H. & HODGKIN, J. H. 1993. Cure kinetics of

elementary reactions of a DGEBA/DDS epoxy resin: 1. Glass transition

temperature versus conversion. Polymer, 34, 4908-12.

MITCHELL, C. A., DOUGLAS, W. H. & CHENG, Y. S. 1999. Fracture toughness of

conventional, resin-modified glass-ionomer and composite luting cements. Dent

Mater, 15, 7-13.

MJÖR, I. A. & GORDAN, V. V. 2002. Failure, repair, refurbishing and longevity of

restorations. Oper Dent, 27, 528-34.

MOHAMMADI, Z. & ABBOTT, P. V. 2009. Antimicrobial substantivity of root canal

irrigants and medicaments: a review. Aust Endod J, 35, 131-9.

MONTICELLI, F., OSORIO, R., MAZZITELLI, C., FERRARI, M. & TOLEDANO,

M. 2008. Limited decalcification/diffusion of self-adhesive cements into dentin.

J Dent Res, 87, 974-9.

MORAES, J. C. S. & GRANDINI, C. R. 2011. The glass transition temperature in

dental composites. Metal, ceramic and polymeric composites for various uses, 7,

669-84.

294

MOSHAVERINIA, A., ANSARI, S., MOSHAVERINIA, M., SCHRICKER, S. R. &

CHEE, W. W. 2011. Ultrasonically set novel NVC-containing glass-ionomer

cements for applications in restorative dentistry. J Mater Sci Mater Med, 22,

2029-34.

MOSZNER, N., FISCHER, U. K., ANGERMANN, J. & RHEINBERGER, V. 2006.

Bis-(acrylamide)s as new cross-linkers for resin-based composite restoratives.

Dent Mater, 22, 1157-62.

MOSZNER, N., SALZ, U. & ZIMMERMANN, J. 2005. Chemical aspects of self-

etching enamel-dentin adhesives: a systematic review. Dent Mater, 21, 895-910.

MUNKSGAARD, E. C., PEUTZFELDT, A. & ASMUSSEN, E. 2000. Elution of

TEGDMA and BisGMA from a resin and a resin composite cured with halogen

or plasma light. Eur J Oral Sci, 108, 341-5.

NAKABAYASHI, N., KOJIMA, K. & MASUHARA, E. 1982. Promotion of adhesion

by the infiltration of monomers into tooth substrates. J Biomed Mater Res, 16,

265-73.

NAKABAYASHI, N. & PASHLEY, D. H. 1998. Hybridization of Dental Hard Tissues.

Quintessence Publishing, Tokyo.

NAKABAYASHI, N. & TAKARADA, K. 1992. Effect of HEMA on bonding to

dentin. Dent Mater, 8, 125-30.

NAKABAYASHI, N., WATANABE, A. & IGARASHI, K. 2004. AFM observation of

collapse and expansion of phosphoric acid-demineralized dentin. J Biomed

Mater Res A, 68, 558-65.

NAKAOKI, Y., NIKAIDO, T., PEREIRA, P. N., INOKOSHI, S. & TAGAMI, J. 2000.

Dimensional changes of demineralized dentin treated with HEMA primers. Dent

Mater, 16, 441-6.

295

NAKAZAWA, Y., SEINO, E., USHIKI, T., OGATA, T., HIRAI, Y., KAWADA, E. &

ODA, Y. 1999. Microhardness evaluations of resin-dentin bonding areas by

nano-indentation. Bull Tokyo Dent Coll, 40, 47-54.

NALLA, R. K., PORTER, A. E., DARAIO, C., MINOR, A. M., RADMILOVIC, V.,

STACH, E. A., TOMSIA, A. P. & RITCHIE, R. O. 2005. Ultrastructural

examination of dentin using focused ion-beam cross-sectioning and transmission

electron microscopy. Micron, 36, 672-80.

NEILL, J. 1838. The Hydrostatic Pressure Doctrine. Transactions of the College of

Physicians of Philadelphia, 111.

NGO, H. 2010. Glass-ionomer cements as restorative and preventive materials. Dent

Clin North Am, 54, 551-63.

NGO, H. C., MOUNT, G., MC INTYRE, J., TUISUVA, J. & VON DOUSSA, R. J.

2006. Chemical exchange between glass-ionomer restorations and residual

carious dentine in permanent molars: An in vivo study. J Dent, 34, 608-13.

NIEDERBERGER, M. & CÖLFEN, H. 2006. Oriented attachment and mesocrystals:

non-classical crystallization mechanisms based on nanoparticle assembly. Phys

Chem Chem Phys, 8, 3271-87.

NIELSEN-MARSH, C. M., HEDGES, R. E. M., MANN, T. & COLLINS, M. J. 2000.

A preliminary investigation of the application of differential scanning

calorimetry to the study of collagen degradation in archaeological bone.

Thermochim Acta, 365, 129-39.

NIKAIDO, T., KUNZELMANN, K. H., CHEN, H., OGATA, M., HARADA, N.,

YAMAGUCHI, S., COX, C. F., HICKEL, R. & TAGAMI, J. 2002. Evaluation

of thermal cycling and mechanical loading on bond strength of a self-etching

primer system to dentin. Dent Mater, 18, 269-75.

296

NIKOLAENKO, S. A., LOHBAUER, U., ROGGENDORF, M., PETSCHELT, A.,

DASCH, W. & FRANKENBERGER, R. 2004. Influence of c-factor and

layering technique on microtensile bond strength to dentin. Dent Mater, 20, 579-

85.

NIMNI, M. E., CHEUNG, D., STARTES, B., KODAMA, M. & SHEIKH, K. 1998.

Bioprosthesis derived from crosslinked and chemically modified collagenous

tissues. Nimni ME, editor. Collagen, Vol. III. Boca Raton, FL: CRC Press, 1-38.

NISHITANI, Y., YOSHIYAMA, M., HOSAKA, K., TAGAMI, J., DONNELLY, A.,

CARRILHO, M., TAY, F. R. & PASHLEY, D. H. 2007. Use of Hoy's solubility

parameters to rank water sorption/solubility of experimental self-etching primers

and adhesives. Eur J Oral Sci, 115, 81-6.

NISHITANI, Y., YOSHIYAMA, M. & WADGAONKAR, B. 2006. Activation of

gelatinolytic/collagenolytic activity in dentin by self-etching adhesives. Eur J

Oral Sci, 114, 160-6.

NOVINEC, M., KOVACIC, L., LENARCIC, B. & BAICI, A. 2010. Conformational

flexibility and allosteric regulation of cathepsin K. Biochem J, 429, 379-89.

NUDELMAN, F., PIETERSE, K., GEORGE, A., BOMANS, P. H., FRIEDRICH, H.,

BRYLKA, L. J., HILBERS, P. A., DE WITH, G. & SOMMERDIJK, N. A.

2010. The role of collagen in bone apatite formation in the presence of

hydroxyapatite nucleation inhibitors. Nat Mater, 9, 1004-9.

OBERMAJER, N., JEVNIKAR, Z., DOLJAK, B. & KOS, J. 2008. Role of cysteine

cathepsins in matrix degradation and cell signalling. Connect Tissue Res, 49,

193-6.

297

OGAWA, K., YAMASHITA, Y., ICHIJO, T. & FUSAYAMA, T. 1983. The

ultrastructure and hardness of the transparent layer of human carious dentin. J

Dent Res, 62, 7-10.

OLIVEIRA, S. S., PUGACH, M. K., STROTHER, J., DARLING, C. L., FRIED, D.,

GANSKY, S. A., MARSHALL, S. J. & MARSHALL JR, G. W. 2003. The

influence of the dentin smear layer on adhesion: a self-etching primer vs. a total-

etch system. Dent Mater, 19, 758-67.

ONG, J. L. & CHAN, D. C. N. 2000. Hydroxyapatite and their use as coatings in dental

implants: a review. Crit Rev Biomed Eng, 2000, 667a-707a.

ONUMA, K. & ITO, A. 1998. Cluster growth model for hydroxyapatite. Chem Mater,

10, 3346-51.

OONISHI, H., KUSHITANI, S., YASUKAWA, E., IWAKI, H., HENCH, L. L.,

WILSON, J., TSUJI, E. & SUGIHARA, T. 1997. Particulate bioglass compared

with hydroxyapatite as a bone graft substitute. Clin Orthop Relat Res, 334, 316-

25.

OSORIO, E., TOLEDANO, M., AGUILERA, F. S., TAY, F. R. & OSORIO, R. 2010.

Ethanol wet-bonding technique sensitivity assessed by AFM. J Dent Res, 89,

1264-9.

OSORIO, R., TOLEDANO, M., OSORIO, E., AGUILERA, F. S. & TAY, F. R. 2005.

Effect of load cycling and in vitro degradation on resin-dentin bonds using a

self-etching primer. J Biomed Mater Res A, 72, 399-408.

OSORIO, R., YAMAUTI, M., OSORIO, E., ROMÁN, J. S. & TOLEDANO, M. 2011.

Zinc-doped dentin adhesive for collagen protection at the hybrid layer. Eur J

Oral Sci, 119, 401-10.

298

OSTWALD, W. F. 1897. Studies on formation and transformation of solid materials. Z

Phys Chem, 22, 289-302.

OWENS, S. E. J. & MILLER, B. H. 2000. A comparison of shear bond strengths of

three visible light-cured orthodontic adhesives. Angle Orthod, 70, 352-6.

PALMER, L. C., NEWCOMB, C. J., KALTZ, S. R., SPOERKE, E. D. & STUPP, S. I.

2008. Biomimetic systems for hydroxyapatite mineralization inspired by bone

and enamel. Chem Rev, 108, 4754-83.

PANIGHI, M. & G’SELL, C. 1993. Effect of the tooth microstructure on the shear bond

strength of a dental composite. J Biomed Mater Res, 27, 975-81.

PARIROKH, M. & TORABINEJAD, M. 2010. Mineral trioxide aggregate: a

comprehensive literature review—part III clinical applications, drawbacks and

mechanism of action. J Endod, 36, 400-13.

PARK, J., YE, Q., TOPP, E. M., MISRA, A., KIEWEG, S. L. & SPENCER, P. 2010.

Effect of photoinitiator system and water content on dynamic mechanical

properties of a light-cured bisGMA/HEMA dental resin. J Biomed Mater Res A,

93, 1245-51.

PARK, J. G., YE, Q., TOPP, E. M., KOSTORYZ, E. L., WANG, Y., KIEWEG, S. L. &

SPENCER, P. 2008. Preparation and properties of novel dentin adhesives with

esterase resistance. J Appl Polym Sci, 107, 3588-97.

PASHLEY, D., OKABE, A. & PARHAM, P. 1985. The relationship between dentin

microhardness and tubule density. Endod Dent Traumatol, 1, 176-9.

PASHLEY, D. H. 1989. Dentin, a dynamic substrate - a review. Scanning Microsc, 3,

161-76.

PASHLEY, D. H. 1991. Dentin bonding: overview of the substrate with respect to

adhesive material. J Esthet Dent, 3, 46-50.

299

PASHLEY, D. H., CIUCCHI, B., SANO, H. & HOMER, J. A. 1993. Permeability of

dentin to adhesive agents. Quintessence Int, 24, 618-31.

PASHLEY, D. H., SANO, H., CIUCCHI, B., CARVALHO, R. M. & RUSSELL, C. M.

1995a. Bond strength versus dentin structures: a modeling approach. Arch Oral

Biol, 40, 1109-18.

PASHLEY, D. H., SANO, H., CIUCCHI, B., YOSHIYAMA, M. & CARVALHO, R.

M. 1995b. Adhesion testing of dentin bonding agents: a review. Dent Mater, 11,

117-25.

PASHLEY, D. H., TAO, L., BOYD, L., KING, G. E. & HORNER, J. A. 1988.

Scanning electron microscopy of the substructure of smear layers in human

dentine. Arch Oral Biol, 33, 265-70.

PASHLEY, D. H., TAY, F. R., BRESCHI, L., TJÄDERHANE, L., CARVALHO, R.

M., CARRILHO, M. & TEZVERGIL-MUTLUAY, A. 2011. State of the art

etch-and-rinse adhesives. Dent Mater, 27, 1-16.

PASHLEY, D. H., TAY, F. R., CARVALHO, R. M., RUEGGEBERG, F. A., AGEE,

K. A., CARRILHO, M., DONNELLY, A. & GARCÍA-GODOY, F. 2007. A

review of the interactions between dentin matrix and solvated resins using a

macromodel of the hybrid layer. Am J Dent, 20, 7-20.

PASHLEY, D. H., TAY, F. R., YIU, C., HASHIMOTO, M., BRESCHI, L.,

CARVALHO, R. M. & ITO, S. 2004. Collagen degradation by host-derived

enzymes during aging. J Dent Res, 83, 216-21.

PASHLEY, E. L., ZHANG, Y., LOCKWOOD, P. E., RUEGGEBERG, F. A. &

PASHLEY, D. H. 1998. Effects of HEMA on water evaporation from water-

HEMA mixtures. Dent Mater, 14, 6-10.

300

PASSALINI, P., FIDALGO, T. K., CALDEIRA, E. M., GLEISER, R., NOJIMA, M. D.

C. & MAIA, L. C. 2010. Mechanical properties of one and two-step fluoridated

orthodontic resins submitted to different pH cycling regimes. Braz Oral Res, 24,

197-203.

PELKA, M. 1999. Elution parameters and HPLC-detection of single components from

resin composite. Clin Oral Investig, 3, 194-200.

PELLENQ, R. J. M., KUSHIMA, A. & SHAHSAVARI, R. 2009. A realistic molecular

model of cement hydrates. Proc Natl Acad Sci USA, 106, 16102-7.

PERINKA, L., SANO, H. & HOSODA, H. 1992. Dentin thickness, hardness, and Ca-

concentration vs bond strength of dentin adhesives. Dent Mater, 8, 229-33.

PERIS, A. R., MITSUI, F. H., LOBO, M. M., BEDRAN-RUSSO, A. K. & MARCHI,

G. M. 2007. Adhesive systems and secondary caries formation: Assessment of

dentin bond strength, caries lesions depth and fluoride release. Dent Mater, 23,

308-16.

PETERS, M. C. 2010. Strategies for noninvasive demineralized tissue repair. Dent Clin

North Am, 54, 507-25.

PETERS, M. C., BRESCIANI, E., BARATA, T. J. E., FAGUNDES, T. C.,

NAVARRO, R. L., NAVARRO, M. F. L. & DICKENS, S. H. 2010. In vivo

dentin remineralization by calcium-phosphate cement. J Dent Res, 89, 286-91.

PETERS, M. C. & MCLEAN, M. E. 2001. Minimally invasive operative care. I.

Minimal intervention and concepts for minimally invasive cavity preparations. J

Adhes Dent, 3, 7-16.

PEUMANS, M., DE MUNCK, J., VAN LANDUYT, K., LAMBRECHTS, P. & VAN

MEERBEEK, B. 2007. Five-year clinical effectiveness of a two-step self-

etching adhesive. J Adhes Dent, 9, 7-10.

301

PEUMANS, M., KANUMILLI, P., DE MUNCK, J., VAN LANDUYT, K.,

LAMBRECHTS, P. & VAN MEERBEEK, B. 2005. Clinical effectiveness of

contemporary adhesives: a systematic review of current clinical trials. Dent

Mater, 21, 864-81.

PIOCH, T., STAEHLE, H. J., DUSCHNER, H. & GARCIA-GODOY, F. 2001.

Nanoleakage at the composite-dentin interface: a review. Am J Dent, 14, 252-8.

PISKOUNOVA, S., FORSGREN, J., BROHEDE, U., ENGQVIST, H. & STRØMME,

M. 2009. In vitro characterization of bioactive titanium dioxide/hydroxyapatite

surfaces functionalized with BMP-2. J Biomed Mater Res B Appl Biomater, 91,

780-7.

PLASSARD, C., LESNIEWSKA, E., POCHARD, I. & NONAT, A. 2005. Nanoscale

experimental investigation of particle interactions at the origin of the cohesion of

cement. Langmuir, 21, 7263-70.

POITEVIN, A., DE MUNCK, J., CARDOSO, M. V., MINE, A., PEUMANS, M.,

LAMBRECHTS, P. & VAN MEERBEEK, B. 2010. Dynamic versus static

bond-strength testing of adhesive interfaces. Dent Mater, 26, 1068-76.

POSNER, A. S. 1985. The mineral of bone. Clin Orthop Relat Res, 200, 87-99.

PRATI, C., FAVA, F., DI GIOIA, D., SELIGHINI, M. & PASHLEY, D. H. 1993.

Antibacterial effectiveness of dentin bonding systems. Dent Mater, 9, 338-43.

PRICE, R. B., DERAND, T., ANDREOU, P. & MURPHY, D. 2003. The effect of two

configuration factors, time, and thermal cycling on resin to dentin bond

strengths. Biomaterials, 24, 1013-21.

PROFETA, A. C., MANNOCCI, F., FOXTON, R. M., THOMPSON, I., WATSON, T.

F. & SAURO, S. 2012. Bioactive effects of a calcium/sodium phosphosilicate

302

on the resin-dentine interface: a microtensile bond strength, scanning electron

microscopy, and confocal microscopy study. Eur J Oral Sci, 120, 353-62.

QI, Y. P., LI, N., NIU, L. N., PRIMUS, C. M., LING, J. Q., PASHLEY, D. H. & TAY,

F. R. 2012. Remineralization of artificial dentinal caries lesions by

biomimetically modified mineral trioxide aggregate. Acta Biomater, 8, 836-42.

RADOVIC, I., MONTICELLI, F., GORACCI, C., VULICEVIC, Z. R. & FERRARI,

M. 2008. Self-adhesive resin cements: a literature review. J Adhes Dent, 10,

251-8.

RAHN, B. A. & PERREN, S. M. 1970. Calcein blue as a fluorescent label in bone.

Experientia, 26, 519.

RAHN, B. A. & PERREN, S. M. 1971. Xylenol orange, a fluorochrome useful in

polychrome sequential labelling of calcifying tissues. Stain Technol, 46, 125-9.

RAHN, B. A. & PERREN, S. M. 1972. Alizarinkomplexon, Fluorochromzur

Markierung von Knocken und Dentinanbau. Experientia, 28, 180.

RAO, A. & MALHOTRA, N. 2011. The role of remineralizing agents in dentistry: a

review. Compend Contin Educ Dent, 32, 26-33.

RAY, S., LUKYANOV, P. & OCHIENG, J. 2003. Members of the cystatin superfamily

interact with MMP-9 and protect it from autolytic degradation without affecting

its gelatinolytic activities. Biochim Biophys Acta, 1652, 91-102.

REINKE, S. M., LAWDER, J. A., DIVARDIN, S., RAGGIO, D., REIS, A. &

LOGUERCIO, A. D. 2012. Degradation of the resin-dentin bonds after

simulated and inhibited cariogenic challenge in an in situ model. J Biomed

Mater Res B Appl Biomater, 100, 1466-71.

303

REIS, A., GRANDE, R. H., OLIVEIRA, G. M., LOPES, G. C. & LOGUERCIO, A. D.

2007. A 2-year evaluation of moisture on microtensile bond strength and

nanoleakage. Dent Mater, 23, 862-70.

REIS, A. F., ARRAIS, C. A., NOVAES, P. D., CARVALHO, R. M., DE GOES, M. F.

& GIANNINI, M. 2004. Ultramorphological analysis of resin-dentin interfaces

produced with water-based single-step and two-step adhesives: nanoleakage

expression. J Biomed Mater Res B Appl Biomater, 71, 90-8.

RHEE, S. H., LEE, J. D. & TANAKA, J. 2000. Nucleation of hydroxyapatite crystal

through chemical interaction with collagen. J Am Ceram Soc, 83, 2890-2.

RITTER, A. V., SWIFT, E. J. & YAMAUCHI, M. 2001. Effects of phosphoric acid and

glutaraldehyde-HEMA on dentin collagen. Eur J Oral Sci, 109, 348-53.

ROOT, M. J. 1990. Inhibition of the amorphous calcium phosphate phase

transformation reaction by polyphosphates and metal ions. Calcif Tissue Int, 47,

112-6.

ROSALES-LEAL, J. I., OSORIO, R., HOLGADO-TERRIZA, J. A., CABRERIZO-

VÍLCHEZ, M. A. & TOLEDANO, M. 2001. Dentin wetting by four adhesive

systems. Dent Mater, 17, 526-32.

ROST, F. W. D. 1995. Fluorescence Microscopy II. Cambridge: Cambridge University

Press.

RUEGGEBERG, F. A., HASHINGER, D. T. & FAIRHURST, C. W. 1990. Calibration

of FTIR conversion analysis of contemporary dental resin composites. Dent

Mater, 6, 241-9.

RYOU, H., NIU, L. N., DAI, L., PUCCI, C. R., AROLA, D. D., PASHLEY, D. H. &

TAY, F. R. 2011. Effect of biomimetic remineralization on the dynamic

nanomechanical properties of dentin hybrid layers. J Dent Res, 90, 1122-8.

304

SADAT-SHOJAI, M., ATAI, M., NODEHI, A. & KHANLAR, L. N. 2010.

Hydroxyapatite nanorods as novel fillers for improving the properties of dental

adhesives: Synthesis and application. Dent Mater, 26, 471-82.

SADEK, F. T., BRAGA, R. R., MUENCH, A., LIU, Y., PASHLEY, D. H. & TAY, F.

2010a. Ethanol wet-bonding challenges current anti-degradation strategy. J Dent

Res, 89, 1499-504.

SADEK, F. T., MAZZONI, A., BRESCHI, L., TAY, F. R. & BRAGA, R. R. 2010b.

Six-month evaluation of adhesives interface created by a hydrophobic adhesive

to acid-etched ethanol-wet bonded dentine with simplified dehydration

protocols. J Dent, 38, 276-83.

SADEK, F. T., PASHLEY, D. H., FERRARI, M. & TAY, F. R. 2007. Tubular

occlusion optimizes bonding of hydrophobic resins to dentin. J Dent Res, 86,

524-8.

SADEK, F. T., PASHLEY, D. H., NISHITANI, Y., CARRILHO, M. R., DONNELLY,

A., FERRARI, M. & TAY, F. R. 2008. Application of hydrophobic resin

adhesives to acid-etched dentin with an alternative wet bonding technique. J

Biomed Mater Res A, 84, 19-29.

SALIAN, S., THAKUR, S., KULKARNI, S. & LATORRE, G. 2010. A randomized

controlled clinical study evaluating the efficacy of two desensitizing dentifrices.

J Clin Dent, 21, 82-7.

SALZ, U., ZIMMERMANN, J., ZEUNER, F. & MOSZNER, N. 2005. Hydrolytic

stability of self-etching adhesive systems. J Adhes Dent, 7, 107-16.

SANCHEZ, F. & ZHANG, L. 2008. Molecular dynamics modeling of the interface

between surface functionalized graphitic structures and calcium–silicate–

305

hydrate: interaction energies, structure, and dynamics. J Colloid Interf Sci, 323,

349-58.

SANO, H., CIUCCHI, B., MATTHEWS, W. G. & PASHLEY, D. H. 1994. Tensile

properties of mineralized and demineralized human and bovine dentin. J Dent

Res, 73, 1205-11.

SANO, H., TAKATSU, T., CIUCCHI, B., HORNER, J. A., MATTEWS, W. G. &

PASHLEY, D. H. 1995a. Nanoleakage: leakage within the hybrid layer. Oper

Dent, 20, 18-35.

SANO, H., YOSHIYAMA, M., EBISU, S., BURROW, M. F., TAKATSU, T. &

CIUCCHI, B. 1995b. Comparative SEM and TEM observations of nanoleakage

within the hybrid layer. Oper Dent, 20, 160-7.

SANTERRE, J. P., SHAJII, L. & LEUNG, B. W. 2001. Relation of dental composite

formulations to their degradation and the release of hydrolyzed polymeric-resin-

derived products. Crit Rev Oral Biol Med, 12, 136-51.

SANTOS, C., LUKLINSKA, Z. B., CLARKE, R. L. & DAVY, K. W. M. 2001.

Hydroxyapatite as a filler for dental composite materials: mechanical properties

and in vitro bioactivity of composites. J Mater Sci Mater Med, 12, 565-73.

SARKAR, N. K., CAICEDO, R., RITWIK, P., MOISEYEVA, R. & I., K. 2005.

Physicochemical basis of the biologic properties of mineral trioxide aggregate. J

Endod, 31, 97-100.

SARR, M., MINE, A., DE MUNCK, J., CARDOSO, M. V., KANE, A. W., VREVEN,

J., VAN MEERBEEK, B. & VAN LANDUYT, K. L. 2010. Immediate bonding

effectiveness of contemporary composite cements to dentin. Clin Oral Investig,

14, 569-77.

306

SAURO, S., MANNOCCI, F., TAY, T., PASHLEY, D. H., COOK, R., CARPENTER,

G. H. & WATSON, T. F. 2009a. Deproteinization effects of NaOCl on acid-

etched dentin in clinically-relevant vs prolonged periods of application. A

confocal and environmental scanning electron microscopy study. Oper Dent, 34,

166-73.

SAURO, S., OSORIO, R., WATSON, T. F. & TOLEDANO, M. 2012a. Therapeutic

effects of novel resin bonding systems containing bioactive glasses on mineral-

depleted areas within the bonded-dentine interface. J Mater Sci Mater Med, 23,

1521-32.

SAURO, S., PASHLEY, D. H., MONTANARI, M., CHERSONI, S., CARVALHO, R.

M., TOLEDANO, M., OSORIO, R., TAY, F. R. & PRATI, C. 2007. Effect of

simulated pulpal pressure on dentin permeability and adhesion of self-etch

adhesives. Dent Mater, 23, 705-13.

SAURO, S., THOMPSON, I. & WATSON, T. F. 2011a. Effects of common dental

materials used in preventive or operative dentistry on dentin permeability and

remineralization. Oper Dent, 36, 222-30.

SAURO, S., TOLEDANO, M., AGUILERA, F. S., MANNOCCI, F., PASHLEY, D.

H., TAY, F. R., WATSON, T. F. & OSORIO, R. 2010. Resin-dentin bonds to

EDTA-treated vs. acid-etched dentin using ethanol wet-bonding. Dent Mater,

26, 368-79.

SAURO, S., TOLEDANO, M., AGUILERA, F. S., MANNOCCI, F., PASHLEY, D.

H., TAY, F. R., WATSON, T. F. & OSORIO, R. 2011b. Resin-dentin bonds to

EDTA-treated vs. acid-etched dentin using ethanol wet-bonding. Part II: Effects

of mechanical cycling load on microtensile bond strengths. Dent Mater, 27, 563-

72.

307

SAURO, S., WATSON, T. F., MANNOCCI, F., MIYAKE, K., HUFFMAN, B. P.,

TAY, F. R. & PASHLEY, D. H. 2009b. Two-photon laser confocal microscopy

of micropermeability of resin-dentin bonds made with water or ethanol wet

bonding. J Biomed Mater Res B Appl Biomater, 90, 327-37.

SAURO, S., WATSON, T. F., MANNOCCI, F., TAY, F. R. & PASHLEY, D. H.

2009c. Prevention of water contamination of ethanol-saturated dentin and

hydrophobic hybrid layers. J Adhes Dent, 11, 271-8.

SAURO, S., WATSON, T. F., TAY, F. R., CHERSONI, S., BRESCHI, L. &

BERNARDI, F. 2006. Water uptake of bonding systems applied on root dentin

surfaces: a SEM and confocal microscopic study. Dent Mater, 22, 671-80.

SAURO, S., WATSON, T. F., THOMPSON, I., TOLEDANO, M., NUCCI, C. &

BANERJEE, A. 2012b. Influence of air-abrasion executed with polyacrylic

acid-Bioglass 45S5 on the bonding performance of a resin-modified glass

ionomer cement. Eur J Oral Sci, 120, 168-77.

SAY, E. C., NAKAJIMA, M., SENAWONGSE, P., SOYMAN, M., OZER, F.,

OGATA, M. & TAGAMI, J. 2006. Microtensile bond strength of a filled vs

unfilled adhesive to dentin using self-etch and total-etch technique. J Dent, 34,

283-91.

SCAFFA, P. M., VIDAL, C. M., BARROS, N., GESTEIRA, T. F., CARMONA, A. K.,

BRESCHI, L., PASHLEY, D. H., TJÄDERHANE, L., TERSARIOL, I. L.,

NASCIMENTO, F. D. & CARRILHO, M. R. 2012. Chlorhexidine inhibits the

activity of dental cysteine cathepsins. J Dent Res, 91, 420-5.

SCHWARTZ, R. S., MAUGER, M., CLEMENT, D. J. & WALKER, W. A. 1999.

Mineral trioxide aggregate: a new material for endodontics. J Am Dent Assoc,

130, 967-75.

308

SCOTT, J. E. & THOMLINSON, A. M. 1998. The structure of interfibrillar

proteoglycan bridges (‘shape modules’) in extracellular matrix of fibrous

connective tissues and their stability in various chemical environments. J Anat,

192, 391-405.

SELA-PASSWELL, N., ROSENBLUM, G., SHOHAM, T. & SAGI, I. 2010. Structural

and functional bases for allosteric control of MMP activities: Can it pave the

path for selective inhibition? Biochim Biophys Acta, 1803, 29-38.

SEPULVEDA, P., JONES, J. R. & HENCH, L. L. 2002. In vitro dissolution of melt-

derived 45S5 and sol-gel derived 58S bioactive glasses. J Biomed Mater Res,

61, 301-11.

SHANNON, I. L. & KEUPER, J. B. 1976. Microhardness of human dentin: Baseline

values and effects of fluorides. J Miss Dent Assoc, 56, 11-8.

SHEN, P., CAI, F., NOWICKI, A., VINCENT, J. & REYNOLDS, E. C. 2001.

Remineralization of enamel subsurface lesions by sugar-free chewing gum

containing casein phosphopeptide-amorphous calcium phosphate. J Dent Res,

80, 2066-70.

SHIMADA, Y., YAMAGUCHI, S. & TAGAMI, J. 2002. Micro-shear bond strength of

dual-cured resin cement to glass ceramics. Dent Mater, 18, 380-8.

SHIN, T. P., YAO, X., HUENERGARDT, R., WALKER, M. P. & WANG, Y. 2009.

Morphological and chemical characterization of bonding hydrophobic adhesive

to dentin using ethanol wet bonding technique. Dent Mater, 25, 1050-7.

SHOKATI, B., TAM, L. E., SANTERRE, J. P. & FINER, Y. 2010. Effect of salivary

esterase on the integrity and fracture toughness of the dentin-resin interface. J

Biomed Mater Res B Appl Biomater, 94, 230-7.

309

SHONO, Y., TERASHITA, M., SHIMADA, J., KOZONO, Y., CARVALHO, R. M. &

RUSSELL, C. M. 1999. Durability of resin–dentin bonds. J Adhes Dent, 1, 211-

8.

SIDHU, S. K. & WATSON, T. F. 1998. Interfacial characteristics of resin-modified

glass-ionomer materials: a study on fluid permeability using confocal

fluorescence microscopy. J Dent Res, 77, 1749-59.

SILVER, F. H. 1994. Wound dressings and skin replacement. Silver FH, editor.

Biomaterials medical devices and tissue engineering an integral approach.

London: Chapman & Hall, 46-91.

SIMONSEN, R. J. 2002. Pit and fissure sealant: review of the literature. Pediatr Dent,

24, 393-414.

SKINNER, L. B., CHAE, S., BENMORE, C. J., WENK, H. R. & MONTEIRO, P. J. M.

2010. Nanostructure of calcium silicate hydrates in cements. Phys Rev Lett, 104,

Article ID 195502.

SKOVRON, L., KOGEO, D., GORDILLO, L. A., MEIER, M. M., GOMES, O. M.,

REIS, A. & LOGUERCIO, A. D. 2010. Effects of immersion time and

frequency of water exchange on durability of etch-and-rinse adhesive. J Biomed

Mater Res B Appl Biomater, 95, 339-46.

SKRTIC, D., HAILER, A. W., TAKAGI, S., ANTONUCCI, J. M. & EANES, E. D.

1996. Quantitative assessment of the efficacy of amorphous calcium

phosphate/methacrylate composites in remineralizing caries-like lesions

artificially produced in bovine enamel. J Dent Res, 75, 1679-86.

SMALL, B. W. 2008. Reparative dentistry or restorative dentistry? Gen Dent, 56, 126-

8, 130.

310

SMITH, A. J. 2003. Vitality of the dentin-pulp complex in health and disease: growth

factors as key mediators. J Dent Educ, 67, 678-89.

SOARES, C. J., SOARES, P. V., SANTOS-FILHO, P. C. & ARMSTRONG, S. R.

2008. Microtensile specimen attachment and shape-finite element analysis. J

Dent Res, 87, 89-93.

SODERHOLM, K. J., ZIGAN, M., RAGAN, M., FISCHLSCHWEIGER, W. &

BERGMAN, M. 1984. Hydrolytic degradation of dental composites. J Dent Res,

63, 1248-54.

SOLES, C. F. & YEE, A. F. 2000. A discussion of the molecular mechanisms of

moisture transport in epoxy resins. J Polym Sci Part B Polym Phys, 38, 792-802.

SOMASUNDARAN, P., OFORI AMANKONAH, J. & ANANTHAPADMABHAN,

K. P. 1985. Mineral---solution equilibria in sparingly soluble mineral systems.

Colloids Surf, 15, 309-33.

SOUZA, P. P., ARANHA, A. M., HEBLING, J., GIRO, E. M. & COSTA, C. A. 2006.

In vitro cytotoxicity and in vivo biocompatibility of contemporary resin-

modified glass-ionomer cements. Dent Mater, 22, 838-44.

SPENCER, P. & WANG, Y. 2002. Adhesive phase separation at the dentin interface

under wet bonding conditions. J Biomed Mater Res, 62, 447-56.

SPENCER, P., YE, Q., PARK, J., TOPP, E. M., MISRA, A., MARANGOS, O.,

WANG, Y., BOHATY, B. S., SINGH, V., SENE, F., ESLICK, J., CAMARDA,

K. & KATZ, J. L. 2010. Adhesive/Dentin interface: the weak link in the

composite restoration. Ann Biomed Eng, 38, 1989-2003.

STANINEC, M., KIM, P., MARSHALL, G. W., RITCHIE, R. O. & MARSHALL, S. J.

2008. Fatigue of dentin–composite interfaces with four-point bend. Dent Mater,

24, 799-803.

311

SUH, B. I. 1991. All-Bond: fourth-generation dentin bonding system. J Esthet Dent, 3,

139-47.

SULKALA, M., WAHLGREN, J., LARMAS, M., SORSA, T., TERONEN, O., SALO,

T. & TJÄDERHANE, L. 2001. The effects of MMP inhibitors on human

salivary MMP activity and caries progression in rats. J Dent Res, 80, 1545-9.

SUNG, H.-W., HUANG, D.-M., CHANG, W.-H., HUANG, R.-N. & HSU, J.-C. 1999.

Evaluation of gelatin hydrogel crosslinked with various crosslinking agents as

bioadhevises: In vitro study. J Biomed Mater Res, 46, 520-30.

SWETHA, M., SAHITHI, K., MOORTHI, A., SARANYA, N., SARAVANAN, S.,

RAMASAMY, K., SRINIVASAN, N. & SELVAMURUGAN, N. 2012.

Synthesis, characterization, and antimicrobial activity of nano-hydroxyapatite-

zinc for bone tissue engineering applications. J Nanosci Nanotechnol, 12, 167-

72.

SWIFT, E. J., PERDIGAO, J. & HEYMANN, H. O. 1995. Bonding to enamel and

dentin: a brief story and state of the art. Quintessence Int, 26, 95-110.

SWIFT, E. J., PERDIGAO, J. & HEYMANN, H. O. 1998. Enamel bond strengths of

"one-bottle" adhesives. Pediatr Dent, 20, 259-62.

TADDEI, P., MODENA, E., TINTI, A., SIBONI, F., PRATI, C. & GANDOLFI, M. G.

2011. Vibrational investigation on the in vitro bioactivity of commercial and

experimental calcium-silicate cements for root-end endodontic therapy. J Mol

Struct, 993, 367-75.

TAKEUCHI, T. & HISHIYA, T. 2008. Molecular imprinting of proteins emerging as a

tool for protein recognition. Org Biomol Chem, 6, 2459-67.

312

TALLANT, C., MARRERO, A. & GOMIS-RÜTH, F. X. 2010. Matrix

metalloproteinases: fold and function of their catalytic domains. Biochim

Biophys Acta, 1803, 20-8.

TANAKA, K., TAIRA, M., SHINTANI, H., WAKASA, K. & YAMAKI, M. 1991.

Residual monomers (TEGDMA and Bis-GMA) of a set visible-light-cured

dental composite resin when immersed in water. J Oral Rehabil, 18, 353-62.

TANG, R. K., NANCOLLAS, G. H. & ORME, C. A. 2001. Mechanism of dissolution

of sparingly soluble electrolytes. J Am Chem Soc, 123, 5437-43.

TAY, F. R., CARVALHO, R. M. & PASHLEY, D. H. 2004a. Water movement across

bonded dentin - too much of a good thing. J Appl Oral Sci, 12, 12-25.

TAY, F. R., GWINNETT, A. I., PONG, K. U. & WEI, S. H. Y. 1995. The overwet

phenomenon, an optical, micromorphological study of surface moisture in the

acid etched, resin-dentin interface. Am J Dent, 9, 43-8.

TAY, F. R., GWINNETT, A. J., PANG, K. M. & WEI, S. H. 1996. Resin permeation

into acid-conditioned, moist, and dry dentin: a paradigm using water-free

adhesive primers. J Dent Res, 75, 1034-44.

TAY, F. R., HASHIMOTO, M., PASHLEY, D. H., PETERS, M. C., LAI, S. C. N.,

YIU, C. K. Y. & CHEONG, C. 2003. Aging affects two modes of nanoleakage

expression in bonded dentin. J Dent Res, 82, 537-41.

TAY, F. R. & PASHLEY, D. H. 2003a. Have dentin adhesives become too hydrophilic?

J Can Dent Assoc, 69, 726-31.

TAY, F. R. & PASHLEY, D. H. 2003b. Water-treeing – a potential mechanism for

degradation of dentine adhesives. Am J Dent, 16, 6-12.

TAY, F. R. & PASHLEY, D. H. 2008. Guided tissue remineralisation of partially

demineralised human dentine. Biomaterials, 29, 1127-37.

313

TAY, F. R. & PASHLEY, D. H. 2009. Biomimetic remineralization of resin-bonded

acid-etched dentin. J Dent Res, 88, 719-24.

TAY, F. R., PASHLEY, D. H., GARCIA-GODOY, F. & YIU, C. K. 2004b. Single-

step, self-etch adhesives behave as permeable membranes after polymerization.

Part II. Silver tracer penetration evidence. Am J Dent, 17, 315-22.

TAY, F. R., PASHLEY, D. H., KAPUR, R. R., CARRILHO, M. R., HUR, Y. B.,

GARRETT, L. V. & TAY, K. C. 2007. Bonding BisGMA to dentin--a proof of

concept for hydrophobic dentin bonding. J Dent Res, 86, 1034-9.

TAY, F. R., PASHLEY, D. H., LOUSHINE, R. J., WELLER, R. N., MONTICELLI, F.

& OSORIO, R. 2006. Self-etching adhesives increase collagenolytic activity in

radicular dentin. J Endod, 32, 862-8.

TAY, F. R., PASHLEY, D. H., SUH, B. I., CARVALHO, R. M. & ITTHAGARUN, A.

2002a. Single-step adhesives are permeable membranes. J Dent, 30, 371-82.

TAY, F. R., PASHLEY, D. H. & YOSHIYAMA, M. 2002b. Two modes of

nanoleakage expression in single-step adhesives. J Dent Res, 81, 472-6.

TAYLOR, H. F. W. 1997. Cement Chemistry, 2nd edn. Thomas Telford Ltd., London

(1997). Thomas Telford Ltd., London.

TEN CATE, J. M. & FEATHERSTONE, J. D. 1991. Mechanistic aspects of the

interactions between fluoride and dental enamel. Crit Rev Oral Biol Med, 2,

283-96.

TEN CATE, J. M. & VAN DUINEN, R. N. B. 1995. Hypermineralization of dental

lesions adjacent to glass-ionomer cement restorations. J Dent Res, 74, 1266-71.

TENO, N. & MASUYA, K. 2010. Orally bioavailable cathepsin K inhibitors with

pyrrolopyrimidine scaffold. Curr Top Med Chem, 10, 752-66.

314

TERSARIOL, I. L., GERALDELI, S., MINCIOTTI, C. L., NASCIMENTO, F. D.,

PÄÄKKÖNEN, V., MARTINS, M. T., CARRILHO, M. R., PASHLEY, D. H.,

TAY, F. R., SALO, T. & TJÄDERHANE, L. 2010. Cysteine cathepsins in

human dentin-pulp complex. J Endod, 36, 475-81.

TEZVERGIL-MUTLUAY, A., AGEE, K. A., HOSHIKA, T., CARRILHO, M.,

BRESCHI, L., TJÄDERHANE, L., NISHITANI, Y., CARVALHO, R. M.,

LOONEY, S., TAY, F. R. & PASHLEY, D. H. 2010a. The requirement of zinc

and calcium ions for functional MMP activity in demineralized dentin matrices.

Dent Mater, 26, 1059-67.

TEZVERGIL-MUTLUAY, A., AGEE, K. A., HOSHIKA, T., TAY, F. R. &

PASHLEY, D. H. 2010b. The inhibitory effect of polyvinylphosphonic acid on

functional matrix metalloproteinase activities in human demineralized dentin.

Acta Biomater, 6, 4136-42.

TEZVERGIL-MUTLUAY, A., MUTLUAY, M. M., GU, L. S., ZHANG, K., AGEE, K.

A., CARVALHO, R. M., MANSO, A., CARRILHO, M., TAY, F. R.,

BRESCHI, L., SUH, B. I. & PASHLEY, D. H. 2011. The anti-MMP activity of

benzalkonium chloride. J Dent, 39, 57-64.

THOMPSON, S. K., HALBERT, S. M., BOSSARD, M. J., TOMASZEK, T. A.,

LEVY, M. A., ZHAO, B., SMITH, W. W., ABDEL-MEGUID, S. S., JANSON,

C. A., D'ALESSIO, K. J., MCQUENEY, M. S., AMEGADZIE, B. Y.,

HANNING, C. R., DESJARLAIS, R. L., BRIAND, J., SARKAR, S. K.,

HUDDLESTON, M. J., LJAMES, C. F., CARR, S. A., GARNES, K. T., SHU,

A., HEYS, J. R., BRADBEER, J., ZEMBRYKI, D., LEE-RYKACZEWSKI, L.,

JAMES, I. E., LARK, M. W., DRAKE, F. H., GOWEN, M., GLEASON, J. G.

315

& VEBER, D. F. 1997. Design of potent and selective human cathepsin K

inhibitors that span the active site. Proc Natl Acad Sci USA, 94, 14249-54.

TJÄDERHANE, L., LARJAVA, H., SORSA, T., UITTO, V. J., LARMAS, M. &

SALO, T. 1998. The activation and function of host matrix metalloproteinases in

dentin matrix breakdown in caries lesions. J Dent Res, 77, 1622-9.

TOLEDANO, M., NIETO-AGUILAR, R., OSORIO, R., CAMPOS, A., OSORIO, E. &

TAY, F. R. 2010. Differential expression of matrix metalloproteinase-2 in

human coronal and radicular sound and carious dentine. J Dent, 38, 635-40.

TOLEDANO, M., OSORIO, R., ALBALADEJO, A., AGUILERA, F. S. & OSORIO,

E. 2006. Differential effect of in vitro degradation on resin-dentin bonds

produced by self-etch versus total-etch adhesives. J Biomed Mater Res A, 77,

128-35.

TOLEDANO, M., PERDIGÃO, J., OSORIO, E. & OSORIO, R. 2002. Influence of

NaOCl deproteinization on shear bond strength in function of dentin depth. Am J

Dent, 15, 252-5.

TORABINEJAD, M. & CHIVIAN, N. 1999. Clinical applications of mineral trioxide

aggregate. J Endod, 25, 197-205.

TORABINEJAD, M., HONG, C. U., MCDONALD, F. & PITT FORD, T. 1995.

Physical and chemical properties of a new root-end filling material. J Endod, 21,

349-53.

TORABINEJAD, M., WATSON, T. F. & PITT FORD, T. R. 1993. Sealing ability of a

mineral trioxide aggregate when used as a root-end filling material. J Endod, 19,

591-5.

316

TRAUB, W., ARAD, T. & WEINER, S. 1989. Three-dimensional ordered distribution

of crystals in turkey tendon collagen fibers. Proc Natl Acad Sci USA, 86, 9822-

6.

TSAI, T. R., TSENG, T. Y., CHEN, C. F. & TSAI, T. H. 2002. Identification and

determination of geniposide contained in gardenia jasminoides and in two

preparations of mixed traditional chinese medicines. J Chromatogr A, 961, 83-8.

TÜRKÜN, S. L. 2003. Clinical evaluation of a self-etching and a one-bottle adhesive

system at two years. J Dent, 31, 527-34.

TURUNEN, T., PELTOLA, J., YLI-URPO, A. & HAPPONEN, R. P. 1994. Effect of

bioactive glass granules and polytetrafluoroethylene membrane on repair of

cortical defects of rabbit tibia. J Mater Sci Mater Med, 6, 639-41.

TYAS, M. J. & BURROW, M. F. 2004. Adhesive restorative materials: a review. Aust

Dent J, 49, 112-21.

UNEMORI, M., MATSUYA, Y., MATSUYA, S., AKASHI, A. & AKAMINE, A.

2003. Water absorption of poly(methyl methacrylate) containing 4-

methacryloxyethyl trimellitic anhydride. Biomaterials, 24, 1381-7.

UNI EN - 2007. Cement - Part 1: Composition, Specifications And Conformity Criteria

For Common Cements.

VÄLIMÄKI, V. V., YRJANS, J. J., VUORIO, E. I. & ARO, H. T. 2005. Molecular

biological evaluation of bioactive glass microspheres and adjunct bone

morphogenetic protein 2 gene transfer in the enhancement of new bone

formation. Tissue Eng, 11, 387-94.

VAN DIJKEN, J. W., SUNNEGÅRDH-GRÖNBERG, K. & LINDBERG, A. 2007.

Clinical long-term retention of etch-and-rinse and self-etch adhesive systems in

non-carious cervical lesions. A 13 years evaluation. Dent Mater, 23, 1101-7.

317

VAN GAALEN, S. M., KRUYT, M. C., GEUZE, R. E., DE BRUIJN, J. D., ALBLAS,

J. & DHERT, W. J. 2010. Use of fluorochrome labels in in vivo bone tissue

engineering research. Tissue Eng Part B Rev, 16, 209-17.

VAN HASSEL, H. J. 1971. Physiology of the human dental pulp. Oral Surg Oral Med

Oral Pathol, 32, 126-34.

VAN LANDINGHAM, M. R., EDULJEE, R. F. & GILLESPIE, J. W. 1999. Moisture

diffusion in epoxy systems. J Appl Polym Sci, 71, 787-98.

VAN LANDUYT, K. L., DE MUNCK, J., SNAUWAERT, J., COUTINHO, E.,

POITEVIN, A., YOSHIDA, Y., INOUE, S., PEUMANS, M., SUZUKI, K.,

LAMBRECHTS, P. & VAN MEERBEEK, B. 2005. Monomer-solvent phase

separation in one-step self-etch adhesives. J Dent Res, 84, 183-8.

VAN LANDUYT, K. L., SNAUWAERT, J., DE MUNCK, J., PEUMANS, M.,

YOSHIDA, Y. & POITEVIN, A. 2007. Systematic review of the chemical

composition of contemporary dental adhesives. Biomaterials, 28, 3757-85.

VAN LANDUYT, K. L., YOSHIDA, Y., HIRATA, I., SNAUWAERT, J., DE

MUNCK, J. & OKAZAKI, M. 2008. Influence of the chemical structure of

functional monomers on their adhesive performance. J Dent Res, 87, 757-61.

VAN MEERBEEK, B., CONN, L. J., DUKE, E. S., EICK, J. D., ROBINSON, S. J. &

GUERRERO, D. 1996. Correlative transmission electron microscopy

examination of nondemineralized and demineralized resin-dentin interfaces

formed by two dentin adhesive systems. J Dent Res, 75, 879-88.

VAN MEERBEEK, B., DE MUNCK, J., YOSHIDA, Y., INOUE, S., VARGAS, M.,

VIJAY, P., VAN LANDUYT, K., LAMBRECHTS, P. & VANHERLE, G.

2003. Buonocore memorial lecture. Adhesion to enamel and dentin: current

status and future challenges. Oper Dent, 28, 215-35.

318

VAN MEERBEEK, B., KANUMILLI, P., DE MUNCK, J., VAN LANDUYT, K.,

LAMBRECHTS, P. & PEUMANS, M. 2005. A randomized controlled study

evaluating the effectiveness of a two-step self-etch adhesive with and without

selective phosphoric-acid etching of enamel. Dent Mater, 21, 375-83.

VAN MEERBEEK, B., PERDIGÃO, J., LAMBRECHTS, P. & VANHERLE, G. 1998.

The clinical performance of adhesives. J Dent, 26, 1-20.

VAN MEERBEEK, B., PEUMANS, M., POITEVIN, A., MINE, A., VAN ENDE, A.,

NEVES, A. & DE MUNCK, J. 2010. Relationship between bond-strength tests

and clinical outcomes. Dent Mater, 26, 100-21.

VAN MEERBEEK, B., WILLEMS, G., CELIS, J. P., ROOS, J. R., BRAEM, M.,

LAMBRECHTS, P. & VANHERLE, G. 1993. Assessment by nanoindentation

of the hardness and elasticity of the resin-dentin bonding area. J Dent Res, 72,

1434-42.

VAN MEERBEEK, B., YOSHIHARA, K., YOSHIDA, Y., MINE, A., DE MUNCK, J.

& VAN LANDUYT, K. L. 2011. State of the art of self-etch adhesives. Dent

Mater, 27, 17-28.

VAN NOORT, R. 1994. Clinical relevance of laboratory studies on dental materials:

strength determination--a personal view. J Dent, 22, S4-8.

VARGAS, M. A., COBB, D. S. & ARMSTRONG, S. R. 1997. Resin-dentin shear bond

strength and interfacial ultrastructure with and without a hybrid layer. Oper

Dent, 22, 159-66.

VERDELIS, K., LUKASHOVA, L., WRIGHT, J. T., MENDELSOHN, R.,

PETERSON, M. G. E., DOTY, S. & BOSKEY, A. L. 2007. Maturational

changes in dentin mineral properties. Bone, 40, 1399-407.

319

VOLLENWEIDER, M., BRUNNER, T. J., KNECHT, S., GRASS, R. N., ZEHNDER,

M., IMFELD, T. & STARK, W. J. 2007. Remineralization of human dentin

using ultrafine bioactive glass particles. Acta Biomater, 3, 936-43.

WAKABAYASHI, Y., KONDOU, Y., SUZUKI, K., YATANI, H. & YAMASHITA,

A. 1994. Effect of dissolution of collagen on adhesion to dentin. Int J

Prosthodont, 7, 302-6.

WANG, L. & NANCOLLAS, G. H. 2009. Pathways to biomineralization and

biodemineralization of calcium phosphates: the thermodynamic and kinetic

controls. Dalton Trans, 15, 2665-72.

WANG, L. J. & NANCOLLAS, G. H. 2008a. Biomineralization. From Nature to

Application. Sigel A, Sigel H, Sigel RKO, editors. Metal Ions in Life Sciences

Series 4. John Wiley & Sons; Chichester, 413-56.

WANG, L. J. & NANCOLLAS, G. H. 2008b. Calcium orthophosphates: crystallization

and dissolution. Chem Rev, 108, 4628-69.

WANG, Y. & SPENCER, P. 2003. Hybridization efficiency of the adhesive/dentin

interface with wet bonding. J Dent Res, 82, 141-5.

WANG, Y. & SPENCER, P. 2005. Continuing etching of an all-in-one adhesive in wet

dentin tubules. J Dent Res, 84, 350-4.

WANG, Z., JIANG, T., SAURO, S., WANG, Y., THOMPSON, I., WATSON, T. F.,

SA, Y., XING, W., SHEN, Y. & HAAPASALO, M. 2011. Dentine

remineralization induced by two bioactive glasses developed for air abrasion

purposes. J Dent, 39, 746-56.

WANG, Z., JIANG, T., SAURO, S., WANG, Y., XING, W., LIANG, S., SA, Y.,

ZHANG, C., SHEN, Y. & HAAPASALO, M. 2012. Nerve-targeted

320

desensitizing toothpastes occlude dentin tubules and induce mineral

precipitation. Am J Dent, 25, 26-30.

WATANABE, I., NAKABAYASHI, N. & PASHLEY, D. H. 1994. Bonding to ground

dentin by a phenyl-P self-etching primer. J Dent Res, 73, 1212-20.

WATSON, T. F. 1997. Fact and artefact in confocal microscopy. Adv Dent Res, 11,

433-41.

WEI, Y. J., SILIKAS, N., ZHANG, Z. T. & WATTS, D. C. 2011. Hygroscopic

dimensional changes of self-adhering and new resin-matrix composites during

water sorption/desorption cycles. Dent Mater, 27, 259-66.

WEINER, S. & WAGNER, H. D. 1998. The material bone: Structure-Mechanical

Function Relations. Annu Rev Mater Sci, 28, 271-98.

WELCH, K., CAI, Y., ENGQVIST, H. & STRØMME, M. 2010. Dental adhesives with

bioactive and on-demand bactericidal properties. Dent Mater, 26, 491-9.

WIEGAND, A., BUCHALLA, W. & ATTIN, T. 2007. Review on fluoride-releasing

restorative materials--fluoride release and uptake characteristics, antibacterial

activity and influence on caries formation. Dent Mater, 23, 343-62.

WILSON, A. D. 1991. Glass-ionomer cement--origins, development and future. Clin

Mater, 7, 275-82.

WILSON, J., PIGOTT, G. H., SCHOEN, F. J. & HENCH, L. L. 1981. Toxicology and

biocompatibility of bioglasses. J Biomed Mater Res, 15, 805-17.

WONG, T. S., BROUGH, B. & HO, C. M. 2009. Creation of functional micro/nano

systems through top-down and bottom-up approaches. Mol Cell Biomech, 1, 1-

55.

321

WU, W. J. & NANCOLLAS, G. H. 1998. The dissolution and growth of sparingly

soluble inorganic salts: A kinetics and surface energy approach. Pure Appl

Chem, 70, 1867-72.

XIA, Y., ZHANG, F. M., XIE, H. F. & GU, N. 2008. Nanoparticle-reinforced resin-

based dental composites. J Dent, 36, 450-5.

XU, H. H. K., SMITH, D. T., JAHANMIR, S., ROMBERG, E., KELLY, J. R.,

THOMPSON, V. P. & REKOW, E. D. 1998. Indentation damage and

mechanical properties of human enamel and dentin. J Dent Res, 77, 472-80.

XU, Z., NEOH, K. G. & KISHEN, A. 2010. A biomimetic strategy to form calcium

phosphate crystals on type I collagen substrate. Mater Sci Eng C, 30, 822-6.

YAMAUCHI, M. 2000. Collagen biochemistry: an overview. Advances in tissue

banking. Vol. 6. Phillips GO, editor. New Jersey: World Scientific Publishing

Co, 455-500.

YAMAUCHI, M. & SHIIBA, M. 2002. Lysine hydroxylation and crosslinking of

collagen. Methods Mol Biol, 194, 277-90

YAMAUTI, M., HASHIMOTO, M., SANO, H., OHNO, H., CARVALHO, R. M.,

KAGA, M., TAGAMI, J., OGUCHI, H. & KUBOTA, M. 2003. Degradation of

resin-dentin bonds using NaOCl storage. Dent Mater, 19, 399-405.

YANNAS, I. V. 1992. Tissue regeneration by use of collagen-glycosaminoglycan

copolymers. Clin Mater, 9, 179-87.

YAP, A. & LEE, C. M. 1997. Water sorption and solubility of resin-modified

polyalkenoate cements. J Oral Rehabil, 24, 310-4.

YE, Q., PARK, J., TOPP, E. & SPENCER, P. 2009a. Effect of photoinitiators on the in

vitro performance of a dentin adhesive exposed to simulated oral environment.

Dent Mater, 25, 452-8.

322

YE, Q., WANG, Y. & SPENCER, P. 2009b. Nanophase separation of polymers

exposed to simulated bonding conditions. J Biomed Mater Res B Appl Biomater,

88, 339-48.

YILDIRIM, S., TOSUN, G., KOYUTÜRK, A. E., SENER, Y., SENGÜN, A. & OZER,

F. 2008. Microtensile and microshear bond strength of an antibacterial self-

etching system to primary tooth dentin. Eur J Dent, 2, 11-7.

YIP, H. K., TAY, F. R., NGO, H. C., SMALES, R. G. & PASHLEY, D. H. 2001.

Bonding of contemporary glass ionomer cements to dentin. Dent Mater, 17,

456-70.

YIU, C. K., KING, N. M., PASHLEY, D. H., SUH, B. I., CARVALHO, R. M.,

CARRILHO, M. R. & TAY, F. R. 2004. Effect of resin hydrophilicity and water

storage on resin strength. Biomaterials, 25, 5789-96.

YOSHIDA, E., HASHIMOTO, M., HORI, M., KAGA, M., SANO, H. & OGUSHI, H.

2004a. Deproteinizing effects on resin-tooth bond structures. J Biomed Mater

Res B Appl Biomater, 68, 29-35.

YOSHIDA, Y., NAGAKANE, K., FUKUDA, R., NAKAYAMA, Y., OKAZAKI, M. &

SHINTANI, H. 2004b. Comparative study on adhesive performance of

functional monomers. J Dent Res, 83, 454-8.

YOSHIDA, Y., VAN MEERBEEK, B., NAKAYAMA, Y., SNAUWAERT, J.,

HELLEMANS, L., LAMBRECHTS, P., VANHERLE, G., WAKASA, K. &

YOSHIYAMA, M. 2000. Evidence of chemical bonding at biomaterial-hard

tissue interfaces. J Dent Res, 79, 709-14.

YOSHIDA, Y., VAN MEERBEEK, B., NAKAYAMA, Y., YOSHIOKA, M.,

SNAUWAERT, J., ABE, Y., LAMBRECHTS, P., VANHERLE, G. &

323

OKAZAKI, M. 2001. Adhesion to and decalcification of hydroxyapatite by

carboxylic acids. J Dent Res, 80, 1565-9.

YOSHIYAMA, M., SANO, H., EBISU, S., TAGAMI, I., CIUCCHI, B. &

CARVALHO, R. M. 1996. Regional tensile bond strengths of cervical sclerotic

root dentin. J Dent Res, 75, 1404-13.

YOSHIYAMA, M., TAY, F. R., DOI, J., NISHITANI, Y., YAMADA, T., ITOU, K.,

CARVALHO, R. M., NAKAJIMA, M. & PASHLEY, D. H. 2002. Bonding of

self-etch and total-etch adhesives to carious dentin. J Dent Res, 81, 556-60.

YOURTEE, D. M., SMITH, R. E., RUSSO, K. A., BURMASTER, S., CANNON, J.

M., EICK, J. D. & E.L., K. 2001. The stability of methacrylate biomaterials

when enzyme challenged: kinetic and systematic evaluations. J Biomed Mater

Res, 57, 522-31.

YU, B., AHN, J. S., LIM, J. I. & LEE, Y. K. 2009. Influence of TiO2 nanoparticles on

the optical properties of resin composites. Dent Mater, 25, 1142-73.

YUAN, H., LI, Y., DE BRUIJN, J. D., DE GROOT, K. & ZHANG, X. 2000. Tissue

responses of calcium phosphate cement: a study in dogs. Biomaterials, 21, 1283-

90.

ZEHNDER, M., SODERLING, E., SALONEN, J. & WALTIMO, T. 2004. Preliminary

evaluation of bioactive glass S53P4 as an endodontic medication in vitro. J

Endod, 30, 220-4.

ZEIGER, D. N. 2008. The effects of bioactive glass on bonding to dentin. Doctoral

Thesis / Dissertation, Chapters 2, 3, and 4, UMI Dissertation Publishing -

ProQuest University of California, San Francisco.

324

ZHANG, S. C. & KERN, M. 2009. The role of host-derived dentinal matrix

metalloproteinases in reducing dentin bonding of resin adhesives. Int J Oral Sci,

1, 163-76.

ZHANG, X., NEOH, K. G., LIN, C. C. & KISHEN, A. 2012. Remineralization of

partially demineralized dentine substrate based on a biomimetic strategy. J

Mater Sci Mater Med, 23, 733-42.

ZHENG, L., PEREIRA, P. N., NAKAJIMA, M., SANO, H. & TAGAMI, J. 2001.

Relationship between adhesive thickness and microtensile bond strength. Oper

Dent, 26, 97-104.

ZHONG, J. P., LA TORRE, G. P. & HENCH, L. L. 1994. The kinetics of bioactive

ceramics part VII: Binding of collagen to hydroxyapatite and bioactive glass.

Bioceramics, 7, 61-6.

ZHOU, J., TAN, J., YANG, X., CHENG, C., WANG, X. & CHEN, L. 2010. Effect of

chlorhexidine application in a self-etching adhesive on the immediate resin-

dentin bond strength. J Adhes Dent, 12, 27-31.

ZICARI, F., COUTHINO, E., DE MUNCK, J., POITEVIN, A., SCOTTI, R. &

NAERT, I. 2008. Bonding effectiveness and sealing ability of fiber-post

bonding. Dent Mater, 24, 967-77.

325

List of publications in international peer-reviewed

journals as a result of this work

Bioactive effects of a calcium/sodium phosphosilicate on the resin-dentine

interface: a microtensile bond strength, scanning electron microscopy, and

confocal microscopy study.

Profeta AC, Mannocci F, Foxton RM, Thompson I, Watson TF, Sauro S.

Eur J Oral Sci. 2012 Aug;120(4):353-62. doi: 10.1111/j.1600-

0722.2012.00974.x. Epub 2012 Jul 3.

PMID: 22813227 [PubMed - indexed for MEDLINE]

Experimental etch-and-rinse adhesives doped with calcium silicate-base micro-

fillers to generate therapeutic bioactive resin-dentin interfaces.

Profeta AC, Mannocci F, Foxton RM, Watson TF, Feitosa VP, De Carlo B,

Mongiorgi R, Valdré G , Sauro S.

Dent Mater. 2013 Jul;29(7):729-41. doi: 10.1016/j.dental.2013.04.001. Epub

2013 Apr 29.

PMID: 23639454 [PubMed - in process]

326

List of abstracts in international conferences of dental

research from this work

Bioactivity and adhesion of an experimental bioactive glass-containing bonding

system.

Profeta AC, Sauro S, Mannocci F, Foxton RM, Festy F, Watson TF.

89th General Session & Exhibition of the International Association for Dental

Research (IADR), San Diego, Calif., USA; March 16-19, 2011

Bioactivity and adhesion of an experimental bioactive glass-containing bonding

agent.

Profeta AC, Sauro S, Mannocci F, Foxton RM, Thompson I, Watson TF.

British Society for Oral and Dental Research (BSODR) Annual Meeting 2011,

Sheffield, England, September 13-15, 2011

Therapeutic effects of two experimental etch-and-rinse adhesives containing

bioactive micro-fillers.

Profeta AC, Mannocci F, Foxton RM, Thompson I, Watson TF, Sauro S.

Sixth International Association for Dental Research Pan-European Region

Meeting (IADR/PER), Helsinki, Finland, September 12 -15, 2012

327

Appendix

Bioactive effects of a calcium/sodiumphosphosilicate on the resin–dentineinterface: a microtensile bondstrength, scanning electronmicroscopy, and confocal microscopystudy

Profeta AC, Mannocci F, Foxton RM, Thompson I, Watson TF, Sauro S. Bioactiveeffects of a calcium/sodium phosphosilicate on the resin–dentine interface: a microten-sile bond strength, scanning electron and confocal microscopy study.Eur J Oral Sci 2012; 120: 353–362. © 2012 Eur J Oral Sci

This study evaluated, through microtensile bond strength (lTBS) testing, the bioac-tive effects of a calcium/sodium phosphosilicate (BAG) at the resin–dentine interfaceafter 6 months of storage in phosphate buffer solution (PBS). Confocal laser scan-ning microscopy (CLSM) and scanning electron microscopy (SEM) were also per-formed. Three bonding protocols were evaluated: (i) RES-Ctr (no use of BAG),(ii) BAG containing adhesive (BAG-AD), and (iii) BAG/H3PO4 before adhesive(BAG-PR). The dentin-bonded specimens were prepared for lTBS testing, whichwas carried out after 24 h or 6 months of storage in PBS. Scanning electron micros-copy ultramorphology analysis was performed after debonding. Confocal laser scan-ning microscopy was used to evaluate the morphological and nanoleakage changesinduced by PBS storage. High lTBS values were achieved in all groups after 24 hof storage in PBS. Subsequent to 6 months of storage in PBS the specimens createdusing the BAG-AD bonding approach still showed no significant reduction inlTBS. Moreover, specimens created using the BAG-AD or the BAG-PR approachshowed an evident reduction of nanoleakage after prolonged storage in PBS. The useof BAG-containing adhesive may enhance the durability of the resin–dentine bondsthrough therapeutic/protective effects associated with mineral deposition within thebonding interface and a possible interference with collagenolytic enzyme activity(matrix metalloproteinases) responsible for the degradation of the hybrid layer.

Andrea C. Profeta, FrancescoMannocci, Richard M. FoxtonIan Thompson, Timothy F. Watson,Salvatore Sauro*

Biomaterials, Biomimetics and BiophotonicsResearch Group (B3), King’s College LondonDental Institute, Guy’s Hospital, London, UK

Salvatore Sauro, Dental Biomaterials Science,King’s College London Dental Institute, Floor17 Guy’s Tower, London SE1 9RT, UK

Telefax: +44-207-1881823E-mail: salvatore.sauro@kcl.ac.uk

Key words: adhesion durability; bioactiveglass; bonded-dentine interface

Accepted for publication May 2012

Bioactive materials are often used in operative den-tistry due to their ability to interact actively withdental hard tissues, inducing calcium-phosphates (Ca/P) deposition in the presence of body fluids or saliva(1–4).

Whereas remineralisation of enamel lesions can beachieved predictably (5, 6), there is little information onwhether it is possible to remineralise specific mineral-deficient areas within the resin–dentine interface (i.e.hybrid layers) (2). Some polyalkanoate cements mayinduce crystal growth within gaps in the bonded interfaceafter long-term storage in water (7). Furthermore, bioac-tive, ion-releasing materials, such as calcium-phosphate(Ca/P) cements, have the potential to encourage dentineremineralisation by mineral precipitations (8–11).

PETERS et al. (12) showed the presence of a highermineral content [determined by electron probe elemen-tal micro-analysis (EPMA) techniques] and an increasein microhardness along the interface of resin-bondedcaries-affected dentine, following the application ofmaterials containing Ca/P cements. Bioactive calcium/sodium (Ca/Na) phosphosilicates, such as Bioglass45S5 (BAG), are able to induce deposition of hydroxy-carbonate apatite (4, 13–15). Although bioactive glasseshave previously been used for dentine remineralisationby direct application onto demineralised dentinal tissuewhen dispersed in water solutions (4, 14), there is littleinformation about the potential therapeutic effects ofBAG on the resin–dentine interface when used duringetch-and-rinse bonding procedures.

Eur J Oral Sci 2012; 120: 353–362DOI: 10.1111/j.1600-0722.2012.00974.xPrinted in Singapore. All rights reserved

! 2012 Eur J Oral Sci

European Journal ofOral Sciences

Therefore, this study was devised to assess the bioac-tive effects of BAG during etch-and-rinse dentine-bond-ing procedures on the resin–dentine interface. This aimwas accomplished by evaluating the microtensile bondstrength (lTBS) of specimens after 24 h and 6 monthsof storage in PBS. Fractographic analysis was also per-formed through scanning electron microscopy (SEM).The ultramorphology and nanoleakage analysis of theresin-bonded dentine was executed using confocal laserscanning microscopy (CLSM).

The null hypotheses to be tested in this study were:(i) the use of BAG employed during bonding proce-dures has no effect on the bond strength, and (ii) thepresence of BAG does not reduce nanoleakage withinthe demineralised ‘poorly-infiltrated’ areas within theresin-dentine interface.

Material and methods

Specimen preparation

Caries-free human third molars, extracted for surgical rea-sons from 20- to 40-yr-old patients, were used in thisstudy. The treatment plan of any of the involved patients,who had given informed consent for use of their extractedteeth for research purposes, was not altered by this study.The study was conducted in accordance with the ethicalguidelines of the Research Ethics Committee (REC) formedical investigations.

The teeth were stored in deionised water (pH 7.1) at 4°C and used within 1 month after extraction. The coronaldentine specimens were prepared by sectioning the roots1 mm beneath the cemento–enamel junction (CEJ) with ahard tissue microtome (Accutom-50; Struers, Copenhagen,Denmark) using a slow-speed, water-cooled diamondwafering saw (330-CA RS-70300; Struers) (Fig. 1). A 180-grit silicon carbide (SiC) abrasive paper mounted on awater-cooled rotating polishing machine (Buehler Meta-Serv 3000 Grinder-Polisher; Buehler, Dusseldorf,Germany) was used (30 s) to remove the diamond sawsmear layer and to replace it with a standard and moreclinically relevant smear layer (16).

Experimental bonding procedures and formulation ofresin adhesives

A resin co-monomer blend was formulated by using ahydrophobic, cross-linking dimethacrylate monomer –bisphenyl-A-glycidyl methacrylate (Bis-GMA; Esstech, Es-sington, PA, USA) – and a hydrophilic monomer –2-hydroxyethyl methacrylate (HEMA; Sigma-Aldrich,Gillingham, UK). In order to obtain a dental bondingsystem with chemical affinity to calcium (present indentine and BAG), an acidic functional monomer – 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylicacid (PMDM; Esstech Essington) – was also includedwithin the composition of the resin blend. Subsequently,the resin blend was made light-curable by a binaryphotoinitiator system based on camphoroquinone (CQ;Sigma-Aldrich) and 1,2-ethyl-dimethyl-4-aminobenzoate(EDAB; Sigma-Aldrich). This resin co-monomer blendwas used to formulate the experimental primer and thebonds used in this study (Table 1).

A BAG (Sylc; OSspray, London, UK) with particlesize < 10 lm was employed in the etch-and-rinse bondingprocedures using two different experimental approaches:(i) BAG-AD (30wt% BAG included within the composi-tion of a resin adhesive as a bioactive microfiller), and (ii)BAG-PR (BAG applied directly onto H3PO4-etched/wetted dentine before bonding procedures). The neat adhe-sive, with no BAG, served as the control (RES-Ctr)(Fig. 1).

In detail, a water wet-bonding dentine substrate wasachieved by water-rinsing, for 15 s, the dentine surfacesacid-etched with 37% phosphoric acid solution (H3PO4)(Sigma-Aldrich) and gently blowing off (for 2 s) excesswater to leave a wet reflective-surface.

The control bonding procedure (RES-Ctr) was accom-plished by applying two consecutive coats of an ethanol-solvated resin primer [50 wt% absolute ethanol (Sigma-Aldrich) and 50 wt% of neat co-monomer resin blend] anda layer of the neat co-monomer resin blend (Table 1) withina period of 20 s. Light-curing was immediately performedfor 30 s (>600 mW/cm!2, Optilux VLC; Demetron,Danbury, CT, USA).

The first experimental bonding procedure (BAG-AD)was performed by applying the same ethanol-solvatedresin primer onto H3PO4-etched dentine, as previouslydescribed, followed by a layer of bonding resin containingBAG (Table 1; Fig. 1). The bonding and the light-curingprocedures were executed as previously described for theRES-Ctr group.

The second experimental bonding procedure (BAG-PR)was performed as follows. The 37% H3PO4 solution(Sigma-Aldrich). was applied onto the dentine surface for15 s. Then, 0.05 g of BAG powder was placed onto theH3PO4-etched wet dentine surface, spread immediately for10 s using a cotton pellet, and finally rinsed with copiousamounts of deionised water for 15 s (Fig. 1). The primer/bond application and the light-curing procedures were per-formed as previously described for the RES-Ctr group.

A final composite build-up (5 mm) was constructed oneach specimen using a light-cured resin composite (Filtek

Fig. 1. Schematic illustrating the experimental study design.Human third molars were used to prepare standardised den-tine surfaces. The three different bonding approaches wereperformed using specific components and application proce-dures. Bis-GMA, bisphenyl-A-glycidyl methacrylate; HEMA,2-hydroxyethyl methacrylate; PMDM, 2,5-dimethacryloyloxy-ethyloxycarbonyl-1,4-benzenedicarboxylic acid.

354 Profeta et al.

Z250; 3M-ESPE, St Paul, MN, USA) in five incrementallayers (of 1 mm thickness). Each layer of composite wasindividually light cured for 20 s. The resin-bonded dentinespecimens were stored in PBS for 24 h or 6 months at 37°C. The PBS was composed of (in g/l) CaCl2 (0.103),MgCl2.6H2O (0.019), KH2PO4 (0.544), KCl (17), andHEPES (acid) buffer (4.77), and the pH was 7.4.

lTBS and SEM fractography and failure analysis

Twenty dentine-bonded specimens from each group weresectioned using a slow-speed water-cooled diamond wafer-ing blade (Struers) mounted on a hard-tissue microtome(Isomet 11/1180; Buehler) in both x and y directionsacross the adhesive interface to obtain matchsticks withcross-sectional areas of 0.9 mm2. By excluding peripheralbeams showing the presence of residual enamel, only theremaining matchsticks (n = 10–15) were selected to createthree groups with the same total number of resin–dentinespecimens in each group (n = 280). The exact width ofeach matchstick was checked using a calliper (MitutoyoCD15; Mitutoyo, Kawasaki, Japan) and half of them(n = 140) were tested after 24 h of storage in PBS and theremaining half (n = 140) were tested after 6 months ofstorage in PBS at 37°C. The lTBS test was performedusing a microtensile jig in a LAL300 linear actuator(SMAC Europe; Horsham, UK) with a LAC-1 high-speedcontroller single axis with a built-in amplifier and at thefollowing settings: stroke length = 50 mm, peakforce = 250 N, displacement resolution = 0.5 mm, andcrosshead speed = 1 mm min!1. Bond-strength data werecalculated and expressed in MPa, the lTBS values ofsticks from the same restored teeth were averaged, and themean bond strength was used as one statistical unit forthe statistical analysis. The lTBS (mean-MPa) data for

each group were analysed using a repeated-measures ANO-

VA and Tukey’s post-hoc test for pairwise comparisons(a = 0.05).

The mode of failure was classified as percentage ofadhesive, mixed, or cohesive. The failed bonds were exam-ined at 930 magnification using a stereoscopic microscope(Leica M205A; Leica Microsystems, Wetzlar, Germany).

Five representative debonded specimens for each groupthat failed in mixed or adhesive modes were selected forultramorphology analysis of the fractured surface (SEMFractography). They were dried overnight and mounted onaluminium stubs with carbon cement, then sputter-coatedwith gold (SCD 004 Sputter Coater; Bal-Tec, Vaduz, Liech-tenstein) and examined using SEM (S-3500; Hitachi, Wok-ingham, UK) with an accelerating voltage of 15 kV and aworking distance of 25 mm at increasing magnifications.

Confocal microscopy ultramorphology andnanoleakage evaluation

A further three dentine specimens from each group werebonded, as previously described, with the primer/bondresins doped with 0.1 wt% rhodamine-B (Rh-B: Sigma-Aldrich, St Louis, MO, USA) and employed for theconfocal microscopy analysis (18, 19). The specimens wereserially sectioned across the adhesive interface to obtainresin–dentine slabs (of 1 mm thickness). The resin–dentineslabs (n = 10 per group) were then divided into two sub-groups based on the period of storage in PBS (24 h or6 months) (Fig. 2). Subsequent to the storage period, thespecimens were coated with two layers of fast-setting nailvarnish applied 1 mm from the resin–dentine interfacesand immersed in 1 wt% aqueous fluorescein (Sigma-Aldrich) solution for 24 h. The specimens were thentreated in an ultrasonic water bath for 2 min and polished

Table 1

Composition of the experimental bonding procedures/adhesive systems used in this study

Experimental bondingprocedures Dentine conditioning

Chemical composition (wt%) of the adhesivesystems

Primer Bond

BAG-AD 37% H3PO4 solution, 15 s immediately followedby the application of a 0.05 g of BAG onH3PO4 wet dentine

20.21 wt% BisGMA15.54 wt% PMDM14.25 wt% HEMA50.00 wt% Absoluteethanol

37.50 wt% BisGMA16.80 wt% PMDM15.70 wt% HEMA30.00 wt% BAG

BAG-PR 37% phosphoric acid solution, 15 s – H3PO4: 20.21 wt% BisGMA15.54 wt% PMDM14.25 wt% HEMA50.00 wt% Absoluteethanol

40.00 wt% Bis-GMA31.50 wt% PMDM28.50 wt% HEMA

RES-Ctr 37% phosphoric acid solution, 15 s – H3PO4: 20.21 wt% BisGMA15.54 wt% PMDM14.25 wt% HEMA50.00 wt% Absoluteethanol

40.00 wt% Bis-GMA31.50 wt% PMDM28.50 wt% HEMA

Bis-GMA, bisphenyl A glycidyl methacrylate; HEMA, hydrophilic 2-hydroxyethyl methacrylate; PMDM, 2,5-dimethacryloyloxy-ethyloxycarbonyl-1,4-benzenedicarboxylic acid.At the end of the formulation of the resins, 0.25 wt% camphoroquinone (CQ) and 1.0 wt% 2-ethyl-dimethyl-4-aminobenzoate(EDAB) were added to the resin mixture.

Effects of a bioactive glass on the resin–dentine interface 355

using SiC abrasive papers of ascending grit (#1200 to#4000) (Versocit; Struers) on a water-cooled rotating pol-ishing machine (Buehler Meta-Serv 3000 Grinder-Polisher;Buehler). A final treatment in an ultrasonic water bath(5 min) completed the specimen preparation for the confo-cal microscopy evaluation (Fig. 2).

The microscopy examination was performed using aconfocal laser scanning microscope (Leica SP2 CLSM; Le-ica, Heidelberg, Germany) equipped with a 63 9 /1.4 NAoil-immersion lens and using 488-nm argon/helium (fluo-rescein excitation) or 568-nm krypton (rhodamine excita-tion) laser illumination. The reflection imaging wasperformed using the argon/helium laser. Confocal laserscanning microscopy reflection and fluorescence imageswere obtained with a 1-lm z-step to section optically thespecimens to a depth up to 20 lm below the surface (18).The z-axis scans of the interface surface were arbitrarilypseudo-coloured by two selected operators and compiledinto single projections using the Leica image-processingsoftware (Leica). The configuration of the system wasstandardised and used at the same settings for the entireinvestigation. Each resin–dentine interface was completelyinvestigated and then five optical images were randomlycaptured. Micrographs representing the most common fea-tures of nanoleakage observed along the bonded interfaceswere captured and recorded (19).

Results

lTBS and SEM fractography and failure analysis

The BAG-bonding technique vs. storage time was statisti-cally significant only for the BAG-AD group (P = 0.001);no significant reduction of the lTBS values was observedafter 6 months of storage in PBS (P > 0.05). On the otherhand, significant lTBS reductions were observed in boththe BAG-PR and RES-Ctr groups (P < 0.05) after pro-longed storage in PBS (6 months). The lTBS results(expressed asMean and SD) are presented in Table 2.

High lTBS values were achieved in all groups after24 h of storage in PBS, with failures occurring mainlyin cohesive mode in all groups; in contrast, importantchanges in the lTBS were observed after 6 months ofstorage in PBS. For instance, the lTBS of specimens inthe RES-Ctr group (no BAG) showed a significant(P < 0.05) decrease after 6 months of storage in PBSand failed mostly in adhesive mode (66%). The speci-mens stored for 24 h in PBS that fractured in mixedmode were characterised by the presence of exposeddentinal tubules with spare extruded resin tags(Fig. 3A2). Conversely, the surface of the specimensthat failed in adhesive mode after 6 months of storagein PBS presented several ‘funnelled’ dentinal tubuleswith no exposed collagen fibrils (Fig. 3A3). The resin–dentine specimens of the BAG-AD group maintained ahigh lTBS (P > 0.05) after 6 months of storage in PBS(23.89 ± 7.74 MPa). In these specimens the failure wasprevalent in cohesive (43%) and mixed (40%) modes(Fig. 3B1) and the SEM analysis of the fractured sur-face revealed a dentine surface predominantly coveredby residual resin and mineral crystals embedded withina resin/collagen network (Fig. 3B3).

The specimens of the BAG-PR group, where theBAG powder was applied onto acid-etched/wetted

Fig. 2. Schematic illustrating the composite-tooth matchsticks(1 mm) prepared using a water-cooled diamond saw, stored inPBS for 24 h or 6 months, and then subjected to microtensilebond strength (lTBS) testing and scanning electronmicroscopy failure analysis. This schematic also illustrateshow composite-tooth slabs were prepared, stored in PBS for24 h or 6 months, and evaluated by confocal laser scanningmicroscopy.

Table 2

Mean and standard deviation (SD) of microtensile bondstrength values (MPa) obtained for the different experimentalgroups and percentage distribution of failure mode after micro-tensile bond strength testing; total number of beams (tested

stick/pre-load failure)

lTBS – mean ± SD(N of tested/pre-failed beams)

% Failure [A/M/C]24 h test 6 month test

BAG-AD A126.91 ± 3.43(140/0)[0/10/90]

B123.89 ± 7.75(135/5)[17/40/43]

BAG-PR A127.20 ± 3.92(140/0)[0/9/91]

A213.35 ± 5.32(134/6)[56/11/33]

RES-Ctr A129.12 ± 4.75(140/0)[0/4/96]

A218.18 ± 5.66(133/7)[66/10/24]

For each horizontal row: values with identical numbers indi-cate no significant difference.For each vertical column: values with identical letters indicateno significant difference using Student-Newman–Keuls test(P > 0.05).

356 Profeta et al.

dentine before application of the adhesive system,showed a significant decrease in lTBS values (P < 0.05)after prolonged storage in PBS (table 2). These speci-mens failed mainly in adhesive mode (56%) after6 months of storage in PBS, and the SEM fractograph-ic analysis showed that the fracture during lTBS test-ing occurred along the intertubular dentine, leaving anintact peritubular dentine and a consistent precipitationof mineral inside the dentinal tubules (Fig. 3C3).

Confocal microscopy ultramorphology andnanoleakage evaluation

The CLSM investigation showed that all the bondingprocedures used in this study were able to create a

resin diffusion within the demineralised dentine(hybrid layer 7–9 lm) and several resin tags into thedentinal tubules (Fig. 4A). Nevertheless, the resin–den-tine interfaces of the specimens created in the threegroups showed evident fluorescein penetration (nano-leakage) within the hybrid layer and along the den-tinal tubules after 24 h of storage in PBS (Fig. 4B,C).The experimental bonding approach used to bond thespecimens of the BAG-PR group created resin–dentineinterfaces characterised by the presence of mineraldeposits inside the dentinal tubules and within thehybrid layer (Fig. 4D).

The prolonged storage in PBS induced importantchanges in terms of ultramorphology and nanoleakage.For instance, the resin–dentine interface of the RES-

A1 A2 A3

B1 B2 B3

C1 C2 C3

Fig. 3. Scanning electron microscopy images of failure modes of the resin-bonded specimens created using the three differentbonding approaches tested. (A) Micrograph of the failure mode (cohesive) of the resin control bonded to etched dentine (37%H3PO4) after 24 h of storage in PBS (A1). At higher magnification (A2) it was possible to observe the presence of some exposeddentinal tubules, but most remained obliterated by resin tags. No exposed collagen fibrils were visible on the dentine surface, anda well resin-hybridised hybrid layer was present (pointer). At 6 months (A3), the resin–dentine interfaces created with the controlresin (RES-Ctr; containing no bioactive filler) showed only a few resin tags inside the dentinal tubules and no collagen fibrils werevisible on a dentine surface characterised by funnelled dentinal tubules (pointer). (B) Micrograph of the failure mode (mixed) ofthe calcium/sodium phosphosilicate-containing adhesive (BAG-AD) bonded to dentine, after 6 months of storage in PBS (B1). Athigher magnification (B2) no exposed dentinal tubules or exposed collagen fibrils were observed; the dentine surface was wellresin-hybridised (pointer). After 6 months of storage in PBS (B3), the debonded resin–dentine interface showed the presence ofresin tags remaining inside the dentinal tubules and mineral crystals embedded within a preserved collagen network (pointer).(C) Micrograph of the failure mode (adhesive) of BAG applied directly onto H3PO4-etched/wetted dentine before bonding (BAG-PR) after 24 h of storage in PBS (C1). At higher magnification (C2) it was possible to observe the presence of some exposeddentinal tubules, while most remained obliterated by resin tags containing few BAG particles. No exposed collagen fibrils werepresent on the dentine surface (pointer). At 6 months testing (C3), the resin–dentine interface created with the BAG-PR showed adentine surface characterised by the presence of remineralised dentinal tubules obliterated by mineral crystals. It is interesting tonote how the fracture occurred along the intertubular dentine leaving an intact peritubular dentine around the mineral-obliterateddentinal tubule (pointer). rt, resin tags; t, dentinal tubules.

Effects of a bioactive glass on the resin–dentine interface 357

Ctr group specimens was affected by severe nanoleak-age within the hybrid layer and the presence of a con-tinuous gap between dentine and composite (Fig. 5A).Conversely, the specimens of the BAG-AD groupshowed the presence of a strong reflective mineralmaterial and partial dye penetration within the hybridlayer (Fig. 5B). The resin–dentine interface of speci-mens in the BAG-PR group was affected by partial dyepenetration within a crystallised hybrid layer. However,gaps were also observed between the hybrid and adhe-sive layers (Fig. 5C), probably caused by the samplepreparation procedure before the CLSM analysis.

Discussion

Hybrid layers created using etch-and-rinse adhesivesinclude water-rich, resin-sparse regions that account for2–3% of their entire volume, which increase subsequent

to prolonged aging in fluids (20). The water-rich, resin-sparse regions represent essentially the nanoporositieswithin the demineralised collagen fibrils, created duringadhesive application as a result of incomplete replace-ment of water by resin infiltration (21). This undis-placed water may act as a functional medium for thehydrolysis of suboptimally polymerised resin matricesby esterases and denaturation of collagen via the acti-vation of host-derived matrix metalloproteinases(MMPs), jeopardizing the durability of the resin–dentine interfaces (21–23).

Several methods have been advocated to increase thelongevity of these resin–dentine interfaces, including theinhibition of the MMPs within the hybrid layer (22, 23)and enhancement of the resin infiltration within thedemineralised collagen fibril using more hydrophobicresin monomers and ethanol wet-bonding (21).

Based on the results obtained in this study, the firstnull hypothesis must be partially rejected because the

A B

C D

Fig. 4. Confocal laser scanning microscopy (CLSM) images showing the interfacial characterisation and nanoleakage, after 24 hof storage in PBS, of the resin–dentine interfaces created using the three different bonding approaches tested. (A) Confocal laserscanning microscopy three-dimensional (3D) single-projection (fluorescence mode) image exemplifying the interfacial characteris-tics of the resin–dentine interface created using the control adhesive system (RES-Ctr) applied onto H3PO4-etched dentine. It ispossible to observe a clear hybrid layer (approximate thickness 9 lm) located underneath a thick adhesive layer and long resintags. (B) This CLSM 3D single-projection (fluorescence/reflection mode) image of the resin–dentine interface created using thebioactive calcium/sodium phosphosilicate-containing adhesive (BAG-AD) shows an intense nanoleakage signal from the hybridlayer (pointer) located underneath a thick adhesive layer characterised by the presence of BAG microfiller. The presence of longresin tags is also evident. (C) The resin–dentine interface created using the bonding procedure where the BAG is applied directlyonto H3PO4-etched/wetted dentine (BAG-PR) shows evident dye penetration within the hybrid layer (pointer). Short resin tagsare visible underneath a thick adhesive layer. The reason why only short resin tags could be created during this type of bondingprocedure is shown in (D) where it is possible to observe a strong reflective signal from the demineralised dentine layer (pointer)and inside the dentinal tubules, indicating the presence of mineral particles. a, adhesive layer; c, composite; fl, BAG microfiller; rt,resin tags; t, dentinal tubules.

358 Profeta et al.

use of BAG produced bioactive/protective effects onthe bond strength only when used as resin microfillerwithin the adhesive composition. The second nullhypothesis must be totally rejected as both the experi-mental bonding approaches based on the use of BAGwere able to reduce the nanoleakage within the demin-eralised ‘poorly infiltrated’ areas within the resin–den-tine interface.

In detail, the control bonding procedure (RES-Ctr)and the two experimental bonding approaches used(BAG-AD and BAG-PR) to bond the acid-etched den-tine produced comparably high lTBS values after 24 hof storage in PBS (Table 2). Conversely, a significantdecrease in lTBS (P > 0.05) occurred in all groupsafter storage in PBS for 6 months, except for the speci-mens bonded using the resin adhesive containing BAGmicrofiller (BAG-AD).

The SEM analysis of the fractured specimens of theRES-Ctr group showed, after 24 h of storage in PBS, adentine surface characterised by a hybrid layer that washighly hybridised with resin and no presence of demi-neralised collagen fibrils exposed (Fig. 3A2). Con-versely, these resin–dentine specimens stored for6 months in PBS had a fractured surface characterisedby ‘funnelled’ dentinal tubules, indicating degradationof the demineralised peritubular dentine (Fig. 3A3). Incontrast, the bonded-dentine specimens of the BAG-ADgroup immersed in PBS for 6 months had a frac-tured (adhesive mode) dentine surface, with mineralcrystals embedded within a preserved collagen networkand no evidence of ‘funnelled’ dentinal tubules(Fig. 3B3).

The SEM ultramorphology analysis of the fracturedspecimens (adhesive mode) of the RES-PR groupstored for 24 h in PBS demonstrated the presence ofdentinal tubules obliterated by resin tags and noexposed collagen fibrils (Fig. 3C2). Interestingly, when

this type of dentine-bonded specimen was immersed inPBS for 6 months it was possible to detect a fractureddentine surface characterised by dentinal tubules oblit-erated by mineral crystals and a distinctive fracturealong the intertubular dentine, which left an intact peri-tubular dentine (Fig. 3C3).

Possible explanations for such longevity attained indentine-bonded specimens created using BAG-AD afterprolonged storage in PBS may be as follows:

(i) The presence of BAG within the resin–dentineinterface may have induced the release of a silicicacid, such as Si(OH)4, and a subsequent polycon-densation reaction between the silanols com-pounds and the demineralised collagen viaelectrostatic, ionic, and/or hydrogen bonding (13,24, 25), which interfered with the ability of MMPs– although BAG is not a direct MMP inhibitor –to execute their collagenolytic and gelatinolyticactivities. A study by OSORIO et al. (26) showedthat it is possible to reduce the collagen-degrada-tion process by using specific chemical com-pounds, such as zinc oxide, which interfere withthe zinc-binding and calcium-binding catalyticdomains of MMPs.

(ii) The precipitation of an amorphous calcium phos-phate (ACP) on the polycondensate SiO2-rich tem-plate of nucleation (3, 5, 13, 18) induced by thedissolution and immediate reaction between Ca2+

and PO43! species from BAG may have also

favoured the formation of a high-molecular-weightcomplex (Ca/P–MMPs), which restricted theactivities of MMP-2 and MMP-9 within thehybrid layer (27). However, the ability of specificbioactive glass, such as Bioglass 45S5, to modulateand/or reduce the presence of collagens I, II, andIII, osteocalcin, osteonectin, and osteopontin, has

A B C

Fig. 5. Confocal laser scanning microscopy (CLSM) images showing the interfacial characterisation and nanoleakage, after6 months of storage in PBS, of the resin–dentine interfaces. (A) Confocal laser scanning microscopy three-dimensional single-pro-jection (fluorescence/reflection mode) image of the resin–dentine interface created using the control adhesive system (RES-Ctr)applied onto H3PO4-etched dentine. It is possible to note the presence of evident dye diffusion (nanoleakage) within the hybridlayer and inside the dentinal tubules (pointer). A gap is present between the dentine and the composite. (B) The resin–dentineinterface created using the bonding approach where the bioactive calcium/sodium phosphosilicate-containing adhesive (BAG-AD)is applied onto H3PO4-etched dentine shows partial dye diffusion within a hybrid layer characterised by a strong reflective signal(pointer). (C) The resin–dentine interface created using the bonding procedure where the BAG is applied directly onto H3PO4-etched/wet dentine (BAG-PR) shows a crystallised reflective layer (pointer) characterised by low dye penetration (nanoleakage). Apronounced gap can be seen between the adhesive layer and the composite. It is also possible to observe the remaining reflectivemineral materials on the fractured edge of the adhesive layer (arrows). a, adhesive layer; c, composite; g, gap.

Effects of a bioactive glass on the resin–dentine interface 359

also been demonstrated in bone-regenerationstudies (24).

(iii) The release of Na+ and Ca2+ ions from BAG,and the incorporation of H3O

+ protons into theglass particles, may have created an optimal alka-line environment (5, 18) within the resin–dentineinterface that interfered with the activity ofMMPs, which are very acidic-pH dependent (22,23).

(iv) The bioactive remineralisation induced by BAGmay have decreased the distribution of the water-rich, resin-sparse regions within the hybrid layer(2, 18) via silanols polycondensation and subse-quent ACP/AH remineralisation, which probablyinterfered with the water-dependent hygroscopicand hydrolytic degradation of the polymer net-work (28).

The confocal microscopy evaluation performed after6 months of storage in PBS indicated that both theexperimental bonding approaches used in this study(BAG-AD and BAG-PR) created a resin–dentine inter-face affected only by partial dye penetration (nanoleak-age) within a hybrid layer characterised by thedeposition of a strong reflective mineral (Fig. 5B,C).

Whereas it is reasonable to believe that the hybridlayer of the specimens created using the BAG-ADapproach remineralised as a result of the bioactive/bi-omimetic activity of Bioglass 45S5 after prolonged(6 months) storage in PBS (17, 18, 25), a completelydifferent bioactive phenomenon may have occurredwithin the resin–dentine interface created by directlyapplying the BAG on the demineralised H3PO4-wetteddentine, as a significant decrease of lTBS was attainedafter prolonged storage in PBS (Table 2).

In this case, a possible explanation for the reducedconfocal nanoleakage may be due to the chemical nat-ure of mineral precipitation that occurred within theresin–dentine interface created as a result of the experi-mental bonding procedures (BAG-PR). Our hypothesisis that the chemical reaction between BAG and H3PO4

solution (Fig. 4D) may have induced the precipitationof dicalcium-phosphate salts (i.e. brushite and mone-tite). BAKRY et al. (17, 29) showed that the acid–basechemical reaction between BAG and H3PO4 mayinduce the formation of brushite via combination ofthe phosphate (released from the BAG and H3PO4)and calcium ions (released from BAG and etched den-tine). The precipitation reaction of the brushite may beresponsible for the creation of an acidic environment(30), which may have evoked the activation of MMPs(22, 23); this situation is also aggravated by the factthat BAG no longer has the ability to create a localised‘protective’ alkaline pH within the resin–dentine inter-face.

Moreover, it is also possible that the BAG/H3PO4

reaction may have altered the chemical and/or physicalcharacteristics of BAG, in particular those responsiblefor the polycondensation of silanols and ACP/HA pre-cipitation (13, 24, 25), which may be fundamental inaltering the activity of MMPs (27), as previously

described. However, even supposing that the reactionbetween hydroxyl ions and Si(OH)4 formed the silanolscompounds and induced the polycondensation reaction,they may have been washed out by application of theair-water jet before application of the primer and bond(31).

Furthermore, a slightly acidic environment may haveremained in loco within the resin–dentine interface dur-ing the prolonged storage in PBS as a result of therelease of H+ from the acidic monomer (2,5-dimethac-ryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylicacid) contained within the resin adhesives (32–34), caus-ing a long-standing, MMP-mediated degradation ofcollagen in both the RES-Ctr and BAG-PR groups. Inaddition, a durable acidic environment may haveinduced supplementary precipitation of dicalcium oroctocalcium phosphates (30, 35) during buffered condi-tion (replacement of PBS) within the microporositiesgenerated by the degradation of the dentine collagenfibrils (Fig. 5C). Indeed, as a result of this probableadditional precipitation of mineral over time, the inter-face created using the BAG-PR bonding technique mayhave achieved mechanical characteristics similar tothose created using glass ionomer cements (GICs)applied onto polyacrylic acid-etched dentine (18, 36,37). This is probably why bond strength reduction andgap formation were observed in the BAG-PR speci-mens. The GIC-bonded interfaces can reach a tensile orshear bond strength of approximately 5 MPa andfrequently prefail during specimen preparation (37, 38).YIP et al. (39) affirmed that the results obtained fromtensile testing of GICs bonded to dentine do not repre-sent the actual strength of such stiff bonded interfacesand that only an accurate ultramorphology analysisusing electron microscopy may reveal the proper bond-ing ability of such restorative materials.

However, it is also important to consider that thehydrophilic characteristics conferred by specific resinmonomers, such as HEMA and PMDM, within thetested adhesives (Fig. 1) may have compromised themechanical properties (i.e. modulus of elasticity) of thehybrid layers (40, 41) as a result of polymer hydrolysisand swelling tensions generated within the polymerchains. In contrast, the BAG microfiller containedwithin the adhesive used in the BAG-AD group mayhave absorbed and used the water not required by thehydrophilic/acid monomers for the bioactive processesof conversion into apatite (18), thus preventing thepolymer network from considerable hygroscopic/hydro-lytic degradation (28).

In conclusion, this study provided preliminary evi-dence for the use of bioactive Ca/Na phosphosilicate,such as Bioglass 45S5, in dentine-bonding proceduresin order to enhance the durability of the resin–dentineinterfaces. However, further in vitro (i.e. transmissionelectron microscopy and atomic force microscopy-nanoindentation examination) and long-term clinicalstudies are required to confirm the protective/therapeu-tic effects of BAG on the resin-dentine interface. Con-focal Raman analysis will be also necessary to confirmthe chemical nature of the mineral precipitates observed

360 Profeta et al.

within the bonded-dentine interfaces created with thetwo experimental BAG-bonding procedures.

Acknowledgements – This article presents independent researchcommissioned by the National Institute for Health Research(NIHR) under the i4i programme and the Comprehensive Bio-medical Research Centre at Guy’s & St Thomas’ Trust. The viewsexpressed in this publication are those of the author(s) and notnecessarily those of the NHS, the NIHR or the Department ofHealth. The authors also acknowledge support from the Centreof Excellence in Medical Engineering funded by the WellcomeTrust.

Conflicts of interest – None.

References

1. KIM YK, GU LS, BRYAN TE, KIM JR, CHEN L, LIU Y, YOON

JC, BRESCHI L, PASHLEY DH, TAY FR. Mineralisation ofreconstituted collagen using polyvinylphosphonic acid/polyac-rylic acid templating matrix protein analogues in the presenceof calcium, phosphate and hydroxyl ions. Biomaterials 2010;31: 6618–6627.

2. RYOU H, NIU LN, DAI L, PUCCI CR, AROLA DD, PASHLEY

DH, TAY FR. Effect of biomimetic remineralization on thedynamic nanomechanical properties of dentin hybrid layers.J Dent Res 2011; 90: 1122–1128.

3. HENCH LL, ANDERSSON O. Bioactive glasses. In: Wilson J, ed.Introduction to bioceramics. Singapore: World Scientific;1993;45–47.

4. SAURO S, THOMPSON I, WATSON TF. Effects of common dentalmaterials used in preventive or operative dentistry on dentinpermeability and remineralization. Oper Dent 2011; 36:222–230.

5. SHEN P, CAI F, NOWICKI A, VINCENT J, REYNOLDS EC. Remin-eralization of enamel subsurface lesions by sugar-free chewinggum containing casein phosphopeptide-amorphous calciumphosphate. J Dent Res 2001; 80: 2066–2070.

6. LAGERWEIJ MD, TEN CATE JM. Remineralisation of enamellesions with daily applications of a high-concentration fluo-ride gel and a fluoridated toothpaste: an in situ study. CariesRes 2002; 36: 270–274.

7. HASHIMOTO M, NAKAMURA K, KAGA M, YAWAKA Y. Crystalgrowth by fluoridated adhesive resins. Dent Mater 2008; 24:457–463.

8. NGO HC, MOUNT G, MCINTYRE J, TUISUVA J, VON DOUSSA

RJ. Chemical exchange between glass-ionomer restorationsand residual carious dentine in permanent molars: an in vivostudy. J Dent 2006; 34: 608–613.

9. DICKENS SH, FLAIM GM, TAKAGI S. Mechanical propertiesand biochemical activity of remineralizing resin-basedCa-PO4 cements. Dent Mater 2003; 19: 558–566.

10. DICKENS SH, FLAIM GM. Effect of a bonding agent on invitro biochemical activities of remineralizing resin-basedcalcium phosphate cements. Dent Mater 2008; 24: 1273–1280.

11. ARENDS J, RUBEN JL, INABA D. Major topics in quantitativemicroradiography of enamel and dentin: R parameter, min-eral distribution visualization, and hyper-remineralization.Adv Dent Res 1997; 11: 403–414.

12. PETERS MC, BRESCIANI E, BARATA TJ, FAGUNDES TC, NAVAR-

RO RL, NAVARRO MF, DICKENS SH. In vivo dentin remineral-ization by calcium-phosphate cement. J Dent Res 2010; 89:286–291.

13. VOLLENWEIDER M, BRUNNER TJ, KNECHT S, GRASS RN, ZEHN-

DER M, IMFELD T, STARK WJ. Remineralization of humandentin using ultrafine bioactive glass particles. Acta Biomater2007; 3: 936–943.

14. EFFLANDT SE, MAGNE P, DOUGLAS WH, FRANCIS LF. Interac-tion between bioactive glasses and human dentin. J Mater SciMater Med 2002; 13: 557–565.

15. HENCH LL, PASCHALL HA. Direct chemical bond of bioactiveglass-ceramic materials to bone and muscle. J Biomed MaterRes 1973; 7: 25–42.

16. OLIVEIRA SS, PUGACH MK, HILTON JF, WATANABE LG, MAR-

SHALL SJ, MARSHALL GW Jr. The influence of the dentinsmear layer on adhesion: a self-etching primer vs. a total-etchsystem. Dent Mater 2003; 19: 758–767.

17. BAKRY AS, TAKAHASHI H, OTSUKI M, SADR A, YAMASHITA K,TAGAMI J. CO2 laser improves 45S5 bioglass interaction withdentin. J Dent Res 2011; 90: 246–250.

18. SAURO S, OSORIO R, WATSON TF, TOLEDANO M. Therapeuticeffects of novel resin bonding systems containing bioactiveglasses on mineral-depleted areas within the bonded-dentineinterface. J Mater Sci Mater Med 2012; 23: 1521–1532.

19. SAURO S, WATSON TF, THOMPSON I, TOLEDANO M, NUCCI C,BANERJEE A. Influence of air-abrasion executed with polyacry-lic acid-Bioglass 45S5 on the bonding performance of a resin-modified glass ionomer cement. Eur J Oral Sci 2012; 120: 168–177.

20. REIS A, GRANDE RH, OLIVEIRA GM, LOPES GC, LOGUERCIO

AD. A 2-year evaluation of moisture on microtensile bondstrength and nanoleakage. Dent Mater 2007; 23: 862–870.

21. PASHLEY DH, TAY FR, BRESCHI L, TJADERHANE L, CARVALHO

RM, CARRILHO M, TEZVERGIL-MUTLUAY A. State of the artetch-and-rinse adhesives. Dent Mater 2011; 27: 1–16.

22. PASHLEY DH, TAY FR, YIU C, HASHIMOTO M, BRESCHI L,CARVALHO RM, ITO S. Collagen degradation by host-derivedenzymes during aging. J Dent Res 2004; 83: 216–221.

23. BRESCHI L, MAZZONI A, NATO F, CARRILHO M, VISINTINI E,TJADERHANE L, RUGGERI AJR, TAY FR, DORIGO S, PASHLEY

DH. Chlorhexidine stabilizes the adhesive interface: a 2-yearin vitro study. Dent Mater 2010; 26: 320–325.

24. VALIMAKI VV, YRJANS JJ, VUORIO EI, ARO HT. Molecularbiological evaluation of bioactive glass microspheres andadjunct bone morphogenetic protein 2 gene transfer in theenhancement of new bone formation. Tissue Eng 2005; 11:387–394.

25. ZHONG JP, LATORRE GP, HENCH LL. The kinetics of bioactiveceramics part VII: binding of collagen to hydroxyapatite andbioactive glass. In: Andersson OH, Yli-Urpo A, eds. Bioce-ramics 7. Oxford: Butterworth-Heinemann, 1994; 61–66.

26. OSORIO R, YAMAUTI M, OSORIO E, ROMAN JS, TOLEDANO M.Zinc-doped dentin adhesive for collagen protection at thehybrid layer. Eur J Oral Sci 2011; 119: 401–410.

27. KREMER EA, CHEN Y, SUZUKI K, NAGASE H, GORSKI JP.Hydroxyapatite induces autolytic degradation and inactiva-tion of matrix metalloproteinase-1 and -3. J Bone Miner Res1998; 13: 1890–1892.

28. FERRACANE JL. Hygroscopic and hydrolytic effects in dentalpolymer networks. Dent Mater 2006; 22: 211–222.

29. BAKRY AS, TAMURA Y, OTSUKI M, KASUGAI S, OHYA K,TAGAMI J. Cytotoxicity of 45S5 bioglass paste used for den-tine hypersensitivity treatment. J Dent 2011; 39: 599–603.

30. MANDEL S, TAS AC. Brushite (CaHPO4·2H2O) to octacalciumphosphate (Ca8(HPO4)2(PO4)4·5H2O) transformation inDMEM solutions at 36.5°C. Mat Sci Mat Sci Eng C-Bio S2010; 30: 245–254.

31. BUNKER BC, TALLANT DR, HEADLEY TJ, TURNER GL, KIRK-

PATRICK RJ. The structure of leached sodium borosilicateglass. Phys Chem Glasses 1988; 29: 106–120.

32. WANG Y, SPENCER P. Continuing etching of an all-in-oneadhesive in wet dentin tubules. J Dent Res 2005; 84: 50–354.

33. SAURO S, TOLEDANO M, AGUILERA FS, MANNOCCI F, PASHLEY

DH, TAY FR, WATSON TF, OSORIO R. Resin-dentin bonds toEDTA-treated vs. acid etched dentin using ethanol wet-bond-ing. Part II: effects of mechanical cycling load on microtensilebond strengths. Dent Mater 2011; 27: 563–572.

34. BAYLE MA, GREGOIRE G, SHARROCK P. The role of acrylo-phosphonic acid monomers in the formation of hybrid layersbased on self-etch adhesives. J Dent 2007; 35: 302–308.

35. JAYARAMAN M, SUBRAMANIAN MV. Preparation andcharacterization of two new composites: collagen-brushite

Effects of a bioactive glass on the resin–dentine interface 361

and collagen octa-calcium phosphate. Med Sci Monit 2002; 8:481–487.

36. LIU G, ZHAO D, TOMSIA AP, MINOR AM, SONG X, SAIZ E.Three-dimensional biomimetic mineralization of dense hydro-gel templates. J Am Chem Soc 2009; 29: 9937–9939.

37. HEWLETT ER, CAPUTO AA, WROBET DC. Glass ionomerbond strength and treatment of dentin with polyacrylic acid.J Prosthet Dent 1991; 66: 767–772.

38. BERRY EA, POWERS JM. Bond strength of glass ionomers tocoronal and radicular dentin. Oper Dent 1994; 19: 122–126.

39. YIP HK, TAY FR, NGO HC, SMALES RG, PASHLEY DH. Bond-ing of contemporary glass ionomer cements to dentin. DentMater 2001; 17: 456–470.

40. GIANNINI M, SOARES CJ, DE CARVALHO RM. Ultimatetensile strength of tooth structures. Dent Mater 2004; 20: 322–329.

41. BEDRAN-RUSSO AK, PEREIRA PN, DUARTE WR, DRUMMOND

JL, YAMAUCHI M. Application of crosslinkers to dentin colla-gen enhances the ultimate tensile strength. J Biomed MaterRes B Appl Biomater 2007; 80: 268–272.

362 Profeta et al.

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Available online at www.sciencedirect.com

jo ur nal home p ag e: www.int l .e lsev ierhea l th .com/ journa ls /dema

Experimental etch-and-rinse adhesives doped withbioactive calcium silicate-based micro-fillers to generatetherapeutic resin–dentin interfaces

A.C. Profetaa, F. Mannoccib, R. Foxtonb, T.F. Watsona, V.P. Feitosaa,c, B. De Carlod,R. Mongiorgid, G. Valdréd, S. Sauroa,∗

a Biomaterials, Biomimetics and Biophotonics, King’s College London Dental Institute, Guy’s, King’s College and St. Thomas’ Hospital,London SE1 9RT, UKb Department of Conservative Dentistry, King’s College London Dental Institute, London, UKc Department of Restorative Dentistry, Dental Materials Division, Piracicaba Dental School, State University of Campinas, Limeira Av.901, 13414-903 Piracicaba, Brazild Department of Scienze della Terra e Geologico-Ambientali, University of Bologna, Bologna, Italy

a r t i c l e i n f o

Article history:Received 2 October 2012Received in revised form17 March 2013Accepted 4 April 2013

Keywords:Bioactive micro-fillersResin–dentin interfaceBond strengthDurabilityDentin remineralization

a b s t r a c t

Objectives. This study aimed at evaluating the therapeutic bioactive effects on the bondstrength of three experimental bonding agents containing modified Portland cement-basedmicro-fillers applied to acid-etched dentin and submitted to aging in simulated body fluidsolution (SBS). Confocal laser (CLSM) and scanning electron microscopy (SEM) were alsoperformed.Methods. A type-I ordinary Portland cement was tailored using different com-pounds such as sodium–calcium–aluminum–magnesium silicate hydroxide (HOPC),aluminum–magnesium–carbonate hydroxide hydrates (HCPMM) and titanium oxide(HPCTO) to create three bioactive micro-fillers. A resin blend mainly constituted by Bis-GMA,PMDM and HEMA was used as control (RES-Ctr) or mixed with each micro-filler to createthree experimental bonding agents: (i) Res-HOPC, (ii) Res-HCPMM and (iii) Res-HPCTO. Thebonding agents were applied onto 37% H3PO4-etched dentin and light-cured for 30 s. Afterbuild-ups, they were prepared for micro-tensile bond strength (!TBS) and tested after 24 hor 6 months of SBS storage. SEM analysis was performed after de-bonding, while CLSMwas used to evaluate the ultra-morphology/nanoleakage and the mineral deposition at theresin–dentin interface.Results. High !TBS values were achieved in all groups after 24 h. Only Res-HOPC and Res-HCPMM showed stable !TBS after SBS storage (6 months). All the resin–dentin interfacescreated using the bonding agents containing the bioactive micro-fillers tested in this studyshowed an evident reduction of nanoleakage and mineral deposition after SBS storage.Conclusion. Resin bonding systems containing specifically tailored Portland cement micro-fillers may promote a therapeutic mineral deposition within the hybrid layer and increasethe durability of the resin–dentin bond.

© 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

∗ Corresponding author. Tel.: +44 207 188 3874; fax: +44 020 71881823.E-mail address: salvatore.sauro@kcl.ac.uk (S. Sauro).

0109-5641/$ – see front matter © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.dental.2013.04.001

730 d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

1. Introduction

The durability of resin–dentin interface represents one of themain concerns in adhesive dentistry as it is affected by severedegradation processes. Bond degradation occurs mainly viawater sorption [1], hydrolysis of monomer methacrylatesester bonds caused by salivary esterases [2], and hydroly-sis of collagen fibrils, which may be enhanced by activationof endogenous dentin matrix metalloproteinases (MMPs) [3].Regarding these different mechanisms of degradation, exper-imental strategies to preserve the hybrid layer such asethanol-wet bonding [4,5] and the use of MMP inhibitors [6]have been proposed. Nevertheless, current attempts to extendthe longevity of resin–dentin bonds via incorporation of morehydrolytically stable resin monomers [7] and/or the use ofmatrix metalloproteinase inhibitors [8] fail to address two fun-damental issues: (1) replacement of the mineral phase withinthe demineralized dentin collagen; and (2) protection of thecollagen from biodegradation through fossilization of MMPs[9].

The use of bioactive materials which promptly inter-act with dental hard tissues through therapeutic/protectiveeffects may provide a feasible means to extend the longevityof resin–dentin interface [10]. Experimental resin-basedcalcium-phosphate cements have been advocated as poten-tial therapeutic restorative base-liner materials due to theirability to induce remineralization of caries-affected dentin[11]. Nonetheless, alternative strategies are being developedin order to enhance calcium (Ca2+), hydroxyl (OH−), andphosphate (PO4

−3) ions delivery within and beneath theresin–dentin hybrid layer. Calcium-silicate Portland-derivedcements are able to release Ca2+ and OH−, so creating favor-able conditions for the remineralization of dental hard tissues(i.e. dentin and enamel) [12,13]. These materials possess abioactive activity since they are able to induce the formationof apatite-like crystals on their surface in a short inductionperiod [14] eliciting a positive response at the interface fromthe biological environment [15]. However, the use of the Port-land cements in operative dentistry is still debated due toclinical limitations related to their long setting time [14,16],high dissolution rate and “specific” mechanical properties[17]. In contrast, the incorporation of resin specific monomerssuch as 2-hydroxyethyl methacrylate (HEMA), triethylenegly-col dimethacrylates (TEGDMA) and urethane dimethacrylates(UDMA) in silicate-based materials has been proposed toimprove the mechanical properties, bond strength to dentaltissues and reduce the setting time (light-curable systems)[14,18].

Since there is little information concerning the use of such“hybrid” resin-based photo-polymerizable dental adhesives,this study was purposed to assess the therapeutic/bioactiveeffects of three innovative bonding agents containingtailored Portland cement-based micro-fillers on theresin–dentin interface. This aim was accomplished byevaluating the micro-tensile bond strength (!TBS) aftersimulated body fluid solution (SBS) storage (24 h or 6 months).Fractography scanning electron microscopy (SEM) of the de-bonded specimens, ultra-morphology confocal microscopy(CLSM) and nanoleakage of the resin–dentin interface were

also executed. The null hypotheses to be tested were that theinclusion of the tested micro-fillers within the compositionof the experimental bonding agents induces: (i) no effect onthe bond strength durability; and (ii) no mineral precipitationand nanoleakage reduction within the demineralized ‘poorlyresin-infiltrated’ areas within the resin–dentin interface.

2. Materials and methods

2.1. Preparation of the experimental bioactiveresin-base bonding agents

A type I ordinary Portland cement (82.5 wt%) (OPC: Ital-cementi Group, Cesena, Italy) mainly consisting of tri-calcium silicate (Alite: 3CaO × SiO2), di-calcium silicate (Belite:2CaO × SiO2), tri-calcium aluminate (3CaO × Al2O3) and gyp-sum (CaSO4 × 2H2O) was mixed with 7.5 wt% of phyllosilicateconsisting of sodium–calcium–aluminum–magnesium silicatehydroxide hydrate [(Na,Ca)(Al,Mg)6(Si4O10)3(OH)·6H2O; AcrosOrganics, Fair Lawn, NJ, USA] in deionized water (ratio 2:1)to create the first experimental filler (HOPC). The secondfiller (HCPMM) was created by mixing 90 wt% of type I OPC,7.5 wt% phyllosilicate and 2.5 wt% of hydrotalcite consist-ing of aluminum–magnesium–carbonate hydroxide hydrate[Mg6Al2(CO3)(OH)16·4(H2O); Sigma–Aldrich, Gillingham, UK].The third filler (HPCTO) used in this study was created bymixing OPC (80 wt%), phyllosilicate (7.5 wt%), hydrotalcite(2.5 wt%) and 10 wt% titanium oxide (TiO2: Sigma–Aldrich).The three modified Portland-based silicates were mixed withdeionized water (ratio 2:1) and allowed to set in incubator at37 C for 24 h. Subsequently, they were ground in an agate jarand sieved to obtain <30 !m micro-filler particles.

A resin co-monomer blend was prepared as a typicalthree-step, etch-and-rinse bonding agent including a neatresin blend as bond and a 50 wt% ethanol–solvated resin mix-ture as primer (Res-Ctr – no filler). The neat resin blend wasformulated by using 40 wt% of a hydrophobic cross-linkingdimethacrylate 2,2-bis[4(2-hydroxy-3-methacryloyloxy-propyloxy)-phenyl]-propane (Bis-GMA; Esstech, Essington,PA, USA) and 28.75 wt% of hydrophilic 2-hydroxyethylmethacrylate (HEMA; Sigma–Aldrich). An acidic functionalmonomer Bis(2-Methacryloyloxyethyl) Pyromellitate (PMDM;Esstech Essington) was also added (30 wt%) to the blendsolution to obtain a dental bonding system with chem-ical affinity to the calcium present in the micro-fillers(Fig. 1). The neat resin was made light-curable by adding0.25 wt% camphoroquinone (CQ; Sigma–Aldrich), 0.5 wt%2-ethyl-dimethyl-4-aminobenzoate (EDAB; Sigma–Aldrich)and 0.5% diphenyliodonium hexafluorophosphate (PIHF;Sigma–Aldrich).

The resin co-monomer blend was used as control filler-freeor mixed with each micro-filler in order to formulate threeexperimental resin-base bonding agents (GB patent applica-tion no. 1118138.5 – filed on 20th October 2011): (i) Res-HOPC:60 wt% of neat resin and 40 wt% of HOPC; (ii) Res-HCPMM:60 wt% of neat resin and 40 wt% of HCPMM; and (iii) Res-HPCTO: 60 wt% of neat resin and 40 wt% of HPCTO filler(Table 1). The hybrid calcium silicate-based bonding agentswere prepared by mixing the neat resin and the fillers for

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741 731

Fig. 1 – Chemical structures of the methacrylate monomers used in the tested resin blends. Abbreviations: BisGMA:2,2-bis[4(2-hydroxy-3-methacryloyloxy-propyloxy)-phenyl]-propane; HEMA: 2-hydroxyethyl methacrylate; TEGDMA:triethylene-glycoldimethacrylate; PDMD: Bis(2-Methacryloyloxyethyl) Pyromellitate.

Table 1 – Chemical composition (wt%) and application mode of the experimental adhesive system used in this study.

Group Primer Bond Bonding procedures

Res-CtrpH (4.6)a

20 wt% Bis-GMA14.35 wt% HEMA14.4 wt% PMDM50 wt% ethanol

40 wt% Bis-GMA28.75 wt% HEMA30 wt% PMDM0.25 wt% camphoroquinone0.5 wt%2-ethyl-dimethyl-4-aminobenzoate0.5% diphenyliodoniumhexafluorophosphate

(1) Dentin conditioning with37% H3PO4 for 15 s(2) Copious rinse withdeionized water(3) Air-drying for 2 s(4) Application of a first layer ofeach experimental primer for20 s(5) Air-drying for 5 s atmaximum stream power(6) Application of a secondlayer of each experimentaladhesive for 20 s(7) Gently air-drying for 2 s(8) Light-curing for 30 s(9) Resin composite applicationand light-curing

Res-HOPCpH (8.4)a

20 wt% Bis-GMA14.35 wt% HEMA14.4 wt% PMDM50 wt% ethanol

24 wt% Bis-GMA17.25 wt% HEMA18 wt% PMDM0.15 wt% camphoroquinone0.3 wt%2-ethyl-dimethyl-4-aminobenzoate0.3 wt% diphenyliodoniumhexafluorophosphate40 wt% HOPC

Res-HCPMMpH (8.1)a

20 wt% Bis-GMA14.35 wt% HEMA14.4 wt% PMDM50 wt% ethanol

24 wt% Bis-GMA17.25 wt% HEMA18 wt% PMDM0.15 wt% camphoroquinone0.3 wt%2-ethyl-dimethyl-4-aminobenzoate0.3 wt% diphenyliodoniumhexafluorophosphate40 wt% HCPMM

Res-HPCTOpH (8.3)a

20 wt% Bis-GMA14.35 wt% HEMA14.4 wt% PMDM50 wt% ethanol

24 wt% Bis-GMA17.25 wt% HEMA18 wt% PMDM0.15 wt% camphoroquinone0.3 wt%2-ethyl-dimethyl-4-aminobenzoate0.3 wt% diphenyliodoniumhexafluorophosphate40 wt% HPCTO

Bis-GMA: bisphenyl A glycidyl methacrylate; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid; HOPC: set Portland cement and smectite; HPCMM: Portland cement, smectite and hydrotalcite; HPTCO: setPortland cement, smectite, hydrotalcite and titanium oxide.a Three discs for each experimental resin-base material (6 mm in diameter and 1 mm thick) and were light-cured for 30 s immersed in 25 ml of

H2O (pH 6.7) at 37 C and maintained for 30 days; the pH/alkalinizing activity was evaluated using a professional pH electrode (Mettler-Toledo,Leicester, UK) at room temperature (∼24 C).

732 d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Fig. 2 – Schematic illustrating the resin–dentin match-sticks prepared using a water-cooled diamond saw, stored in SBS for24 h or 6 months, and then subjected to micro-tensile bond strength (!TBS) testing and scanning electronmicroscopy/fractography. This schematic also illustrates how composite-tooth slabs were prepared, stored in SBS for 24 h or6 months, immersed in fluorescein (nanoleakage) or xylenol orange (calcium-binding dye) and finally analyzed usingconfocal laser scanning microscopy (CLSM).

30 s on a glass plate to form a homogeneous paste prior thebonding procedures.

2.2. Specimen preparation and bonding procedures

Caries-free human molars (age 20–40 years), extracted for peri-odontal reasons were used in this study. The treatment plan ofany of the involved patients, who had given informed consentthat their extracted teeth could be used for research pur-poses, was not altered by this investigation. This study wasconducted in accordance with the ethical guidelines of theResearch Ethics Committee (REC) for medical investigations.

The teeth were stored in deionized water (pH 7.1) at4 C and used within 1 month after extraction. A flat mid-coronal dentin surface was exposed using a hard tissuemicrotome (Accutom-50; Struers, Copenhagem, Denmark)equipped with a slow-speed, water-cooled diamond waferingsaw (330-CA/RS-70300; Struers). A 180-grit silicon carbide (SiC)abrasive paper mounted on a water-cooled rotating polishingmachine (Buehler Meta-Serv 3000; Grinder-Polisher, Düssel-dorf, Germany) was used (30 s) to remove the diamond sawsmear layer and to replace it with a standard and more clin-ically related smear layer [19]. The specimens were dividedinto four groups (n = 5/group) based on the tested materials(Table 1).

The specimens were etched using a 37% phosphoric acidsolution (H3PO4; Aldrich Chemical) for 15 s followed by copiouswater rinse. The etched-dentin surfaces were air-dried for 2 sto remove the excess of water. The control (Res-Ctr) and exper-imental adhesives (Res-HOPC; Res-HCPMM; Res-HPCTO) wereapplied within a period of 20 s. The specimens were imme-diately light-cured for 30 s using a quartz–tungsten–halogen(QTH) system (>600 mW/cm2, Optilux VLC; Demetron, CT,

USA). Five 1-mm-thick incremental build-up were performedusing a resin composite (Filtek Z250; 3M-ESPE, St. Paul, MN,USA), light-activated for 20 s each step with a final curing of60 s (Fig. 2). The specimens were finally stored in SBS solutions(Oxoid, Basingstoke, Hampshire, UK) for 24 h and 6 months at37 C.

2.3. !TBS test and SEM analysis of the failed bonds

The specimens were sectioned perpendicular to the adhesiveinterface with a slow speed water-cooled diamond waferingblade (Accutom-50; Struers) mounted on a hard tissue micro-tome (Isomet 11/1180; Buehler). Subsequently, match-stickswith cross-sectional adhesive area of 0.9 mm2 were created(Fig. 2). As each tooth yielded 16 beams, there were a total of 80match-sticks in each group. Half of these match-sticks (n = 40)were tested after 24 h and the remaining half (n = 40) after 6months of static SBS storage (37 C). Each resin–dentin match-stick was attached to a testing apparatus with a cyanoacrylateadhesive (Zapit; Dental Ventures, CA, USA). A tensile load wasapplied with a customized micro-tensile jig in a LAL300 lin-ear actuator (SMAC Europe; Horsham, West Sussex, UK) withLAC-1 high speed controller single axis with built-in ampli-fier, that has a stroke length of 50 mm, peak force of 250 N,displacement resolution of 0.5 mm and crosshead speed of1 mm/min [20]. The load (N) at failure and the cross-sectionalarea of each failed beam (Digital micrometer Mitutoyo CD15;Mitutoyo, Kawasaki, Japan) permitted calculation of the !TBSin MPa. The !TBS (mean-MPa) data for each group were sub-jected to the repeated measures ANOVA and Tukey’s posthoc test for pair-wise comparisons ( = 0.05). Fisher’s leastsignificant difference (LSD) test was used to isolate and com-pare the significant differences (P < 0.05) between the groups.

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741 733

Premature failures were included in the statistical analysis aszero values.

Modes of failure were classified as percentage of adhe-sive (A), mixed (M), or cohesive (C) when the failed bondswere examined at 30× using a stereoscopic microscope (LeicaM205A; Leica Microsystems, Wetzlar, Germany). For eachgroup, five representative de-bonded specimens, depictingthe most frequent failure modes, were chosen for SEM ultra-morphology analysis of the fractured surfaces. They weredried overnight and mounted on aluminum stubs with carboncement. They were sputter-coated with gold (SCD 004 Sput-ter Coater; Bal-Tec, Vaduz, Liechtenstein) and examined usingan SEM (S3500; Hitachi, Wokingham, UK) with an acceleratingvoltage of 15 kV and a working distance of 25 mm at increasingmagnifications from 60× to 5000×.

2.4. Dye-assisted CLSM evaluation

Three further dentin-bonded specimens were prepared as pre-viously described for each group with the primer/bond resinsdoped with 0.05 wt% Rhodamine B (Rh-B: Sigma–Aldrich)and then serially sectioned across the adhesive interfaceto obtain resin–dentin slabs (n = 12 per group) with a thick-ness of approx. 1 mm (Fig. 2). The resin–dentin slabs werethen allocated to two subgroups (n = 6/group) based on theperiod of static storage in SBS (24 h or 6 months). Follow-ing each aging period, the specimens were coated with twolayers of fast-setting nail varnish applied 1 mm away from theresin–dentin interfaces. Three specimens from each subgroupwere immersed in 1 wt% aqueous fluorescein (Sigma–Aldrich)and the other three specimens in 0.5 wt% xylenol orangesolution (XO: Sigma–Aldrich) for 24 h at 37 C (pH 7.2). Thelatter is a calcium-chelator fluorophore commonly used inbone remineralization studies [21], due to its ability to formcomplexes with divalent calcium ions. The specimens werethen treated in an ultrasonic water bath for 2 min and pol-ished using ascending (#1200–4000) grit SiC abrasive papers(Versocit; Struers) on a water-cooled polishing device (BuehlerMeta-Serv 3000 Grinder-Polisher; Buehler). A final ultrasonictreatment (5 min) concluded the specimen preparation for theconfocal microscopy analysis which was performed using aconfocal laser scanning microscope (DM-IRE2 CLSM; Leica,Heidelberg, Germany) equipped with a 63×/1.4 NA oil immer-sion lens. The fluorescein was excited at 488-nm, while XO at514-nm using an argon laser. The ultra-morphology evalua-tion (resin-diffusion) was executed using a 568-nm krypton(rhodamine excitation) laser. CLSM images were obtainedwith a 1 !m z-step to optically section the specimens to adepth up to 20 !m below the surface [22]. The z-axis scansof the interface surface were arbitrarily pseudo-colored bythe same operator for better exposure and compiled intosingle projections using the Leica image-processing software(Leica). The configuration of the system was standardizedand used at the same settings for the entire investigation.Each resin–dentin interface was completely investigated andthen five optical images were randomly captured. Micro-graphs representing the most common features observedalong the bonded interfaces were captured and recorded[10].

Table 2 – Mean and standard deviation (SD) of the !TBS(MPa) to dentin.

Group 24 h 6 m

Res-Ctr29.2 ± 9.9 A1(0/40)

18.5 ± 10.4 B2(5/35)

[5/15/80] [40/43/17]

Res-HOPC32.2 ± 9.4 A1(0/40)

30.3 ± 11.5 A1(3/37)

[2/36/62] [36/57/7]

Res-HCPMM28.2 ± 11.4 A1(0/40)

25 ± 12.7 A1(0/40)

[3/20/77] [40/54/6]

Res-HPCTO29 ± 11.1 A1(0/40)

5.7 ± 8.1 C2(23/17)

[3/25/72] [31/31/38]

Values are mean ± SD in MPa. In each row, same numbers indi-cate no differences (p > 0.05) after 24 h and 6 m of SBS storage. Incolumns, same capital letter indicates no statistically significantdifferences between each group (p > 0.05). Premature failures wereincluded in the statistical analysis as zero values and are indicatedin parentheses (for instance 5/35 means that there were 5 pre-mature failures and 35 testable beams). The modes of failure areexpressed in percentage in the brackets [adhesive/mix/cohesive].

3. Results

3.1. !TBS test and SEM analysis of the failed bonds

The interaction bonding system vs SBS storage was statis-tically significant only for the Res-HOPC and Res-HCPMMgroups (P = 0.001); no significant reduction of the !TBS valueswas observed after 6 months of SBS aging (P > 0.05). Conversely,significant drops in the !TBS value were observed in both Res-HPCTO and Res-Ctr groups (P < 0.05) after prolonged storage inSBS (6 months). The !TBS results (expressed as mean and SD)and modes of failures obtained for each group are summarizedin Table 2.

In details, all the tested materials showed high !TBS val-ues after 24 h of SBS storage with failures occurring mainly incohesive mode. However, only the resin–dentin specimens ofthe Res-HOPC and Res-HCPMM groups maintained high !TBSvalues (P > 0.05) after 6 months of storage in SBS (31.3 ± 11.5and 24 ± 12.7 MPa, respectively); the failure mode was preva-lently mixed (57% and 54%, respectively). The SEM analysisof the fractured surfaces at 24 h of SBS storage revealed theabsence of both patent dentinal tubules and collagen fib-rils, so indicating a good hybridization of acid-etched dentin(Fig. 3A and B, respectively). After 6 months of SBS storage, thede-bonded dentin surfaces were characterized by embeddedmineral crystals and remnant resin presenting filler lacunas(Res-HOPC: Fig. 3A1 and A2; Res-HCPMM: Fig. 3B1 and B2).In contrast, a significant drop (P < 0.05) in !TBS was observedafter 6 months of storage in SBS, with the specimens createdusing the Res-Ctr group (filler-free) and with those createdwith the Res-HTCPO. These latter specimens showed, after24 h, a well-hybridized de-bonded surface embedding severalmicro-fillers (Fig. 3C). On the contrary, specimens de-bondedafter 6 months SBS aging showed a de-bonded surface with afew dentinal tubules but with no sign of clear degradation and

734 d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Fig. 3 – SEM failure analysis of de-bonded specimens. (A) SEM micrograph (1000×) of an adhesively fractured stick bondedwith Res-HOPC after 24 h of SBS storage. Note the dentin entirely covered with adhesive resin (ra) with some fillers’ lacunas(pointer) and few patent dentinal tubules (dt). (A1) It was also possible to see the presence of resin adhesive (ra) and nounprotected collagen fibrils. Some filler particles were detached during testing procedures (pointer) but initial mineralprecipitation may be observed (white asterisk). (A2) After 6 months, the de-bonded dentin surface at higher magnification(2500×) was covered with adhesive resin (ra). Mineral crystals (asterisk) embedded within a preserved collagen networkwere vastly encountered albeit some fillers were detached (pointer). (B) SEM micrograph of a specimen bonded withRes-HCPMM after 24 h. Note the presence of adhesive resin (ra) covering the dentin and dentinal tubules (dt); some fillers’lacunas are also observed (pointer). (B1) In another portion of the same specimen, more lacunas are observed, but alsoinitial mineral crystallization was depicted (asterisk). (B2) After 6 months storage, at higher magnification (2500×), very lowbonding degradation can be observed. More mineral precipitation was detected (asterisk) and the fillers’ lacunas (pointer)are wider probably due to the expansion of the fillers when exposed to water. (C) Micrograph of a de-bonded stick fromgroup Res-HPCTO after 24 h showing dentin completely covered with rare filler detachment (pointer). (C1) In other region,note the presence of few exposed dentinal tubules as well as intact resin tags (rt). (C2) After 6 months, the fractured dentinsurface bonded with Res-HPCTO showed a de-bonding within the hybrid layer and/or at its bottom. Several resin tags (rt)were observed well-hybridized with peritubular dentin (black pointer); some funneled dentinal tubules were alsoencountered (black asterisk). (D) SEM micrograph from a specimen bonded with Res-Ctr and presenting a well-hybridizedresin-covered dentin surface. (D1) A fractured stick from the same group showed an adhesive failure at the bottom of hybridlayer with remnants of resin adhesive (ar) and some resin tags (rt). (D2) The control adhesive after 6 months showed manysigns of degradation since most collagen fibrils were degraded, the funneled dentinal tubules (asterisk) were often foundalong with poorly hybridized resin tags (pointer). White finger: filler lacuna; white asterisk: mineral precipitation; blackasterisk: funneled degraded peritubular dentin; black finger: hybridization between tags and peritubular.

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741 735

well-hybridized peritubular dentin with resin tags (Fig. 3C1and C2). The Res-Ctr specimens tested after 24 h of SBS stor-age and analyzed with SEM presented few exposed dentinaltubules but mostly were obliterated by resin tags or covered byresin remnants (Fig. 3D). Conversely, the surface of the speci-mens de-bonded after 6 months of SBS exhibited no collagenfibrils on the dentin surface with rare resin tags and degradedfunneled dentinal tubules (Fig. 3D1 and D2).

3.2. Dye-assisted CLSM evaluation

CLSM imaging of the bonded-dentin interfaces subsequentto 24 h of SBS storage showed relevant ultra-morphology andnanoleakage information for all groups. It was observed thatall tested materials were able to diffuse within the dem-ineralized dentin, creating a hybrid layer 8–14 !m thick, witha multitude of resin tags penetrating the dentinal tubules(Fig. 4). Nevertheless, all these interfaces were affected byconspicuous fluorescein penetration (nanoleakage) throughdentinal tubules into a porous hybrid layer (Fig. 4). Fur-thermore, the resin–dentin interface created using theexperimental bonding agents containing the bioactive micro-fillers showed the presence of XO dye within the hybrid,adhesive layers and inside the dentinal tubules (Fig. 5). On thecontrary, the acid-etched dentin bonded using the resin con-trol (Res-Ctr, filler-free) showed no presence of XO along theinterface.

Significant ultra-morphology changes were observed sub-sequent prolonged SBS storage. For instance, the CLSManalysis revealed no gap and limited fluorescein penetra-tion (nanoleakage) within the resin–dentin interfaces createdusing the Res-HOPC and Res-HCPMM (Fig. 6A and B). In addi-tion, OX-dye produced a clearly outlined fluorescence due to aconsistent Ca-minerals deposited within the bonding inter-face and inside the dentinal tubules (Fig. 5A and B). Theresin–dentin interfaces created using Res-HPCTO showed lessnanoleakage within the hybrid layer; evident porosities werealso observed (Fig. 6C). Intense nanoleakage and constant gapsaffected the resin–dentin interfaces created using the Res-Ctr (Fig. 6D). When the same interfaces were investigatedemploying OX, only the walls of the dentinal tubules werehighlighted (Fig. 5E).

4. Discussion

The resin–dentin interfaces created during bonding proce-dures when using contemporary “simplified” etch-and-rinsebonding agents are affected by bond strength reduction sub-sequent to prolonged (3–6 months) water aging [23]. Thisphenomenon occurs mainly due to the inability of such mate-rials to completely replace loosely bound and bulk-free waterfrom the apatite-depleted dentin collagen matrix during bond-ing procedures. This residual water within the resin–dentininterface causes hygroscopic swelling effects and hydrolyticdegradation of polymer networks and favors metallopro-teinases (MMPs)-mediated collagenolytic degradation [23–27].

In specific circumstances, the presence of water withinthe hybrid layer may be essential to facilitate a biomimeticapatite nucleation within the gap zones of collagen fibrils

and to fossilize host-derived, collagen-bound MMPs [24,25,28].Indeed, recent investigations have demonstrated that it is pos-sible to reduce the nanoleakage and micropermeability withinthe resin–dentin interface and maintain the bond strength[10] of bioactive resin-base materials applied to H3PO4 acid-etched dentin subsequent to simulated body fluid storage for3–6 months [18,22]. Ryou et al. [29] demonstrated that usinga biomimetic remineralization approach it is feasible to rem-ineralize the dentin collagen within the resin–dentin interfacevia slow release of calcium ions from set white Portlandcement and subsequent interaction of these ions with phos-phate species from SBS or dentin substrate. Portland cementsdesigned for dental applications, also known as hydraulic sili-cate cements or MTA, exhibit outstanding biological propertiesand high bioactivity when immersed in SBF [18,29,30].

In the present study, modified Portland cement-basedmicro-fillers were created and included within the compo-sition of a representative three-step/etch-and-rinse bondingagent in order to create a material with therapeuticremineralizing effects on the mineral-deficient areas along thebonding interface.

Based on the results obtained in this study, the first nullhypothesis that the inclusion of bioactive micro-fillers withinthe composition of the experimental bonding agent has noeffect on the bond strength durability must be must be rejectedas the use of the experimental Res-HOPC and Res-HCPMM sys-tems preserved the bond strength over prolonged SBS storage.The second null that no mineral precipitation and nanoleak-age reduction would be observed within the demineralized‘poorly resin-infiltrated’ areas within the resin–dentin inter-face must be also rejected.

In detail, the three experimental bonding agents con-taining the bioactive micro-fillers (Res-HOPC, Res-HCPMM,Res-HPCTO) and the control co-monomer blend (RES-Ctr) usedin this study to bond the acid-etched dentin produced com-parably high !TBS values (P > 0.05) following 24 h of storagein SBS (Table 2). Conversely, after 6 months of storage inSBS a significant decrease in !TBS (P < 0.05) was observedfor the RES-Ctr and Res-HPCTO groups, while the speci-mens bonded using Res-HOPC or Res-HCPMM maintainedconsistent long-term bond strength values (P > 0.05). The spec-imens of the Res-HOPC and Res-HCPMM groups de-bondedafter 6 months of SBS storage showed, during SEM frac-tography examination, residual resin presence and mineralbodies on the fractured surface (Fig. 3A1, A2, B1 and B2).Important morphological differences were observed in thespecimens of the Res-HPCTO group which presented a de-bonded surface characterized by very few partially exposeddentinal tubules and an important presence of mineral crys-tals after 6 months of SBS storage (Fig. 3C1 and C2). TheSEM analysis revealed that the de-bonded dentin surface ofthe specimens in RES-Ctr group was well resin-hybridizedafter 24 h of SBS storage (Fig. 3D). In contrast, the pro-longed SBS storage (6 months) induced radical changes; thedentin surface presented funneled dentinal tubules as anessential sign of degradation of the “poorly resin-infiltrated”demineralized and peritubular dentin (Fig. 3D1 and D2). Theseresults were also supported by the CLSM analysis performedto evaluate the nanoleakage and the presence of calcium-compounds within the resin–dentin interface subsequent to

736 d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Fig. 4 – Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterization andnanoleakage, after 24 h of storage in SBS. Images (1) indicate the fluorescein dye (nanoleakage) within the resin dentininterface, whereas the images (2) disclose the diffusion (Rhodamine B) of the bonding systems within the demineralized(acid-etched) dentin. The images (3) are the combined projections of both dyes. (A1–A3) CLSM images showing theinterfacial characteristics of the bonded-dentin interface created using Res-HOPC. It is possible to observe a clear hybridlayer (hl) with long resin tags (rt) penetrating the dentinal tubules (dt) underneath the adhesive layer (ad) presenting evidentmineral fillers (FL). Intense fluorescein uptake was observed within the entire resin–dentin interface as well as the adhesivelayer. (B1–B3) Characteristics of the bonded-dentin interface created using Res-HCPMM. Similarly to images from Res-HOPC,these images presented high dye uptake (nanoleakage throughout the entire resin–dentin interface as well as the adhesivelayer). (C1–C3) The resin–dentin interface created using the Res-HPCTO bonding system was characterized by a clear hybridlayer (hl) located underneath the adhesive layer (ad) containing the experimental micro-filler (FL). Long resin tags (rt)penetrating the dentinal tubules (dt) were observed as well as evident nanoleakage and dye uptake along the entireinterface and adhesive layer. (D1–D3) These images show the control bonded-dentin interface (RES-Ctr) characterized by athick hybrid layer (hl) (approximately 8 !m thickness) located underneath an adhesive layer (ad) devoid of fillers.

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741 737

Fig. 5 – CLSM single-projection image disclosing the fluorescent calcium-chelators dye xylenol orange. All images wereobtained from specimens immersed in simulated body-fluid solution for 24 h or 6 months. (A) Resin–dentin interfacecreated with Res-HOPC after 24 h of SBS storage. Mineral deposition can be visualized within the adhesive layer (ad), thehybrid layer (hl) along the walls of dentinal tubules (dt) and the filler inside the resin tags (rt). (B) Resin–dentin interfacecreated with Res-HCPMM and immersed in SBS for 6 months showing a clear fluorescence signal due to a consistentpresence of Ca-deposits within the adhesive layer (ad), hybrid layer, walls of the dentinal tubules (dt) and resin tags (rt). (C)Resin–dentin interface created with Res-HPCTO and immersed in SBS for 24 h. Xylenol orange was able to stain theCa-minerals within adhesive layer, hybrid layer and dentinal tubule (dt). Note the intense calcium deposition at bottom ofhybrid layer. (D) Resin–dentin interface created with Res-HPCTO and immersed in SBS for 6 months showing also in thiscase Ca-mineral presence at the bottom and within the hybrid layer, dt and rt. (E) Resin–dentin interface created withRes-Ctr (no filler) in which it is possible to note absence of calcium deposition both within the hybrid (hl) and adhesive layer(ad). Only the walls of the dentinal tubule tubules (dt) were stained by the fluorescent dye.

738 d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Fig. 6 – Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterization andnanoleakage after 6 months of SBS storage. (A) Resin–dentin interface created using Res-HOPC characterized by reducednanoleakage within the hybrid layer (hl). Note the absence of fluorescein uptake within the adhesive layer. (B) Imageshowing the interfacial features of the bonded-dentin interface created using the Res-HCPMM. Note the low overallnanoleakage with very little fluorescein uptake in hybrid and adhesive layers. (C) Resin–dentin interface created usingRes-HPCTO. Despite the mineral deposition and reduced nanoleakage within adhesive layer, a gap was probably created bythe cutting procedures. As the low presence of fluorescein within the interface, it is possible to assume that the resindegradation was replaced by mineral precipitation creating an interface with low elasticity (high stiffness properties). (D)Resin–dentin interface created using the control adhesive system (RES-Ctr). Note the presence of intense dye uptake(nanoleakage) within the hybrid layer and at the bottom of adhesive layer. In this case the presence of gaps frequentlyobserved between hybrid and adhesive was very likely due to hybrid layer degradation but with no clear mineralprecipitation within the interface. Dt, dentinal tubules; rt, resin tags; ad, adhesive layer; c, composite; hl, hybrid layer.

SBS storage (24 h or 6 months). Indeed, further evidencesof the therapeutic bioactivity of the experimental bond-ing agents containing the tailored Portland cement-basedmicro-fillers were attained; reduced fluorescent-dye uptake(nanoleakage) was observed along the entire resin–dentininterface after 6 months of storage in SBS (Fig. 5A–C).These latter observations along with the strong xylenolorange (XO-dye) signal from the hybrid layer and the denti-nal tubules (Fig. 5), clearly indicated the remineralizationof those areas which were previously detected as mineral-deficient/poor-resin infiltrated zones of the resin–dentininterface.

We hypothesize that the therapeutic remineralizing effectsobserved within the mineral-depleted resin–dentin interfacewere essentially due to the bioactivity of the experimentalmicro-fillers. Indeed, the reaction mechanism of the Portland

cement-based micro-fillers may have involved the reactionof the polymerized calcium-silicate hydrate gel with waterto release calcium hydroxide and to the consequent increaseof the alkalinity of the surrounding environment [31]; thisincrease of pH was confirmed in this study (Table 2). Thislocalized increase in pH within the resin–dentin interface mayhave also interfered with the activity of MMPs [3,23]. Further-more, the interaction between the phosphate ions present inthe aging solution (SBF) or in the dentin substrate, and thecalcium released from the Portland-based micro-fillers mayhave enhanced the formation of new mineral deposits uponexisting mineral constituents within the dentin matrix (bio-catalyzation) [32]. However, it is well known that the increasein environmental pH and the presence of free OH− mayfacilitate apatite nucleation and reduce the solubility of inter-mediate Ca/P species formed during the aging period [33]. The

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741 739

most appropriate pH to support the formation of stoichiomet-ric hydroxyapatite (HA) in vitro [34] and in vivo [35] falls in arange between 8 and 9. At higher pH it is common to obtain aCa-deficient HA (lower solubility then stoichiometric hydroxy-apatite) characterized by higher concentrations of PO4

3− andlower Ca2+ ions [36].

Moreover, we hypothesize the release of carboxylic species(R-COO−) from the PMDM which may have interacted withthe remnant calcium present along the front of demineral-ization at the bottom of the hybrid layer acting as a sort ofsequestering agent for Ca/P cluster, promoting the precipita-tion of Ca-compounds [18,22]. All these mineralizing processesjust mentioned may also have reduced the distribution ofwater-rich regions within the hybrid layer at the resin–dentininterface [18,37]. However, the water absorbed overtime dur-ing SBS storage may have been responsible for the hydrolyticand hygroscopic mechanisms involved in the degradation ofdental polymers [38]. In addition to the formation of min-erals precipitants within the interface, the nanostructure ofthe calcium-silicate hydrate may also have contributed toseal the dentinal tubules due to the small-scale volume ofthe forming gels, along with a slight expansion of the cal-cium silicate-based materials once immersed in SBS [39]. Inparticular, the phyllosilicates (i.e. smectite) and hydrotalcite,which were contained in the micro-fillers used in this study,have the ability to expand considerably following water sorp-tion into the interlayer molecular spaces [40]. The amount ofexpansion is due largely to the type of exchangeable cationcontained in the micro-filler; the uptake kinetics of cationexchange is fast and the presence of Na+, as the predominantexchangeable cation, can result in material swelling. In thiscondition, the exceeding water is removed, thereby preventinghygroscopic effects and hydrolytic degradation of the polymerchains [41]. Also, it is reasonable to expect that the metallicions intercalated on phyllosilicate were easily released by ion-exchange with cations present in the surrounding solutionsand acted as effective antibacterial substances in the longterm [38]; further studies are necessary to confirm this latterhypothesis.

In contrast, the bond strength reduction observed in theresin–dentin interfaces created using the Res-HPCTO bond-ing agent after prolonged immersion in SBS (Table 2) may bedue to the high hydrophilicity of the TiO2. Micro-fine tita-nium oxide (TiO2) has been used as inorganic additive ofresin composites to match the opaque properties of teeth[42] and as nano-particles to increase the microhardnessand flexural strength of dental composites [40]. However,TiO2 has been advocated as a super-hydrophilic component,in particular under ultraviolet (UV) light irradiation [43–45].Therefore, a possible explanation for the !TBS reduction maybe attributed to this high hydrophilicity which may have per-mitted excessive water adsorption and induced severe resindegradation as well as the extraction of water-soluble un-reacted monomers or oligomers from the resin-matrix [46];also in this case further studies are necessary to clarify theseparticular processes of degradation. Moreover, the replace-ment of the degraded resin by “over” mineral crystallizationwithin the Res-HPCTO bonded-dentin interface (Figs. 4 and 5)during prolonged SBS storage may have conferred mechanicalcharacteristics related to bond strength comparable to those

created by conventional glass-ionomer cements (GICs) appliedonto polyacrylic acid-etched dentin and submitted to tensiletests [47,48]. Indeed, several studies indicated that the bondstrength of GICs when tested using tensile or shear methodswas approximately 5 MPa; these values results do not reflectthe true adhesive strength to dentin [49]. These factors mayhave been also responsible for the formation of gaps withinthe resin–dentin interface created by the Res-HPCTO duringthe sample preparation [10].

Finally, a further important issue regarding the micro-fillers used in this study is their biocompatibility. Severalstudies showed the adequate biocompatibility of Portlandcement-based fillers [50–52]. The cytotoxicity of degradationproducts from methacrylate-based resins is far worse thanthose from the micro-fillers. Therefore, the addition of 40 w%micro-fillers reduced the resin percentage and, consequently,the negative effects of its degradation products. Furthermore,the products of hydrolysis of the therapeutic micro-fillersare essentially mineral phases of calcium compounds whichprecipitate within the resin–dentin interface (Figs. 4 and 5)promoting remineralization.

As the results of this study demonstrated that theresin–dentin bond may be maintained overtime by inducinga therapeutic remineralization of the bonding interface, spe-cific experimental resin bonding systems containing bioactivemicro-fillers, such as Res-HOPC, Res-HCPMM or Res-HPCTOmay offer the possibility to improve the durability of theresin–dentin interfaces. The characteristic of promoting bioac-tivity should also open up the potential to create therapeuticrestorative materials able to reduce the incidence of secondarycaries. Indeed, it is important to consider that restorativematerials containing bioactive fillers may be effective in killinga wide selection of aerobic bacteria due to the increase of thelocal pH and concentration of alkali ions [53,54]. The antibac-terial properties are potentially of great importance as theinfiltration of microorganisms may cause secondary carieswhich jeopardize the longevity of resin–dentin interface lead-ing to the replacement of dental restorations [55,56]. Furtherstudies are ongoing in order to evaluate the species-specificantibacterial effects and biocompatibility of the materialstested in this study.

5. Conclusion

The present outcomes showed that the inclusion of cal-cium silicate-based micro-fillers within the composition ofresin bonding agent may promote a therapeutic mineraldeposition within the resin–dentin interface. However, onlythe resin–dentin interfaces created using Res-HOPC andRes-HCPMM showed prolonged bond durability. Therefore,these two micro-fillers may represent potential bioac-tive components to be included during the formulationof new innovative and smart therapeutic adhesive resinsystems. The addition of titanium oxide within the com-position of bioactive micro-fillers for resin-based systemsmay enhance the remineralization effects. Thus, this typeof micro-filler should be avoided for bonding purposes butconsidered for healing treatments in step-wise restorativeprocedures [57].

740 d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Acknowledgments

This work was partially supported by the Centre of Excellencein Medical Engineering funded by the Wellcome Trust and bythe independent research commissioned by the National Insti-tute for Health Research under the Comprehensive BiomedicalResearch Centre at Guy’s & St. Thomas’ Trust. The viewsexpressed in this publication are those of the author(s) andnot necessarily those of the NHS, the NIHR or the Departmentof Health. The authors also acknowledge the laboratory sup-port offered by Dr. Silvano Zanna in the formulation of themodified Portland-based micro-fillers used in this study. Theauthors have no financial affiliation or involvement with anycommercial organization with direct financial interest in thematerials discussed in this manuscript. Any other potentialconflict of interest is disclosed.

r e f e r e n c e s

[1] Ito S, Hashimoto M, Wadgaonkar B, Svizero N, Carvalho RM,Yiu CK, et al. Effects of resin hydrophilicity on watersorption and changes in modulus of elasticity. Biomaterials2005;26:6449–59.

[2] Moszner N, Salz U, Zimmermann J. Chemical aspects ofself-etching enamel–dentin adhesives: a systematic review.Dental Materials 2005;21:895–910.

[3] Pashley DH, Tay FR, Yiu CK, Hashimoto M, Breschi L,Carvalho RM, et al. Collagen degradation by host-derivedenzymes during aging. Journal of Dental Research2004;83:216–21.

[4] Sadek FT, Pashley DH, Ferrari M, Tay FR. Tubular occlusionoptimizes bonding of hydrophobic resins to dentin. Journalof Dental Research 2007;86:524–8.

[5] Tay FR, Pashley DH, Kapur RR, Carrilho MR, Hur YB, GarrettLV, et al. Bonding BisGMA to dentin – a proof of concept forhydrophobic dentin bonding. Journal of Dental Research2007;86:1034–9.

[6] Carrilho MR, Carvalho RM, de Goes MF, di Hipolito V,Geraldeli S, Tay FR, et al. Chlorhexidine preserves dentinbond in vitro. Journal of Dental Research 2007;86:90–4.

[7] Moszner N, Fischer UK, Angermann J, Rheinberger V.Bis-(acrylamide)s as new cross-linkers for resin-basedcomposite restoratives. Dental Materials 2006;22:1157–62.

[8] Carrilho MR, Carvalho RM, Sousa EN, Nicolau J, Breschi L,Mazzoni A, et al. Substantivity of chlorhexidine to humandentin. Dental Materials 2010;26:779–85.

[9] Brackett MG, Li N, Brackett WW, Sword RJ, Qi YP, Niu LN,et al. The critical barrier to progress in dentine bonding withthe etch-and-rinse technique. Journal of Dentistry2011;39:238–48.

[10] Profeta AC, Mannocci F, Foxton RM, Thompson I, Watson TF,Sauro S. Bioactive effects of a calcium/sodiumphosphosilicate on the resin–dentin interface: amicrotensile bond strength, scanning electron microscopy,and confocal microscopy study. European Journal of OralSciences 2012;120:353–62.

[11] Peters MC, Bresciani E, Barata TJ, Fagundes TC, Navarro RL,Navarro MF, et al. In vivo dentin remineralization bycalcium-phosphate cement. Journal of Dental Research2010;89:286–91.

[12] Huan Z, Chang J, Huang XH. Self-setting properties andin vitro bioactivity of Ca2SiO4/CaSO4·1/2H2O composite bone

cement. Journal of Biomedical Materials Research Part B:Applied Biomaterials 2008;87:387–94.

[13] Coleman NJ, Awosanya K, Nicholson JW. Aspects of thein vitro bioactivity of hydraulic calcium (alumino)silicatecement. Journal of Biomedical Materials Research Part A2009;90:166–74.

[14] Taddei P, Modena E, Tinti A, Siboni F, Prati C, Gandolfi MG.Vibrational investigation on the in vitro bioactivity ofcommercial and experimental calcium-silicate cements forroot-end endodontic therapy. Journal of Molecular Structure2011;993:367–75.

[15] Parirokh M, Torabinejad M. Mineral trioxide aggregate: acomprehensive literature review. Part III: clinicalapplications, drawbacks and mechanism of action. Journalof Endodontics 2010;36:400–13.

[16] Camilleri J. Evaluation of selected properties of mineraltrioxide aggregate sealer cement. Journal of Endodontics2009;35:1412–7.

[17] Bogen G, Kuttler S. Mineral trioxide aggregate obturation: areview and a case series. Journal of Endodontics2009;35:777–90.

[18] Sauro S, Osorio R, Watson TF, Osorio E, Toledano M. Novellight-curable materials containing experimental bioactivemicro-fillers remineralise mineral-depleted bonded-dentineinterfaces. Journal of Biomaterials Science, Polymer Edition.http://dx.doi.org/10.1080/09205063.2012.727377

[19] Koibuchi H, Yasuda N, Nakabayashi N. Bonding to dentinwith a self-etching primer: the effect of smear layers. DentalMaterials 2001;17:122–6.

[20] Sauro S, Watson TF, Thompson I, Toledano M, Nucci C,Banerjee A. Influence of air-abrasion executed withpolyacrylic acid-Bioglass 45S5 on the bonding performanceof a resin-modified glass ionomer cement. European Journalof Oral Sciences 2012;120:168–77.

[21] Rahn BA, Perren SM. Xylenol orange, a fluorochrome usefulin polychrome sequential labelling of calcifying tissues.Stain Technology 1971;46:125–9.

[22] Sauro S, Osorio R, Watson TF, Toledano M. Therapeuticeffects of novel resin bonding systems containing bioactiveglasses on mineral-depleted areas within the bonded-dentininterface. Journal of Materials Science Materials in Medicine2012;23:1521–32.

[23] Breschi L, Mazzoni A, Ruggeri A, Cadenaro M, Di Lenarda R,De Stefano Dorigo E. Dental adhesion review: aging andstability of the bonded interface. Dental Materials2008;24:90–101.

[24] Kim J, Gu L, Breschi L, Tjaderhane L, Choi KK, Pashley DH,et al. Implication of ethanol wet bonding in hybrid layerremineralization. Journal of Dental Research 2010;89:575–80.

[25] Pashley DH, Tay FR, Breschi L, Tjaderhane L, Carvalho RM,Carrilho MR, et al. State of the art etch-and-rinse adhesives.Dental Materials 2011;27:1–16.

[26] Breschi L, Mazzoni A, Nato F, Carrilho MR, Visintini E,Tjaderhane L, et al. Chlorhexidine stabilizes the adhesiveinterface: a 2-year in vitro study. Dental Materials2010;26:320–5.

[27] Liu Y, Li N, Qi Y, Niu LN, Elshafiy S, Mao J, et al. The use ofsodium trimetaphosphate as a biomimetic analog of matrixphosphoproteins for remineralization of artificial caries-likedentin. Dental Materials 2011;27:465–77.

[28] Tay FR, Pashley DH. Biomimetic remineralization ofresin-bonded acid-etched dentin. Journal of Dental Research2009;88:719–24.

[29] Ryou H, Niu LN, Dai L, Pucci CR, Arola DD, Pashley DH, et al.Effect of biomimetic remineralization on the dynamicnanomechanical properties of dentin hybrid layers. Journalof Dental Research 2011;90:1122–8.

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741 741

[30] Darvell BW, Wu RC. MTA – an hydraulic silicate cement:review update and setting reaction. Dental Materials2011;27:407–22.

[31] Lawrence CD. Physicochemical and mechanical properties ofPortland cements. In: Hewlett PC, editor. LEA’s chemistry ofcement and concrete. 4th edition Woburn, MA, USA:Butterworth & Heinemann; 1998. p. 343–420.

[32] Gandolfi MG, Van Landuyt KL, Taddei P, Modena E, VanMeerbeek B, Prati C. Environmental scanning electronmicroscopy connected with energy dispersive X-rayanalysis and Raman techniques to study ProRoot mineraltrioxide aggregate and calcium silicate cements in wetconditions and in real time. Journal of Endodontics 2010;36:851–7.

[33] Somasundaran P, Ofori Amankonah J, AnanthapadmabhanKP. Mineral-solution equilibria in sparingly soluble mineralsystems. Colloids and Surfaces 1985;15:309–33.

[34] Kikuchi M, Itoh S, Ichinose S, Shinomiya K, Tanaka J.Self-organization mechanism in a bone-likehydroxyapatite/collagen nanocomposite synthesizedin vitro and its biological reaction in vivo. Biomaterials2001;22:1705–11.

[35] Recillas S, Rodríguez-Lugo V, Montero ML, Viquez-Cano S,Hernandez L, Castano VM. Studies on the precipitationbehavior of calcium phosphate solutions. Journal of CeramicProcessing Research 2012;13:5–10.

[36] Durucan C, Brown PW. Alpha-tricalcium phosphatehydrolysis to hydroxyapatite at and near physiologicaltemperature. Journal of Materials Science Materials inMedicine 2000;11:365–71.

[37] Qi YP, Li N, Niu LN, Primus CM, Ling JQ, Pashley DH, et al.Remineralization of artificial dentinal caries lesions bybiomimetically modified mineral trioxide aggregate. ActaBiomaterialia 2012;8:836–42.

[38] Ferracane JL. Hygroscopic and hydrolytic effects indental polymer networks. Dental Materials 2006;22:211–22.

[39] Skinner LB, Chae SR, Benmore CJ, Wenk HR, Monteiro PJM.Nanostructure of calcium silicate hydrates in cements.Physical Review Letters 2010;104:1955021–4.

[40] Bhattacharyya KG, Gupta SS. Adsorption of a few heavymetals on natural and modified kaolinite andmontmorillonite: a review. Advances in Colloid andInterface Science 2008;140:114–31.

[41] Malachová K, Praus P, Pavlickova Z, Turicova M. Activity ofantibacterial compounds immobilised on montmorillonite.Applied Clay Science 2009;43:364–8.

[42] Yu B, Ahn JS, Lim JI, Lee YK. Influence of TiO2 nanoparticleson the optical properties of resin composites. DentalMaterials 2009;25:1142–7.

[43] Xia Y, Zhang F, Xie H, Gu N. Nanoparticle-reinforcedresin-based dental composites. Journal of Dentistry2008;36:450–5.

[44] Chadwick RG, Traynor N, Moseley H, Gibbs N. Blue lightcuring units – a dermatological hazard? British DentalJournal 1994;176:17–21.

[45] Bonding N, Graem N, Rygaard J, Dabelsteen E. Effects ofirradiation with dental light curing units on Langerhanscells in human stratified epithelium in heterotransplantedskin. Scandinavian Journal of Dental Research 1987;95:463–6.

[46] Karuppuchamy S, Jeong JM. Super-hydrophilic amorphoustitanium dioxide thin film deposited by cathodicelectrodeposition. Materials Chemistry and Physics2005;93:251–4.

[47] Spencer P, Ye Q, Park J, Topp EM, Misra A, Marangos O, et al.Adhesive/dentin interface: the weak link in the compositerestoration. Annals of Biomedical Engineering2010;38:1989–2003.

[48] Hewlett ER, Caputo AA, Wrobel DC. Glass ionomer bondstrength and treatment of dentin with polyacrylic acid.Journal of Prosthetic Dentistry 1991;66:767–72.

[49] Berry 3rd EA, Powers JM. Bond strength of glass ionomers tocoronal and radicular dentin. Operative Dentistry1994;19:122–6.

[50] Camilleri J, Montesin FE, Di Silvio L, Pitt Ford TR. Thechemical constitution and biocompatibility of acceleratedPortland cement for endodontic use. InternationalEndodontic Journal 2005;38:834–42.

[51] Hwang YC, Kim DH, Hwang IN, Song SJ, Park YJ, Koh JT, et al.Chemical constitution, physical properties, andbiocompatibility of experimentally manufactured Portlandcement. Journal of Endodontics 2011;37:58–62.

[52] Ribeiro DA, Duarte MA, Matsumoto MA, Marques ME,Salvadori DM. Biocompatibility in vitro tests of mineraltrioxide aggregate and regular and white Portland cements.Journal of Endodontics 2005;31:605–7.

[53] Yip HK, Tay FR, Ngo HC, Smales RJ, Pashley DH. Bonding ofcontemporary glass ionomer cements to dentin. DentalMaterials 2001;17:456–70.

[54] Hu S, Chang J, Liu M, Ning C. Study on antibacterial effect of45S5 Bioglass. Journal of Materials Science Materials inMedicine 2009;20:281–6.

[55] Boeckh C, Schumacher E, Podbielski A, Haller B.Antibacterial activity of restorative dental biomaterials invitro. Caries Research 2002;36:101–7.

[56] Sauro S, Watson TF, Tay FR, Chersoni S, Breschi L, Bernardi F,et al. Water uptake of bonding systems applied on rootdentin surfaces: a SEM and confocal microscopic study.Dental Materials 2006;22:671–80.

[57] Bjørndal L, Larsen T, Thylstrup A. A clinical andmicrobiological study of deep carious lesions duringstepwise excavation using long treatment intervals. CariesResearch 1997;31:411–7.