Synergies in Biolubrication - kth.diva-portal.org1073157/FULLTEXT01.pdf · Synergies in...
Transcript of Synergies in Biolubrication - kth.diva-portal.org1073157/FULLTEXT01.pdf · Synergies in...
Synergies in Biolubrication AKANKSHA RAJ Doctoral Thesis 2017 KTH Royal Institute of Technology School of Chemical Science and Engineering Division of Surface and Corrosion Science SE-10044 Stockholm, Sweden
Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan i Stockholm framlägges till offentlig granskning för avläggande av tekniska doktorsexamen fredagen den 17 March kl. 10:00 i Kollegiesalen, KTH, Brinellvägen 8, Stockholm. Avhandlingen försvaras på engelska Synergies in Biolubrication
Akanksha Raj ([email protected])
Doctoral Thesis
KTH Royal Institute of Technology
School of Chemical Science and Engineering
Surface and Corrosion Science
Drottning Kristinas Väg 51
SE-100 44 Stockholm
Sweden
TRITA-CHE Report 2017:8 ISSN 1654-1081 ISBN 978-91-7729-268-5 Denna avhandling är skyddad enligt upphovsrättslagen. Alla rättigheter förbehålles. Copyright © 2017 Akanksha Raj. All rights reserved. No part of this thesis may be reproduced by any means without permission from the author. The following items are printed with permission: PAPER II: © 2017 Elsevier. PAPER III: © 2015 Elsevier. PAPER IV: © 2016 Royal Society of Chemistry (ACS). PAPER V: © 2016 Elsevier. Printed at Universitetsservice US-AB, Stockholm 2017
Mr. and Mrs. Raj,
this one’s for you.
i
Abstract The principal objective of this thesis was to advance the level of understanding in the field of biolubrication, finding inspiration from the human synovial joints. This was addressed specifically by investigating the synergistic association of key biolubricants and the resulting lubrication performance. A range of techniques was employed during the course of this thesis work. Atomic force microscopy (AFM) was used to carry out force and friction measurements. Further information on the topography, nature, and structure of the adsorbed layers of biolubricants was obtained by using instruments such as Quartz crystal microbalance with dissipation monitoring (QCM-D), X-ray reflectivity (XRR), and AFM imaging. Key synovial fluid and cartilage components have been used as biolubricants in the investigations, namely dipalmitoylphosphatidylcholine (DPPC), hyaluronan (HA), lubricin, and cartilage oligomeric matrix protein (COMP). Focus was directed towards two lubrication couples, i.e. DPPC-hyaluronan and COMP-lubricin. DPPC-hyaluronan mixtures were probed on hydrophilic silica model surfaces whereas COMP-lubricin association structures were explored on weakly hydrophobic poly (methyl methacrylate) (PMMA) model surfaces. Both the systems included salt solution at a concentration of ≈ 150 mM. Investigations of the COMP-lubricin pair revealed that individually these components are unable to reach the desired level of lubrication. However, when they associate synergistically, COMP facilitates firm attachment of lubricin to the PMMA surface in a favourable confirmation that imparts low friction coefficient (≈ 0.06).
The interplay between DPPC and hyaluronan imparts a lubrication advantage over lone DPPC bilayers, wherein hyaluronan provides a reservoir of DPPC on the model surface and thereby imparts self-healing of the lubricating layer. The system resulted in very low friction coefficient (< 0.01). Other factors such as temperature, presence of calcium ions, molecular weight of hyaluronan, and pressure were also explored. DPPC bilayers at higher temperature (liquid disordered phase) had higher load bearing capacity due to higher flexibility (as the chains were flexible enough to patch defects and self-heal). Association between DPPC Langmuir layers and hyaluronan was enhanced in the presence of calcium ions. Further, lower molecular weight hyaluronan had a stronger tendency to bind to DPPC, and it thereby affects the packing and organisation more strongly. Finally by subjecting the model system to high pressures, it was found that DPPC-hyaluronan composite layers were more stable and robust compared to lone DPPC bilayers.
Keywords: Biolubrication, Synergies, Adsorption, Surface Force, Friction, Load Bearing Capacity, Self Healing, Phospholipids, DPPC, Hyaluronan, COMP, Lubricin, QCM-D, AFM, XRR.
ii
Sammanfattning Huvudsyftet med det här avhandlingsarbetet var att öka förståelsen för biosmörjning, vilket inspirerades av egenskaper hos våra synovialleder. Specifikt så angreps problemet genom studier av synergistisk association mellan några av nyckelkomponenterna och dess inverkan på den smörjande förmågan. Ett antal huvudtekniker användes vid studierna. Atomkraftsmikroskopi (AFM) användes för att mäta ytkrafter och friktionskrafter. Ytterligare information om topografi, egenskaper och struktur hos adsorberade skikt erhölls från teknieker som kvartskristallmikrovåg med dissipationsmätning (QCM-D), röntgenstråle reflektivitet (XRR), och AFM avbildning.
Några nyckelkomponenter hos synovialvätska och brosk användes som smörjande komponenter, specifikt dipalmitoyl fosfokolin (DPPC), hyaluronan (HA), lubricin och “cartilage oligomeric matrix protein” (COMP). Fokus riktades mot synergistiska par, dvs DPPC-hyaluronan och COMP-lubricin. DPPC-hyaluronan blandningar studerades på hydrofila modellytor av kiseldioxid, medan COMP-lubricin undersöktes på svagt hydrofoba poly(metyl metakrylat) (PMMA) modellytor. Båda systemen inkluderade saltlösningar med en koncentration av 150 mM.
Undersökningarna av COMP-lubricin paret visade att dessa två komponenter var för sig inte kunde uppnå önskad smörjningseffekt. Emellertid, när de associerade synergistiskt så att COMP möjliggjorde stark förankring av lubricin till PMMA ytan i gynnsam konformation erhölls en låg friktionskoefficient (≈ 0.06).
Växelverkan mellan DPPC och hyaluronan ger också upphov till förbättrade smörjningsegenskaper över enbart DPPC bilager, där hyaluronan möjliggör förankring av en reservoir av DPPC på modellytan och därigenom ger upphov till en självläkande förmåga hos det smörjande skiktet. Det här systemet gav en mycket låg friktionskoefficient (< 0.01).
Andra aspekter undersöktes också genom att studera inverkan av faktorer så som temperatur, närvaro av kalcium joner, hyaluronans molekylvikt och tryck. DPPC bilager vid högre temperatur (i flytande kristallin fas) hade högre lastbärande förmåga på grund av högre rörlighet, vilket möjliggjorde själv-läkning av defekter. Associationen mellan DPPC Langmuir skikt och hyaluronan förstärktes i närvaro av kalcium joner. Dessutom hade lågmolekylär hyaluronan en strakare tendens att associera med DPPC, vilket påverkade packning och skiktstrukur mer. Slutligen så visade vi att DPPC-hyaluronan kompositskikt var mer stabila än DPPC biskikt under höga tryck.
iii
List of appended papers The papers listed below have been appended to the thesis.
Paper I Molecular synergy in biolubrication: The role of cartilage oligomeric matrix protein (COMP) in surface-structuring of lubricin Akanksha Raj, Min Wang, Chao Liu, Liaquat Ali, Niclas G. Karlsson, Per M. Claesson, Andra Dėdinaitė Manuscript accepted, Journal of Colloid and Interface Science, 2017
Paper II Lubrication synergy: Mixture of hyaluronan and dipalmitoylphosphatidylcholine (DPPC) vesicles Akanksha Raj, Min Wang, Thomas Zander, D.C. Florian Wieland, Xiaoyan Liu, Junxue An, Vasil M. Garamus, Regine Willumeit-Römer, Matthew Fielden, Per M. Claesson, Andra Dėdinaitė Journal of Colloid and Interface Science 488 (2017) 225–233
Paper III The effect of temperature on supported dipalmitoylphosphatidylcholine (DPPC) bilayers: Structure and lubrication performance Min Wang, Thomas Zander, Xiaoyan Liu, Chao Liu, Akanksha Raj, D.C. Florian Wieland, Vasil M. Garamus, Regine Willumeit-Römer, Per Martin Claesson, Andra Dėdinaitė Journal of Colloid and Interface Science 445 (2015) 84–92 Paper IV Structure of DPPC–hyaluronan interfacial layers – effects of molecular weight and ion composition D. C. Florian Wieland, Patrick Degen, Thomas Zander, Soren Gayer, Akanksha Raj, Junxue An, Andra Dėdinaitė, Per Claesson, Regine Willumeit- Römer Soft Matter, 12, (2016), 729 Paper V The influence of hyaluronan on the structure of a DPPC—bilayer under high pressures Thomas Zander, D.C. Florian Wieland, Akanksha Raj, Min Wang, Benedikt Nowak, Christina Krywka, Andra Dėdinaitė, Per Martin Claesson, Vasil M. Garamus, Andreas Schreyer, Regine Willumeit-Römer Colloids and Surfaces B: Biointerfaces, 14,2 (2016) 230–238
iv
Contribution by the respondent:
Paper I. Major part of experimental work (except for the AFM measurements), planning, part of data interpretation and analysis, part of manuscript preparation.
Paper II. Major part of experimental work (except for the SAXS and DLS measurements), planning, major part of data interpretation and analysis, and major part of manuscript preparation.
Paper III. Part of experimental work, planning, part of data interpretation and analysis (except XRR data analysis), and part of manuscript preparation.
Paper IV. Part of experimental work, planning, part of data interpretation and analysis, and part of manuscript preparation. I did not participate in the XRR, BAM, and GID measurements.
Paper V. Part of experimental work, planning, and part of data interpretation. I did not participate in XRR data analysis.
Other papers not included in this thesis:
Influence of high hydrostatic pressure on solid supported DPPC bilayer with HA in the presence of divalent Ca2+ ions
Thomas Zander, D.C. Florian Wieland, Akanksha Raj, Paul Salmen, Susanne Dogan, Andra Dėdinaitė, Per Martin Claesson, Vasil M. Garamus, Andreas Schreyer, and Regine Willumeit-Römer
Manuscript
Lubricin binds cartilage proteins, cartilage oligomeric matrix protein, fibronectin and collagen II at the cartilage surface
Sarah A Flowers, Akanksha Raj, Chao Liu, Agata Zieba, Jessica Örnros, Liaqat Ali, Lena I Björkman, Thomas Eisler, Sebastian Kalamajski, Masood Kamali-Moghaddam, Andra Dėdinaitė, and Niclas Karlsson.
Manuscript
v
Summary of papers
The synergistic association of COMP and lubricin on model hydrophobic (PMMA) surfaces was studied in Paper I. These biolubricants were first studied individually and then in combination. We infer from the results that COMP and lubricin associate on the model surfaces employed. Friction measurements highlighted that the friction force was high in the case of adsorbed lubricin, and it was easy to wear away the lubricin. COMP resulted in lower friction force but was still not as low as that experienced in the joints. Interestingly, when lubricin was adsorbed on the COMP-covered PMMA surface, the resulting friction was low. COMP prevents wearing away of lubricin by ensuring firm attachment in a favourable confirmation and this results in efficient lubrication.
It has previously been shown that DPPC and hyaluronan build composite layers upon sequential injections onto a model hydrophilic (silica) surface. However under physiological conditions it is more likely that these biolubricants first self assemble in bulk and then attach to the surface. Paper II investigated the structure, adsorption and lubrication performance of these aggregates from a premixed solution of DPPC vesicles and hyaluronan. We observed that in the bulk solution, hyaluronan binds to the outer shell of the DPPC vesicles. When the mixture solution was injected on the silica surface, a DPPC bilayer was rapidly formed followed by a slow adsorption of DPPC hyaluronan aggregates. Overall, the layer formed was heterogeneous in nature, and in spite of this, it resulted in low friction force. The aggregates are easily compressed and disrupted and the layers are transformed to structures facilitating low friction and high load bearing capacity. Even though disruption of aggregate structures causes transient high friction peaks, the accumulation of DPPC (build-up provided by hyaluronan) imparts a self-healing ability and therefore the return to very low friction force. This self-healing ability is a clear synergistic advantage over lone DPPC bilayers.
Results from Paper III highlight DPPC bilayer morphology and lubrication performance at different temperatures, ranging from 25 °C to 52 °C. This was done by the XRR technique, AFM imaging, and AFM force and friction measurements. The DPPC bilayer was in the gel phase at low temperature and fluid phase at higher temperature. The structural changes associated with the phase transition were noticed at a relatively lower temperature in the case of AFM as compared to XRR. This is believed to be due to the kinetic energy that is transferred by the tapping
vi
AFM tip. The friction force generated by the DPPC bilayers was low in both the phases. However, it was observed that bilayers in the fluid phase were more stable under the combined action of load and shear, and therefore the load bearing capacity was higher. This is because increased flexibility of the chains and higher surface mobility at higher temperature allows for repair of defects and therefore provides a self-healing ability.
Paper IV discusses (i) the effect of molecular weight of hyaluronan on DPPC Langmuir layers and (ii) the influence of calcium ions on the association structures. The data suggest that hyaluronan binds to the DPPC Langmuir layers and this interaction is influenced by the molecular weight of hyaluronan. Strong interaction is observed in the case of lower molecular weight hyaluronan, and the packing and organisation of the structures are affected most in this case. The association between DPPC and hyaluronan was further enhanced in the presence of calcium ions.
The influence of hyaluronan on DPPC bilayers subjected to high pressures was investigated in Paper V. The structure of the hyaluronan layer that adsorbed to solid-supported DPPC bilayers was characterised by the XRR technique. It was found that hyaluronan mostly adsorbs to the DPPC headgroup. Thereafter the response of the system to high pressures was evaluated. Data reveal that at high pressure, addition of hyaluronan to DPPC bilayers results in a composite layer, which is more stable than lone DPPC bilayers. Phase transitions associated with changes in pressure were also observed and it was found that the pressure-induced phase transition in DPPC-hyaluronan composite layers was fully reversible. This is in contrast to lone DPPC bilayers, wherein the pressure-induced phase transitions were not completely reversible. This is of importance in the case of joint lubrication as only unimpaired layers ensure low friction.
Table of contents
Abstract i
Sammanfattning ii
List of appended papers iii
Summary of papers v
List of abbreviations
1. Introduction 1
1.1 The synovial joint 1
1.2 Lubrication synergies 4
1.3 Model systems 5
1.3.1 COMP-lubricin model system 5
1.3.2 Phospholipid-hyaluronan model system 8
1.4 Surface interaction and friction forces 11
1.4.1 Van der Waals force and electrical double layer force (DLVO-theory)
11
1.4.2 Attractive forces 14
1.4.3 Repulsive forces 15
1.4.4 Friction force 16
1.4.5 Hydration lubrication 18
2. Materials 20
3. Instruments 23
3.1 Atomic force microscopy (AFM) 23
3.1.1 Force measurements 24
3.1.2 Friction measurements 26
3.2 Quartz crystal microbalance with dissipation monitoring (QCM-D)
27
3.3 Table of other instruments 29
4. Key results and discussion 30
4.1 COMP-lubricin synergy system 30
4.1.1 COMP on PMMA 30
4.1.2 Lubricin on PMMA 32
4.1.3 COMP-lubricin synergy on model surfaces 32
4.2 Phospholipid-hyaluronan synergy system 34
4.2.1 DPPC bilayer morphology at different temperatures
34
4.2.2 Lubrication performance of DPPC bilayers at different temperatures
36
4.2.3 Lubrication synergy: Mixture of hyaluronan and DPPC vesicles
39
4.2.3.1 DPPC and hyaluronan association structures in bulk
40
4.2.3.2 Adsorption process of DPPC-hyaluronan association structures
41
4.2.3.3 Morphology of adsorbed layer from DPPC- hyaluronan mixture
43
4.2.3.4 Surface force measurements 44
4.2.3.5 Lubrication performance of DPPC-hyaluronan association structures
46
4.3 Effect of molecular weight and ions on DPPC-hyaluronan interactions
48
4.3.1 Effect of molecular weight of hyaluronan 49
4.3.2 Effect of Ca2+ ions 53
4.4 Effect of high pressures on DPPC bilayers and DPPC-hyaluronan structures
54
5. Conclusions and impact 57
6. Acknowledgements 60
7. References 62
List of abbreviations
1. DPPC Dipalmitoylphosphatidylcholine
2. COMP Cartilage oligomeric matrix protein
3. HA Hyaluronan
4. PBS Phosphate buffered saline
5. PMMA Poly (methyl methacrylate)
6. AFM Atomic force microscopy
7. QCM-D Quartz crystal microbalance with
dissipation monitoring
8. XRR X-ray reflectivity
9. BAM Brewster angle microscopy
10. DLS Dynamic light scattering
11. SAXS Small-angle X-ray scattering
12. CMC Critical micelle concentration
13. Mw Weight average molecular weight
Chapter 1
1
1. Introduction
The mammalian synovial joints are remarkable constructs of nature that
allow free movement under varying pressures with friction coefficients in
the range of 0.001–0.01 [1]. These are elegantly designed to last for, in
best case, over a century. This is illustrated by the likes of athletes
Harriette Thompson, Fauja Singh, Stanislow Kowalski running well past
their ninth to tenth decades. The optimal performance of the synovial
joints relies on an effective lubrication system. Failure of the lubrication
system leads to acute pain, disabilities, reduced quality of life, and is an
economic burden on the healthcare system. For example, according to the
World Health Organisation, ten years after the onset of rheumatoid
arthritis, at least 50% of patients in the developed countries are unable to
retain a full-time job [2]. Considered estimates by the European League
against Rheumatism suggest that rheumatic and musculoskeletal diseases
cost upwards of € 200 billion of public spending in Europe alone [3]. All
this along with the fact that we are veering towards an ageing population,
adds impetus to study and better understand biolubrication and
biolubricants, which work together to achieve healthy functioning and
admirable nanotribological performance, specifically in the context of the
human synovial joints. In this, nature also provides us with a template
that continues to inspire and which may be extended to other cutting-
edge technologies and environmentally friendly lubricants.
1.1 The synovial joint
The excellent lubrication of the synovial joints is attributed to the
cartilage that lines the synovial joints, the surrounding synovial fluid, key
biolubricants, and associating self-assembly structures.
The cartilage is a dynamic tissue that bestows the joint surfaces with low
friction, high load bearing capacity, shock absorption, and wear
Chapter 1
2
resistance. In spite of being ≈ 2 mm in thickness [4] and regularly
subjected to high loads, it has lasting durability.
The cartilage is avascular and aneural. It has very low stiffness with a
Young’s modulus in the range of a few MPa [5]. The cartilage comprises
both cellular and extracellular components. The cellular components are
called chondrocytes. They are present at 2 vol% in the cartilage and are
responsible for producing, organising, and maintaining [6] the other 98%
of the extracellular material. The extracellular matrix of the cartilage
primarily consists of water (≈ 80%), collagen (50–75% dry weight)
proteoglycans (15–30% dry weight) and lipids (10% dry weight) [7].
Collagen fibres impart tensile strength to the cartilage and enclose
proteoglycans that confer elasticity and load distribution.
The structure of the cartilage is best described (refer figure 1) by four
zones; superficial, transitional, radial, and calcified. In the superficial
zone the collagen fibres are oriented parallel to the surface. The
outermost part of the superficial zone consists of lamina splendens,
which face towards the synovial fluid (200-500 nm thick, devoid of
collagen fibres) [8] and can be easily peeled off from the rest of the
cartilage [9]. Chondrocytes are of an elongated shape and lie parallel to
the surface. The proteoglycan content here is at its lowest, whereas the
water and collagen content is at its highest. In the transitional zone the
collagen fibres are hemispherical, ill defined in shape, and chondrocytes
are more round. The thickness of the collagen fibres increases towards
the bone and in the radial zone is the thickest and perpendicularly
oriented. Chondrocytes appear spherical and align themselves
perpendicularly in a columnar pattern. The calcified zone is separated
from the radial zone by a tidemark. This transitional area helps anchor
cartilage to the bone.
Chapter 1
3
Figure 1: Schematic of the knee joint (above) and cartilage (below)
Synovial fluid
Bone
Synovial membrane
Chapter 1
4
Given that the cartilage is avascular, the neighbouring synovial fluid is
responsible for providing nourishment with nutrients and removing
waste products. The synovial fluid (ionic strength ≈ 150 mM [10], pH 7.8
[11]) surrounds the joints and provides lubrication by virtue of reduced
friction and wear. It is composed of key biolubricants such as hyaluronan,
lubricin, and phospholipids. The cartilage surface forms a boundary layer
with these key molecules, which when depleted, may be replenished via
the synovial fluid and diffusion from the cartilage surface. Several modes
of lubrication are considered to play a role in cartilage lubrication but
boundary lubrication has been the focus of this study. It relies on
characteristic molecular and biochemical properties of the lubricant layer
in preventing solid-solid contact and consequently wear of the cartilage
surface.
1.2 Lubrication synergies
The ability of joint surfaces to self-repair is limited. These surfaces
respond to various factors including load and shear. While they are
constructed of molecules, which themselves are soft and compliant, the
pressures borne by them can be continual, abrupt, and high. It is hard to
believe that a single component, and one which works individually, is
capable of performing all the necessary functions and utilising all
necessary features to impart the required level of lubricity to the joints for
extended periods of time. Rather, we believe that meaningful insights can
be gained for better understanding, by exploring the association of these
components and the resulting synergism in lubrication. Therefore, we
have pursued the pathway of investigating possible synergies between the
key synovial fluid components. In this work we have focused on two
lubrication couples (i) Cartilage Oligomeric Matrix Protein (COMP) and
lubricin (ii) Dipalmitoylphosphatidylcholine (DPPC) and hyaluronan.
Nonetheless we would like to impress on the fact that the natural
Chapter 1
5
biolubrication system is far more complex, and there are believed to be
other biolubricants, mechanisms, and factors, which come into play but
are beyond the subject of this work.
For the COMP-lubricin system, hydrophobic PMMA model surfaces were
used, whereas for the DPPC-hyaluronan system hydrophilic silica model
surfaces were used. The fact that the respective surfaces allow strong
adsorption of the corresponding molecules, reasons their selection.
The overall objective of this work was to advance understanding in the
field of biolubrication pertaining to human synovial joints. As mentioned,
this has specifically been addressed by studying the synergy between key
synovial components and its effect on the lubricity of model surfaces.
This has been elucidated and detailed in papers I and II. Further, other
aspects of joints were tapped into by investigating the effect of
temperature, high pressure, calcium ions, and molecular weight of a key
component on the model system. This has been described in papers III,
IV, and V.
1.3 Model Systems
1.3.1 COMP-lubricin model system
As mentioned in the previous section, lubricin is postulated to be a key
biolubricant for the knee joint [12-14]. It is known to impart very good
lubricity at the cartilage surface, however, to be able to do so it is equally
important for the boundary lubricant to prevent wear (which it has been
able to achieve modestly on model surfaces [12, 15]). The boundary
lubrication conditions experienced by the cartilage surface under the
combined action of load and shear, require the surface lubricin layer to be
capable of strongly adhering to the cartilage surface. COMP and lubricin
are known to associate in bulk and in one of the recent works of my
Chapter 1
6
collaborators [16], both were found to co-localise at the cartilage surface.
In light of these novel findings, I wished to investigate the interaction of
this lubrication couple.
COMP
COMP has piqued the interest of many researchers, especially in the
recent past. It plays a role in maintaining the integrity of the knee joint
[17]. Mutations in the COMP gene induce skeletal disorders such as
Pseudoachondroplasia (PSACH) and Multiple Epiphyseal Dysplasia
(MED) [18]. Even though the concentration of COMP in the cartilage is
relatively low [19], the crucial functions it has and it’s ability to form
synergistic structures, can not in the least be ignored.
Figure 2: Schematic of a native COMP molecule
COMP is a non-collagenous, extra cellular matrix protein found in the
cartilage, ligaments, tendons, synovium, meniscus, and osteoblasts [20].
It is a 524 kDa pentameric glycoprotein having multiple domains, and is
Chapter 1
7
stabilised by disulphide bonds [21] as shown in figure 2. The chains of
this bouquet-like protein are connected at its amino terminus by a coiled-
coil domain. Further, it comprises four small globular epidermal growth
factor (EGF) like domains, type III (calcium binding) domain, and is
terminated by a large carboxyl globular domain [21].
Fun facts: COMP mutation associated with disorders such as dwarfism
and early onset osteoarthritis, mainly occurs in the type III domain of
COMP molecule [18].
Lubricin
As the name implies, lubricin is a mucinous glycoprotein, which
contributes to the lubricating ability of the cartilage. The absence and
mutation of the lubricin gene have been linked to decreased joint
lubrication, increased cartilage wear [22, 23], and the rare
camptodactyly-arthropathy-coxa vara-pericarditis (CACP) syndrome
[23]. Lubricin is found in the superficial zone of the articular cartilage,
synovial membrane, and synovial fluid [24]. The concentration of lubricin
is 0.052–0.450 mg/mL in healthy synovial fluid [25]. In humans, the
proteoglycan 4 (PRG4) gene encodes it.
Figure 3: Schematic of a lubricin molecule
The average length of lubricin first observed by Swann et al. was 222 nm
and a diameter of 1-2 nm. It has a molecular weight of 227.5 kDa [26].
Lubricin is primarily a flexible rod [26] constituting a large mucin-like
domain. This central domain has anionic side chains that participate in
O-linked glycosylation (nearly 50% w/w) with NeuAc-Gal-GalNAc. Two-
Chapter 1
8
thirds of these sugar groups are capped by charged sialic acid [12, 13].
This region accumulates most of the negative charge of the protein, and
due to heavy glycosylation renders it hydrophilic. Two non-glycosylated
end domains lie adjacent to the central domain. These subdomains at the
left and right flanks are similar to two globular proteins; Somatomedin-B
(SMB) at the N terminus, and Hemopexin (PEX) at the right terminus.
Positive charge is mainly located in these domains. Due to the presence of
cysteine residues in the end domains lubricin can form intra and
intermolecular disulphide bonds, thereby enabling it to form
supramolecular structures [22].
Fun facts: When 54% or less of β (Gal-GalNAc) was removed from
human lubricin, the lubricating ability was reduced by almost 80% [13].
1.3.2 Phospholipid-hyaluronan model system
Hyaluronan has long been implied to play a major role in the lubrication
of the synovial joints. However, it has been shown that hyaluronan alone
can not provide low friction and therefore, is a rather poor boundary
lubricant [27]. Phospholipids so far are the only key biolubricant to have
provided both low friction and high load bearing capacity on artificial
surfaces [28]. It has been concluded after assessment of the lubricating
abilities of DPPC and hyaluronan on an osteoarthritis (OA) damaged
cartilage, that greatest improvements are seen when both are
administered together [29]. Both DPPC and hyaluronan are in plenty in
the knee joint, and are known to form association structures [30, 31].
Given these reasons, we link the appreciable lubrication performance of
this model system to the interplay between these components.
Chapter 1
9
DPPC
The total amount of phospholipids in human synovial fluid ranges from
0.1–0.2 mg/mL [32]. Three major classes of phospholipids have been
identified from bovine articular cartilage surfaces. These were
phosphatidylcholine (41%), phosphatidylethanolamine (27%), and
sphingomyelin (32%). Modern lipodimic analysis also showed that
phosphatidylcholine was the predominant phospholipid [33]. Among the
phosphatidylcholines, unsaturated acyl chains were predominant. At 11%
and 15%, DPPC was the most abundant among the saturated lipids
obtained from analysis of the respective polymeric and metallic
component of retrieved implants [34]. As DPPC is saturated and not
prone to degradation and oxidation, it is easier to handle. These reasons,
along with the fact that it has been well-characterised in the literature,
has helped me choose DPPC to conduct various investigations for the
better part of my studies.
Figure 4: Sketch of a (C 16:0) DPPC molecule
DPPC consists of a hydrophilic zwitterionic headgroup (comprising a
positively charged choline group and a negatively charged phosphate
group) linked by a glycerol group to two saturated hydrocarbon chains as
shown in figure 4. It has limited solubility in water with a CMC of 0.46
nM [35]. DPPC molecules tend to assemble spontaneously and form
bilayer structures in aqueous solutions and on surfaces, where the
hydrophilic head group interacts freely with water and hydrocarbon
chains are facing inwards, away from water. In the course of this work
DPPC bilayers were formed on silica surfaces from DPPC vesicles
Chapter 1
10
solution via spontaneous vesicle rupture. Details on preparation of the
vesicle solution can be found in the papers appended to the thesis. Above
≈ 41 °C the bilayers are in the liquid disordered phase where the chains
are fluid-like, and below this temperature they are in the gel phase where
the chains are solid-like. In the subsequent sections of this thesis we shall
visit this aspect further and discuss its relevance to our model system.
Hyaluronan
Out of all the components that I have considered for my doctoral studies,
hyaluronan is the most abundantly available in the synovial joints. In
healthy knee joints, it is present both at the cartilage surface and in the
synovial fluid at a concentration of 1.45–3.12 mg/mL [36]. Hyaluronan
was used across a range of molecular weights in my work; 10 kDa to 1500
kDa. As shown in figure 5 hyaluronan is a linear polysaccharide
consisting of repeating units of D-glucuronic acid and N-acetyl-
glucosamine. The pKa of D-glucuronic acid is about 3.3 [37] and
dissociation of the acid makes hyaluronan negatively charged in
physiological solution. Hyaluronan has an intrinsic persistence length of
about 9 nm [38]. It has a unique viscoelastic nature; wherein it can form
a viscoelastic gel at higher concentrations, which is broken down under
high shear forces and then reversibly restored as shear is reduced [39].
Figure 5: Molecular structure of hyaluronan
Chapter 1
11
Hyaluronan is one of the most highly hydrated molecules in nature [40]
and capable of retaining water. However it is important to note that in its
secondary structure, hyaluronan has a hydrophobic patch [41]. This
property allows hyaluronan to associate with itself and other molecules
that are of relevance in my model system, such as DPPC. Thus,
hyaluronan is highly hydrophilic with hydrophobic patches, making it
amphiphilic in nature.
Fun facts: The consistency, shear thinning, biocompatibility and water
retaining ability (up to 1000 times its own weight) of hyaluronan make it
a popular choice for skin care products like moisturisers [40] and for
achieving anti-wrinkle effects in injectable fillers [42].
1.4 Surface interaction and friction forces
1.4.1 Van der Waals force and electrical double layer force (DLVO-theory)
In order to analyse our model system, it is important to interpret the
surface forces that come into play when two opposing surfaces interact.
This sheds light into the behaviour of the system and can complement
dynamic interactions such as those that have been studied by conducting
friction measurements. The following section describes friction,
lubrication, and (at the outset) the main characteristics of the surface
forces that are of primary importance to the model systems investigated
in this work. These surface forces, depending on their nature, have been
classified as attractive or repulsive.
Van der Waals force
The van der Waals force originates due to interaction between fluctuating
electromagnetic waves extending from the surface of a material. It is a
Chapter 1
12
result of (i) Orientation force (ii) Induction force and (iii) London force,
with London force being the dominant one as it acts between all
molecules and atoms [43].
The van der Waals force is attractive between similar surfaces and may be
repulsive between dissimilar ones. It is always present and relatively
short ranged in nature. It is not simple to calculate this force, but it can
be best described by the Lifshitz theory [44]. This theory ignores the
molecular nature of the interacting bodies and the intervening medium;
rather it employs a continuum approach.
The van der Waals interaction of two media across a third intervening
medium, can be calculated depending on the geometry being considered.
In my work, we consider the case of a sphere against a flat surface. By
convention, negative sign denotes attraction and a positive sign denotes
repulsion. The van der Waals force in this case is given as follows [45]:
𝐹 (𝑑)𝑣𝑑𝑊 = − 𝐴𝐻𝑅6𝐷2
(1)
Where 𝑅 is the radius of the sphere, 𝐷 is the distance between the surface
and the sphere, 𝐴𝐻 is the Hamaker constant and can be calculated by
[45]:
𝐴𝐻 = 34
𝑘𝐵𝑇 (𝜀1 − 𝜀3𝜀1 + 𝜀3
) (𝜀2 − 𝜀3𝜀2 + 𝜀3
) +
3ℎ𝑣𝑒8√2
(𝑛12 − 𝑛32)(𝑛22 − 𝑛32)√(𝑛12 + 𝑛32)√(𝑛22 + 𝑛32)[√(𝑛12 + 𝑛32)+ √(𝑛22 + 𝑛32)]
(2)
Where 𝜀𝑖 is the static dielectric constant of medium 𝑖, 𝑛𝑖 is the refractive
index of medium 𝑖, ℎ is Planck’s constant, 𝑣𝑒 is the dominant electronic
absorption frequency, 𝑘𝐵 is Boltzmann’s constant, and 𝑇 is the absolute
temperature.
Chapter 1
13
Electrical double layer force
Most surfaces in water obtain a surface charge either by surface
dissociation or adsorption of charged species. This surface charge is
countered by oppositely charged ions distributed outside the surface.
Together the surface charge and these ions form the electric double layer.
When another such charged surface is brought close together, the
concentration of charged ions in the gap between the surfaces increases.
This generates an osmotic repulsion and thus a repulsive force, which at
large separations decays exponentially. The interaction free energy at
large separations is given by:
𝑊 (𝑑)𝐸𝐷𝐿 = 𝐶𝑒−𝜅𝐷 (3)
Here, 𝐷 is the distance between the two surfaces, and 𝐶 is a constant that
depends on surface charge density and solution conditions. The range of
the interaction is characterised by the Debye length, 𝜅−1, and depends on
the salt concentration. The Debye length decreases with an increase in
salt concentration. In water at 25 °C, for a monovalent electrolyte the
Debye length is [43]:
𝜅−1 = 0.304
√𝑐
(4)
Here, 𝑐 is the salt concentration and given in molar, and the Debye length
in nm. In my experiments, a salt concentration of 155 mM was used.
Plugging this in the equation above results in a Debye length of 0.77 nm.
This means that the double layer force is of short range and consequently
contribution of other forces (mentioned subsequently) would be more in
my studies.
Chapter 1
14
I note that in my studies similar surfaces were always considered, and in
this case the double layer force is repulsive. However, for surfaces with
opposite charge the double layer force is attractive.
The van der Waals force and the electrical double layer force together are
considered in the DLVO theory. The interplay of these forces describes
the net interaction between two surfaces. However, in a system several
other forces may contribute significantly. Some of these non-DLVO forces
include hydration, protrusion, and hydrophobic forces and these will now
be discussed.
1.4.2 Attractive forces
Hydrophobic attraction
Hydrophobic interactions are prevalent and of importance in nature. For
example they are of relevance to cell membranes, protein folding, and
DNA double helix stabilisation. When water molecules come in contact
with a hydrophobic molecule, the water molecules undergo
rearrangement and restructuring to maintain hydrogen bonding [46].
However, such a reorientation and restructuring of water molecules is
entropically unfavourable. Thus, in order to minimise this disturbance
hydrophobic molecules tend to cluster, and the resultant force between
them is known as the hydrophobic interaction [47]. In water, as a result
of the hydrophobic interactions, hydrophobic surfaces attract each other
as the coming together of such surfaces is entropically favourable.
In my work hydrophobic interaction explained self-assembly structures
formed by DPPC bilayers. It also possibly causes the association of DPPC
and hyaluronan, wherein the hydrophobic patches of hyaluronan
associate with the hydrophobic chain of the DPPC molecules. Another
example would be the adsorption of COMP and lubricin driven by
Chapter 1
15
hydrophobic interaction between these molecules and the PMMA model
surface.
Bridging attraction
Bridging forces can influence interaction of two surfaces coated with a
polymer. When opposing surfaces are close enough and have vacant
binding sites, the polymer chains may bridge both the surfaces. This
makes it possible for the polymer chain to adopt many favourable
conformations, which leads to an increase in entropy. Upon separation of
these surfaces, detachment is resisted and attractive bridging interaction
persists. In my work I have encountered such forces in the case of
adsorption of lubricin on the PMMA surface. Interestingly, this
phenomenon is witnessed in other biological systems as well. Examples
include polymer bridges that were found to connect myelin membranes
and prevented them from escaping by keeping them at an appropriate
distance [48], and surface bound polymers containing ligand that exposes
the binding sites to bring it to the receptor [49].
1.4.3 Repulsive forces
While the important attractive forces that are most relevant to my work
have been briefly discussed above, one can imagine what would follow if
these alone were to exist, especially in the physiological context.
Repulsive forces help to keep a system stable and safely separated.
Protrusion force
As two opposing amphiphilic surfaces approach each other, protrusion
forces become important. Thermal motion of the molecules drives this
force. The protruding segments are forced into the surface and their
motion is perpendicular to the surface in an in and out manner [50].
When two surfaces are close enough, motion of protruding segments is
Chapter 1
16
restricted, which leads to a decrease in entropy and thus this short-range
steric force is repulsive in nature. Protrusion forces have been
encountered in my work in relation to DPPC phospholipid bilayers and
that too at different temperatures.
Hydration force
Water molecules are known to bind strongly to surfaces with hydrophilic
groups. When opposing surfaces approach each other, it costs energy to
dehydrate the surfaces, and thus short-range repulsive hydration forces
arise. The strength of this hydration force depends on the energy required
to dehydrate the surface. In my studies, these forces were seen at < 5 nm
(between 1-3 nm) in the case of both silica and DPPC-coated surfaces.
Steric force
When two surfaces covered with polymers/polyelectrolytes approach
each other and their segments begin to overlap, a repulsive force is
generated due to the unfavourable entropy associated with the confining
of the chains. Further, this force depends on factors such as coverage of
polymer on the surface, type of adsorption of polymer to the surface, and
quality of the solvent [50]. This force has often been seen in my studies.
Examples include the steric repulsive force resulting from the adsorption
of COMP on the PMMA surface, and that of premixed DPPC-hyaluronan
on the silica surface.
1.4.4 Friction force
Friction is the resistance to relative motion of surfaces sliding against
each other. Depending on the requirement of a system, friction may be
regulated. In gripping things or applying brakes of a vehicle, friction is
desirable, whereas in knee joints, implants or contact lenses, it needs to
Chapter 1
17
be small. Friction is often described by Amonton’s first rule, which is
mathematically defined as:
𝐹𝑓 = 𝜇𝐹𝑛 (5)
Here 𝐹𝑓 is the friction force, 𝐹𝑛 the applied load, and 𝜇 is the friction
coefficient. This is an empirical rule, wherein the friction coefficient
depends on the system, and the friction force does not depend on the
macroscopic area. In some of the systems that I investigated Amonton’s
first rule applied, whereas in others it did not. In cases when it did not
apply, it was due to new energy dissipative mechanisms (e.g. adhesion)
coming in to play. Such attractive interactions then modify the rule to:
𝐹𝑓 = 𝜇𝐹𝑛 + 𝐶 (6)
Where 𝐶 is the friction force at zero load. Furthermore, when both these
rules do not hold, it is appropriate to define an effective friction
coefficient as:
𝜇𝑒𝑓𝑓 = 𝐹𝑓
𝐹𝑛
(7)
Load bearing capacity is a recurrent term in this thesis. We define it as
the load at which the lubricating layer starts to wear off. This is seen by a
superlinear increase in friction force with load. For polymer-coated
surfaces several energy dissipation mechanisms can contribute to friction
[51]. These include formation and breakage of attractive bonds between
segments on opposing surfaces, formation and breakage of polymer-
surface bonds, and polymer sliding past polymer in the interpenetration
zone. The energy dissipated, 𝑊, due to sliding a distance 𝑙 is then given
by:
Chapter 1
18
𝑊 = 𝐹𝑓𝑙 (8)
Friction measurements have been conducted in my work using the AFM
technique, which will be discussed further in section 3.1.2.
Biolubricants and biolubrication helps in maintaining low friction,
characteristic to our biological model system. This low friction was
attained by virtue of various different mechanisms and phenomena. In
my studies lubricin provided low friction by (i) being well anchored to
COMP-coated PMMA surface and thus avoiding breakage and formation
of segment-surface attachment points (ii) having side chains that reduced
interpenetration and (iii) having side chains that interact favourably with
water and thus avoid segment-segment attraction. Phospholipid bilayers
have demonstrated very low friction upon shearing. The reduction in
friction is attributed largely to hydration lubrication and this mechanism
of significance will be discussed below.
1.4.5 Hydration lubrication
Hydration lubrication has been of critical importance in my studies.
Repulsive hydration/protrusion forces are able to withstand high loads
and large pressures. At the same time, the fluid-like manner in which the
surrounding fluid behaves when the opposing surfaces are sheared,
allows for the fluid to be easily sheared and thereby accounts for the low
friction and excellent lubrication. This is particularly relevant in
biological systems as in ours, where the natural medium is water. High
hydration in case of phospholipid bilayers and lubricin in my model
systems, are able to bear high normal loads and easily shear the
surrounding water layers. This may reason for the very low friction
witnessed in both the systems. The plane of shear between two opposing
bilayers surrounded by water is depicted below. In case there is a water
layer present between the phospholipid headgroup and substrate, an
Chapter 1
19
additional shear plane will exist there. This is a possibility even though
we could not find evidence for water between the silica surface and the
DPPC bilayer in our X-ray reflectivity data.
Figure 6: Interpretation of opposing phospholipid bilayers indicating position of shear plane
Chapter 2
20
2. Materials The following materials were used for the investigations carried out in my thesis work:
Material Purchased/Received Comments
DPPC Avanti Polar Lipids, Inc.,
Alabama, USA
Powder form
Hyaluronan Mw 10 kDa, 250 kDa, 750
kDa, 1500 kDa, 2500 kDa
Creative PEG works, USA
Mw 620 kDa kindly gifted by
Novozymes, Denmark
Powder form
Recombinant
COMP
R&D Systems, Inc., USA COMP used in this work
was in monomeric form
(Mw 81.8 kDa), while
native COMP is in
pentameric form.
Lubricin and
reduced
lubricin
Kind gift from Dr. Niclas
Karlsson, Gothenburg
University, Sweden
Collected and isolated
from synovial fluid from a
pool of patients diagnosed
with rheumatoid arthritis
and spondyloarthritis (for
reduced lubricin).
Dialysed and lyophilised
protein powder was
provided and used.
Chapter 2
21
Sodium
chloride
Sigma Aldrich ACS reagent, assay > 99%
Phosphate
Buffered
Saline
Sigma Aldrich Sigma P4417
Calcium
chloride
dihydrate
Sigma Aldrich ACS reagent, assay > 99%
Water Purified using Millipore
system comprising Milli-Q
system with 0.22 μm
Millipak filter.
Resistivity of purified
water was 18.2 MΩcm at
25 °C and total organic
carbon content < 3 ppb.
QCM crystals AT cut silica crystals,
AT cut PMMA crystals
Q-sense, Gothenburg,
Sweden
QSX 303
QSX 999
Silica wafer Wafer net, Inc., USA In paper 3, wafer surfaces
were roughened to
amplify difference in
friction measurement
between bare and coated
surfaces
Cantilevers Mikromasch, Germany Rectangular tipless
cantilever for force and
friction measurements,
Chapter 2
22
silicon nitride tip for AFM
imaging
Probes Silica colloidal probes, Bang
Laboratories, Fishers, USA
PMMA colloidal probes,
Kisker, Germany
All chemicals were used without further purification while colloidal
probes and surfaces were cleaned prior to use as specified in each
publication.
Chapter 3
23
3. Instruments
3.1 Atomic force microscopy (AFM)
AFM is a type of scanning probe microscopy technique, which provides
high resolution imaging of the topography of a sample surface. Sample
surface interactions can also be measured using an AFM. I have used the
AFM to measure both normal and friction forces. The key elements of an
AFM include a sample surface, laser beam probe, cantilever spring
(rectangular), piezo scanner, and a photo diode. The working of an AFM
for normal force and friction measurements in my project is such that the
probe is glued at the end of the cantilever and scans the sample surface,
which is mounted on a scanner (as shown in figure 7). The scanner allows
for movement in the x-y-z direction. A laser beam is focused on the
backside of the cantilever and reflected off to the detector system, i.e. the
photodiode. The photodiode is split into four quadrants and detects
changes associated with the deflection of the cantilever. Vertical shifts are
related to force measurements and horizontal shifts are associated with
friction measurements. In this work for normal force and friction
measurements, the colloidal probe microscopy technique was employed
[52]. A spherical silica or PMMA particle (size range 5 to 20 μm) was
glued to the end of the cantilever using a micromanipulator connected to
an optical microscope. This imparts advantages over a sharp tip
including, a well-defined geometry, Derjaguin’s approximation being
valid, larger forces between probe and surface and therefore more
accurate measurement.
Chapter 3
24
The AFM can be used in different modes according to the information
required. Peak force tapping mode has been used in this work to perform
imaging. In this mode, the sharp tip and sample are in contact
intermittently and due to the short period of contact, lateral forces are
reduced. There is a feedback loop that keeps the maximum force
imparted on the tip constant, thereby protecting both the tip and sample
from damage. This is beneficial for soft samples such as ours, which are
easy to damage.
Figure 7: A schematic illustration of the main components of an AFM. The colloidal probe is
used for normal force and friction measurements, while a sharp tip is used for imaging.
3.1.1 Force measurements
Surface force interactions were investigated with the help of AFM
colloidal probe force measurements. These measurements are made by
ramping the probe (normal direction) with respect to a surface and
monitoring the cantilever deflection as a function of piezo expansion. The
probe and surface approach each other from a large distance where no
sample surface
probe
piezo scanner
laser
photodiode
Chapter 3
25
force acts on the probe. At shorter distances, attractive and repulsive
forces are experienced depending on the type of interactions. Once the
probe and surface come into physical contact, the probe deflects the same
amount as the surface moves. This corresponding linear region of the
force-displacement curve gives us the constant compliance region. The
surfaces are then snapped apart as they are retracted. In actuality it is the
sample that is moved. The force curve consists of an approach and
retraction cycle. The deflection of the cantilever as a function of piezo
expansion is converted to force as a function of probe-sample separation.
This is done by the AFM Force IT software (Force IT, Sweden). In theory:
(i) The cantilever deflection measured in volts 𝑍 (𝑉) is converted to
metres 𝑍 (𝑚) by multiplying it with deflection sensitivity 𝛿 (inverse of
slope corresponding to constant compliance region) as [53]:
𝑍 (𝑚) = 𝑍 (𝑉) × 𝛿 (𝑚𝑉 ) (9)
(ii) This deflection in metres is then converted to force (𝑁) by Hooke’s
law:
𝐹𝑛 = 𝑘𝑛 𝑍 (𝑚) (10)
Where 𝑘𝑛 is the normal spring constant that is determined from a
separate calibration step. The normal spring constant was calibrated
using the Sader method [54]. In order to make comparisons of two
different experiments, the force 𝐹 is normalised by radius of the probe 𝑅
with the help of Derjaguin’s approximation using [55]:
𝐹(𝐷)𝑅 = 2𝜋𝑊(𝐷)
(11)
Where 𝑊 (𝐷) is the interaction free energy per unit area between two
surfaces separated by a distance 𝐷.
Chapter 3
26
(iii) The separation of surfaces is then obtained from the combined
movement of the piezo and cantilever. It should be noted though that the
constant compliance provides only a relative zero of separation especially
in the case of soft samples where physical contact occurs before constant
compliance is reached. Thus to minimise error, the deflection sensitivity
used is measured between hard substrates before the adsorption of soft
layers.
3.1.2 Friction measurements
While normal force measurements are done by the vertical movement of
the probe with respect to the sample, friction measurements are done by
sliding the probe laterally with respect to the sample surface. During a
friction measurement a certain load is applied and the probe and sample
are brought into contact. The generation of friction between the probe
and surface during the lateral movement, results in the twisting of the
cantilever. This twisting of the cantilever (both back and forth) is
recorded by the lateral deflection of the signal. The lateral deflection is
converted to friction force using the following equation:
𝐹𝑓 = ∆𝑉𝑙𝑎𝑡𝑘𝑡
2ℎ𝑒𝑓𝑓𝛿 (12)
Where 𝑘𝑡 is the torsional spring constant (determined by the hybrid
method [56]), ℎ𝑒𝑓𝑓 is the effective height of the probe (diameter of probe
plus half the thickness of the cantilever), 𝛿 is the torsional detector
sensitivity, and ∆𝑉𝑙𝑎𝑡 is the lateral output voltage.
The data was processed using the Friction (IT) software (Force IT,
Sweden) mentioned earlier. The friction force can then be measured at
different loads, and for a predetermined number of loops at the same
load (typically ten in my work), which can then be averaged. The average
Chapter 3
27
is then plotted as a function of the applied normal load. When converting
the applied load to pressure, the following formula is used:
𝑃 = 𝐹𝑛
𝐴 (13)
Here 𝐹𝑛 is the applied load and 𝐴 is the contact area that can be calculated
by using the JKR or Hertz theory [57]; whichever is more appropriate to
the given system. The JKR method is preferred for soft and adhesive
surfaces, and the Hertz method for hard and non-adhesive surfaces.
3.2 Quartz crystal microbalance with dissipation (QCM-D)
The QCM-D is able to detect both frequency and dissipation changes due
to e.g. adsorption and is therefore a good tool to characterise samples of
adsorbed layers. The QCM-D consists of a quartz crystal sensor
sandwiched between electrodes. An AC voltage is applied to the crystal,
which causes it to oscillate at its resonance frequency. In case of
adsorption of material on the quartz surface, there will be an associated
change in mass. This change in mass is in the simplest case directly
related to the accompanied change in frequency of the oscillating crystal.
It should be noted that the mass we refer to, is actually the sensed mass,
which includes the mass of the absorbed material and the coupled water.
When the driving voltage is shut off, energy from the oscillating crystal is
dissipated from the system. Dissipation is measured by monitoring the
amplitude decay profile of the oscillator. It is defined as [58]:
𝐷 = 𝐸𝑙𝑜𝑠𝑡
2𝜋𝐸𝑠𝑡𝑜𝑟𝑒𝑑
(14)
𝐸𝑙𝑜𝑠𝑡 is the energy dissipated per oscillation, and 𝐸𝑠𝑡𝑜𝑟𝑒𝑑 is the total energy
stored in the system.
Chapter 3
28
The change in frequency and dissipation were measured using the
programme Q-tools (from Q sense, Gothenburg). These were then used to
calculate changes in mass. There are a few models that can be adopted to
do so. The Sauerbrey equation is used to calculate changes in mass for
films that are rigid, and sufficiently thin. The linear relationship between
changes in frequency, ∆𝑓 and mass, ∆𝑚 is given by [59]:
∆𝑚 = − 𝐶 ∆𝑓𝑛
𝑛 (15)
Where 𝐶 is a constant based on the crystal property, and for the crystal
used in this work it is 17.7 ng cm-2 Hz-1, and 𝑛 is the overtone number.
However, if the adsorbed layer is viscoelastic in nature, the Sauerbrey
relation underestimates the mass, and another model is then used to fully
characterise the film to take into account the energy losses. In my studies
the Voigt model [60] was used. The viscoelastic response of the film was
represented by an elastic component coupled in parallel to a viscous
component. Again the software Q-tools (Q sense) was used to analyse the
data using Voigt viscoelastic modelling. The programme uses data from
different overtones together to make the calculations. In my work the
third, fifth, and seventh overtones were used. The following equations are
used to calculate changes in frequency and dissipation:
∆𝑓 = −
12𝜋𝜌0ℎ0
{𝜂2
𝛿2+ ℎ1𝜌1𝜔 − 2ℎ1 (
𝜂2
𝛿2)
2 𝜂1𝜔2
𝜇12 + 𝜂1
2𝜔2} (16)
∆𝐷 =
1𝜋𝑓𝜌0ℎ0
{𝜂2
𝛿2+ ℎ1𝜌1𝜔 + 2ℎ1 (
𝜂2
𝛿2)
2 𝜂1𝜔2
𝜇12 + 𝜂1
2𝜔2} (17)
Where 𝜌 is density of quartz, ℎ is thickness of quartz, 𝜂 is the shear
viscosity, 𝛿 is the viscous penetration depth, 𝜇 is the shear elasticity, and
Chapter 3
29
𝜔 is angular frequency of oscillation. The subscripts correspond to the
respective layers, i.e. 0 for the crystal, 1 for the adsorbed layer, and 2 for
the bulk solution.
3.3 Table of other instruments
Other techniques that provided useful information in this work are listed
below. The relevant information obtained in the context of this thesis has
also been summarised and included in the following table.
Instrument and measurement Information obtained
X-ray reflectivity (XRR) [61] Vertical structure of adsorbed bilayers and composite layers at silica - water interface (normal to the surface), and Langmuir layers at water - air interface. One can also learn of the structural variations, thickness, and roughness of a film.
Small angle X-ray scattering (SAXS) [61]
The self-assembly structure at the nano scale formed in mixed hyaluronan and DPPC vesicle solutions, presence of double bilayer vesicles, repeat distance of lamellae.
Electrophoretic mobility [62] The sign of the charge of the aggregates and the mobility with which they travel towards oppositely charged electrodes.
Structural organisation of Langmuir monolayers
Lateral interactions between and phase behaviour of the molecules in the monolayers
Chapter 4
30
4. Key results and discussion
As mentioned earlier, it is of importance to learn if COMP and lubricin
associate favourably in our model system, what impact COMP has in
retaining lubricin on the surface, and consequently on the lubrication
performance. Studying the biolubricants in question on hydrophobic
PMMA model surfaces, first individually and then synergistically followed
this line of thought.
4.1 COMP-lubricin synergy system
4.1.1 COMP on PMMA
As seen from QCM-D measurements (figure 8), COMP adsorbs on the
PMMA surface from a 100 μg/mL solution of COMP in a 150 mM PBS
solution. The resulting layer is about 23 nm thick and has a sensed mass
of 26.5 mg/m2. Additionally we know from the AFM normal force
measurements, that when COMP adsorbs to two opposing surfaces that
slide against each other, the force generated is purely repulsive in nature.
Figure 8: Adsorption of COMP on PMMA at 25°C from 100 μg/mL solution dissolved in 150 mM PBS, pH 7.5. Rinsing was done with 150 mM PBS (marked by ↓ arrows). Solid and open symbols are frequency change and dissipation change, respectively. For clarity, only every 20th point is plotted.
Chapter 4
31
This means that the lubricating layer should be sheared rather easily, and
would likely result in low friction. Compared to PMMA-PMMA and
lubricin-PMMA surfaces, the resulting friction for COMP covered PMMA
surfaces is low and the friction coefficient is ≈ 0.3 (figure 9 bottom).
Figure 9: Friction force as a function of applied load between (top) PMMA surfaces carrying a weakly adsorbed lubricin layer on loading (solid squares) and unloading (open squares) (bottom) PMMA surfaces coated with COMP (solid circles) and COMP-lubricin (solid triangles).
Chapter 4
32
4.1.2 Lubricin on PMMA
The lubrication performance of the lubricin layer formed on PMMA
surface was worse than that offered by COMP alone. From surface
interaction measurements, repulsive surface forces were seen (on
approach) at large separations. However upon compression, structural
rearrangements ensued. Some bridging was also seen in the force curves
on retrace. The adsorption of lubricin on PMMA was seemingly driven by
hydrophobic interactions between the non-glycosylated parts of lubricin
and the hydrophobic PMMA surface. Due to the weak attachment of
lubricin to PMMA, the resulting friction is significantly higher as
compared to that of COMP on PMMA (figure 9). Also there is clear
removal of some lubricin from the surface due to the combined action of
shear and load, since friction is higher on unloading compared to loading.
This is in agreement with what has been reported in literature about
lubricin being able to adsorb to model surfaces, but unable to individually
impart lubrication close to that experienced in the joints [15, 22].
4.1.3 COMP-lubricin synergy on model surfaces
As can be seen from the results above, it is evident that while COMP is
more resistant than lubricin to failing under the combined action of load
and shear, the friction levels attained are still not close to those
experienced physiologically in the joints. Thus we probed into the
lubrication performance resulting from the combined effect of COMP and
lubricin, i.e. with COMP forming the underlying layer and attached
lubricin being exposed to the aqueous solution. The results were notably
better. The resulting structure imparts very low friction with a friction
coefficient of 0.06, and load bearing capacity of minimum 7 MPa (refer
once again figure 9). Additionally, the combined COMP-lubricin layer has
high resistance to wear. COMP and lubricin associate specifically by non-
Chapter 4
33
covalent and covalent interactions via cysteine residues present at the C
terminal of COMP and the N terminal of lubricin. This in turn leads to
exposure of the glycosylated region of lubricin to the aqueous solution.
Figure 10: Adsorption of lubricin on COMP-coated PMMA at 25°C from 100 μg/mL solution in 150 mM PBS, pH 7.5. Rinsing was done with 150 mM PBS (marked by ↓ arrows). Solid and open symbols are frequency change and dissipation change, respectively. For clarity, only every 30th point is plotted.
QCM-D studies give insights into the nature of the formed layer (figure
10). 100 μg/mL human lubricin dissolved in 150 mM PBS was injected on
COMP-covered PMMA surface, and the lubricin layer on top of the COMP
layer was found to be viscoelastic with a thickness of about 20 nm and
sensed mass of 23 mg/m2.
These results underscore the importance of synergism when discerning
mechanisms that govern joint lubrication. The desirable lubrication
performance, which lubricin and COMP are unsuccessful in achieving
individually, is attained through their joined efforts. Lubricin associates
with COMP specifically and requires it as an underlying support. Zappone
et al. have also demonstrated favourable resistance to shear offered by
the tenacious anchoring of end domains of lubricin boundary layer [22].
As a consequence of the firm attachment, the structure in this system is
Chapter 4
34
better able to resist wear and the specific association allows lubricin to
expose parts of its structure in a manner that renders superb lubricity to
sliding model surfaces covered with these biolubricants.
4.2 Phospholipid-hyaluronan synergy system
The other lubrication couple that has been investigated at length in my
thesis is DPPC and hyaluronan. Naturally, before I began to look into
their association structures, it was imperative to characterise DPPC
bilayers. Adsorption of DPPC on a hydrophilic silica support and the
subsequent formation of bilayers in fluid phase have been elaborated by
our group earlier [30, 63]. Free hydrated DPPC bilayers undergo a main
phase transition at about 41 °C [64]. It is improbable that in physiological
scenario the complex mixture of phospholipids (including unsaturated
phospholipids) found in the synovial joints is in the gel state with frozen
acyl chains. To elucidate how the state of the DPPC layer affected
lubrication, I have studied this lipid across a range of temperatures and
phase transition on silica surfaces [65].
4.2.1 DPPC bilayer morphology at different temperatures
The DPPC bilayer structure was investigated by the XRR technique.
Figure 11 depicts the modelled electron density profiles, which were fitted
to experimental data using a six-layer model. The figure also illustrates
the corresponding structure and orientation of the bilayer; the left side of
the bilayer is next to the silicon wafer and the right side is oriented
towards the aqueous phase. It can be inferred from the profiles, that at 25
°C (and also 39 °C) the thickness of the bilayer is 5.6 nm, whereas at 55
°C it is 4.8 nm. This decrease in bilayer thickness indicates that a
transition occurred from the gel phase to the liquid disordered phase.
Chapter 4
35
Figure 11: Electron density profiles of DPPC bilayers at the silica liquid interface at different temperatures, ‘z’ denotes the distance from the silicon surface. A sketch of the bilayer structure is also shown.
Information on lateral variation across bilayer adsorption on the silica
surface was obtained using AFM imaging. DPPC bilayers were adsorbed
on silica surface at 52 °C and temperature variations were made by
following a sequence of 52 °C o 47 °C o 37 °C o 32 °C o 25 °C, and
then reversed to 32 °C o 37 °C o 47 °C o 52 °C.
Figure 12: AFM PeakForce height images of a DPPC bilayer on a silica surface, taken at different temperatures. The images were recorded in 150 mM PBS buffer solution, and flattened to remove tilt. The image size is 1x1 µm2, and the height scale bar is 3 nm in each case. The arrows show the order in which the images were recorded, and the temperature is provided below each image.
Chapter 4
36
Images recorded at 52 °C, 47 °C, and 37 °C were similar and in the liquid
disordered phase. A visible change in the form of small and grainy
structures started to occur at 32 °C, below which the bilayer was in gel
phase. This difference in the phase transition occurring at a lower
temperature of 32 °C in the case of AFM imaging and at a higher
temperature of 39 °C in the case of XRR, may be accounted to the kinetic
energy transferred by the tapping AFM tip, which resultantly causes a
shift in phase transition temperature.
4.2.2 Lubrication performance of DPPC bilayers at different temperatures
Force and friction measurements were performed on DPPC bilayer
covered silica surfaces at varying temperatures to get insights into the
lubrication performance of DPPC bilayers in the different phases.
There were no long-range forces observed when force measurements
were conducted on DPPC covered silica surfaces. Only short-range
repulsive forces arising from hydration and protrusion forces were
observed. Further, these repulsive forces increased with temperature.
Friction forces were measured at different temperatures on silica surfaces
covered with the DPPC bilayer, and it was seen that low friction forces
prevailed across the range of temperatures. The friction coefficient at <
47 °C is low (< 0.03) and nearly independent of temperature (figure 13).
The appreciable lubrication performance is ascribed to protrusion and
hydration forces, which makes shearing of the water layers - separating
the opposing bilayers easy.
Chapter 4
37
Figure 13: Friction force as a function of load between DPPC bilayer-coated silica surfaces immersed in 150 mM PBS at different temperatures.
The maximum load that the bilayer is subjected to in this system is 20
nN. When converted by the JKR theory [57] this corresponds to 42 MPa
(nearly double that which the cartilage can sustain [66]). However, it was
also noted that in some cases the bilayer was destroyed, particularly at
lower temperatures. An example of a friction versus load cycle where this
was observed at 25 °C is presented in figure 14.
Chapter 4
38
Figure 14: Example of a friction vs. load cycle measured between DPPC bilayers at 25 °C in 150 mM PBS. In this case the bilayer structure was compromised at a load of about 18 nN. Friction forces measured on loading and unloading are shown with filled (●) and unfilled (○) symbols, respectively.
The bilayers were intact and able to withstand loads up to about 18 nN.
As the load is increased further, friction increases significantly and the
bilayers give in. This is followed by the advent of a new energy dissipative
mechanism that increases the friction force. One such energy dissipative
process may arise from attractive hydrocarbon-hydrocarbon contacts of
phospholipids chains between the sliding surfaces. As the load was
decreased, the friction force failed to return to its original value indicating
that the DPPC bilayers do not heal.
Since the load bearing capacity at a given temperature differed between
different experiments, a statistical evaluation was used to assess the load
bearing capacity. At 25 °C half of the experiments showed bilayer failure
at < 20 nN (figure 15), whereas no such failure was seen at 52 °C.
Chapter 4
39
Figure 15: The percentage of friction experiments performed at a given temperature where the load bearing capacity was found to be less than 20 nN. The numbers in the bar columns indicate the total number of experiments performed at the respective temperatures.
This evaluation leads us to draw the inference that the stability of the
bilayer under the combined action of load and shear increases with
temperature. Increased fluidity seems to impart flexibility to the bilayer
and thereby induces a self-healing ability wherein the chains are flexible
enough to heal the defects. This results in the high load bearing capacity
that is witnessed at higher temperatures.
4.2.3 Lubrication synergy: Mixture of hyaluronan and DPPC vesicles
That DPPC and hyaluronan associate on surfaces upon sequential
injections, has been shown in one of the earlier works of our group [30].
A composite layer was formed whereby phospholipids and hyaluronan
were accumulated on model surfaces. However, in a physiological
environment the likelihood is that these biolubricants, which are present
in synovial fluid, form self-assembly aggregates and then attach to the
surface. Therefore, in order to study association of hyaluronan and DPPC
vesicles in a premixed solution, and consecutively its adsorption and
Chapter 4
40
lubrication performance on model surfaces, we have carried out the
investigations summarised below [31].
4.2.3.1 DPPC-hyaluronan association structures in bulk
SAXS data of DPPC-hyaluronan structures associating in bulk indicates
the presence of unilamellar and double bilayer vesicles. Electron density
profiles (figure 16) were obtained from the fitting of scattering curves,
and a satisfactory fit was achieved with an additional layer of hyaluronan.
As vesicles are closed structures, additional material would only adsorb to
one side of the bilayer (as depicted by the hump in the figure below). The
head-to-head distance of the bilayers remained the same in the presence
and absence of hyaluronan; this evidently indicates that hyaluronan
binds to the outer shell of the DPPC vesicles. Furthermore,
complementary to our current findings, previously reported DLS
measurements suggested that addition of hyaluronan to a DPPC vesicle
solution resulted in an increase in hydrodynamic size of the vesicles [30].
Figure 16: Electron density profiles of the single DPPC bilayer structure (top) and DPPC double bilayer structure (bottom) in absence (blue) and presence (green) of hyaluronan.
Chapter 4
41
4.2.3.2 Adsorption process of DPPC-hyaluronan association structures
A mixed hyaluronan/DPPC vesicle solution was injected on a silica
surface, and adsorption of the associating structures was studied with the
QCM-D technique (figure 17 a). Following the first injection, the initial
peak in frequency and dissipation change indicates rupturing of DPPC
vesicles and formation of a DPPC bilayer, consistent with that observed in
our previous studies [30]. This part of the adsorption step incurred small
changes in dissipation and thus was modelled using the Sauerbrey model.
Thereafter, adsorption was allowed to continue and the magnitude of Δf
and ΔD continued to increase in nearly a linear manner. Surface
saturation was not observed even after 100 minutes. This was followed by
rinsing and limited desorption was seen. A second injection again
resulted in a linear change in Δf and ΔD with time and limited desorption
upon rinsing. As indicated in the figure, the last two parts of the
adsorption were modelled using the Voigt model due to large changes in
ΔD.
Figure 17 b reveals the nature of the resulting layer after initial formation
of a DPPC bilayer. The close to linear relation between ΔD and Δf
indicates that there was no significant stiffening of the outer layer, and
the structures formed by adsorption from the DPPC-hyaluronan mixture
are extended and viscoelastic. When this is compared to previous work
where hyaluronan was added sequentially to a DPPC bilayer, the relation
between ΔD and Δf was seen to be sublinear. This means that the layer
formed in case of sequential addition was stiffer than that formed by
adsorption from a DPPC-hyaluronan mixture.
Chapter 4
42
Figure 17: (a) Adsorption from a solution containing 0.5 mg/mL DPPC and 0.5 mg/mL hyaluronan (start marked by ↑) on a silica surface in 155 mM NaCl solution at 55°C monitored by QCM-D for the 3rd overtone (frequency (black line) and dissipation (grey line) change). Rinsing (start marked by ↓) was done with 155 mM NaCl solution at 55 °C. For clarity every 30th point is plotted. (b) Dissipation change versus frequency change during the first adsorption step illustrated in a (black line) and data for hyaluronan adsorption on a pre-adsorbed DPPC bilayer (grey line) re-plotted from ref [30]. For clarity, every 7th point is plotted.
Chapter 4
43
4.2.3.3 Morphology of adsorbed layer from DPPC-hyaluronan mixture
Figure 18: AFM topography image of the adsorbed layer formed after 40 minutes adsorption from solutions containing 0.5 mg/mL DPPC and 0.5 mg/mL hyaluronan in 155 mM NaCl on a silica surface. After 40 minutes the solution was exchanged with pure 155 mM NaCl solution, and the resulting layer was imaged in this solution at a temperature of 47 °C. Image size 1 x 1 μm2. Image given in reference [30] from sequential adsorption of hyaluronan on DPPC bilayers on a silica surface can be used for comparison.
The morphology of the adsorbed layer from the DPPC-hyaluronan
mixture can give insights into the structures discussed above. The
adsorbed layer was not homogeneous and aggregates of varying sizes
were present. Some were smaller than the hydrodynamic diameter of
vesicles found in bulk solution [30], whereas others would correspond to
the size of flattened vesicles. It may be suggested that DPPC-hyaluronan
aggregates form on top of the DPPC bilayer, and these also include intact
vesicles. This is further evidenced by the force curves that report the
presence of vesicle-like structures on the surface.
Chapter 4
44
4.2.3.4 Surface force measurements
As the layer imaged above was heterogeneous, the force curves measured
on this layer also resulted in variations from spot to spot. Typical surface
force curves demonstrated long-range repulsion that increased
monotonically with decreasing separation as depicted in figure 19 a. This
is different from the short-range repulsion observed between lone DPPC
bilayers. This suggests that hyaluronan constitutes the outermost layer
and complements earlier findings of hyaluronan binding to the outer
surface of DPPC vesicles.
Chapter 4
45
Figure 19: Force normalised by radius as a function of separation between silica surfaces carrying an adsorbed layer formed from solutions containing 0.5 mg/mL DPPC and 0.5 mg/mL hyaluronan in 155 mM NaCl at 47 °C. The adsorption was allowed to proceed for 40 minutes before rinsing with 155 mM NaCl, and the forces were recorded after rinsing. (a) Typical force curves recorded before friction (approach (■) and separation (□)). (b) Very long range repulsive forces noted on the first approach on one spot (■) before friction measurements. The forces recorded on first separation (□) are also shown. The inset shows the 10th force curve measured on approach (■) and separation (□) on this spot.
The presence of aggregates observed in the AFM image above affects
surface interactions. An example is shown in figure 19 b. The presence of
a large aggregate was indicated by a very long-range repulsive force,
Chapter 4
46
which was observed on the approach curve of the first force measurement
(up to 300 nm). The force continued to increase and then suddenly
returned to zero. This can be interpreted as deformation of the aggregate,
which is followed by structural rearrangements. Ten consecutive force
measurements were performed and the final force measurement is shown
in the inset of figure 19 b. Following compressions, a few small steps were
observed; each step size roughly corresponding to the thickness of a
DPPC bilayer. Thus it can be understood from the results above that large
aggregates, which are present on the surface are disrupted upon
compression and transformed into multilayer structures. It can be
proposed that the presence of these DPPC-hyaluronan aggregates on the
surface allows build-up of a reservoir of biolubricants.
4.2.3.5 Lubrication performance of DPPC-hyaluronan association structures
It is impressive that the heterogeneous nature of the adsorbed layer does
not compromise its lubrication performance. Friction forces measured
between the inhomogeneous layers formed from the DPPC hyaluronan
mixture were very low. Friction measurements were carried out at three
different spots. The best fit, which corresponds to the red line in figure 20
a resulted in a friction coefficient value of 0.006. Such low friction force is
comparable to that reported for synovial joints [1]. The inhomogeneous
layers rearrange under the combined action of load and shear into
phospholipid bilayer or multilayer structures, which are separated by
easily sheared water layers. As discussed in an earlier section of this
thesis it is the presence of strong hydration and entropic protrusion
forces between DPPC bilayers [67] that facilitates sliding with a low
friction force. The maximum load used in this study corresponded to a
pressure of 23 MPa, close to the critical pressure that a cartilage surface
can withstand before it breaks under axial load [66].
Chapter 4
47
Figure 20: (a) Friction force as a function of load between two silica surfaces carrying an adsorbed layer formed after 40 minutes adsorption from a 155 mM NaCl solution containing 0.5 mg/mL DPPC and 0.5 mg/mL hyaluronan, followed by rinsing (■). The temperature was 47 oC. The solid red line (follows Amonton’s rule) has a slope of 0.006 and the slope of the dashed line (reference line) is 0.01. (b) Illustration of one of the three friction measurements depicting high friction force upon loading at a normal force of 10 nN.
An observation of interest that occurred in some cases is illustrated in
figure 20 b. At a load of 10 nN, there was a sharp and transient increase
in the value of the fiction force. This may be attributed to the energy
dissipation arising from structural rearrangements induced by load and
Chapter 4
48
shear. It is of importance to note that the layer continued to provide low
friction even after the friction peak. Thus, the DPPC hyaluronan mixture
provides lubrication synergy; permits accumulation of large quantities of
phospholipids at the interface and furthers the return to a low friction
state even after the initial layer structure is disrupted. The same does not
hold good in the case of lone DPPC bilayers (as we saw earlier, figure 14),
which are compromised under high loads [65].
The interplay of DPPC and hyaluronan presents a clear lubrication
advantage over that of lone DPPC bilayer. Hyaluronan is not essential for
achieving low friction force, yet it provides build-up of a reservoir of
phospholipid lubricant on the model surface, and consequently the
DPPC-hyaluronan structures possess a self-healing ability. Clearly this
excellent lubrication and self-healing ability will be advantageous if such
layers are present on joint surfaces. Considering B.A. Hills had assumed
and managed to demonstrate through morphological evidence that
phospholipids form the outermost lubricating lining of the surface of
joints [68], such a conjecture does not seem far-fetched.
4.3 Effect of molecular weight and ions on DPPC-hyaluronan interactions
Ca2+ ions are present in the synovial fluid [69] and are also known to bind
to DPPC [70]. Thus, they may have an influence on the structure and
interactions of DPPC and hyaluronan, and resultantly affect the system
under investigation. In the context of joint diseases it would be
compelling to also examine the effect of molecular weight of hyaluronan
on this model system. To the best of our knowledge, no direct
measurements have been conducted to study lubrication performance of
cartilage by varying the molecular weight of hyaluronan. However low
molecular weight hyaluronan has been associated with rheumatoid
Chapter 4
49
arthritis [36, 71]. Langmuir monolayers of DPPC were chosen as model
system in this part of the study as they were more suitable for the
required techniques of the investigations.
4.3.1 Effect of molecular weight of hyaluronan
Information on lateral interactions between molecules and phase changes
in Langmuir monolayers was obtained from the surface pressure versus
mean molecular area isotherms (π/A isotherms) [72]. As can be seen
from figure 21 a, the π/A isotherm for DPPC with low molecular weight
hyaluronan in the subphase is more expanded than the other subphases.
This implies that low molecular weight hyaluronan affects lateral
interactions greatly and therefore the packing of monolayers is most
severely influenced in its presence.
Chapter 4
50
Figure 21: (top) Π/A-isotherms of DPPC on aqueous subphases containing 155 mM sodium chloride. The concentration of HA was 0.5 mg/mL. (bottom) Π/A-isotherms of DPPC on subphases containing 155 mM sodium chloride and 10 mM calcium chloride. The concentration of HA was 0.5 mg/mL.
BAM images complement the findings above and provide information on
changes in the domain structures at the micrometre scale. Again, addition
of low molecular weight hyaluronan caused the most apparent
morphological changes in the domain structures; wherein clearly
separated small-condensed regions with a round circumference were
seen. However, the morphology in presence of high molecular weight
hyaluronan and that in the absence of hyaluronan remained similar, with
bigger multilobed-structures.
Chapter 4
51
Figure 22: Brewster angle microscopy images of DPPC monolayers at the liquid air interface on salt solutions with (a) HA of MW 10 kDa (left) 155 mM sodium chloride (right) 155 mM sodium chloride with 10 mM calcium chloride (b) HA of MW 1500 kDa (left) 155 mM sodium chloride (right) 155 mM sodium chloride with 10 mM calcium chloride.
The images of subphase containing low molecular weight hyaluronan also
point to the presence of some material in between the condensed regions,
which is presumably due to the presence of two phases at the interface; a
condensed phase and another more disordered and thin layer. In the case
of the disordered phase, a possible route of interaction of hyaluronan is
through it’s hydrophobic patches to the alkyl chains of DPPC [73, 74] as
sketched in figure 23. In the case of the ordered phase, it is likely that
hyaluronan interacts with the DPPC headgroup, via its negatively charged
carboxylic acid groups and the positively charged region of the DPPC
headgroup [75, 76].
Chapter 4
52
This was further explored through XRR measurements. It was concluded
from the XRR findings that there was significant accumulation of low
molecular weight hyaluronan in the hydrophilic headgroup region of
DPPC, which prevented the headgroups from closely packing together.
On the other hand, higher molecular weight hyaluronan mostly resided
underneath (away from the liquid air interface) the polar DPPC
headgroup.
It has previously been shown in our group that the addition of
hyaluronan to supported DPPC bilayers somewhat reduces the load
bearing capacity [30]. It can be speculated from the above-mentioned
results that the increase in distance between headgroups resulting from
the addition of low molecular weight hyaluronan, disturbs the lateral
packing of the layers. This results in poorer load bearing capacity as only
well ordered bilayers should be capable of providing appreciable load
bearing capacity. This may provide insights into an aspect of lubrication
failure in the event of arthritis.
Chapter 4
53
Figure 23: Schematic of the interaction of HA with DPPC at the water-air interface
4.3.2 Effect of Ca2+ ions
Electrophoretic mobility measurements indicated that the positive value
of electrophoretic mobility of DPPC increased with an increase in calcium
chloride concentration, which confirmed adsorption of divalent ions to
the DPPC phospholipid headgroup [77].
From the π/A isotherms it was noted that addition of 10 mM CaCl2 affects
both DPPC monolayer, and DPPC-hyaluronan interactions significantly.
Take for instance the isotherm of DPPC without hyaluronan in figure 21
b. The surface pressure at a given area per molecule is lower when
calcium ions are present (especially for high surface pressures). This
means that the lipids pack better in the presence of calcium ions [78]. The
interactions between DPPC and hyaluronan, when calcium was present in
the subphase, were influenced greatly as the isotherms changed for both
low and high molecular weight hyaluronan. In the absence of calcium
Chapter 4
54
ions the monolayer expanded only with low molecular weight hyaluronan
in the subphase, whereas in the presence of calcium ions, it always
expanded. This suggests enhanced interactions take place between DPPC
and hyaluronan owing to the presence of calcium ions. The observations
mentioned above were once again supported by BAM images. In all the
different subphases, condensed regions were more densely packed in the
presence of calcium ions.
Further, XRR measurements highlighted that when calcium ions were
present in the subphase hyaluronan was amassed next to DPPC, which
explains the significant changes that occured in π/A isotherms in
presence of calcium ions.
4.4 Effect of high pressures on DPPC bilayers and DPPC-hyaluronan structures
Human joints are subjected to different loads throughout the day
depending on activities that are partaken. For daily activities joints have
been reported in vivo to bear physiological loads up to 18 MPa (180 bar)
[79]. In light of improved lubrication performance, which is imparted by
the synergistic effect of important biomolecules in the joint, it will be of
relevance to examine the behaviour of DPPC-hyaluronan composite
layers under high pressures. As a first step towards this we investigated
pressure effects on supported DPPC layers in absence of hyaluronan
using XRR [80].
XRR studies were performed at different pressures ranging from those
experienced during every day activities and during extreme conditions.
As discussed in an earlier section of this thesis silica-supported DPPC
bilayers are in the fluid phase at 55°C. Phase transitions can also be
induced by pressure and for DPPC bilayers at 55 °C it occurs at 500 bar
[81]. Pressure cycle measurements were performed from 60 bar to 2 kbar
Chapter 4
55
by increasing the pressure in steps. In order to verify the reversibility of
pressure-induced changes, the final measurement was brought down and
conducted at 60 bar. In case of lone DPPC bilayers the reflectivity curves
indicate that the position of the first minima shifted to lower q-values
(increase in bilayer thickness) when there was an increase in pressure
from 100 bar to 1 kbar; indicating a phase transition from fluid to gel
phase. Thereafter on return from a pressure of 2 kbar to 60 bar the final
reflectivity curve was found not to be identical to the first measurement
at 60 bar. This demonstrates that the shift was not reversible, and that
DPPC bilayers by themselves are vulnerable to high hydrostatic
pressures.
The same pressure cycle measurements were then performed for DPPC-
hyaluronan composite layers as were done for lone DPPC bilayers and
striking differences were noticed. It was found that unlike DPPC bilayers,
structural changes induced by high pressures in DPPC-hyaluronan
composite layers were reversible (depicted in fig 24). This stresses the
fact that with addition of hyaluronan the stability and robustness of
mixed DPPC-hyaluronan composites is improved over that of DPPC alone
at high hydrostatic pressures.
These results point to the likelihood that hyaluronan contributes to
preserving the integrity of the lipid bilayers when subjected to high
pressures. This is of importance while considering boundary lubrication
conditions in the joints, as only unimpaired lipid layers would ensure low
friction.
Chapter 4
56
Figure 24: Fresnel normalized reflectivity curves of (top) DPPC at 55 °C and different pressures up to 2 kbar (b) DPPC/HA1500 composite layers at 55 °C and different pressures up to 2 kbar. Solid black lines show the fits. Blue bars indicate the position of the first minimum (A: 1 & 2kbar, B: 60, 100 & 1bar).
Chapter 5
57
5. Conclusions and impact
The optimal functioning of joints necessitates low friction, high load
bearing capacity, wear protection, and a self-healing ability. This is
ensured by several mechanisms and biolubricants, i.e. not by any one
component working alone. With this work, we have achieved a better
understanding of the association, layer structure, and resulting
lubrication performance of some major biolubricants implicated in
biolubrication.
From the first lubrication couple (COMP-lubricin) we learnt that even
though the favourable confirmation of one component (lubricin) could
result in low friction, it was imperative that it stayed strongly attached to
the model surface to have a load-bearing capacity of significance. This
was achieved with the help of the other component (COMP). COMP
associates specifically with lubricin, ensures lubricin’s anchoring, and
exposes its structure in a favourable confirmation that results in low
friction. To the best of our knowledge this synergistic behaviour has not
been experimentally studied elsewhere.
We have progressed from our previous work by building a more
physiologically relevant model of the second lubrication couple (DPPC-
hyaluronan), wherein DPPC and hyaluronan were injected from a
premixed solution. The friction force generated was extremely low. More
importantly, it was inferred that even though hyaluronan is not important
for reaching low friction force values, it bestows an important self-healing
ability by providing a reservoir of the DPPC biolubricants at the sliding
surfaces. This inference holds importance in the understanding of a few
biological scenarios such as, the depletion of biolubricants in certain
areas following trauma/suffering. Components like hyaluronan can
Chapter 5
58
complement its synergy partner and may replete the required
biolubricants and hence provide self-repair.
By studying additional factors (effect of temperature, pressure, molecular
weight of hyaluronan, presence of calcium ions) a deeper understanding
of the topic was gained and advances were made to the existing
knowledge.
It was found that fluidity (and therefore flexibility and mobility) of the
DPPC chains was needed to repair and heal areas that included defects.
This was seen to occur only when the phospholipid was present in the
fluid phase and not in the more rigid gel phase.
Low molecular weight hyaluronan is linked to rheumatoid arthritis. Our
studies on the effect of molecular weight of hyaluronan on DPPC-
hyaluronan interfacial layers, shed light on the fact that low molecular
weight hyaluronan affects the packing and organisation of such layers
more than high molecular weight hyaluronan. If such layers exist on the
joint surfaces, then a disturbance of this sort may unfavourably affect the
intactness of the layers, and consequently its load bearing capacity. The
same study examined the presence of calcium ions and concluded that
interactions between DPPC and hyaluronan were enhanced by the
presence of these ions. Given that calcium ions are present in the synovial
fluid, this information is consequential to gain insights on the influence
of calcium ions on DPPC-hyaluronan structures and interactions.
Finally, by subjecting the DPPC-hyaluronan system to high pressures, it
was found that hyaluronan helps in stabilising DPPC bilayers. DPPC
bilayers by themselves are vulnerable to such high pressures. As we
constantly subject our knees to varying and also high pressures, a better
understanding of this aspect of joint lubrication was achieved with the
help of such a synergy study.
Chapter 5
59
The above synergy studies potentially layout the basis for future work,
which can build on existing knowledge with additional biolubricants.
Successful findings in this field can also be related to biolubrication in the
eyes and lungs. Further, such studies would also be of valuable clinical
relevance for joints, possibly resulting in better repair strategies, longer
lasting prosthesis, artificial cartilage, and synthetic lubricants.
Additionally, it would be useful in the context of tissue engineering,
biomedical devices, and environmentally friendly lubricants.
We are still far from grasping the ways of Nature with regards to joint
lubrication. Nevertheless, encouraging progress is being made in this
field. The quest for a magic bullet molecule has been veered towards key
biolubricants and their synergistic workings. This field holds the potential
for contribution to alleviating the suffering of millions, and at the same
time has tremendous business potential. I will be eagerly awaiting
advancements and strides that will be made in this field.
60
6. Acknowledgements
It is the gentle guidance, constant support, and unflinching faith of so many people that has led me to this new milestone in life. I am grateful for the opportunities that were provided to me at the right time and right place. This has helped me realise my endeavours, and thus acknowledgements are in order.
Before I begin to thank members of this PhD journey, I would like to thank extended family, friends, teachers, supervisors, and colleagues, whose guidance and motivation I have received in the past, well before my PhD days. Especially the ones in India and Switzerland. I am ever so indebted for your generous help during my stay, without which, reaching this far would have been out of question.
Dr. Andra Dėdinaitė and Prof. Per Claesson it has been an honour and my good fortune to have known and worked with you. You have always encouraged an atmosphere that is conducive to learning and questioning. Andra, I have been inspired by your dedication to and vast knowledge of this field. I ardently hope for continued progress and success in this field. No question was ever a silly question for you. I have been amazed fairly regularly in the past four years by how you have always made time for me. Always! Additionally, I could freely discuss any problem outside the work front with you (very helpful when one is thousands of miles away from home). You have fostered team spirit, and that makes me feel we have achieved this together. All this has been invaluable for my scientific and personal growth. Per, seldom does one come across consideration and kindness to other people’s needs to the degree you have shown. Your constructive criticism, keen eye for details, clinical approach and reassuring manner, has taught me to deconstruct complex problems and helped me improve. I have walked out of your office having learnt something new every time, and enjoyed intellectually stimulating moments. You hold a genuine interest in the well being of your student. I am glad I had both of you to learn from as one of my early teachers, and would like to carry these work ethics henceforth with me.
I thank financial support from the People Programme of the European Union’s Seventh Framework Programme (Marie Curie Actions). Collaborations have been instrumental to my studies and so I take this opportunity to thank all of my collaborators at the University of Gothenburg, Lund University, HZG Hamburg, and SP Stockholm. Particularly, Florian and Thomas, your commitment to work, well past the end of the project has been commendable! In hindsight, I will fondly look back at the long beam times and night shifts at the synchrotron facilities. It has been a pleasure to work with you. The entire Marie Curie network is appreciated. I also give my thanks to all my co-authors. Thanks to Master’s student, Yushi for working so hard when the situation demanded it.
To all the past and current members of this division, thank you! I have made some friends for life. Your warm welcome helped me settle in. Your friendships have helped me enjoy the good times and endure the challenging ones. Thanks (in no particular order) to Marie, Katerina, Jie, Rachel, Jonathan, Yousef, Golrokh, Maziar, Chao, Min, Junxue, Olga, Eleonora, Mattias, Krishnan, Xiaoyan, Tingru, Maria (SP), Elizaveta, Neda, Eric, Erik, Hui, Sara, Celine, Angelika, Georgia, and
61
Illia. Forgive me in case I have forgotten someone. Working late in the lab was more fun when there was company, and there is something exclusive about such labmate-ship. Thank you Laetitia, Adrien, Deb, Elin, and Rasmus.
All badminton partners are acknowledged for helping me smash away my worries.
Indebted to the very talented-Ana Filipa for the cover page illustration.
The Chauhan’s, Sulena and family, Ashu and family, Anu and family, Divya and family, Richa, Zahra, Yamini, Neha, Deeksha, Ramya, you have been home away from home. Grateful to Param, Sukhda, Mitra, Shweta, Sonal, Ramaa, Gautami, Kateryna, Floor, Leila, Roya, Mina and Bhavya.
Gajendra, thank you for everything. This would not have been possible without you. To those who know you, also know that I mean this quite literally!
Mamma and papa, it is not becoming to thank you in words. Nevertheless, I would like you to know that I have deep gratitude and admiration for the both of you. I have been very lucky with all the opportunities you have provided, and for the happy, healthy, stimulating, and enterprising environment you have always maintained at home. Anu, when I fall short of strength I lean on you, and Amrit, when I begin to lose perspective I tap into your wisdom.
It couldn’t have been better!
Despite everyone’s generous contribution, I take full responsibility for any inadvertent errors that may have escaped my attention.
आकाांक्षा राज AKANKSHA RAJ जनवरी २०१७ JANUARY 2017 स्टाक्होम, स्वीडन STOCKHOLM, SWEDEN
62
7. References [1] J. Klein, Molecular mechanisms of synovial joint lubrication, Journal of engineering tribology
220(8) (2006) 691-710.
[2] W.H. Organisation, Chronic rheumatic conditions. 2017.
[3] E.L.A. Rheumatism, Questions and answers on rheumatic diseases.
[4] J.M. Mansour, Biomechanics of Cartilage, in: C.A. Oatis (Ed.), Kinesiology: The Mechanics and
Pathomechanics of Human Movement, Lippincott Williams & Wilkins2004, pp. 68-77.
[5] V. Wright, D. Dowson, Lubrication and cartilage, Journal of Anatomy 121(1) (1976) 107-118.
[6] E.B. Hunziker, The structure of articular cartilage, in: C. Archer, J. Ralphs (Eds.), Regenerative
Medicine and Biomaterials for the Repair of Connective Tissues, Woodhead Publishing2010, pp. 83-
105.
[7] B. Loret, F.M.F. Simoes, Articular cartilage with intra- and extrafibrillar waters: a chemo-
mechanical model, Mech Mater 36(5-6) (2004) 515-541.
[8] J.S. Jurvelin, D.J. Muller, M. Wong, D. Studer, A. Engel, E.B. Hunziker, Surface and subsurface
morphology of bovine humeral articular cartilage as assessed by atomic force and transmission
electron microscopy, Journal of structural biology 117(1) (1996) 45-54.
[9] L.C. Hughes, C.W. Archer, I. ap Gwynn, The ultrastructure of mouse articular cartilage: collagen
orientation and implications for tissue functionality. A polarised light and scanning electron
microscope study and review, Eur Cell Mater 9 (2005) 68-84.
[10] K.L. Yielding, D. Platt, H.L. Holley, Synovial fluid. I. Comparison of sodium and potassium
concentrations in normal and diseased joint fluid, Proc Soc Exp Biol Med 85(4) (1954) 665-7.
[11] E.H. Jebens, M.E. Monk-Jones, On the viscosity and pH of synovial fluid and the pH of blood,
The Bone & Joint Journal 41-B(2) (1959) 388-400.
[12] B. Zappone, M. Ruths, G.W. Greene, G.D. Jay, J.N. Israelachvili, Adsorption, lubrication, and
wear of lubricin on model surfaces: polymer brush-like behavior of a glycoprotein, Biophys J 92(5)
(2007) 1693-708.
[13] G.D. Jay, D.A. Harris, C.J. Cha, Boundary lubrication by lubricin is mediated by O-linked beta(1-
3)Gal-GalNAc oligosaccharides, Glycoconjugate Journal 18(10) (2001) 807-815.
[14] D.A. Swann, F.H. Silver, H.S. Slayter, W. Stafford, E. Shore, The molecular structure and
lubricating activity of lubricin isolated from bovine and human synovial fluids, Biochem J 225(1)
(1985) 195-201.
[15] D.P. Chang, N.I. Abu-Lail, J.M. Coles, F. Guilak, G.D. Jay, S. Zauscher, Friction force microscopy
of lubricin and hyaluronic acid between hydrophobic and hydrophilic surfaces, Soft Matter 5(18)
(2009) 3438-3445.
[16] L. Ali, Bio-lubrication: Structural Investigation of Lubricin and its Glycosylation, Department of
Medical Biochemistry, University of Gothenburg, Gothenburg, 2014.
[17] K.L. Posey, P.M. Liu, H.R. Wang, A.C. Veerisetty, J.L. Alcorn, J.T. Hecht, RNAi Reduces Expression
and Intracellular Retention of Mutant Cartilage Oligomeric Matrix Protein, Plos One 5(4) (2010).
[18] J. Zaia, R.E. Boynton, A. McIntosh, D.R. Marshak, H. Olsson, D. Heinegard, F.P. Barry, Post-
translational modifications in cartilage oligomeric matrix protein - Characterization of the N-linked
oligosaccharides by matrix-assisted laser desorption ionization time-of-flight mass spectrometry,
Journal of Biological Chemistry 272(22) (1997) 14120-14126.
[19] L.S. Lohmander, T. Saxne, D.K. Heinegard, Release of Cartilage Oligomeric Matrix Protein
(Comp) into Joint Fluid after Knee Injury and in Osteoarthritis, Annals of the Rheumatic Diseases
53(1) (1994) 8-13.
[20] E. Hedbom, P. Antonsson, A. Hjerpe, D. Aeschlimann, M. Paulsson, E. Rosa-Pimentel, Y.
Sommarin, M. Wendel, A. Oldberg, D. Heinegård, Cartilage matrix proteins. An acidic oligomeric
protein (COMP) detected only in cartilage, Journal of Biological Chemistry 267(9) (1992) 6132-
61366.
63
[21] A. Oldberg, P. Antonsson, K. Lindblom, D. Heinegard, Comp (Cartilage Oligomeric Matrix
Protein) Is Structurally Related to the Thrombospondins, Journal of Biological Chemistry 267(31)
(1992) 22346-22350.
[22] B. Zappone, G.W. Greene, E. Oroudjev, G.D. Jay, J.N. Israelachvili, Molecular aspects of
boundary lubrication by human lubricin: effect of disulfide bonds and enzymatic digestion,
Langmuir 24(4) (2008) 1495-508.
[23] J. Marcelino, J.D. Carpten, W.M. Suwairi, O.M. Gutierrez, S. Schwartz, C. Robbins, R. Sood, I.
Makalowska, A. Baxevanis, B. Johnstone, R.M. Laxer, L. Zemel, C.A. Kim, J.K. Herd, J. Ihle, C.
Williams, M. Johnson, V. Raman, L.G. Alonso, D. Brunoni, A. Gerstein, N. Papadopoulos, S.A.
Bahabri, J.M. Trent, M.L. Warman, CACP, encoding a secreted proteoglycan, is mutated in
camptodactyly-arthropathy-coxa vara-pericarditis syndrome, Nat Genet 23(3) (1999) 319-22.
[24] C.R. Flannery, C.E. Hughes, B.L. Schumacher, D. Tudor, M.B. Aydelotte, K.E. Kuettner, B.
Caterson, Articular cartilage superficial zone protein (SZP) is homologous to megakaryocyte
stimulating factor precursor and is a multifunctional proteoglycan with potential growth-promoting,
cytoprotective, and lubricating properties in cartilage metabolism, Biochem Bioph Res Co 254(3)
(1999) 535-541.
[25] Z. Pawlak, Z.A. Figaszewski, A. Gadomski, W. Urbaniak, A. Oloyede, The ultra-low friction of the
articular surface is pH-dependent and is built on a hydrophobic underlay including a hypothesis on
joint lubrication mechanism, Tribology International 43(9) (2010) 1719-1725.
[26] D.A. Swann, H.S. Slayter, F.H. Silver, The molecular structure of lubricating glycoprotein-I, the
boundary lubricant for articular cartilage, The Journal of Biological Chemistry 256(11) (1981) 5921-
5925.
[27] R. Tadmor, N. Chen, J.N. Israelachvili, Thin film rheology and lubricity of hyaluronic acid
solutions at a normal physiological concentration, Journal of Biomedical Materials Research 61(4)
(2002) 514-523.
[28] R. Goldberg, A. Schroeder, G. Silbert, K. Turjeman, Y. Barenholz, J. Klein, Boundary Lubricants
with Exceptionally Low Friction Coefficients Based on 2D Close-Packed Phosphatidylcholine
Liposomes, Advanced Materials 23(31) (2011) 3517-3521.
[29] R.W. Forsey, J. Fisher, J. Thompson, M.H. Stone, C. Bell, E. Ingham, The effect of hyaluronic acid
and phospholipid based lubricants on friction within a human cartilage damage model, Biomaterials
27(26) (2006) 4581-4590.
[30] M. Wang, C. Liu, E. Thormann, A. Dėdinaitė, Hyaluronan and Phospholipid Association in Biolubrication, Biomacromolecules 14(12) (2013) 4198-4206.
[31] A. Raj, M. Wang, T. Zander, D.C. Wieland, X. Liu, J. An, V.M. Garamus, R. Willumeit-Romer, M.
Fielden, P.M. Claesson, A. Dedinaite, Lubrication synergy: Mixture of hyaluronan and
dipalmitoylphosphatidylcholine (DPPC) vesicles, J Colloid Interface Sci 488 (2017) 225-233.
[32] T.A. Schmidt, N.S. Gastelum, Q.T. Nguyen, B.L. Schumacher, R.L. Sah, Boundary lubrication of
articular cartilage: role of synovial fluid constituents, Arthritis & Rheumatology 56(3) (2007) 882-
891.
[33] M.K. Kosinska, G. Lochnit, J. Wilhelm, H. Klein, U. Kaesser, G. Lasczkowski, M. Rickert, G.
Schmitz, J. Steinmeyer, A lipidomic study of phospholipid classes and species in human synovial
fluid, Arthritis & Rheumatology 65(9) (2013) 2323-2333.
[34] L.R. Gale, Y. Chen, B.A. Hills, R. Crawford, Boundary lubrication of joints: characterization of
surface-active phospholipids found on retrieved implants, Acta orthopaedica 78(3) (2007) 309-14.
[35] R. Smith, C. Tanford, The critical micelle concentration of l-α-dipalmitoylphosphatidylcholine in
water and water/methanol solutions, J Mol Biol 67(1) (1972) 75-83.
[36] E.A. Balazs, D. Watson, I.F. Duff, S. Roseman, Hyaluronic acid in synovial fluid. I. Molecular
parameters of hyaluronic acid in normal and arthritic human fluids, Arthritis & Rheumatism 10(4)
(1967) 357-376.
[37] R. Kohn, P. Kovac, Dissociation constants of D-galacturonic and D-glucuronic acid and their O-
methyl derivatives, Chemical Papers 32(4) (1978) 478-485.
64
[38] E. Buhler, F. Boue, Chain persistence length and structure in hyaluronan solutions: Ionic
strength dependence for a model semirigid polyelectrolyte, Macromolecules 37(4) (2004) 1600-
1610.
[39] A. Maleki, A.-L. Kjøniksen, B. Nyström, Anomalous Viscosity Behavior in Aqueous Solutions of
Hyaluronic Acid, Polym. Bull. 59(2) (2007) 217-226.
[40] J. Necas, L. Bartosikova, P. Brauner, J. Kolar, Hyaluronic acid (hyaluronan): a review, Vet Med-
Czech 53(8) (2008) 397-411.
[41] J.E. Scott, Secondary structures in hyaluronan solutions: chemical and biological implications,
Ciba Found Symp 143 (1989) 6-15; discussion 15-20, 281-5.
[42] T.M. Tamer, Hyaluronan and synovial joint: function, distribution and healing, Interdiscip
Toxicol 6(3) (2013) 111-25.
[43] J. Israelachvili, Intermolecular and Surface Forces, Academic Press, 1992.
[44] E.M. Lifshitz, The theory of molecular attractive forces between solids, Soviet Physics 2(1)
(1956) 73-83.
[45] J.N. Israelachvili, Chapter 13 - Van der Waals Forces between Particles and Surfaces, in: J.N.
Israelachvili (Ed.), Intermolecular and Surface Forces (Third Edition), Academic Press, San Diego,
2011, pp. 253-289.
[46] J.N. Israelachvili, 8 - Special Interactions: Hydrogen-Bonding and Hydrophobic and Hydrophilic
Interactions, in: J.N. Israelachvili (Ed.), Intermolecular and Surface Forces (Third Edition), Academic
Press, San Diego, 2011, pp. 151-167.
[47] D. Chandler, Interfaces and the driving force of hydrophobic assembly, Nature 437(29) (2005)
640-647.
[48] J.N. Israelachvili, 21 - Interactions of Biological Membranes and Structures, in: J.N. Israelachvili
(Ed.), Intermolecular and Surface Forces (Third Edition), Academic Press, San Diego, 2011, pp. 577-
616.
[49] D. Leckband, J. Israelachvili, Intermolecular forces in biology, Quarterly Reviews of Biophysics
34(2) (2001) 105-267.
[50] J.N. Israelachvili, Chapter 16 - Steric (Polymer-Mediated) and Thermal Fluctuation Forces, in:
J.N. Israelachvili (Ed.), Intermolecular and Surface Forces (Third Edition), Academic Press, San Diego,
2011, pp. 381-413.
[51] J. Klein, Shear, Friction, and Lubrication Forces Between Polymer-Bearing Surfaces, Annual
Review of Materials Science 26 (1996) 581-612.
[52] W.A. Ducker, T.J. Senden, R.M. Pashley, Direct Measurement of Colloidal Forces Using an
Atomic Force Microscope, Nature 353(6341) (1991) 239-241.
[53] T.J. Senden, Force microscopy and surface interactions, Current Opinion in Colloid & Interface
Science 6 (2001) 95-101.
[54] J.E. Sader, J.W.M. Chon, P. Mulvaney, Calibration of rectangular atomic force microscope
cantilevers, Review of Scientific Instruments 70(10) (1999) 3967-3969.
[55] J.N. Israelachvili, 11 - Contrasts between Intermolecular, Interparticle, and Intersurface Forces,
in: J.N. Israelachvili (Ed.), Intermolecular and Surface Forces (Third Edition), Academic Press, San
Diego, 2011, pp. 205-222.
[56] R. Alvarez-Asencio, E. Thormann, M.W. Rutland, Note: Determination of torsional spring
constant of atomic force microscopy cantilevers: Combining normal spring constant and classical
beam theory, Review of Scientific Instruments 84(9) (2013).
[57] K.L. Johnson, K. Kendall, A.D. Roberts, Surface Energy and the Contact of Elastic Solids,
Proceedings of the Royal Society of London. A. Mathematical and Physical Sciences 324(1558)
(1971) 301-313.
[58] A. Heptonstall, G. Cagnoli, J. Hough, S. Rowan, Characterisation of mechanical loss in synthetic
fused silica ribbons, Physics Letters A 354(5–6) (2006) 353-359.
[59] G. Sauerbrey, Verwendung von Schwingquarzen zur Wägung dünner Schichten und zur
Mikrowägung, Z. Physik 155(2) (1959) 206-222.
65
[60] M.V. Voinova, M. Jonson, B. Kasemo, ‘Missing mass’ effect in biosensor's QCM applications, Biosensors and Bioelectronics 17(10) (2002) 835-841.
[61] J. Als-Nielsen, D. McMorrow, Elements of modern X-ray physics, 2nd ed., Wiley, Hoboken,
2011.
[62] A.V. Delgado, F. Gonzalez-Caballero, R.J. Hunter, L.K. Koopal, J. Lyklema, Measurement and
interpretation of electrokinetic phenomena, J Colloid Interf Sci 309(2) (2007) 194-224.
[63] C. Liu, M. Wang, J. An, E. Thormann, A. Dėdinaitė, Hyaluronan and phospholipids in boundary lubrication, Soft Matter 8(40) (2012) 10241-10244.
[64] R. Koynova, M. Caffrey, Phases and phase transitions of the phosphatidylcholines, Biochimica
et Biophysica Acta 1376(1) (1998) 91-145.
[65] M. Wang, T. Zander, X. Liu, C. Liu, A. Raj, D.F. Wieland, V.M. Garamus, R. Willumeit-Römer,
P.M. Claesson, A. Dėdinaitė, The effect of temperature on supported dipalmitoylphosphatidylcholine (DPPC) bilayers: Structure and lubrication performance, J Colloid
Interf Sci 445 (2015) 84-92.
[66] G. Spahn, R. Wittig, Biomechanical properties (compressive strength and compressive pressure
at break) of hyaline cartilage under axial load, zentralblatt fur chirurgie 128(1) (2003) 78-82.
[67] J. Israelachvili, H. Wennerstrom, Role of hydration and water structure in biological and
colloidal interactions, Nature 379(6562) (1996) 219-225.
[68] B.A. Hills, Surface-active phospholipid: a Pandora's box of clinical applications. Part II. Barrier
and lubricating properties, Internal Medicine Journal 32(5-6) (2002) 242-251.
[69] V. Perman, 19 - Synovial Fluid A2 - KANEKO, JIRO J, Clinical Biochemistry of Domestic Animals
(Third Edition), Academic Press1980, pp. 749-783.
[70] J. Seelig, Interaction of Phospholipids with Ca-2+ Ions - on the Role of the Phospholipid Head
Groups, Cell Biol Int Rep 14(4) (1990) 353-360.
[71] J.M. Cyphert, C.S. Trempus, S. Garantziotis, Size Matters: Molecular Weight Specificity of
Hyaluronan Effects in Cell Biology, International Journal of Cell Biology 2015 (2015) 8.
[72] D.C.F. Wieland, P. Degen, T. Zander, S. Gayer, A. Raj, J.X. An, A. Dedinaite, P. Claesson, R.
Willumeit-Romer, Structure of DPPC-hyaluronan interfacial layers - effects of molecular weight and
ion composition, Soft Matter 12(3) (2016) 729-740.
[73] J.E. Scott, C. Cummings, A. Brass, Y. Chen, Secondary and tertiary structures of hyaluronan in
aqueous solution, investigated by rotary shadowing-electron microscopy and computer simulation.
Hyaluronan is a very efficient network-forming polymer, Biochemical Journal 274(3) (1991) 699-705.
[74] P. Ghosh, N. Hutadilok, N. Adam, A. Lentini, Interactions of Hyaluronan (Hyaluronic-Acid) with
Phospholipids as Determined by Gel-Permeation Chromatography, Multi-Angle Laser-Light-
Scattering Photometry and H-1-Nmr Spectroscopy, Int J Biol Macromol 16(5) (1994) 237-244.
[75] A. Taglienti, F. Cellesi, V. Crescenzi, P. Sequi, M. Valentini, N. Tirelli, Investigating the
interactions of hyaluronan derivatives with biomolecules. The use of diffusional NMR techniques,
Macromol Biosci 6(8) (2006) 611-622.
[76] I. Pasquali-Ronchetti, D. Quaglino, G. Mori, B. Bacchelli, P. Ghosh, Hyaluronan–phospholipid
interactions, Journal of structural biology 120(1) (1997) 1-10.
[77] C. Altenbach, J. Seelig, Ca-2+ Binding to Phosphatidylcholine Bilayers as Studied by Deuterium
Magnetic-Resonance - Evidence for the Formation of a Ca-2+ Complex with 2
Phospholipidmolecules, Biochemistry 23(17) (1984) 3913-3920.
[78] J. Marra, J. Israelachvili, Direct Measurements of Forces between Phosphatidylcholine and
Phosphatidylethanolamine Bilayers in Aqueous-Electrolyte Solutions, Biochemistry 24(17) (1985)
4608-4618.
[79] K.C. Morrell, W.A. Hodge, D.E. Krebs, R.W. Mann, Corroboration of in vivo cartilage pressures
with implications for synovial joint tribology and osteoarthritis causation, P Natl Acad Sci USA
102(41) (2005) 14819-14824.
66
[80] T. Zander, D.C.F. Wieland, A. Raj, M. Wang, B. Nowak, C. Krywka, A. Dedinaite, P.M. Claesson,
V.M. Garamus, A. Schreyer, R. Willumeit-Romer, The influence of hyaluronan on the structure of a
DPPC-bilayer under high pressures, Colloid Surface B 142 (2016) 230-238.
[81] L.F. Braganza, D.L. Worcester, Hydrostatic pressure induces hydrocarbon chain interdigitation
in single-component phospholipid bilayers, Biochemistry 25(9) (1986) 2591-6.