Influence of Hyperbranched Multi-methacrylates for Dental NeatResins on Proliferation of Human Gingival Fibroblasts

Qichun Wan, Deborah Rumpf, Scott R. Schricker, Angelo Mariotti, and Bill M. Culbertson*

College of Dentistry, The Ohio State University, 305 West 12th Street, P.O. Box 182357,Columbus, Ohio 43218-2357

Received October 2, 2000; Revised Manuscript Received November 13, 2000

We have previously shown that hyperbranched multi-methacrylate (H-MMA)-modified dental resins haveVLC activities, lower polymerization shrinkage, and improved mechanical properties, compared to the 2,2-bis[4-(2-hydroxy-3-methacryloyolxypropoxy)phenyl]propane/triethyleneglycol dimethacrylate (BisGMA/TEGDMA) neat resin. The results are due to the unique molecular structure and high molecular weight ofH-MMA intermediates. The purpose of this study was to evaluate the biocompatibility of H-MMA-modifieddental neat resins. The cell proliferation of three human gingival fibroblast strains on either H-MMA, BisGMA/TEGDMA, or a polystyrene disk was examined. Following 10 days of cell proliferation, there was no statisticaldifference in cell number between H-MMA-modified and unmodified resin disks. H-MMA-modified resinshad less free monomer leaching than the unmodified resin but showed similar properties in water sorptionand contact angle values. All these results suggest that the biocompatibility of H-MMA-modified dentalneat resins is as good as that of commercially used BisGMA/TEGDMA resin and H-MMA has potentialapplications in dental composites.

Introduction

The current widely used 2,2-bis[4-(2-hydroxy-3-meth-acryloyolxypropoxy)phenyl]propane (BisGMA)- and trieth-yleneglycol dimethacrylate (TEGDMA)-based dental com-posite resins have exhibited satisfactory clinical performance,especially for anterior restoratives. However, there are somedeficiencies associated with using this resin matrix, such assusceptibility to water sorption, incomplete conversion ofdouble bonds, and high polymerization shrinkage. Over thepast several decades, extensive efforts have been made todevelop new resin systems, including BisGMA derivatives,1-3

fluorinated dimethacrylates,4-6 dimethacrylates with highmolecular weight and rigid structure,7-9 urethane dimethacryl-ate,10-12 and spiro-ortho-ester based monomers.13-16 Amongthem, high molecular weight multi-methacrylates have beenconsidered as one of the most promising resin system toreduce the polymerization shrinkage and improve the me-chanical properties. Significantly, recently developed hyper-branched multi-methacrylates have been proved to be usefulas comonomer candidates in formulating dental resinsystems.17-19

In recently published papers, we reported the syntheses,characterization, and copolymerization of hyperbranchedmulti-methacrylates and evaluation of these hyperbranchedpolyesters as modifiers for formulating dental composites.20,21

It is generally considered that modification of dentalcomposites with high molecular weight polymers will reducetheir polymerization shrinkage and improve their mechanicalproperties, but at the expense of increasing formulationviscosities and decreasing the polymerization activities.

Hyperbranched polymers could improve the mechanicalproperties and decrease the polymerization shrinkage whilemaintaining lower viscosity due to their unique molecularstructure. However, little is known about their influences onthe biocompatibility with the vital tissues, especially gingivalfibroblasts, since such dental composites often come intointimate contact with periodontal tissue.

There are three stages for the biocompatibility evalua-tion: initial screening test; small animal testing; long-termmammals or human subject testing.22 Among them, the invitro cell culture technique is the easiest and cheapest methodand has been used to evaluate the cellular behavior on dentalcomposites.23-25 Generally, the biocompatibility of dentalcomposites is good, especially when they are highly filled.26,27

However, the release of unreacted monomers and compo-nents or degradation products of dental materials is consid-ered a major factor causing tissue inflammation such as pulpsensitivity and death or cell lipid metabolism activity.28-32

Hanks et al.29 studied the inhibitory effects of resin compo-nents on DNA and protein synthesis of Balb/c 3T3 mousefibroblasts. They found that initiators such as camphorquino-ne and ethoxylated Bisphenol A were the most cytotoxiccomponents, while BisGMA, UDMA, TEGDMA, and Bisphe-nol A showed moderate cytotoxicity. Ratanasathien et al.33

found that the cytotoxicity ranking of the unpolymerizedmonomers were BisGMA> UDMA > TEGDMA >>>HEMA. It was also found that the monomethyl ether ofhydroquinone (MEHQ) and DMAEMA could inhibit thehamster oral epithelial growth and BPO and DMAEMAcould affect cell lipid synthesis.32

The cytotoxicity of acrylates was correlated to theirlipophilicity, indicating that monomer might incorporate into

* To whom correspondence should be addressed. Tel. (614) 292-0777.Fax: ext. 9422. E-mail: culbertson.4@osu.edu.

217Biomacromolecules 2001,2, 217-222

10.1021/bm000101p CCC: $20.00 © 2001 American Chemical SocietyPublished on Web 12/21/2000

the lipid bilayer of the cell membrane. It was also foundthat acrylates were more toxic than the methacrylates withsimilar structure and the presence of a hydroxyl group alsoincreased the cytotoxicity.34

The impurity in commercial BisGMA may have anestrogenesis effect on the development, growth, and normalfunction of estrogen-sensitive tissues.35 It was found thatBisGMA exerts a small but detectable effect on estrogen-sensitive tissues. This may be due to existence of a knownxenoestrogene Bisphenol A, a starting compound for makingBisGMA.

For the visible-light-cured dental resins, Tiba comparedthe biocompatibility of a multi-methacrylate-containing resinwith the conventional BisGMA/TEGDMA, using cell culturetechniques of human gingival fibroblasts. The results revealedthat the multi-methacrylate-based resins favored the humangingival fibroblast cell growth.36

Young37 evaluated the effect of the amount of TEGDMAin dental resin on the cell proliferation by using a fluoro-metric cell proliferation assay and a radiolabeled proteinassay. It was found that materials without TEGDMA resultedin the greater cell proliferation.37

The objectives of this initial study were to examine theinfluence of dental neat resins with H-MMA modifiers onproliferation of human gingival fibroblasts. The acetoneextraction, water sorption, and contact angle have also beenstudied. It is hypothesized that hyperbranched multi-meth-acrylates, with less free monomer leaching, could be usedas modifiers in dental applications. BisGMA/TEGDMAresins modified with hyperbranched, multi-methacrylates

should have less free monomer(s) in the polymerizedformulations because the H-MMA-based multi-methacrylateshave significantly higher molecular weight than BisGMAor TEGDMA dimethacrylates and the large number ofmethacrylate double bonds on the hyperbranched polymersshould tether more of the BisGMA and TEGDMA moleculesinto the cross-linked matrix.

Materials and Methods

Materials. 2,2-Bis[4-(2-hydroxy-3-methacryloyolxypro-poxy)phenyl]propane (BisGMA) and triethyleneglycol dimeth-acrylate (TEGDMA) were received from Cook Compositesand Polymers (Kansas, MO) and Aldrich Chemical Co.(Milwaukee, WI), respectively. Camphorquinone (CQ) and2-(N,N-dimethylamino)ethyl methacrylate (DMAEMA) werealso used as received from Aldrich Chemical Co. Hyper-branched multi-methacrylates (H20-MMA and H30-MMA)were synthesized from commercial (Boltorn) hyperbranchedpolyesters, supplied by Perstorp Corp. In the publishedprocedures,20,21 the Boltorn polyesters were esterified withmethacryloyl chloride or methacrylic anhydride, with thepolymers purified and characterized by FT-IR, NMR (1H and13C), and gel permeation chromatography (GPC).20,21 Mo-lecular weights (MW) of the multi-methacrylates, estimatedby GPC, were 3075 and 5340, compared to BisGMA having512. The general structure of the monomers used or preparedin this study are shown in Figure 1.

Preparation of VLC Disk Samples. All of the resinformulation contained 0.5 wt % of initiator CQ and 1.0 wt

Figure 1. Structures of BisGMA, TEGDMA, and H-MMA monomers used in this study

218 Biomacromolecules, Vol. 2, No. 1, 2001 Wan et al.

% of co-initiator DMAEMA. Two experimental resins wereformulated by mixing 10% of hyperbranched multi-meth-acrylates H20-MMA or H30-MMA with BisGMA/TEGD-MA (50:50 wt/wt). BisGMA/TEGDMA (50:50 wt/wt) servedas control. In addition, tissue culture polystyrene disks wereused as a negative control.

To make the polymer disks, a stainless steel mold withan inner diameter of 15 mm and a thickness of 2 mm wasused. The mold was put against a clean glass slide. Afterthe above mixture was added to the mold, another glass slidewas placed on top of the mold. It was then cured for 1 minby using a light curing unit (Belle Glass HP TEGLITE, 400mW/cm2) on each side and further cured for another 4 minusing a Triad 2000 light oven. In the disk preparation, carewas given to avoid any surface defects since they wouldinfluence the cell attachment and behavior.

Isolation and Culture of Human Gingival Fibroblasts.Fibroblasts used in these experiments were obtained fromthree different maxillary gingival papilla biopsies (see Table1). Cells were cultured and propagated as previouslydescribed.38 Cells used for these experiments were thawedfrom The College of Dentistry Cryogenic Storage Facilityand grown in minimal essential media (MEM) supplementedwith 10% fetal bovine serum,L-glutamine, penicillin (100units/mL), and streptomycin (100 mg/mL). Only fibroblastsbetween the third and fifth passages were used in theexperiments described.

Cell Proliferation Assay. Cell proliferation was measuredover a 10-day period on dental neat resins. By use of aCoulter counter to determine cell number, cells were platedat 20 000 cells per disk in 1 mL of MEM supplemented with10% fetal bovine serum,L-glutamine, penicillin (100 units/mL), and streptomycin (100 mg/mL) on BisGMA/TEGDMA,H20-MMA, H30-MMA, or polystyrene control disks. Eachdisk was used multiple times, and disks were cleanedbetween uses as follows. Sixteen disks were put into a 250-mL beaker with 100 mL of 1% Alconox detergent solutionand ultrasonically cleaned (Jelenko Ultrasonic Cleaner,Armonk, NY) at 44-48 kHz for 20 min. The Alconox wasthen removed and 100 mL of distilled water was added,followed by ultrasonic treatment for another 20 min. Diskswere rinsed with 50 mL of distilled water approximately 15times to make sure all the Alconox detergent was removed.The clean disks were pop-dried by using Kleenex tissue paperand stored in a clean container. The silicon O-rings werecleaned using the same procedure.

Prior to use, all disks and silicon O-rings were sterilizedovernight in ultraviolet light. The disks were then placedinto the wells of 24-well cell culture plates, and siliconeO-rings were used to hold disks in place. At days 1, 3, 5, 7and 10, the cell number was determined using the AlamarBlue assay. Alamar Blue (AB) is a nondestructive methodmeasuring cell metabolism activities.39 After incubation with

cells, the Alamar Blue is removed, a new fresh medium isadded. Cells are still alive for the next period testing. Thismethod is shown very effective in testing the cell metabolismover a period of time intervals. The AB dye was diluted fromthe stock solution with growth medium in a 1:10 dilution.After the cell plating incubated for the desired time period,it was taken out from the incubation. The growth mediumwas removed from each well, and 1 mL of AB-mediasolution was added. Then the cell culture plates wereincubated for 3 h. During the same time, the dilution plateswere prepared. That is, 1.9-mL of double distilled water waspipetted into a clean 24-well cell culture plate. Afterincubation, 100µL of liquid was drawn from each well andadded to the above dilution plates, containing 1.9 mL ofdistilled water. This gave a dilution of 1:20. Next, 100µLfrom each dilution well was placed into a corresponding wellof an opaque 96-well plate. The fluorescence intensities weremeasured using a Perkin-Elmer luminescence spectrometerLS 50 B (Perkin-Elmer, USA). The excitation wavelengthwas 560 nm with a slit width of 2.5 nm, and the emissionwavelength was 590 nm with a slit width of 5 nm. Relativefluorescence intensity is the intensity of samples with cellsminus that of the Alamar Blue blank with no cells.

Contact Angle Measurement.The contact angle wasmeasured by placing a 2-µL drop of double distilled wateronto the surface of the material using a NRL contact anglegoniometer (Rame-Hart, Inc., Mountain View, NJ). Thepreparation and cleaning procedure used for the disk sampleswere the same as those for cell proliferation testing statedabove. Three disk specimens were tested with each on fourdifferent locations. Thus, 12 readings were made for eachmaterial.

Water Sorption. For each resin, three visible-light-cureddisks (15 mm in diameter and 0.8 mm in thickness) wereconditioned to a constant weight in a desiccator and thenimmersed in distilled water at 37°C. At different timeintervals, the specimens were removed from the water, lightlyblotted with a paper tissue, and weighed. After immersionfor about 1 year, the specimens were removed and recon-ditioned to a constant weight in a desiccator. The watersorption for each specimen was determined from the differ-ence in weight between the specimen immersed for thepredetermined time intervals and the reconditioned specimen.The BisGMA/TEGDMA neat resin served as control.

Acetone Extraction. Disk samples (15 mm in diameterand 0.8 mm in thickness) were photopolymerized for a totalof 5 min, weighed (W0), and then extracted by acetone for 4days. Since the matrix is organic, it has some affinity withorganic solvents such as acetone. After the extraction, thesamples were immediately weighed (Wa). The weight in-crease from the starting weight due to absorption of acetonewas calculated. Then the samples were put into vacuum ovenat room temperature for 3 days and weighed again. Afterone more day in a vacuum oven, samples were weighedagain. Repeated dry and weight processes were done untilthe difference of the last two weights was less than 0.0005g. The weight was recorded asWf. The percentage of

Table 1. Cell Strain Information

strain donor sex donor age passages used

A M 39 T3B M 75 T3C F 36 T3

Biocompatibility of Modified Dental Composites Biomacromolecules, Vol. 2, No. 1, 2001 219

absorption is calculated by (Wa - Wf) × 100%/Wf and thepercentage weight loss is calculated by (W0 - Wf) × 100%/W0.

Results and Discussion

Since AB is relatively new for testing the effects of dentalmaterials on gingival fibroblast cells, the repeatability of theAlamar Blue fluorescence intensity procedure used here wasfirst examined. It was found that the AB fluorescent readinghad very good repeatability. In addition, a very goodassociation exists between the cell number and fluorescentreading for each cell strain. That is, for each cell strain, thefluorescent reading could be explained to a great extent bythe appearance of cell numbers. The results of the averagerelative fluorescence intensity of three cell strains over theperiod of 10 days are presented graphically in Figure 2.

It can be seen from Figure 2 that cell growth occurred onall materials over the 10-day period. Generally, for the first5 days, only a slight increase was found. Then, dramaticchanges occurred for days 7 and 10. This trend held for everymaterial. However, different materials showed a differentextent of change. Polystyrene (Pst) control always hadhighest fluorescent intensity of the four materials. Hyper-branched H20-MMA- or H30-MMA-modified resins hadfluorescent intensity similar to that of the unmodifiedBisGMA/TEGDMA.

Figure 3 provides a comparison of relative fluorescentintensity of four materials contacted with three strains at day10. For all strains, the Pst showed significantly higherfluorescent intensity than the other three experimentalmaterials, indicating that Pst had the best interaction for thecell growth. The fluorescent intensities of modified andunmodified experimental resins were not statistically differ-ent. The results suggested that hyperbranched multi-meth-

acrylate-modified resins have cell interactions or biocom-patibility at least as good as the BisGMA/TEGDMA control.While it has become commonly accepted that hybridizeddentin acts as a barrier to protect dentin and pulp fromresidual monomers,40 there remains a worthy goal of havingno leached monomers in the oral cavity and restoratives thatallow for normal cell growth. Thus, biocompatibility, as usedabove, says only that the hyperbranched-based materialsprepared in this work allows cell growth as good or betterthan the commonly used BisGMA/TEGDMA control, animportant consideration in the possible use of such materals.

Many factors will influence the interactions of materialswith the cells, including the material surface physical orchemical characteristics, amount and properties of leachableor degradable materials, cell properties, etc. The surfaceproperties include the surface roughness, wettability, andelectronic charge and van der Waals forces.41 It is knownthat surface charges could promote cell attachment. Forexample, commercial polystyrene tissue culture plates havebeen plasma or corona discharge treated to give a surface

Figure 2. Cell growth curves for four materials over 10 days (average of three cell strains).

Figure 3. Relative fluorescent intensity of four materials contactedwith three cell strains at day 10.

220 Biomacromolecules, Vol. 2, No. 1, 2001 Wan et al.

with a negative charge, which is suitable for most cellattachment and growth.42,43In the present study, disk sampleswere prepared against the smooth glass surface and washedultrasonically by distilled water. Presumably all the sampledisks had similar surface topography and features such ashills or pits. Another important factor for cell interaction isthe surface chemical characteristics including functionalgroups and hydrophilicity. Some functional groups such as-OH, -COOH, -COOR, -CHO, -CONH2, and -NH2

will influence the cell attachment or proliferation. For ourmaterials, BisGMA contains some hydroxyl groups, whileothers (TEGDMA and H-MMAs) have only ether or estergroups. Modified and unmodified resins have similar chemi-cal features. This could be seen from the contact angle test(Table 2). Modified and unmodified resins have similarsurface chemical features.

Another major consideration may be the amount orproperties of materials leaching to the media, especially forthe VLC neat resins, which are polymerized at roomtemperature with some free monomers trapped in the polymermatrix. When contacted with water or other media, the neatresins will absorb water or other liquids. This processfacilitates free monomers leaching out of the resins, espe-cially for small molecules such as TEGDMA and otheradditives.44 In general, the amounts of leachable freemonomers and water sorption play important roles in thebiocompatibility.

The results of percentage weight loss and absorption aregiven in Table 3. It can be seen from Table 3 that all thesamples absorbed acetone with the amount of absorptionabout 10%. Materials with addition of 10% of H-MMAsshowed less absorption than the BisGMA/TEGDMA. How-ever, compared to the high acetone absorption, all materialshad a smaller amount of weight loss. H-MMA-modifiedresins had about 20% less weight loss than the BisGMA/TEGDMA. This may be due to the high functionality ofH-MMAs, which could tether more monomers into thepolymer network. Low free monomers usually produces lesstoxicity, locally or systemically.

When dental composites are in contact with water or oralfluids, two processes, absorption of water and leaching ofsmall molecules, occur simultaneously. This is shownthrough the water sorption curves (Figure 4). First, neat resinsquickly absorbed water during the first few days with aweight increase. Small molecules then began to leach out.When the leaching rate became higher than the absorptionrate, the weight began to decrease. Finally these twoprocesses reached a steady state and the percentage weightchange remained constant. In our study, H-MMA-modifiedor unmodified resins showed similar water sorption trends

Figure 4. Water sorption of VLC neat resins in distilled water at 37 °C for about 1 year.

Table 2. Contact Angle Test for Neat Resins

resins contact angle (deg)

BisGMA/TEGDMA 63.2 (2.5)10% H20-MMA 65.1 (2.2)10% H30-MMA 65.3 (3.0)polystyrene 78.9 (3.6)

Table 3. Extraction of Neat Resins of BisGMA/TEGDMA (50/50,wt/wt) Modified by 10% H-MMAa

weight loss (%) absorption (%)

BisGMA/TEGDMA control 0.92 (0.37) 9.52 (0.10)10% H20-MMA 0.82 (0.38) 9.28 (0.28)!0% H30-MMA 0.79 (0.45) 7.12 (0.34)10% H40-MMA 0.73 (0.51) 1.00 (0.47)

a BisGMA/TEGDMA 50/50 (w/w) with 0.5 wt % initiator (CQ) and 1.0wt % co-initiator (DMAEM). Resin specimens were 15.0 mm in diameter× 0.7 mm in thickness; each entry is the mean value (standard deviation)for a group of five specimens (N ) 5).

Biocompatibility of Modified Dental Composites Biomacromolecules, Vol. 2, No. 1, 2001 221

over the period of 1 year and had a similar amount of watersorption. This could partly explain why modified andunmodified resins had similar cell interactions.

Conclusions

Hyperbranched multi-methacrylate-modified resins had asmaller amount of free monomer leaching out than theunmodified resin but showed similar properties in watersorption and contact angle. In addition, in contact with threehuman gingival fibroblasts over the 10 days, the H-MMA-modified resins had similar cellular responses or proliferationto the commercially used BisGMA/TEGDMA resin systems.On the basis of this preliminary testing, it could be concludedthat hyperbranched polymers have acceptable biocompat-ibility and have the potential to be used in dental applications.

All these results are in favor of our initial hypothesis.Future studies will include developing a better way tosynthesize and purify the hyperbranched multi-methacrylates.In addition, the modified mixtures will be combined withsuitable fillers to form composites and their properties willbe tested. Moreover, a longer-term biocompatibility testshould be conducted.

Acknowledgment. The authors express appreciation toNIH (DE 13433) and OSU (an Interdisciplinary Seed Grant)for partial support of this work.

References and Notes

(1) Kalachandra, S.; Taylor, D. F.; DePorter, C. D.; Grubbs, H. J.;McGrath, J. E.Polymer1993, 34, 778.

(2) Kawaguchi, M.; Fukushima, T.; Horibe, T.Dent. Mater. J. 1988, 7,174.

(3) Holter, D.; Frey, H.; Mulhaupt, R.Polym. Prepr. 1997, 38 (2), 84.(4) Tanaka, J.; Inoue, K.; Masamura, H.; Matsumura, K.; Nakai, K.;

Inoue, K.Dent. Mater. J.,1993, 12 (1),1.(5) Stansbury, J. W.; Choi, K. M.; Antonucci, J. M.Polym. Prepr. 1997,

38 (2), 96.(6) Wang, G.; Culbertson, B. M.; Xie, D.; Seghi, R. R. J. Macromol.

Sci., Pure Appl. Chem. 1999, A36 (2), 225.(7) Culbertson, B. M.; Tong, Y.; Wan, Q.Polym. Prepr.1997, 38 (1),

217.(8) Culbertson, B. M.; Xu, J.; Tiba, A.Polym. AdV. Technol. 1999, 10

(4), 206.(9) Sankarapandian, M.; Shobha, H. K.; Kalachandra, S.; Taylor, D. F.;

Shultz, A. F.; McGrath, E.Polym. Prepr.1997,38 (2), 92.(10) Antonucci, J. M.; Brauer, G. M.; Termini, D. J.J. Dent. Res.1980,

59 (1), 35.(11) Matsukawa, S.; Hayakawa, T.; Nemoto, K.Dent. Mater. 1994, 10,

343.

(12) Burke, T. C.; Dickens, S. H.; Floyd, C. J. E.J. Dent. Res. 2000, 79,no. 1786, 367.

(13) Chappelow, C. C.; Pinzino, C. S.; Power, M. D.; Eick, J. D.Polym.Prepr. 1997, 38 (2), 90.

(14) Moszner, N.; Volkel, T.; Rheinberger, V.; Klemm, E.Macromol.Chem. Phys. 1997,198 (3), 749.

(15) Zeuner, F.; Moszner, N.; Rheinberger, A.Macromol. Chem. Phys.1996,197 (9), 2745.

(16) Stansbury, J. W.J. Dent. Res. 1992, 71, 1408.(17) Klee, J. E.; Walz, U.; Holter, D.; Frey, H.; Mulhaupt, R. Angew.

Macromol. Chem., 1998, 260, 71.(18) Bengt, R.; Shi, W. WO 96/07688, 1996.(19) Klee, J. E.; Walz, U.; Holter, D.; Burgath, A.; Frey, H.; Mulhaupt,

R. WO 98/36729, 1998.(20) Wan, Q.; Schricker, S. R.; Culbertson, B. M.J. Macromol. Sci., Pure

Appl. Chem.2000, A37 (11), 1301-1315.(21) Wan, Q.; Schricker, S. R.; Culbertson, B. M.J. Macromol. Sci., Pure

Appl. Chem.2000, A37 (11), 1317-1331.(22) Craig, R. Biocompatibility of dental materials. InRestoratiVe Dental

Materials, 10th ed.; Mosby: St. Louis, MO, 1997; p 137.(23) Hensten-Petterson, A.; Victorian, L.Acct. Odontol. Scand.1981, 39,

101.(24) Hanks, C. T.; Anderson M.; Craig, R. G.J. Oral Pathol.1981, 10,

101.(25) Wataha, J. C.; Hanks, C. T.; Strawn, S. E.; Fat, J. C.J. Oral Rehabil.

1994, 21, 453.(26) Meryon, S. D.; Browne, R. M.J. Oral Rehabil.1983, 10 (4), 363-

372.(27) Cox, C. F.; Hafez, A. A.; Akimoto, N.; Otsuki, M.; Suzuki, S.; Tarim,

B. Am. J. Dent, 1998, 11, Spec No. S 55-63.(28) Caughman, W. F.; Caughman, G. B.; Dominy, W. T.; Schuster, G.

S. J. Prosthet. Dent., 1990, 63 (5), 513.(29) Hanks, C. T.J. Dent. Res.1991, 70 (11), 1450-1455.(30) Wataha, J. C.; Hanks, C. T.; Strawn, S. E.; Fat, J. C.J. Oral Rehabil.

1994, 21, 453-462.(31) Gerzina, T. M.; Hume, W. R.J. Dent.1996, 24 (1-2), 125-128.(32) Lefebvre, C. A.J. Biomater. Sci.: Polym. Ed.1996, 7 (11), 965-

976.(33) Ratanasathien, S.J. Dent. Res.1995, 74 (9), 1602-1606.(34) Yoshii, E.J. Biomed. Mater. Res.1997, 37 (4), 524-527.(35) Mariotti, A.; Hassell, T.; Jacobes, D.; Manning, C. J.; Hefti, A. F.J.

Oral Pathol. Med.1998, 27, 260-266.(36) Tiba, A.; Culbertson, B. M.J. Macromol. Sci., Pure Appl. Chem.

1999, A36 (9), 1209-1226.(37) Young D. P. Proliferation and adhesion of human gingival fibroblasts

on various dental polymer materials. Ph.D. Dissertation, Ohio StateUniversity, 1999.

(38) Mariotti, A. Eur. J. Oral Sci.1998, 106 (6), 1022-1027.(39) Page, B.; Page, M.; Noel, C.Int. J. Oncol.1993, 3, 473.(40) Nakabayashi, N.; Pashley, D. H.Quintessence(Chicago) l998.(41) Von Recum A. F.; Van Kooten, T. G.J. Biomater. Sci., Polym. Ed.

1995, 7 (2), 181.(42) Ramsey, W. S.; Hertl, W.; Nowlan, E. D.; Binkowski, V. G.In Vitro

1984, 20, 802.(43) Amstein C. F.; Hartman, P. A.J. Clin. Microbiol. 1975, 2, 46.(44) Gerzina, T. M.; Hume, W. R.J. Dent. Res.,1995, 74(1) 369.(45)

BM000101P

222 Biomacromolecules, Vol. 2, No. 1, 2001 Wan et al.