1

Abstract: We used a polymicrobial (PM) biofilm model to examine associations of bacterial adhesive-ness with surface characteristics of various dental materials. Four types of dental materials (apatite pellet, zirconia, ceramic, and composite resin) with rough and mirror surfaces were used. Surface roughness, surface free energy, zeta potential, and colony-forming units (CFUs) of the biofilm forma-tions were measured. Biofilms were cultured for 24 h under anaerobic conditions, plated onto blood agar medium, and anaerobically cultured for 4 days. After culturing, CFU per mm2 was calculated, and samples were observed under a scanning electron microscope. Means and standard deviations of the experimental data were estimated, and one-way ANOVA and Tukey multiple comparison assays were performed. Pearson correlation coefficients were obtained for the CFU

and surface characteristics. Surface roughness and surface free energy appeared to affect generation of PM biofilms on oral materials, and zeta potential was involved in generation of PM biofilms on mirror-ground oral materials.

Keywords: adhesiveness; polymicrobial biofilm; surface free energy; zeta potential.

IntroductionDental plaque, a biofilm present in the oral cavity, causes two of the most common oral infections: caries and gum disease (1,2). The mechanism by which plaque adheres to and forms on the surface of teeth and restoration mate-rials has been extensively studied (3-5). Saliva-derived pellicles immediately form on the surfaces of thoroughly cleaned teeth and attract bacteria through chemical or electrostatic interactions or by antigens on the surfaces of bacterial cells (6). The most common organisms present in oral biofilms are the initial colonizer Strep-tococcus spp., early colonizer Veillonella spp., middle colonizer Porphyromonas gingivalis, and Fusobacterium nucleatum (7,8). The ability of individual bacteria to generate biofilms increases dramatically in the presence

Original

Journal of Oral ScienceJ-STAGE Advance Publication: November 22, 2017

Relevance of surface characteristics in the adhesiveness of polymicrobial biofilms to crown restoration materialsAyako Teranaka1), Kiyoshi Tomiyama2), Katsura Ohashi3), Kaori Miyake3),

Tota Shimizu4), Nobushiro Hamada5), Yoshiharu Mukai2), Satoshi Hirayama1), and Tomotaro Nihei3)

1)Department of Operative Dentistry, Nihon University School of Dentistry at Matsudo, Matsudo, Japan2)Division of Cariology and Restorative Dentistry, Department of Oral Function and Restoration,

Graduate School of Dentistry, Kanagawa Dental University, Yokosuka, Japan3)Division of Clinical Biomaterials, Department of Oral Science, Graduate School of Dentistry,

Kanagawa Dental University, Yokosuka, Japan4)Division of Prosthodontic Dentistry for Function of TMJ and Occlusion,

Department of Oral Function and Restoration, Kanagawa Dental University, Yokosuka, Japan5)Division of Microbiology, Department of Oral Science, Graduate School of Dentistry,

Kanagawa Dental University, Yokosuka, Japan

(Received November 1, 2016; Accepted May 12, 2017)

Correspondence to Dr. Tomotaro Nihei, Division of Clinical Biomaterials, Department of Oral Science, Graduate School of Dentistry, Kanagawa Dental University, 82 Inaoka-cho, Yokosuka, Kanagawa 238-8580, JapanFax: +81-46-822-8864 E-mail: nihei@kdu.ac.jpdoi.org/10.2334/josnusd.16-0758DN/JST.JSTAGE/josnusd/16-0758

2

of the Veillonella genus (9). Increased plaque growth via repeated adhesion and desorption increases anaerobiosis in lower layers of biofilm and makes bacterial flora more complex (10). The resulting thick biofilm is not easily removed, except by physical means. It therefore accumu-lates and causes deliming of dentin and gum irritation. Hence, the formation of oral biofilms must be controlled to prevent caries and gum disease in the oral cavity. Adhesion of plaque and pellicle is reportedly affected by surface free energy (SFE) and the surface roughness of the adherend (11,12). Busscher et al. reported very low plaque adhesion when the SFE of the tooth surface was less than 50 mN/m (13). Thus, a reduction in the SFE of teeth and dental materials may reduce the extent of ongoing caries and gum disease.

The materials used for restorative and prosthetic devices have undergone major developments in recent years. In addition, a great variety of dental materials are now in use. The quantity of plaque generated on the surfaces of these materials has varied substantially in clinical trials. However, the surface characteristics of dental materials, as well as the adhesion and generation of biofilms on these surfaces, are rarely reported (14,15).

The present study used a polymicrobial (PM) biofilm model previously reported by Exterkate et al. to assess bacterial adhesiveness (16) by varying crown restoration materials and their surface roughness. To reveal biofilm adhesion factors, associations of bacterial adhesion with various surface characteristics were investigated in rela-tion to roughness, SFE, and zeta potential. This model is based on next-generation sequencing of biofilm, which comprises at least 21 oral cavity-derived bacteria (17). One bacterium may have multiple strains; thus, an environment approximating a biofilm could contain more than 700 strains of oral cavity bacteria.

Materials and MethodsSample preparationThe experiments were conducted on four materials (Table 1): sintered apatite (HAP, apatite pellet APP-100, HOYA Technosurgical, Inc., Tokyo, Japan), zirconia (Zr,

Lava Plus Zirconia XL, 3M ESPE, Seefeld, Germany), ceramic (VB, VITABLOCS Mark II for CEREC/in Lab, VITA, Bad Säckingen, Germany), and composite resin (CR, Clearfil Majesty ES2, MX, Kuraray Noritake Dental Inc., Tokyo, Japan). HAP and Zr were used at their commercially available sizes (areas and thickness of HAP and Zr were 10 × 10 mm2 and 2 mm, and 10 × 10 mm2 and 0.8 mm, respectively). For VB, a CAD/CAM block (area and thickness of 10 × 12 mm2 and 15 mm, respectively) was cut out into a 1 mm-thick block with a low-speed precision cutting instrument (ISOMET, Buehler, Waukegan, IL, USA). CR was stamped onto a 10 × 10 × 1 mm3 plastic mold and pressure-welded at 2 kg; the upper and lower surfaces were light-irradiated with a quartz-tungsten-halogen light curing unit (Optilux LCT, Sybron Dental Specialties, Inc., Tokyo, Japan) through a cover glass (Micro-slide glass, Matsunami Glass Ind., Ltd., Kishiwada, Japan) for 60 s.

Two sample groups with different surface roughnesses were prepared. All surfaces in the roughly ground surface (RS) group were ground with a #600 waterproof abrasive paper (FUJI STAR, Sankyo Rikagaku Co., Ltd., Okegawa, Japan) under running water; those in the mirror-ground surface (MS) group were ground with a #2000 abrasive paper and a 0.5-µm diamond slurry (Maruto Instru-ment Co., Ltd., Tokyo, Japan) under running water. All samples were adjusted to a thickness of 0.8 mm. The ground samples were ultrasonically cleaned for 30 min to remove all extraneous surface matter, dried at 50°C for 2 h, loaded to a clamp attached inside the stainless steel top cover of a 24-well culture plate, and sterilized at 60°C for 20 min with ethylene oxide gas.

Preparation of the saliva sampleSaliva was collected by mastication of Parafilm (Pechiney Plastic Packaging, Chicago, IL, USA) under ice applica-tion. The subject was a healthy adult who had taken no antibiotics or antibacterial agents during the previous 3 months and possessed a row of natural teeth not infected with caries or gum disease. The collected saliva was immediately filtered through sterilized glass wool (NRK

Table 1 Dental materials used in this studySample Chemical composition Manufacturer Lot No. CodeApatite pellet APP-100 Ca10(PO4)6OH2 HOYA Technosurgical, Inc., Tokyo, Japan 10939210 HAPLava Plus zirconia ZrO2, Y2O3 3M ESPE, Seefeld, Germany 461045 ZrVITABLOCS Mark II for CEREC/in Lab

SiO2, AlO3, NaO, KO, CaO2 VITA, Bad Säckingen, Germany 28300 VB

Clearfil Majesty ES2 Silanated barium glass filler, pre-polymerized organic filler, Bis-GMA, hydrophobic aromatic dimethacrylate, camphorquinone

Kuraray Noritake Dental Inc., Tokyo, Japan 2S0002 CR

3

GRW-10, Nippon Rikagaku Kikai Co., Ltd., Tokyo, Japan) to remove necrotic tissue, diluted to 70 vol% with sterilized glycerin, frozen, and stored at −80°C. Human saliva was used with approval from the Kanagawa Dental University Research Ethical Review Board (No. 206), in adherence with the ethical principles of the Declaration of Helsinki for medical research involving human subjects.

Surface roughnessThe surface roughness of each sample was measured with a contact-type surface profilometer (Surfcom 590A, Tokyo Seimitsu Co., Ltd., Tokyo, Japan) at a measure-ment distance and speed of 4.0 mm and 0.3 mm/s, respectively, with a cutoff value of 0.8 mm. The center-line average roughness (Ra) was obtained and calculated as the surface roughness.

Surface free energyDistilled water and diiodomethane (Wako Pure Chemical Industries, Ltd., Osaka, Japan) were dropped separately onto the surface of the sample, and their contact angles with the surface were measured at 25°C by using an automatic contact angle meter (Model DCA-VZ, Kyowa Interface Science Co., Ltd., Niiza, Japan). SFE was then calculated by using the Owen and Wendt theoretical formula with FAMAS surface free energy calculation software (Kyowa Interface Science Co., Ltd.).

Zeta potentialThe zeta potential of the sample surface was measured with a flow potential-type zeta potential measuring instrument for solid samples (SurPASS, Anton Paar, Graz, Austria). Measurements were performed in a 0.001-mol/L sodium chloride aqueous solution (Wako Pure Chemical Industries, Ltd.) under a pressure of 400 hPa.

Preparation of PM biofilm and measurement of colony-forming units (CFUs)Frozen saliva was diluted to 50 times in buffered McBain semi-defined medium (0.2 wt% sucrose, 50 mmol PIPES, 2.5 g/L mucin, 2.0 g/L Bacto peptone, 2.0 g/L trypticase peptone, 1.0 g/L yeast extract, 0.35 g/L NaCl, 0.2 g/L KCl, 0.2 g/L CaCl2, 0.001 g/L hemin, and 0.0002 g/L vitamin K1), as described by Exterkate et al. (16). The diluted saliva was dispersed onto a 24-well culture plate. Samples were loaded onto the cover of the culture plate, dipped into the culture, and cultivated in an anaerobic atmosphere (CO2: 10.0%, H2: 10.0%, N2: 80.0%) at 37°C for 10 h. As a control, buffered McBain medium without saliva was cultivated in an anaerobic atmosphere for

14 h in the same manner. After cultivation, the samples with their formed PM biofilms were dipped in 2 mL of cysteine peptone water. The PM biofilm was removed from the sample under supersonic vibration (Transonic T780, Elma Electronic GmbH, Stuttgart, Germany) and dispersed with a test tube mixer (Vortex tube mixer VTX-3500, BDL, Turnov, Czech Republic). The dispersed samples were stepwise-diluted with cysteine peptone water. Next, 50 µL of the solution was seeded onto a tryptic soy agar blood plate and cultivated under the same anaerobic conditions at 37°C for 4 days. The average viable bacterial count and its standard deviation per 1 mm2 of sample were determined from the measure-ments. The obtained viable bacterial count per 1 mm2 was log-transformed and analyzed as logCFU/mm2.

Observation of the biofilm surfaceThe biofilm-coated PM sample was dipped into phos-phate buffered saline (0.01 mol) for 30 s, washed with cacodylic acid buffer solution (0.1 mol) for 30 s, and dipped and fixed in a mixed solution of cacodylic acid buffer solution (0.1 mol) and glutaraldehyde (1%) for 1 h. The film was then washed twice (30 s per wash) with cacodylic acid buffer solution (0.1 mol), and dipped stepwise in 50%, 70%, 80%, 90%, and 100% ethanol (15 min per dip at room temperature). After a second dipping in 100% ethanol for 15 min, the film was finally dipped in isoamyl acetate for critical point drying.

The critical point-dried sample was fixed onto a brass sample stage with carbon tape, vacuum-deposited with a 200 Å-thick platinum surface layer, and observed under a scanning electron microscope (SEM; JSM-820, JEOL Ltd., Tokyo, Japan) at an accelerating voltage of 5.0 kV.

Statistical analysisThe means and standard deviations of experimental data were calculated and analyzed with the one-way ANOVA and Tukey multiple comparison assays. The associations of viable bacterial count with surface characteristics were investigated by using the Pearson correlation test at a significance level of 5%. All statistical analyses were performed with IBM SPSS Ver. 20.0 software (IBM Japan, Ltd., Tokyo, Japan).

ResultsCenterline average roughnessTable 2 shows the Ra values of the samples measured with the contact-type surface profilometer. Ra was significantly lower in the MS group than in the RS group (0.024-0.052 µm vs. 0.055-0.249 µm; P < 0.05). In the RS group, Ra was particularly large for CR and

4

significantly lower for Zr as compared with HAP (P < 0.05). There was no significant difference between the Ra values of Zr and VB (P > 0.05). In the MS group, no significant differences in Ra values were noted between HAP, Zr, VB, and CR (P > 0.05).

Surface free energyFigure 1 shows the SFE of the study samples. The SFEs of all samples were less than 50 mN/m and did not differ significantly between the RS and MS groups (P > 0.05), except for CR. The SFE of CR was 41.6 mN/m in the RS group and 38.1 mN/m in the MS group.

Zeta potentialFigure 2 shows the zeta potentials of the samples in

the MS group. Zeta potential was highest in CR (−58.8 mV) and lowest in VB (−39.0 mV). Zeta potential was significantly lower for VB than for the other materials (P < 0.05).

Colony-forming unitsFigure 3 shows the viable counts of bacteria adhered to the samples. In the RS group, viable bacterial counts were significantly lower for Zr and VB than for HAP (P < 0.05) but were similar in HAP and CR (P > 0.05). In the MS group, viable bacterial counts were similar for Zr and VB, as compared with HAP (P > 0.05), but were significantly higher for CR than for the other materials (P < 0.05). The viable bacterial count differed significantly (P < 0.05) between the RS and MS groups regardless

Table 2 Surface roughness of dental materialsRS (S.D.) MS (S.D.)

HAP 0.152 (0.026) a 0.025 (0.007) b Zr 0.055 (0.007) b 0.024 (0.003) b VB 0.153 (0.027) a 0.052 (0.013) b CR 0.249 (0.042) c 0.038 (0.008) b(n = 6)RS: roughness surface, MS: mirror surface. There is no significant difference between values labeled with the same letters for each dental material (one-way ANOVA and post-hoc Tukey multiple comparison tests, P < 0.05). The surface roughness of RS group specimens was 0.055-0.249 µm, and the surface roughness of MS group specimens other than GL was 0.024-0.052 µm.

a bab

0

10

20

30

40

50

60

RS MS RS MS RS MS RS MS

HAP Zr VB CR

ab ab bab

Surf

ace

free

ene

rgy

(mN

/m)

(n = 6)

a

d

c

Col

ony

form

ing

unit

(CFU

/mm

2 )

1.00E+05

1.00E+06

1.00E+07

RS MS RS MS RS MS RS MS

HAP Zr VB CR

105

107

106

ab

cc

dbd

(n = 4)

Zr VBHAP CR

ba

c

d-70

-60

-50

-40

-30

-20

-10

0

Zeta

pot

entia

l (m

V)

Fig. 1 Graph showing the surface free energy of dental mate-rials used in this study (n = 9). There is no significant difference between bars labeled with the same letters for each dental material used (one-way ANOVA and post-hoc Tukey multiple comparison tests, P < 0.05). Surface free energy was less than 50 mN/m in all specimens.

Fig. 3 Graph showing CFU values for bacteria on dental mate-rials. There is no significant difference between bars labeled with the same letters for each dental material (one-way ANOVA and post-hoc Tukey multiple comparison tests, P < 0.05). Significant differences in CFUs between the rough surface (RS) and mirror surface (MS) groups (P < 0.05) were observed for HAP, Zr, VB, and CR. The values for Zr and VB were significantly lower than that for HAP.

Fig. 2 Graph showing the zeta potentials of the MS group for each dental material. There is no significant difference between bars labeled with the same letters for each dental material used (one-way ANOVA and post-hoc Tukey multiple comparison tests, P < 0.05). The zeta potential of CR was significantly higher than that of HAP (P < 0.05).

5

of the material used. Among all samples, counts were remarkably low for the HAP, Zr, and VB materials in the MS group.

Biofilm observationsFigure 4 shows images of PM biofilm adhered to samples in the MS group. The HAP surface was covered with several small bacterial aggregates separated by gaps that revealed the underlying bare surface. The Zr and VB surfaces were sparsely coated with bacteria, which were separated by numerous gaps, exposing parts of the underlying surface of the dental material. However, the CR surface was completely and evenly coated by a layer of bacteria.

Correlations of viable bacterial counts with surface characteristicsThe correlations of viable bacterial count with Ra, SFE, and zeta potential were assessed with the Pearson corre-lation test (Table 3). The correlation coefficients between viable bacterial count and Ra were 0.799 (P = 0.201) and 0.167 (P = 0.833) in the RS and MS groups, respectively, and the correlation coefficients between viable bacterial count and SFE were 0.921 (P = 0.079) and −0.066 (P = 0.934) in the RS and MS groups, respectively. Viable bacterial count was unrelated to zeta potential (correla-tion coefficient −0.835; P = 0.165).

DiscussionDental plaque is a typical biofilm composed of multiple bacterial strains. To generate a biofilm, floating bacterial cells in the saliva first adhere to the tooth surface and form microcolonies. Over time, the bacteria aggluti-

nate, proliferate, and grow into a mushroom-shaped mature biofilm composed of glycocalyx, a capsular polysaccharide synthesized outside the bacterial cells. Biofilm-generating bacteria such as Actinomyces spp., Streptococcus spp., Propionibacterium acnes, and P. gingivalis produce extracellular polysaccharide (EPS) and proliferate or agglutinate, eventually covering the surface with a biofilm (6). Subsequently, within a few minutes, the thickness of the biofilm coating the oral cavity reaches approximately 1 µm (18). The bacteria adhere to the tooth surface through nonspecific interactions such as van der Waals forces, electrostatic interactions, and specific interactions with glycoproteins on the cell surfaces. Non-EPS-producing bacterial strains are also important in biofilm generation (9,19); however, the generation mechanism has not yet been determined.

Previous in vitro biofilm studies used microtiter plates as the substrate (20), mixtures of bacteria and saliva as plantation bacteria (21), and an artificial oral cavity plaque as the biofilm model (22,23). In the present study, we adopted the PM biofilm model (16) described by Exterkate et al., which generates biofilms from various bacteria in the oral cavity. This model also generates and treats multiple independent biofilms in single experi-ments, offers strong adhesion, simultaneously treats different solid phases (substrates), and facilitates easy adjustments of treatment times and drug concentrations (17,24-27). No alternative existing biofilm model satis-fies all these conditions. According to Exterkate et al., their model is superior to conventional biofilm models for several reasons. First, it achieves more active adhe-sion of bacteria to the solid phase than do conventional models, which often exhibit defective adhesion proper-

Table 3 Correlations between surface characteristics and CFU of materials used in the RS and MS groups

Colony-forming units

RS groupSurface roughness 0.799 (P = 0.201)Surface free energy 0.921 (P = 0.079)

MS groupSurface roughness 0.167 (P = 0.833)Surface free energy −0.066 (P = 0.934)Zeta potential −0.835 (P = 0.165)

Fig. 4 Scanning electron micrographs of PM biofilms in the MS groups. Substantially fewer bacteria were present in Zr and VB than in the other groups. Scale bars, 10 µm.

CR

HAP

VB

Zr

6

ties (16). Second, bacterial flora in the generated biofilm reflect the inhabitants of the oral cavity (17,25). The cultivation and identification of biofilms in oral cavities containing various bacteria is difficult in conventional models. Biofilms containing single or only a few strains respond differently to diverse biofilms after treatment with an antibacterial agent (16). Reproduction of various bacterial strains probably contributes to strong adhesion of the biofilm to the solid phase. In the present study, saliva was collected from a test subject, mixed with sterilized glycerin, and stored at −80°C. Consequently, all biofilms in every experiment were identical in quality.

To investigate the effect of surface roughness on plaque adhesion, the material surface was roughly ground with a #600 waterproof abrasive paper. These samples, ground to the carborundum point or white point, formed the roughly ground RS group. The standardized abrasive ground method was used in this experiment because it is difficult to approximate the roughness of a material from its hardness, even in those ground with waterproof abrasive papers of identical roughness.

Consequently, the Ra values in the RS group differed among samples and were much higher than those in the MS group. The standard deviation of the Ra was large in the RS group and small in the MS group, indicating that the Ra values in the latter group were approximately correct. Shintani et al. (3), Filiz et al. (4), and Scarano et al. (14) reported that plaque adhesion increases on rougher surfaces. In contrast, Bollen et al. reported that Ra values less than 0.2 µm do not affect bacterial adhesion (28), and a study by Nascimento et al. showed that bacterial adhesion to titanium and zirconium was insensitive to Ra (29). However, Dorkhan et al. reported high bacterial adhesion on rough surfaces coated with saliva (30) and found that adhesion to composite resin is not influenced by surface characteristics. Yamamoto et al. reported that adhesion to fillers in resin is affected by Ra (31). However, Busscher et al. reported high biofilm adhesion on composite resins with deteriorated surfaces and suggested that conductive materials such as metals increase bacterial adhesion by electron attraction (15). In the present study, viable bacterial count was uncorrelated with Ra in the RS group, despite the increase in counts after an increase in Ra values. The significant differences in viable bacterial counts between the RS and MS groups for all materials (HAP, Zr, VB, and CR) suggests that polishing decreases bacterial adhesion to the surfaces. Moreover, in the MS group, the Ra values of all samples were low and weakly correlated with viable counts, suggesting that factors other than adhesion affect biofilm growth on very smooth surfaces (Ra ≤0.1 µm); therefore,

we presume that bacterial adhesion is influenced by Ra values ≥0.1 µm.

The SFE of all materials was 50 mN/m or less and was not affected by changing Ra values. The SFE is related to the wettability of the material surface. A small contact angle implies high SFE and high surface adsorbability of the material (32). According to Wenzel and Nonomura, the SFE per unit area on rough surfaces is an increasing function of the changing surface area (33,34). However, in the present study, SFE was not significantly associated with roughness. The wettability of solid surfaces depends on the SFE and Ra of the materials. The SFE per unit area, expressed as the surface tension ɤ (Nm−1 or Jm−2), decreases with increasing hydrophobicity. Furthermore, SFEs of solid substances also depend on the chemical structure, such as the functional groups on the surfaces, and can be changed by surface modification or other techniques (35). Thus, on the basis of these factors, we infer that Ra does not affect SFE.

Van Dijk et al. reported that high SFE encourages plaque adhesion (11). Similarly, Quiynen et al. demon-strated low bacterial adhesion on Teflon, with a SFE of 20 mN/m, and concluded that the amount of plaque adhesion is affected more by Ra than by SFE (12). In the present study, because the SFE of all samples was approximately 50 mN/m, we presume that SFE can be altered by bacte-rial adhesion at 20 mN/m. Ionescu et al. reported that SFE is affected more by initial bacterial adhesion than by long-term cultivated biofilms, which is inconsistent with our results (36). This difference may be attributable to the 24 h cultivation time in our study. Moreover, we did not experimentally assess initial bacterial adhesion in this study. The correlation between viable bacterial count and SFE was not significantly different between the RS and MS groups, and the SFE in each group was only approximated. Therefore, we presume that factors other than SFE may have an effect on biofilm generation.

The effect of zeta potential on Ra has not been reported previously; therefore, in the present study zeta potential was measured in the MS group only. Zeta potential indi-cates the charged state on surfaces of colloidal particles and solid substances in solution. Under an applied electric field, charged particles respond to electric force associated with the zeta potential of the surrounding electric double layer. Although zeta potential has been measured and discussed in relation to bacterial adhesion to dental materials (37), many issues remain unresolved. The bacterial cell membrane is affected by proteins and sugar chains, which impart a small negative charge to the bacterial cell surface. In many cases, the adhered surface is also negatively charged in water. When a bacterium

7

approaches a negatively charged surface, it experiences van der Waals forces and electrostatic repulsive forces, which are summed in the total energy calculation. Nakamura et al. reported that the zeta potential is most negative in the tooth crown resin, followed by the base resin and ceramics. The smaller the absolute value of the zeta potential on bacterial surface, the more easily bacteria adhere to a surface. Similarly, the larger the nega-tive zeta potential of a material, the higher the quantity of bacterial adhesion (37). Electrostatic interaction has been reported during the adsorption of oral cavity bacteria to the surfaces of dental materials, and this phenomenon critically depends on zeta potential (37). Scarano et al. attributed the low bacterial adhesion on zirconium to the electrical conductivity of the material (14). In the present study, the correlation between viable bacterial count and zeta potential did not differ significantly among the mate-rials. Moreover, the correlation was moderately negative, indicating that the larger the negative zeta potential of the dental material, the greater the extent of bacterial adhesion.

In the MS group, SEM observations of the PM biofilms revealed different adhesion patterns among the various samples (Fig. 4). In HAP, the bacteria adhered to the entire sample, whereas in Zr and VB, they formed small bacterial aggregates separated by many gaps exposing the underlying bare surface. In CR, the bacteria covered the entire surface of the sample and formed thicker aggregates than in HAP. These differences in adhesion pattern with similar viable bacterial counts suggest that the surface characteristics of the dental materials play a role in bacterial adhesion.

Bacterial adhesion to the surfaces of dental materials has been extensively studied. Busscher et al. reported limited bacterial adhesion on tooth surfaces with SFEs lower than 50 mN/m (13). Previously, we attempted to develop a tooth surface adulterant that reduces the SFEs of tooth and dental material surfaces, while offering acid resistance and prevention of adhesion, plaque generation, and deliming of the dentin. The adulterant was intended to prevent caries and gum disease (38). In the present study, however, viable bacterial counts changed when the SFE was less than 50 mN/m, regardless of the material used. This indicates the involvement of factors other than SFE. Decreasing the SFE by grinding alone is extremely difficult in clinical trials. To limit bacterial adhesion to the surfaces of dental materials, the surface should be smoothed as much as possible, to reduce SFE. Alterna-tively, materials with high zeta potential should be used.

In summary, the present findings suggest that surface roughness affects generation of PM biofilm on oral

materials, that zeta potential is involved in the generation of PM biofilms on mirror-ground oral materials, and that SFE has a role in the generation of PM biofilms.

AcknowledgmentsThis study was supported, in part, by the MEXT-Supported Program for the Strategic Research Foundation at Private Univer-sities (2012 to 2014) and a Grant-in-Aid for Scientific Research (C) (25462973) from the Ministry of Education, Culture, Sports, Science and Technology. The authors have no conflict of interest regarding the materials and methods used in this study.

References 1. Kolenbrander PE, London J (1993) Adhere today, here

tomorrow: oral bacterial adherence. J Bacteriol 175, 3247-3252.

2. Ruby J, Barbeau J (2002) The buccale puzzle: the symbiotic nature of endogenous infections of the oral cavity. Can J Infect Dis 13, 34-41.

3. Shintani H, Satou J, Satou N, Hayashihara H, Inoue T (1985) Effects of various finishing methods on staining and accu-mulation of Streptococcus mutans HS-6 on composite resins. Dent Mater 1, 225-227.

4. Aykent F, Yondem I, Ozyesil AG, Gunal SK, Avunduk MC, Ozkan S (2010) Effect of different finishing techniques for restorative materials on surface roughness and bacterial adhe-sion. J Prosthet Dent 103, 221-227.

5. Egawa M, Miura T, Kato T, Saito A, Yoshinari M (2013) In vitro adherence of periodontopathic bacteria to zirconia and titanium surface. Dent Mater J 32, 101-106.

6. Okuda T, Okuda K, Kokubu E, Kawana T, Saito A, Ishihara K (2012) Synergistic effect on biofilm formation between Fusobacterium nucleatum and Capnocytophaga ochracea. Anaerobe 18, 157-161.

7. Hamada S, Slade HD (1980) Biology, immunology, and cariogenicity of Streptococcus mutans. Microbiol Rev 44, 331-384.

8. Burne RA (1998) Oral streptococci: products of their envi-ronment. J Dent Res 77, 445-452.

9. Periasamy S, Kolenbrander PE (2009) Aggregatibacter acti-nomycetemcomitans builds mutualistic biofilm communities with Fusobacterium nucleatum and Veillonella species in saliva. Infect Immun 77, 3542-3551.

10. Ohsumi T, Takenaka S, Wakamatsu R, Sakaue Y, Narisawa N, Senpuku H et al. (2015) Residual structure of Streptococcus mutans biofilm following complete disinfection favors secondary bacterial adhesion and biofilm re-development. PLoS One 10, e0116647.

11. van Dijk J, Herkströter F, Busscher HJ, Weerkamp A, Jansen H, Arends J (1987) Surface-free energy and bacterial adhe-sion. An in vivo study in beagle dogs. J Clin Periodontol 14, 300-304.

12. Quiynen M, Marechal M, Busscher HJ, Weerkamp AH, Arends J, Darius PL et al. (1989) The influence of surface

8

free-energy on planimetric plaque growth in man. J Dent Res 68, 796-799.

13. Busscher HJ, Weerkamp AH, van der Mei HC, van Pelt AWJ, de Jong HP, Arends J (1984) Measurement of the surface free energy of bacterial cell surfaces and its relevance for adhe-sion. Appl Environ Microbiol 48, 980-983.

14. Scarano A, Piattelli M, Caputi S, Favero GA, Piattelli A (2004) Bacterial adhesion on commercially pure titanium and zirconium oxide disks: an in vivo human study. J Periodontol 75, 292-296.

15. Busscher HJ, Rinastiti M, Siswomihardjo W, van der Mei HC (2010) Biofilm formation on dental restorative and implant materials. J Dent Res 89, 657-665.

16. Exterkate RA, Crielaard W, Ten Cate JM (2010) Different response to amine fluoride by Streptococcus mutans and polymicrobial biofilms in a novel high-throughput active attachment model. Caries Res 44, 372-379.

17. Exterkate RA, Zaura E, Buijs MJ, Koopman J, Crielaard W, ten Cate JM (2014) The effects of propidium monoazide treatment on the measured composition of polymicrobial biofilms after treatment with chlorhexidine. Caries Res 48, 291-298.

18. Frank RM, Steuer P (1985) Transmission electron microscopy of plaque accumulations in denture stomatitis. J Prosthet Dent 53, 115-124.

19. Mashima I, Nakazawa F (2013) A review on the characteriza-tion of a novel oral Veillonella species, V. tobetsuensis, and its role in oral biofilm formation. J Oral Bio 55, 184-190.

20. Guggenheim B, Giertsen E, Schüpbach P, Shapiro S (2001) Validation of an in vitro biofilm model of supragingival plaque. J Dent Res 80, 363-370.

21. MacBain AJ, Bartolo RG, Catrenich CE, Charbonneau D, Ledder RG, Gilbert P (2003) Effects of triclosan-containing rinse on the dynamics and antimicrobial susceptibility of in vitro plaque ecosystems. Antimicrob Agents Chemother 47, 3531-3538.

22. Deng DM, ten Cate JM (2004) Demineralization of dentin by Streptococcus mutans biofilms grown in the constant depth film fermentor. Caries Res 38, 54-61.

23. Hayati F, Okada A, Kitasako Y, Tagami J, Matin K (2011) An artificial biofilm induced secondary caries model for in vitro studies. Aust Dent J 56, 40-47.

24. MacBain AJ (2009) Chapter 4. In vitro biofilm models: an overview. Adv Appl Microbiol 69, 99-132.

25. Huang X, Exterkate RA, ten Cate JM (2012) Factors associ-ated with alkali production from arginine in dental biofilms. J Dent Res 91, 1130-1134.

26. Tomiyama K, Mukai Y, Kumada H, Watanabe K, Hamada N, Teranaka T (2015) Formation of subsurface dentin lesions using a polymicrobial biofilm model. Am J Dent 28, 13-17.

27. Hasegawa H, Tomiyama K, Kumada H, Kawata A, Higashi K, Takahashi O et al. (2015) Antimicrobial effect of carbamide peroxide against a polymicrobial biofilm model. Am J Dent 28, 57-60.

28. Bollen CM, Papaioanno W, van Eldere J, Schepers E, Quirynen M, van Steenberghe D (1996) The influence of abutment surface roughness on plaque accumulation and peri-implant mucositis. Clin Oral Implants Res 7, 201-211.

29. do Nascimento C, da Rocha Aguiar C, Pita MS, Pedrazzi V, de Albuquerque RF Jr, Ribeiro RF (2013) Oral biofilm formation on the titanium and zirconia substrates. Microsc Res Tech 76, 126-132.

30. Dorkhan M, Chávez de Paz LE, Skepö M, Svensäter G, Davies JR (2012) Effects of saliva or serum coating on adher-ence of Streptococcus oralis strains to titanium. Microbiology 158, 390-397.

31. Yamamoto K, Ohashi S, Taki E, Hirata K (1996) Adherence of oral streptococci to composite resin of varying surface roughness. Dent Mater J 15, 201-204.

32. Eliades T, Eliades G, Brantley WA (1995) Microbial attach-ment on orthodontic appliances : I. Wettability and early pellicle formation on bracket materials. Am J Orthod Dento-facial Orthop 108, 351-360.

33. Wenzel RN (1936) Resistance of solid surface to wetting by water. Ind Eng Chem 28, 988-994.

34. Nonomura Y, Komura S, Tsujii K (2006) Adsorption of microstructured particles at liquid-liquid interfaces. J Phys Chem B 110, 13124-13129.

35. Yoshino N, Teranaka T (1997) Synthesis of silane coupling agents containing fluorocarbon chain and applications to dentistry: plaque-controlling surface modifiers. J Biomater Sci Polym Ed 8, 623-653.

36. Ionescu A, Wutscher E, Brambilla E, Schneider-Feyrer S, Giessibl FJ, Hahnel S (2012) Influence of surface properties of resin-based composites on in vitro Streptococcus mutans biofilm development. Eur J of Oral Sci 120, 458-465.

37. Miyake T, Sato T, Maki Y (2013) Effect of zeta potentials on bovine serum albumin adsorption to hydroxyapatite surfaces. Bull Tokyo Dent Coll 54, 97-101.

38. Nihei T, Omoto N, Ohashi K, Kondo Y, Yoshino N, Teranaka T (2013) Effect of enamel surface modification by novel aqueous phosphate-type fluoride surfactants. Dent Mater J 32, 83-87.