A graphene/zinc oxide nanocomposite film protects dental implant surfaces against cariogenic Streptococcus mutans

Oral biofilms play a crucial role in the development of dental caries and other periodontal diseases. Streptococcus mutans is one of the primary etiological agents in dental caries. Implant systems are regularly employed to replace missing teeth. Oral biofilms accumulate on these implants and are the chief cause of dental implant failure. In the present study, the potential of graphene/zinc oxide nanocomposite (GZNC) against the cariogenic properties of Streptococcus mutans was explored and the anti-biofilm behaviour of artificial acrylic teeth surfaces coated with GZNC was examined. Acrylic teeth are a good choice for implants as they are low cost, have low density and can resist fracture. Microscopic studies and anti-biofilm assays showed a significant reduction in biofilm in the presence GZNC. GZNC was also found to be nontoxic against HEK-293 (human embryonic kidney cell line). The results indicate the potential of GZNC as an effective coating agent for dental implants by efficiently inhibiting S. mutans biofilms.


Introduction
Oral biofilms are complex three-dimensional structures with adherent multispecies bacterial communities which can contribute to dental caries and numerous periodontal diseases (Selwitz et al. 2007;Nance et al. 2013). These common infectious diseases can lead to a major public health concerns (Falsetta et al. 2014). Biofilms have a greater tendency to resist antibiotics and create an environment that enhances microbial resistance with respect to their planktonic counterparts. Streptococcus mutans is one of the most frequently detected microorganisms on the tooth surface and is a major etiological agent of human dental caries (Hasan et al. 2014). This bacterial species has also been recognized as a causative agent of endocarditis (Abranches et al. 2011). S. mutans grows in the oral cavity by means of several unique mechanisms (Dmitriev et al. 2011). Acidogenicity and aciduricity play a major role in the increase in the severity of infection along with the ability to produce extracellular polysaccharides (Koo et al. 2003;Krol et al. 2014).
At present implant systems are utilized extensively to replace missing teeth. Oral biofilms consisting mainly of Streptococcus spp. accumulate on implants (Nakazato et al. 1989). The formation of biofilm on these implants is one of the major causes of implant failure (Costerton et al. 2005). The inflammatory changes in the soft tissues surrounding the implant induced by infection give rise to progressive destruction of the supporting bone (Zitzmann & Berglundh 2008). Nanoparticle-based implant coatings may well offer useful antimicrobial and anti-biofilm functionalities to prevent dental implant failure.
Nanoparticle based approaches are expected to open new horizons for preventing biofilm based infections by their unique mode of action (Ruparelia et al. 2008;Raghupati et al. 2011). Zinc oxide nanoparticles (ZnO NP) have already been found to have antibacterial activity against a wide range of microorganisms (Huang et al. 2008;Xie et al. 2011), but aggregation is one of the drawbacks since it makes them toxic against mammalian cells (Yuan et al. 2010).
Graphene oxide (GO) has unique physical and chemical properties . GO contains a single layer of sp 2 carbon atoms with hydroxyl and epoxy functional groups on the surface and the carboxyl groups at the edges (Dai et al. 2014). These functional groups offer active sites for hybridization with metals and metal oxides and thus act as a supporting surface for growing metal and metal oxide nanoparticles (Ocsoy et al. 2013). Graphene in its functionalized state has been used for biosensing, photothermal therapy, drug delivery as well as imaging . Recently the use of graphene and graphene based nanocomposites as potential antimicrobial agents has gained substantial interest in the field of nanomedicine Xu et al. 2011;Tang et al. 2013;Fariaa et al. 2014). Moreover graphene oxide has been reported to show good biocompatibility compared to other nanoparticles (Chang et al. 2011). To the best of the authors' knowledge no study has yet examined the anti-biofilm action of graphene/zinc oxide nanocomposites (GZNC) and GZNC coated tooth surfaces on S. mutans.
The objective of this present study was to evaluate the antimicrobial, anti-biofilm and anti-adherence activity of GZNC against S. mutans, a major cause of caries infection, and to access the applicability of GZNC as a coating for dental implants.

Materials and methods Microorganisms
S. mutans MTCC 497, an ATCC analogue of the UA159 strain of S. mutans (purchased from IMTECH, Chandigarh, India) was used in this study. The clinical isolates of S. mutans used in this study (SM 497,SM 34 and SM 06) were isolated and characterized earlier in the authors' laboratory (Islam et al. 2008). The clinical samples were collected from the Department of Conservative Dentistry, Dental College, AMU, Aligarh, India. CLSI guidelines were followed for the isolation and characterization of S. mutans from samples. The strains of S. mutans were grown in a CO 2 -rich environment using candle jar incubation. The isolates were confirmed by PCR amplification of conserved regions of the GTF B and GTF C genes. All the strains were grown in Brain Heart Infusion broth (BHI) (Hi Media Laboratories, Mumbai, India) at 37°C. The cultures were stored at -80°C in BHI containing 25% glycerol.

Synthesis of GZNC
Graphene oxide (GO) was prepared according to the method described by Hummers and Offman (1958) with little modification. H 2 SO 4 (100 ml) was added to 2 g of graphite powder (Loba Chemie, Mumbai, India) while stirring in an ice-water bath. KMnO 4 (25 g) was slowly added to the solution. The mixture was kept on a stirrer at room temperature until it became pasty brown. It was then diluted with the slow addition of 200 ml of water. Finally, 10 ml of a 30% aqueous solution of H 2 O 2 were added. The impurities were removed from the graphene oxide (GO) using 3% HCl with repeated washing (Hummers & Offman 1958). For the synthesis of the GZNC, 100 mg of zinc acetate and 200 mg of GO were dispersed into 200 ml of absolute ethanol followed by sonication and harvested by centrifugation at 5,000 rpm for 5 min. It was then washed with 80% ethanol. The pellet was vacuum dried and 100 mg of this dried sample were mixed with 100 ml of ethylene glycol using sonication for 10 min. The resulting mixture was heated to 140°C with vigorous stirring on a magnetic stirrer (REMI, Mumbai, India; Model: 1MLH) for 3 h. The synthesized GZNC suspension was centrifuged, washed with 80% ethanol and dried in a vacuum oven at 60°C.

Characterization of GZNC
The synthesis of GZNC in solution was monitored by measuring absorbance using a UV-visible spectrophotometer (Perkin Elmer Life and Analytical Sciences, Shelton, CT, USA) in the wavelength range 200 to 800 nm. Transmission electron microscope (TEM) analysis was performed using a JEM-2100F TEM (Jeol, Tokyo, Japan) operating at 120 kV. The X-ray diffraction (XRD) patterns of the powdered sample were recorded on a Mini-Flex™ II benchtop XRD system (Rigaku Corporation, Tokyo, Japan) operating at 40 kV. For the Fourier transform infrared (FTIR) spectroscopic measurements, graphene/ZnO powder was mixed with spectroscopic grade potassium bromide (KBr) in the ratio of 1:100 and spectra were recorded in the range 400-4,000 wave number (cm −1 ) on a Perkin Elmer FTIR Spectrum BX (PerkinElmer Life and Analytical Sciences) in the diffuse reflectance mode at a resolution of 4 cm −1 in KBr pellets. The thermal stability of the graphene/ZnO was investigated by thermogravimetric analysis (TGA) at a heating rate of 10°C min −1 under a nitrogen atmosphere.

Determination of bacteriostatic (MIC) and bactericidal (MBC) concentrations
The minimum inhibitory concentration of GZNC against the strains MTCC 497 and clinical isolates was determined using the micro-dilution method (Khan et al. 2010). Overnight grown cultures (50 μl) were diluted to 10 5 -10 6 cfu ml −1 and inoculated into fresh BHI (50 μl) containing various concentrations of serially diluted GZNC, starting at an initial concentration of 1 mg ml −1 . The MIC was determined as the lowest concentration that totally inhibited visible bacterial growth. The MBC, on the other hand, was determined by sub-culturing the test dilutions on tryptic soya agar plates and incubating for 24 h. These determinations were performed in triplicate and the means of three independent experiments were calculated.

Effect on adherence to smooth glass surfaces
The effect of sub-MIC concentrations of GZNC on all three strains was performed as the inhibition of adhered cells on a smooth glass surface (Hamada & Torii 1978). The bacteria were grown at 37°C for 24 h at an angle of 30°in a glass tube containing 10 ml of BHI with 5% (w/v) sucrose (Aires et al. 2006) and sub-MIC concentrations of GZNC. After incubation, the glass tubes were slightly rotated and the planktonic cells were decanted. The adhered cells were then removed by adding 0.5 mM sodium hydroxide followed by agitation. The cells were washed and suspended in saline. Adherence was quantified spectrophotometrically at 600 nm. These determinations were performed in triplicate, using untreated BHI medium as control.
The percentage adherence was calculated as the OD of adhered cells divided by the OD of total cells divided by 100.
Biofilm formation assay (crystal violet assay) Biofilm formation was assessed by using the protocol of Loo et al. (2000) with a few modifications. Briefly, 50 ml of overnight grown cultures of S. mutans strains (MTCC 497 and clinical isolates) diluted to 10 5 -10 6 cfu ml −1 were inoculated into fresh BHI (150 ml) with 5% sucrose containing various sub-MIC concentrations of GZNC with respective controls and used to inoculate the wells of microtitre plates. After incubation for 24 h at 37°C, the medium and unattached cells were decanted from the microtitre plates. The remaining planktonic cells were removed by gentle rinsing with sterile water. The wells with adhered biofilms were fixed with formalin (37%, diluted 1:10) plus 2% sodium acetate, and each well was stained with 200 ml of 0.1% crystal violet (CV) for 15 min at room temperature. After two rinses with distilled water, bound dye was removed from the cells using 100 ml of 95% alcohol. Plates were then set on a shaker for 5 min to allow full release of the dye. Biofilm formation was quantified by measuring the OD 630 using a BIO-RAD iMark TM Microplate reader, Gurgaon, India.

Assessment of cellular viability
This assay was performed as described previously with slight modifications as performed earlier (Islam et al. 2008). 2,3-Bis(2-methoxy-4-nitro-5-sulfophenyl)-5-([phenylamino] carbonyl)-2H-tetrazolium hydroxide (XTT) (Sigma-Aldrich, New Delhi, India) was dissolved in phosphate buffer saline (PBS) at a final concentration of 250 mg l −1 . The solution was filter sterilized using a 0.22-mm pore-size filter and stored at −70°C until required. Menadione (Sigma-Aldrich) solution (0.4 mM) was also prepared and filtered immediately before each assay. Adherent cells were washed with 200 μl of PBS, then 158 μl of PBS, and 40 μl of XTT, and 2 μl of menadione were added to each well. After incubation in the dark for 4 h at 37°C, 100 μl of the solution were transferred to a new well and a colorimetric change in the solution was measured using a microtitre plate reader (BIORAD iMark TM Microplate reader) at 490 nm.

Effect on growth
The effect of sub-MIC concentrations of GZNC was tested on the growth of all strains of S. mutans. Overnight grown broth cultures of S. mutans were inoculated into tubes to obtain a final inoculum of 1.56 × 10 4 CFU ml −1 followed by the addition of GZNC. The tubes were incubated at 37°C. Growth was monitored spectrophotometrically (UV mini 1240, UV-VIS Spectrophotometer Shimadzu, New Delhi, India) by taking the absorbance of the culture at 600 nm for 24 h. All determinations were performed as triplicates using untreated growth controls.

Inhibition of water-insoluble glucan synthesis
A crude GTFase was assayed to estimate the effect of GZNC on glucan synthesis (Hasan et al. 2012). Cell-free enzymes were precipitated from culture supernatant of S. mutans by adding solid ammonium sulphate to 70% saturation (an ammonium cut). The mixture was stirred at 4°C for 1 h and allowed to stand for another 1 h under cold conditions. The precipitate was collected by centrifugation at 12,000 g at 4°C for 20 min, dissolved in a minimum volume of 20 mM phosphate buffer (pH 6.8) and then dialysed against 2 mM phosphate buffer (pH 6.8) at 4°C for 24 h. The crude enzyme was stored at -70°C for further experiments. A reaction mixture consisting of 0.25 ml of crude enzyme and varying concentrations of GZNC in 20 mM phosphate buffer (pH 6.8) containing 0.25 ml of 0.4 M sucrose was incubated at 37°C for 18 h. The fluid was removed post incubation and the tube contents were washed with sterile water. The total amount of water-insoluble glucan was measured by the phenol-sulphuric acid method (Dubois et al. 1956). Three replicates were performed for each concentration of the GZNC.

Influence of GZNC on EPS production
The Congo red (CR) binding assay which detects glucose containing polymers was used to evaluate exopolysaccharide (EPS) production, as previously reported (Friedman et al. 2001;López-Moreno et al. 2014). Fifty microlitres of overnight growth culture of S. mutans strain (MTCC 497) diluted to 10 5 -10 6 cfu ml −1 were inoculated into fresh BHI (50 ml) with 5% sucrose containing sub-MIC concentrations of GZNC. These solutions as well as untreated controls were used to inoculate microtitre plates. After incubation for 24 h at 37°C, the medium was removed and biofilms were washed with PBS and then fresh medium (100 μl) was added to each well including the respective controls. Then 50 μl of CR (0.5 mM) were added to each well. Medium (100 μl) along with 50 μl CR were added to another well for blank measurements (Blank CR). Plates were incubated for 60 min. The medium in each well was transferred to 200 μl micro centrifuge tubes and centrifuged at 10,000 × g for 5 min. The supernatant was transferred to empty wells in microtitre plates. Absorbance was measured at 490 nm. The absorbance value of the supernatant was subtracted from the absorbance value of the 'blank CR'. The resultant value represents the amount of bound CR or EPS produced. This experiment was conducted in triplicate.

CR agar method
An alternative method of screening biofilm formation and EPS production as described by Freeman et al. (1989) was also used in this study. Solid agar medium was prepared using BHI (37 g l −1 ), sucrose (5%), agar No. 1 (2%), and CR stain (0.8 g l −1 ). CR was prepared in the form of a concentrated aqueous solution and was autoclaved at 121°C for 15 min, separately. After autoclaving, CR was added to the agar which had been cooled to 55°C. For treated samples, sub-MIC concentrations of nanoparticles were added to the medium. Plates were inoculated and incubated aerobically for 24 h at 37°C. EPS production was indicated by black colonies with a dry crystalline consistency.

Effect on acid production
The effect of GZNC on the acidogenicity of S. mutans and its clinical isolates was assessed using a previously published protocol (Khan et al. 2010). Five millilitres of BHI broth containing 5% (w/v) of sucrose and sub-MIC concentrations of GZNC were inoculated with SM 497, SM 06 and SM 34 and incubated at 37°C for 24 h. The pH of the treated samples and the controls was assessed at 0 h and after incubation for 24 h. All determinations were performed in triplicate.

Scanning electron microscope analysis of biofilms
The effects of GZNC on biofilms was visualized using scanning electron microscopy (SEM) using saliva-coated glass coverslips. Saliva was collected under masticatory stimulation (chewing a piece of parafilm), from a healthy individual who abstained from tooth brushing and eating for 5 h prior to collection. The collected saliva was centrifuged at 8,000 g for 15 min to obtain clarified saliva and was stored at -80°C (Hasan et al. 2012). Clarified saliva (100 ml) was added to each well of a 12-well microtitre plate equipped with glass coverslips. The plate was then incubated at 37°C for 2 h to coat the coverslips with a salivary pellicle. After incubation, these coverslips were rinsed three times with PBS before adding the bacterial culture. The experiment was performed in triplicate. Sub-MIC concentrations of GZNC were used as treatment while the control was untreated S. mutans. The wells were inoculated (10 5 -10 6 cfu ml −1 ) and incubated at 37°C for 24 h. The coverslips were removed after 24 h and washed three times in sterile PBS. The resultant samples were fixed with 2% formaldehyde and 2.5% glutaraldehyde. The samples were rinsed three times with PBS after fixing, followed by an ethanol dehydration series. Samples were then completely dried, coated with gold, and observed using a scanning electron microscope.

Biofilm reduction seen through confocal microscopy
Bacterial biofilm was grown in the presence of sub-MIC concentrations of GZNC at 37°C in covered glass bottom confocal dishes (Genetix Biotech Asia Pvt. Ltd, New Delhi, India) with a dish size of 35 mm, 22 mm coverglass, 9.4 cm 2 growth area and a working volume of 3 ml. The dishes were inoculated (10 5 -10 6 CFU l −1 ) and incubated at 37°C for 24 h. Then medium was removed and attached biofilms were washed with PBS and treated with SYTO-9 (5 μM; excitation and emission wavelength; 488 nm, 498 nm) and propidium iodide (PI) (0.75 μM; excitation and emission wavelength; 536 nm, 617 nm). The stained bacterial biofilm was observed with a FluoView FV1000 (Olympus, Tokyo, Japan) confocal laser scanning microscope equipped with argon and HeNe lasers.
Coating of acrylic teeth with GZNC and biofilm assays GZNC coating on the tooth surface was done using sonochemistry as previously described (Eshed et al. 2012). A GZNC coating on artificial acryl teeth (obtained from Dr Ziauddin Dental College at the Aligarh Muslim University) was performed by placing the tooth in a nanoparticle suspension in a sonicator. The tooth was kept at a constant distance from the sonicator tip throughout the process. The nanoparticle surface coating was characterized by SEM (EVO 40; Zeiss, Jena, Germany). Artificial teeth were assayed using a static biofilm assay. Teeth were placed in a 24-well plate. Each well contained 2 ml of a suspension of S. mutans at a final concentration of1 .5 × 10 8 CFU ml −1 in BHI medium. After incubation for 24 h at 37°C, biofilm formation was assayed using the CV assay as described above. To examine biofilm morphology, teeth samples were further exposed after incubation in fixative (glutaraldehyde + paraformaldehyde) for 4 h. Finally, samples were dehydrated using increasing concentrations of ethanol. Samples were then air dried and imaged by SEM.

Detection of reactive oxygen species (ROS)
The ROS formed by GZNC were identified using 2′,7′dichlorofluorescein diacetate (DCFDA) (Cui et al. 2012). The sub-MIC concentration of GZNC was used and the number of bacterial cells was adjusted to 10 5 CFU ml −1 . All cultures were incubated at 37°C for 4 h and then centrifuged at 4°C for 30 min at 8,000 g. Then each supernatant was treated with 100 μM DCFDA for 1 h. The ROS formed in the sample were detected at a fluorescence excitation wavelength of 485 nm and an emission wavelength of 528 nm. The same procedure was used to determine ROS after incubation for 12 h.

Cytotoxicity assay
A human embryonic kidney cell line (HEK-293) obtained from National Centre for Cell Science (NCCS) Pune, was cultured in Dulbecco's Modified Eagle's Medium (DMEM) (Biological Industries, Beit HaEmek, Israel), supplemented with 10% heat inactivated foetal calf serum and IX Penstrep antibiotic solution, incubated at 37°C and 5% CO 2 . HEK-293 cell viability was measured using an MTT (3-(4,5-dimethythiazol-2yl)-2,5-diphenyltetrazoliumbromide) assay as described earlier (Denizot & Lang 1986). Exponentially growing cells (~10 5 cells well −1 ) were seeded into 96-well culture plates and incubated with various concentrations of GZNC (100, 200, 300 and 400 μg ml −1 ) for 24 and 48 h. Four hours before termination, the supernatants were removed and 90 μl of fresh medium and 10 μl of MTT (1 mg ml −1 ) solution were added to each well. After further incubation for 4 h the formazan crystals formed by the cellular reduction of MTT were dissolved in 150 μl of DMSO and plates were read on an ELISA-reader using a 570 nm filter. All measurements were done in triplicate. The relative cell viability (%) related to control wells containing cells without GZNC was calculated as: where, [A] test is absorbance of the test sample and [A] control is the absorbance of the control sample. By using this non-radioactive assay (Ahmed et al. 1994) for assessing the proliferation of cells, the authors were able to quantify the amount of MTT cleaved, which is directly proportional to the viable cell population.

Intracellular uptake of GZNC in HEK-293 and bioimagining
HEK-293 cells were cultured in DMEM supplemented with 10% heat inactivated foetal calf serum (Biological Industries) and IX Penstrep antibiotic solution (Biological Industries) and then incubated in a fully humidified 5% CO 2 incubator at 37°C. All cells were seeded in culture flasks and were divided into treatment group and control group. When~70% of growth occurred, the cells were washed with 0.1 M PBS (phosphate buffer saline) and old medium was replaced with fresh medium. Culture plates were treated with various concentrations of GZNC (100, 200, 300 and 400 μg ml −1 ) and incubated at 37°C and 5% CO 2 for 24 h. All plates were washed with 0.1 M PBS and cells were collected by trypsinization (0.05% trypsinase). The cell pellets were dissolved in 1 ml of 0.1 M PBS and were imaged under bright field, UV-excitation and blue excitation with an Olympus Fluo View TM FV1000 laser scanning confocal microscope.

Statistical analysis
All experiments were executed in triplicate. For each assay, data are presented as mean ± standard deviation (SD). The values were calculated as the mean of individual experiments in triplicate and compared with those of the control groups. Differences between two mean values were calculated by the Student's test. Data with p-values < 0.05 were considered statistically significant.

Results and discussion
Characterization of GZNC TEM images ( Figure 1A) of GZNC showed ZnO nanocrystals dispersed on the surface of the graphene sheets. The average size of the ZnO nanoparticles in the composite was found to be in the range of 20-40 nm ( Figure 1B). TEM results confirmed the attachment of zinc oxide with GO platelets and ZnO nanoparticles. The UV visible spectrum of graphene, ZnO and GZNC is shown in Figure 1C. The presence of a sharp characteristic absorption peak at~370 nm clearly indicated the formation of pure crystalline ZnO nanostructures. In addition, the red shift in the GZNC curve as compared to pure ZnO shows an increased π-electron concentration and structural ordering and may be ascribed to the chemical bonding (Zn-O-C bond) of zinc oxide and graphene (Liu et al. 2012;Hu et al. 2013). Figure 1D represents the FTIR spectrum of GZNC with strong absorption bands around 3468.5, 2342.07, 1547.47, 1059.65, 648.9 and 482.9 cm −1 . The band at 3468.5 cm -1 may be assigned to O-H bond stretching of adsorbed water molecules. The absorption band at 1547.47 cm −1 can be assigned to the stretching vibration of the C=C of graphene and another at 1059.65 cm −1 can be assigned to the stretching vibration of the C-O of graphene . The presence of a Zn-O bond is further verified by the peak at 482.9 cm -1 (Bora et al. 2013). Powder X-ray diffraction (XRD) analysis confirmed that GZNC consists of cubic ZnO ( Figure S1A). [Supplementary information is available via a multimedia link on the online article web page.] The average crystalline size of ZnO NPs was calculated following the Debye-Scherrer formula (Cullity 1978). The calculated average particle size was found to bẽ 14.76 nm. The thermogravimetric (TG) measurement was carried out in order to determine the mass ratio of ZnO to graphene in the composite. The TGA curve of GZNC is shown in Figure S1B. A schematic representation of nucleation of zinc oxide nanoparticles on functionalized graphene oxide is illustrated in Figure 2.

Enhanced antibacterial activity of ZnO anchored to graphene sheets (GZNC)
The minimum inhibitory concentration (MIC) of GZNC against S mutans and its clinical isolates was found to be 125 μg ml −1 whereas the minimum bacteriocidal concentration (MBC) was found to be 250 μg ml −1 . The antibacterial effect of GZNC is much greater than that of ZnO nanoparticles alone where the MIC and the MBC are reported to be 500 ± 306.18 μg ml −1 and 500 μg ml −1 respectively (Hernández-Sierra et al. 2008), indicating that graphene is enhancing the antibacterial property of ZnO nanoparticles. It is also clear from the results that the killing of planktonic cells of S. mutans is not dependent on clinical vs reference strains. In view of the above results, it is evident that the nanostructure formed by the interaction of ZnO with graphene provides a unique nano-interface for interacting with microbes as compared to ZnO alone.

Effect on cariogenic properties and biofilms of S. mutans
Adherence of bacteria to the tooth surface is an important step in biofilm formation and a reduction in adherence could serve as preventive biofilm formation (Hasan et al. 2012). The effect of sub-inhibitory concentrations of GZNC on sucrose dependent adherence of all three strains is represented in Figure 3A. The sub-inhibitory concentration (62.5 μg ml −1 ) of GZNC reduced the adherence of SM 497, SM 34 and SM 06 by 46, 68 and 69%, respectively (results were statistically significant with p ≤ 0.05). There was a considerable decrease in adherence on treatment with sub-inhibitory concentrations of GZNC in all three strains. S. mutans secretes exopolysaccharide (glucans) in the presence of sucrose which helps in the clumping and adherence of cells. This implies that the composite reduces the polysaccharide mediated adherence of bacterial cells.
The reduction in biofilm and the inhibition of biofilm formation was observed to occur in a dose-dependent manner ( Figure 3B). The inhibition of biofilm in SM 497 was 80% while in the case of SM 34 and SM 06, the reduction was found to be 44% and 29%, respectively (results were statistically significant with p ≤ 0.05). The XTT assay was performed to detect the amount of viable cells present after treatment with the composite. SM 497, SM 34 and SM 06 showed only a 37, 31 and 38% decrease in viability, respectively, when treated with 62.5 μg ml −1 GZNC ( Figure 3C), which is not statistically significant. No significant differences were observed in the growth curves of all strains compared to the control (Figure 4). XTT and growth curve data indicated that the nanocomposite may be inhibiting virulence traits without affecting bacterial viability.

Effect on acid production and glucan formation
Acid production and acid tolerance are considered to be primary physiological factors linked with the cariogenic potential of S. mutans (Krol et al. 2014). GZNC reduced acid production in both the reference and clinical strains (Table 1). Reducing acid production is a cariostatic effect and may also influence the biofilm forming abilities of S. mutans (Welin-Neilands & Svensater 2007). Water insoluble glucans play a significant role in adhesive interactions compared to water soluble glucans. Further, the effects of different concentrations of GZNC on the synthesis of water insoluble glucans were evaluated ( Figure 3D). There were reductions of almost 90, 85 and 60% in the case of SM 497, SM 34 and SM 06, respectively (the results were statistically significant with p ≤ 0.05). A considerable reduction in insoluble glucans was observed in the presence of GZNC. Reducing the amount of insoluble glucans could potentially influence the process of biofilm formation by disturbing the physical integrity and stability, affecting the diffusion properties and reducing the binding sites for S. mutans. The malformed exopolysaccharide (EPS) matrix containing less insoluble glucans may also be more susceptible to the influences of antimicrobials and other environmental attacks .

Significant reduction in EPS
The production of EPS is one of the key virulence factors of cariogenicity as it is produced by bacteria for the formation, spread and maintenance of biofilms. EPS mediates the adhesion of biofilms to surfaces, provides mechanical stability and transiently immobilizes cells (Flemming & Wingender 2010). EPS production in the presence and absence of GZNC was evaluated by using CR dye which binds to glucose containing polymers. Biofilm formation by S. mutans was tested by growing the organism on BHI agar supplemented with CR in the presence and absence of GZNC. When the colonies were grown without GZNC in the medium, the organisms appeared as dry crystalline black colonies, indicating the production of EPS ( Figure 5A). Whereas when the organisms were grown in the presence of sub-MIC concentrations of GZNC, they continued to grow, but GZNC treatment decreased the synthesis of EPS, as indicated by the decrease in dry crystalline black colonies. Further, the reduction in the amount of EPS of attached cells was calculated by the CR binding assay which directly relates to the amount of EPS formed. The results showed a reduction of almost 51% in EPS production on treatment with sub-MIC concentrations of GZNC (the results were statistically significant with p ≤ 0.05). Figure 5B clearly indicates that the nanocomposite is reducing EPS production which is a prerequisite for the formation and maintenance of biofilm.
Microscopic exploration of the effect of GZNC on S. mutans biofilm architecture Scanning electron microscopy (SEM) and confocal laser scanning microscopy (CLSM) were performed to examine the architecture of the biofilm in the presence of GZNC. In SEM images of control samples ( Figure 6A), S. mutans cells can be seen embedded in EPS, while in treated samples ( Figure 6B) cells were highly dispersed,  indicating a reduction in EPS. In CLSM images, the majority of cells were observed to be alive (green) in non-treated samples; a green mat was visible with chains of Streptococcus cells interacting with each other ( Figure 6C). In treated samples, cells were observed to be scattered all around with poor interaction, representing a reduction in biofilm ( Figure 6D, E).

Characterization and anti-biofilm properties GZNC coated tooth surfaces
Dental implants are widely accepted as a replacement for natural teeth. A considerable percentage of medical implants are the cause of device related infections (Costerton et al. 2005). These infections are difficult to eradicate because the bacteria that cause them live in well-developed biofilms. S. mutans biofilm is one of the causes of failure of dental implants (Busscher et al. 2010). Oral implant related biofilms can cause inflammation of peri-implant tissues which may be a direct cause of periodontal disease (Heuer et al. 2007). In general, the most effective way of preventing biofilm formation on implants is to prevent initial bacterial adhesion as biofilms are relatively difficult to remove after formation. Therefore, these infections may be greatly reduced by improving the antimicrobial and anti-biofilm properties of the implant surface by means of surface modification (Zhao et al. 2009).
Acrylic teeth are a good choice for dental prosthesis. The basic ingredient of acrylic teeth is polymethyl methacrylate (PMMA) resin. PMMA resins are a resilient plastic formed by joining multiple methyl methacrylate molecules. A cross linking agent is added which serves as a bridge that unites two polymer chains (Stoia et al. 2011). In this manner it yields a net-like structure that provides increased resistance to deformation, thus the teeth have greater fracture toughness and are easier to adjust.
Thus, the focus of the present study was to observe the ability of nanoparticle coated acrylic teeth to inhibit biofilm formation. Teeth were coated GZNC using sonochemistry. Sonochemical irradiation has been demonstrated to be a successful technique for the synthesis and deposition of nanoparticles on/into glass, polymer supports, and fabrics as well as tooth surfaces (Pol et al.   . Figure 7A and C shows photographs of uncoated and coated teeth. The deposition of GZNC was characterized by SEM. Figure 7B represents the uncoated acrylic tooth surface while in Figure 7D, a uniform coating was observed over the entire tooth surface. Further, biofilm biomass on the tooth surface was quantified using the CV assay. GZNC coating of teeth reduced biofilm formation by 85% ( Figure 7I), compared to the uncoated tooth surface. Photographic images of CV stained tooth surfaces clearly show the inhibition of biofilm formation ( Figure 7J-M). These results were also supported by SEM imaging (Figure 7E, F). A considerable reduction in S. mutans biofilm was observed on coated teeth ( Figure 7E, F) compared to the control tooth in which dense colonization was observed ( Figure 7G, H).

Production of reactive oxygen species
The antibacterial properties of many nanoparticles have been attributed to the production of reactive oxygen species (ROS), such as TiO 2 (Su et al. 2009) and Ag NPs (Vecitis et al. 2010). GZNC produced a noticeable increase in cellular ROS ( Figure S2). The internalization of zinc oxide nanoparticles in bacteria induces the production of ROS and this can affect DNA and may also affect the total cellular machinery of bacteria (Xie et al. 2011). The present study confirms that GZNC is very effective against some of the main caries causing virulence factors of S. mutans. The anti-biofilm properties of GZNC may be attributed to the wrapping of graphene sheets on the bacterial surfaces which reduces cell to cell interaction and further causes the deposition of ZnO nanoparticle on the surfaces, leading to high concentrations of zinc ions in the cell. Moreover, the leaching of Zn 2+ ions from nanoparticles may inhibit the active transport and metabolism of sugars. Zinc has also been reported to reduce acid production by S. mutans and has the ability to inhibit glucosyltransferase activity (Sevinc & Hanley 2010). Based on the above discussion, a mechanism of anti-biofilm activity of GZNC against S. mutans is proposed (Figure 8). Biofilm formation in S. mutans is a complex process and more investigation is needed to further understand the mechanisms of biofilm inhibition in the presence of the GZNC.

Cytotoxicity
Despite potential antibacterial activity, the therapeutic use of nanoparticles is limited because of their cytotoxicity against mammalian cells (Yen et al. 2009). Graphene oxide has already been shown to be non-toxic even at very high concentrations (Chang et al. 2011). On the other hand, due to aggregation and other factors, zinc oxide nanoparticles have been reported to be toxic to mammalian cells even at low concentrations (Yuan et al. 2010). Figure S3 shows the dose dependent effect of GZNC on the viability of HEK 295 cells. There was almost 80% viability during the treatment of cells with 200 μg ml −1 of GZNC, which is three times higher than the concentration used for the anti-biofilm experiment. Even at very high concentrations of GZNC (up to 400 μg ml −1 ), the cell viability was found to be > 50% after incubation for 24 h. The results clearly indicate that GZNC causes much lower toxicity compared with zinc oxide nanoparticle alone. CLSM images of the internalization of GZNC in HEK-293 cells ( Figure S4) showed an increase in blue fluorescence compared with the control with increasing GZNC concentrations. Membrane damage was observed at a much higher concentration of GZNC (400 μg ml −1 ). These observations may indicate the promising antibacterial and anti-biofilm activity of GZNC.

Conclusions
This study indicates that GZNC may be an effective antibacterial and anti-biofilm agent against S. mutans. Moreover, sub-MIC GZNC treatment resulted in a significant reduction in biofilm and the cariogenic properties of S. mutans. This is the first study where GZNC has been investigated as a potential coating material for dental implants. GZNC coated acrylic tooth surfaces successfully inhibited S. mutans biofilm (85%) formation. Furthermore, the decreased toxicity of the nanocomposite makes it an effective coating agent for dental implants. However, future work on the stability and practical consequently blocking all further steps in biofilm formation; (B′) GZNC wraps around S. mutans surface making direct contact with zinc oxide nanoparticles thus inhibiting cell to cell contact; (C′) entry of zinc oxide nanoparticle into cell leads to production of reactive oxygen species, further leaching of Zn 2+ ions from zinc oxide nanoparticles inhibits virulence factors of S. mutans and also assists in anti-biofilm action.
implication of this nanocomposite is clearly required before it can be implemented.