Curcumin as a Promising Antibacterial Agent: Effects on Metabolism and Biofilm Formation in S. mutans

Streptococcus mutans (S. mutans) has been proved to be the main aetiological factor in dental caries. Curcumin, a natural product, has been shown to exhibit therapeutic antibacterial activity, suggesting that curcumin may be of clinical interest. The objective of this study is to evaluate the inhibitory effects of curcumin on metabolism and biofilm formation in S. mutans using a vitro biofilm model in an artificial oral environment. S. mutans biofilms were treated with varying concentrations of curcumin. The biofilm metabolism and biofilm biomass were assessed by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide assay and the crystal violet assay. Confocal laser scanning microscopy was used to analyse the composition and extracellular polysaccharide content of S. mutans biofilm after curcumin treatment. The biofilm structure was evaluated using a scanning electron microscope. The gene expression of virulence-related factors was assessed by real-time PCR. The antibiofilm effect of curcumin was compared with that of chlorhexidine. The sessile minimum inhibitory concentration (SMIC50%) of curcumin against S. mutans biofilm was 500 μM. Curcumin reduced the biofilm metabolism from 5 min to 24 h. Curcumin inhibited the quantity of live bacteria and total bacteria in both the short term (5 min) and the long term. Moreover, curcumin decreased the production of extracellular polysaccharide in the short term. The expression of genes related to extracellular polysaccharide synthesis, carbohydrate metabolism, adherence, and the two-component transduction system decreased after curcumin treatment. The chlorhexidine-treated group showed similar results. We speculate that curcumin has the capacity to be developed as an alternative agent with the potential to reduce the pathogenic traits of S. mutans biofilm.


Introduction
Dental caries, one of the most prevalent infectious diseases worldwide, is a biofilm-mediated, sugar-driven, multifactorial, dynamic disease [1]. Although the pathobiology of dental caries is complicated, it is widely recognized that the formation of dental plaque biofilm is one of the important causes of dental caries [2,3]. Biofilm formation creates an anaerobic and acidic environment that results in the formation and development of caries [4]. Therefore, the eradication of dental biofilm is an effective method of controlling caries.
Streptococcus mutans (S. mutans) has been proved to be the main aetiological factor in dental caries [5,6]. By adhering to solid surfaces, S. mutans can colonize the oral cavity and form a bacterial biofilm [7]. Dental biofilms are microbial aggregates encased in a self-produced extracellular polymer matrix. The phenotype of the bacteria involved in biofilm formation is quite different from the phenotype in the planktonic state [8]. The bacteria in a dental biofilm are far more resistant to unfavourable growth conditions, such as biocides and hydrodynamic shear forces [9,10]. The extracellular polysaccharide (EPS) produced by S. mutans through glycosyltransferases (Gtfs) has been shown to form the matrix of the biofilm [11]. The EPS mediates the irreversible adherence between bacteria to form a high-celldensity biofilm [12], which increases the bacterial resistance to antibiotics.
Currently, many natural products have been considered as alternative or adjunctive anticaries therapies, particularly botanically derived molecules, which offer advantages over synthetic derivatives due to their natural evolution and diminished likelihood of resistance [13]. Curcumin, the major 2 BioMed Research International constituent of Curcuma longa L. or turmeric, has been confirmed as a potential therapeutic antibacterial agent [14]. Curcumin has been reported to inhibit various bacteria: Staphylococcus aureus, Salmonella paratyphi, Trichophyton gypseum, and Mycobacterium tuberculosis [14]. Recent publications have also reported that curcumin is active against a plethora of drug-resistant bacterial strains [14].
Studies have shown the effects of curcumin on oral bacteria associated with dental disease. Song et al. [15] found that curcumin could significantly inhibit the adhesiveness of S. mutans by its effects on collagen and fibronectin. Hu et al. [16,17] reported that curcumin is an S. mutans sortaseA inhibitor and has promising anticaries characteristics based on an antiadhesion-mediated mechanism. Manoil et al. [18] found that blue light-activated curcumin can photoinactivate planktonic S. mutans, but the effect on S. mutans biofilm was poor. These studies proposed the potential use of curcumin as an antibacterial agent and underlined the need to explore the mechanism by which curcumin acts on oral bacteria.
Our research aims to explore the effects of curcumin on the biofilm formed by S. mutans and compares the results with the effects of chlorhexidine. The study contributes to the possibility of a natural medicine with fewer side effects and stronger antibacterial effects, thereby promoting the wider clinical application of curcumin.

Bacterial Strain and Growth
Conditions. S. mutans strain UA159 (ATCC700610) was provided by the Guangdong Microbial Culture Collection Center. The S. mutans growth in brain-heart infusion (BHI, Difco, Detroit, MI, USA) broth medium was measured. For biofilm formation, an overnight culture of bacteria was adjusted to OD = 1.0 (1 * 10 8 CFU/ml). The adjusted bacteria were inoculated into 96-well flatbottom plates. The medium was 1% BHIS, that is, BHI with the addition of 1% (wt/vol) sucrose. The bacteria were grown at 37 ∘ C under anaerobic conditions (5% CO 2 , 10% H 2 , 85% N 2 ).

Minimum Inhibitory Concentration
Assay. The minimum inhibitory concentration (MIC) of curcumin against planktonic S. mutans was determined by microdilution methods following the Clinical Laboratory Standards Institute (CLSI) procedure [19], with some modifications as described below. Curcumin from a 250 mM stock in dimethyl sulfoxide (DMSO) was diluted in BHI. The resulting solution was then serially diluted in sterile BHI broth in a 96-well round-bottom plate to obtain 2000, 1000, 500, 250, 125, 62.5, and 31.25 M curcumin in BHI culture medium. The corresponded concentrations of DMSO were used as control. Each well contained 100 l of serially diluted curcumin, to which 100 l of an overnight culture of S. mutans was added before incubation at 37 ∘ C under anaerobic conditions (5% CO 2 , 10% H 2 , 85% N 2 ) for 24 h. Three parallel samples were prepared at each concentration, and BHI without curcumin was used as a control. The MIC was defined as the lowest concentration at which no visible bacteria grew in the broth.
The sessile minimum inhibitory concentration (SMIC) of curcumin against biofilm S. mutans was determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. Overnight S. mutans UA159 cultures were adjusted with 1% BHIS to an initial optical density at 600 of 0.1 (approximately 10 7 CFU/ml), and equal volumes (200 l) of diluted bacteria were added to 96-well plates and incubated for 24 h at 37 ∘ C in a humidified atmosphere containing 5% CO 2 . After 24 h, the planktonic bacteria were removed carefully, and the biofilms formed on the plates were washed with sterile phosphate buffered saline (PBS). The stock solution of curcumin was diluted in BHIS broth to concentrations ranging from 15.625 M to 1000 M and added to the test wells. The corresponded concentrations of DMSO were used as control. After 24 h of incubation, the BHIS broth with curcumin was removed carefully, and the biofilms formed on the plates were washed again with sterile PBS. To quantify the biofilms, 200 l of 0.5 mg/ml MTT was added to each well and incubated at 37 ∘ C in the dark in a humidified atmosphere for 3 h. Then, the supernatants were removed, and 100 l of 100% DMSO was added to each well. For each well, the absorbance at 570 nm was recorded using a microplate reader. The percentage of biofilm viability was calculated as follows: (absorbance of treated group/absorbance of control group) × 100% [22].

Assays of the Viability and Biomass of S. mutans Biofilm.
The biofilm viability and biomass were tested at 5 min, 10 min, 15 min, 30 min, 60 min, and 24 h using the MTT assay and the crystal violet (CV) assay [20]. The SMIC 50% concentration of curcumin was selected for the assay, and 0.12% (w/t) chlorhexidine was used in the experiment for comparison and the 0.12% (w/t) water was used as control. For the CV assay, the contents of the microplate were removed after incubation, and the wells were washed with PBS, fixed with 95% methanol, washed again, and stained with 0.1% (wt/vol) crystal violet solution for 15 min at room temperature. Subsequently, the microplates were vigorously tapped on napkins to remove any excess liquid and air-dried. The remaining CV was dissolved in 100 l of 100% ethanol for 15 min at room temperature, and, finally, 75 l from each sample was transferred to a new 96-well plate, and the extract was read at 600 nm in a spectrophotometer [20,21].

Analysis of the Bacteria Composition and Extracellular
Polysaccharide ( [23]. After incubation for 24 h at 37 ∘ C in a humidified atmosphere containing 5% CO 2 , treatments consisting of 500 M curcumin, 0.12% chlorhexidine, and 1% BHIS as a control were added to separate dishes. Then, after incubation for 5 min or 24 h, the biofilms were gently washed with normal saline and incubated with 1 M SYTO9 green fluorescent dye at room temperature for 15 min to label the live bacteria. The image collection gates were set to 655-690 nm for Alexa Fluor 647 and 495-515 nm for SYTO9. Images were obtained by CLSM and analysed using the image-processing software COMSTAT [24].

Assay of Gene Expression by Real-Time PCR.
Biofilms grown for 24 h were treated with drugs for 5 min or 24 h. The biofilms were harvested by centrifugation at 12,000 rpm for 5 min, and the total RNA was isolated by ultrasonic crushing and using the RNeasy Mini Kit (QIAGEN, Valencia, CA, USA). The total RNA concentration and purity were determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Pittsburgh, PA, USA). The reverse transcription of the total RNA and quantitative real-time PCR were performed similarl to the procedures described in previous studies [23]. The primer sequences were based on the previous literature and are listed in Table 1 [17,[24][25][26][27]. The 2 −ΔΔCt method was used to calculate the gene expression fold changes in S. mutans.

Statistical Analysis.
Each experiment was independently repeated at least three times. GraphPad Prism version 5.04 (GraphPad Software, San Diego, CA) was used to assess the data. The differences between the experimental group and the untreated control group were statistically analysed using the SPSS 17.0 software. The data were assessed to determine whether they were normally distributed. One-way analysis of variance (ANOVA) and Tukey's test were performed for multiple groups. The unpaired -test was used for two groups. The significance level of the value was <0.05. of 125 M. After treatment with concentrations higher than 125 M, no visible S. mutans growth was observed. The MTT assay showed that, after treatment with serial dilutions of curcumin, the biofilm viability decreased significantly, by 50%, at 500 M. The DMSO has no effect on viability of biofilm (Figure 1(a)). The growth curve at concentrations of 500 M and for the control showed similar trends. The bacteria in both cases reached the stationary phase at 12 h (Figure 1(b)). The variation in biofilm biomass between 500 M and 0 was independent of the growth kinetics. Therefore, a concentration of 500 M was defined as causing 50% inhibition (SMIC 50% ) and selected for the next study.

Inhibition of the Metabolism of Biofilm Formation by
Curcumin. Both MTT and CV assays were conducted to evaluate the alteration in relevant biofilm characteristics of S. mutans under the influence of curcumin. In addition, 0.12% chlorhexidine treatment was used to provide a comparison group. The results of the MTT assay showed that curcumin decreased the biofilm metabolism at both 5 min and 24 h compared to that of untreated biofilm ( < 0.001, Figure 2(a)). Chlorhexidine treatment produced the same trend as curcumin, and the biofilm metabolism decreased significantly more than for curcumin ( < 0.001, Figure 2(a)). The CV assay showed that curcumin and chlorhexidine both resulted in reduced S. mutans biomass at 24 h. Chlorhexidine resulted in a significantly greater decrease than curcumin ( < 0.001, Figure 2(b)). As curcumin decreased the metabolism of the S. mutans biofilm in a short time (5 min), we next compared the short-term (5 min) and long-term (24 h) effects of curcumin on S. mutans biofilm.

Evaluation of the Effect on the Live/Dead Bacteria Ratio in
S. mutans Biofilm. The effects of curcumin and chlorhexidine on the live/dead bacteria ratio in S. mutans biofilm were assayed by CLSM (Figure 3). In the confocal micrographs, green florescence indicates lived cells, while red fluorescence indicates dead cells. The images reflect different green and red florescence intensities from those in the control group. Lower green-stained density than in the control group was observed in both treated groups at 5 min and 24 h (Figures 3(a) and  3(b)). At 5 min, the live bacteria in the curcumin group and chlorhexidine group were reduced, while the total bacteria in the curcumin group were reduced (Figure 3(c)). At 24 h, the live bacteria in the curcumin group were reduced, while the total bacteria were reduced in both groups (Figure 3(d)).

Evaluation of the Effect on the EPS of S. mutans Biofilm.
The three-dimensional reconstruction images obtained from CLSM of the S. mutans biofilm (Figures 4(a) and 4(c)) confirmed that the EPS decreased with both curcumin and chlorhexidine treatment after 5 min ( < 0.05), and this trend continued at 24 h. The thickness of the biofilm showed no difference among the three groups at 5 min but was visibly reduced at 24 h (Figures 4(b) and 4(d)). The SEM image confirmed the change in EPS ( Figure 5). Treatment with curcumin or chlorhexidine was able to reduce the overall quantity of S. mutans EPS compared to the untreated control, as indicated by arrows. Furthermore, long-term treatment (24 h) resulted in a more obvious reduction than short-term treatment (5 min).

Evaluation of the Effect on the mRNA Levels of S. mutans Biofilm Virulence Factors.
Compared with the levels observed in S. mutans biofilms grown in 1% BHIS, the expression levels of gtfB, gtfC, and gtfD in S. mutans biofilms exposed to 500 M curcumin for 5 min were significantly decreased by 0.57-, 0.59-, and 0.59-fold, respectively. At 24 h, the expression levels of gtfB, gtfC, and gtfD in S. mutans were also decreased, and there was no significant difference from the levels at 5 min. Meanwhile, the expression of gtfB, gtfC, and gtfD was also downregulated under the influence of 0.12% chlorhexidine for 5 min and 24 h; there was no significant difference between two time points (Figure 6(a)). In Figure 6(b), we observed no significant difference in the expression of scrA at either 5 min or 24 h. However, the expression of atpH and scrB decreased in the presence of curcumin ( < 0.05). The expression of atpH decreased further over time. In contrast, the expression of scrB increased over time.
The entire set of virulence genes involved in quorumsensing signalling mechanisms was found to be downregulated after treatment with curcumin. Treatment with curcumin for 5 min repressed the expression levels of luxS, comC, comD, and comE to 50.7%, 57.4%, 64.8%, and 54.1%, respectively, as compared to the control values. Likewise, chlorhexidine repressed the gene expression levels to 50.1%, 52.3%, 60.7%, and 52.3%, respectively. After treatment with the drugs for 24 h, only the expression of comDE showed a significant difference.
The expression levels of srtA and spaP in the biofilms treated with curcumin for 5 min were significantly downregulated by approximately onefold compared with the levels in untreated biofilm, and the same tendency was shown by chlorhexidine (Figure 6(d)). After 24 h of treatment with either drug, the expression levels of srtA and spaP showed no significant difference.

Discussion
Dental caries, a common infectious disease worldwide, is caused by biofilm known as dental plaque that result from the adhesion of bacteria to tooth surfaces. The effective eradication of cariogenic biofilm and the pathogenic microorganisms within is the key way to prevent tooth decay [28]. S. mutans is the main aetiological microorganism involved in biofilm [29]. In this study, we used curcumin, a natural foodgrade product, to control the formation of S. mutans biofilm.
Previous studies have shown that curcumin could inhibit the viability of S. mutans with an MIC value of 125-175 M [17]. Similar MIC results were obtained in this study, but the SMIC values here were much higher than the MIC. As is well known, it is more difficult to inhibit the viability of S. mutans biofilm than planktonic bacteria because the planktonic bacteria are more susceptible to the effects of commonly used antimicrobial drugs than the bacteria embedded in a biofilm [30,31].
Accordingly, we evaluated the effect of curcumin on S. mutans biofilm over time. At the SMIC 50 , curcumin was shown to exert both a short-term effect and a long-term effect on the viability of S. mutant biofilm, which was also true for 0.12% chlorhexidine. Previous studies showed the effect of chlorhexidine on S. mutant biofilm. Yang et al. [32] reported that S. mutans biofilm was susceptible to 0.12% chlorhexidine, with a biomass reduction of over 80% observed after treatment with 0.12% chlorhexidine for 1 h. Tamura et al. [33] found that 8 g/ml chlorhexidine was effective in removing biofilm after 24 h. All of these results illustrate that chlorhexidine has antibiofilm effects, which was consistent with our results. The short-term effect of curcumin on S. mutans biofilm also suggested that curcumin has the potential to be a new antibiofilm drug because it exhibited similar effects to chlorhexidine.
S. mutans has a regulatory network to integrate its cellular response to environmental change [6]. This study found that, after treatment with curcumin, the amounts of viable bacteria and total bacteria were reduced, as confirmed by the results of CLSM. Because curcumin exhibited no effect on the growth kinetics of S. mutans, the reduced quantity of live bacteria indicated a direct effect of curcumin on S. mutans.
A biofilm is a highly dynamic and structured community of microbial cells that are enmeshed in an extracellular matrix of polymeric substances such as exopolysaccharides, proteins, and nucleic acids [34]. As reported previously, the production of EPS by S. mutans on a surface enhances the local accumulation and clustering of microbes. In addition, as the biofilm develops, the spatial heterogeneities resulting from EPS synthesis form a complex 3D matrix architecture and create environmental and protective niches within the biofilm that can directly modulate caries pathogenesis [34,35]. Curcumin was shown to destroy the structure of EPS in the short term, decreasing the biomass of EPS, the quantity of total bacteria, and the percentage of lived bacteria. With prolonged curcumin exposure, the structure of the EPS was badly damaged, and the thickness of the biofilm and the number of total bacteria were decreased demonstrably. Khan et al. reported that the extract of Trachyspermum ammi seeds could modulate the expression of specific virulence genes by S. mutans, which disrupted the accumulation and structural organization of EPS, ultimately affecting the cariogenicity of S. mutans [36]. Xu et al. found that the tea catechin epigallocatechin gallate could inhibit the formation of S. mutans biofilm by suppressing the gtfs gene [37]. These studies serve as a reminder that the effects of natural products on the EPS of S. mutans biofilm are usually exerted by regulating cariogenic genes. As previously reported, the regulatory systems in S. mutans, such as the EPS synthesis system, carbohydrate metabolism system, quorum-sensing system, and twocomponent transduction system, are connected to the ecological balance and to bacterial carcinogenesis [6,38]. The results showed that curcumin decreased the expression of cariogenic genes in the plaque of S. mutans. Although the majority of the 24 h qPCR results were not statistically significant, the values between the control group and curcumin group were close to 0.05, and the decrease was pronounced. Thus, curcumin has an inhibitory effect on certain genes in S. mutans at 24 h. Gtfs, encoded by the gtfBCD genes, are indispensable for the utilization of glucose and for EPS synthesis in S. mutans. They are important factors in biofilm formation and the development of caries [24]. GtfB synthesizes water-insoluble polysaccharide [39], which is the main component of the EPS. Downregulation of the expression of gtfB in curcumin might explain the decreased biomass of EPS in the short term. Oral bacterial aggregation is mediated by interactions between surface-associated glucan-binding proteins (GbpBs) that adhere to glucans, thereby promoting plaque formation [40]. The expression of gbpB was decreased in both treatment groups, which may relate to the reduced thickness of the biofilm. Fructosyltransferase (FTF) is an enzyme that converts sucrose to extracellular homopolymers of fructose, the fructans. FTF is the product of the fruA gene and is an exo--d-fructosidase that releases fructose from (2.6)and (2.1)-linked fructans and cleaves fructose from sucrose and raffinose [41][42][43]. Both ftf and fruA expressions were dysregulated after short-term exposure to curcumin.
In S. mutans, the scrA gene in the phosphotransferase system (PTS) encodes a high-affinity permease, which internalizes sucrose. Then, intracellular sucrose-6-PO 4 is first hydrolyzed by ScrB, a sucrose-6-PO 4 hydrolase, to produce fructose and glucose-6-PO 4 . In addition, the downstream of scrA and scrB could encode a regulator to change the expression level of scrAB. Both scrAB genes were downregulated in our study, which might contribute to the reduction in the S. mutans biofilm. In the oral environment, S. mutans is subjected to rapid pH fluctuations arising from the availability of carbohydrates. The atpH gene in S. mutans encodes subunit C of a multisubunit enzyme (F1F0-ATPase) involved in intracellular pH regulation and acid tolerance [44,45]. Gene expression analysis found that atpH was downregulated in S. mutans biofilm after treatment with curcumin for 5 min and 24 h. Thus, curcumin inhibits the acid tolerance ability of S. mutans and probably decreases the cariogenic traits of S. mutans.
The quorum-sensing system is an essential component of entire gene regulation networks responsible for the adaptation of bacteria in biofilms [46]. Two well-studied quorumsensing systems exist in S. mutans: the ComCDE system, which can enable intraspecies cell-cell communication and has been proved to have a positive regulatory effect on the expression of biofilm-related genes such as gtfB, gtfC, and gbpB in S. mutans, and the LuxS system, which can catalyse the formation of the signal peptide AI-2 to mediate interspecies and intraspecies interaction in the multispecies plaque community [6,47]. The findings of the current work suggest that curcumin can potentially contribute to reducing the quantity of lived bacteria in the biofilm and decrease the biomass of EPS, by decreasing the cariogenic potential of S. mutans biofilm, via inhibiting the expression of ComCDE system and LuxS system. The short-term inhibitory effect is more significant than the long-term effect. There is a decreasing tendency in the VicR expression of S. mutans after curcumin treatment, suggesting that curcumin may act by inhibiting the expression of VicR. The regulatory network of cariogenic genes in S. mutans is complex. It was reported that the ComR/ComS is the second quorum-sensing system in S. mutans [48]. The cell-to-cell communication system in S. mutans is not clear. Further research is required in this field.
S. mutans possesses a series of cell-surface proteins such as SpaP (also known as antigen I/II, Pac, P1 and antigen B), which are anchored to the cell envelope by sortase A (SrtA). Sortase A is encoded by the gene srtA and recognizes a motif consisting of leucine, proline, X, threonine, and glycine (LPXTG, where X is any amino acid) [33,49]. Natural products attract substantial attention because they can inhibit srtA in vitro and in vivo but do not significantly decrease the viability of bacteria. The current study demonstrates that curcumin can inhibit the expression of spaP and srtA with an SMIC 50 of 500 M. The present study explored the effect of curcumin on S. mutans UA159. However, S. mutans has heterogeneity in different strains. Much more strains of S. mutans are needed to be test in future.
In conclusion, curcumin has both a short-term and a long-term antibacterial effect. Additionally, it is a food-grade natural product with a similar effect to that of chlorhexidine and it takes effect faster than 0.12% chlorhexidine. Curcumin could be an alternative strategy to treat oral disease. Curcumin is a promising antibacterial agent with potential for wider clinical application in the future.

Conflicts of Interest
The authors declare that they have no conflicts of interest.