Dental plaque or dental biofilm is a dynamic and complex community of microorganisms found on a tooth surface, embedded in a matrix of polymers of host and bacterial origin. This structure forms via an ordered sequence of events, resulting in a structurally- and functionally-organized, species-rich microbial community (Marsh, 2004). Under health conditions, dental plaque plays an essential role in natural host defense mechanisms, protecting the host from invasions by exogenous pathogens; however, when the homeostasis is disrupted, changes in the species composition of dental biofilms can lead to diseases such as caries (Filoche et al., 2010).

During caries development, remarkable changes occur in the microbiota on the tooth surface towards the predominance of acidogenic and aciduric bacterial species such as Streptococcus mutans (Takahashi and Nyvad, 2011; He et al., 2015). It is widely accepted that S. mutans is one of the main caries-related bacteria, since it is responsible for the beginning of the caries process on enamel and root surfaces (Tanzer et al., 2001).

Streptococcus sanguinis is among the most abundant species within the first few hours of biofilm formation of newly cleaned tooth surfaces (Li et al., 2004). This pioneer colonizer is thought to play a beneficial role in the oral cavity (Caufield et al., 2000). It is of interest that both S. mutans and S. sanguinis have been known for their antagonism at the ecological level (Giacaman et al., 2015). Epidemiological studies showed that early colonization by S. sanguinis in infants results in delayed colonization of S. mutans. Conversely, high levels of S. mutans in the oral cavity correlate with low levels of S. sanguinis (Caufield et al., 2000). A possible molecular mechanism underlying these fascinating interspecies interactions relies, at least in part, on antimicrobial compounds such as bacteriocins and hydrogen peroxide, which are produced by S. mutans and S. sanguinis, respectively, ultimately to create an environment that favors the colonization of one group of organisms over the other (Kreth et al., 2005; Giacaman et al., 2015).

Based on the notion that dental plaque is a complex ecosystem constituted by both benign and pathogenic bacteria populations, it is desirable that compounds targeted to treat dental plaque-related diseases be selective in their action, preserving the benign bacteria and inactivating the pathogenic ones in order to lead to the re-establishment of a health-compatible species composition (Filoshe et al., 2010). Many studies have shown that plant products can be promising agents for the prevention of dental caries, especially due to their antimicrobial properties against S. mutans (Limsong et al., 2004; Yatsuda et al., 2005; Percival et al., 2006; Rukayadi and Hwang, 2006; Tomczyk et al., 2011).

Piper aduncum L. (Piperaceae) is a tropical shrub widespread in South and Central America, growing naturally in the Amazon and in the Atlantic Forests of Brazil. This plant has been widely used for medicinal purposes and its antibacterial properties have also been described, including against Streptococcus species (Lentz et al., 1998; Kloucek et al., 2005). Thus, the main aim of this study was to evaluate the in vitro antimicrobial activity of extracts and fractions of P. aduncum on the growth, sucrose-dependent cell adherence and acidogenicity of S. mutans. Furthermore, to assess the possible effects of these plant products towards maintaining or restoring the health-compatible state of dental plaque, their inhibitory activity on growth of S. sanguinis was also examined.

Microorganisms

The microorganisms used in this study were S. mutans IM/UFRJ and S. sanguinis ATCC 10557. The culture was grown in Brain Heart Infusion (BHI) broth at 37°C for 24 h, under anaerobic conditions (anaerobic chamber model 1025; Forma Scientific Company, Marietta, OH, USA, containing an atmosphere of 85% N₂, 10% H₂ and 5% CO₂). The stock organism was stored in BHI broth containing 50% glycerol at -80°C.

Determination of minimum inhibitory concentration (MIC)

The MIC values were determined based on the broth microdilution method, according to Clinical and Laboratory Standards Institute (CLSI) guidelines (CLSI 2006), with modifications. Briefly, the assays were performed in polystyrene microtiter plates with 96 flat-bottomed wells. Two-fold dilution series of extracts and fractions (concentrations ranging from 10 to 0.08 mg/mL) were tested. Diluted suspensions (100 µL) of each bacterial strain were added to 100 µl of various concentrations of vegetal products diluted with the BHI broth to reach a final bacterial count of approximately 5 × 10⁵ CFU/mL. Growth and sterility controls were included for each assay and tests were performed in triplicate in at least three independent experiments. The vehicle (DMSO) served as negative control and was used at the final concentration ≤ 4%. Chloramphenicol MIC for S. aureus ATCC 29213 was included for quality control, and its value (8 μg/mL) was within established ranges as published by the CLSI guidelines. The microdilution plates were incubated at 37°C for 24 h in an anaerobic atmosphere (85% N₂, 10% H₂ and 5% CO₂). The MIC was defined as the lowest concentration of extract that completely inhibited visible growth of microorganisms in the microdilution wells.

Inhibition of bacterial adherence to glass surface

To assess the bacterial adherence of S. mutans to a smooth glass surface, the bacteria (approximately 10⁶ CFU/mL) were grown at 37°C for 6 h at an angle of 30° in a glass test tube with 1ml of BHI containing 1% (weight by volume (wt/vol)) sucrose plus two-fold dilution series of extracts and fractions (concentrations ranging from 0.62 to 0.04 mg/ml), as described earlier (Limsong et al., 2004), with modifications. After incubation, the tubes were washed three times with 5 ml of saline solution (NaCl 0.85%) and attached cells were stained with 1% crystal violet. The concentration for total bacterial adherence inhibition (TBAI) was defined as the lowest concentration that allowed no visible cell adherence on the glass surface. All determinations were performed in triplicate.

Effect of extracts and fractions on acid production

The assay was performed by standard pH drop with dense cell suspensions, according to Belli et al. (1995), with modifications. S. mutans cells from suspension cultures were harvested, washed three times in salt solution (50 mM KCl and 1 mM MgCl₂) and resuspended in 5 ml of the same salt solution, which was titrated to pH 7.2 with 0.1 M KOH. Glucose (100 mM final concentration), with and without different concentrations of extracts and fractions was added and the pH fall was assessed with glass electrode over a period of 1 h. Three independent assays were performed, and a statistical analysis was carried out using the Student’s t-test. Differences between control (no treatment) and treatment with plant products were considered statistically significant if p <0.05.

Time-kill assays

Time-kill curves were obtained according to Santos et al. (2007), with adaptations. Starting inocula of approximately 10⁶ CFU/ml of each bacterial strain was grown anaerobically at 37°C in BHI broth until the middle of the exponential growth phase (approx. 3 h for S. sanguinis and 6 h for S. mutans). Then, the crude ethanol extract (final concentration of 0.31 or 0.16 mg/ml for S. sanguinis and S. mutans, respectively) or its fractions (final concentration of 0.62 mg/ml for J or 0.31 mg/ml for L) were added to each test vial. No vegetal product was added to the control vial. Cultures were then incubated at 37°C under anaerobic conditions. Samples were removed to determine viable counts every 30 min over a 2.5 h period. Ten-fold serial dilutions were prepared in sterile saline and 0.1 ml of each dilution was plated onto BHI agar. The plates were incubated at 37°C for 24 h, at which time the number of colonies was determined.

All four crude extracts inhibited the growth of both bacteria tested with MIC values ranging from 0.16 to 2.5 mg/mL (Table 1). The MICs for S. sanguinis were consistently higher (2 to 4-fold) than those found to S. mutans, and the crude extract yielded by maceration technique (MAC) was the most effective. Thus, MAC was fractionated by gel permeation chromatography, yielding 17 fractions, designated as A to Q (data not shown). Only seven of them (G, I, J, L, M, N, Q) were available in sufficient amounts to test their inhibitory effect on bacterial growth (Table 1). The G, N and Q fractions displayed no inhibitory activity, whereas the J and L fractions exhibited the highest activities against S. mutans and S. sanguinis, respectively. It is noteworthy that the J fraction was approximately eight times more active against S. mutans than on S. sanguinis, showing also a lower MIC than that of the MAC.

MAC and its J and L fractions had the strongest antibacterial activity against planktonic cells, and were thus selected for the time kill test. The results revealed that these agents at MIC concentrations exhibited a bacteriostatic, but not a bactericidal effect against both S. sanguinis and S. mutans in exponential growth, since the decrease in the bacterial count for every species, relative to the starting inoculum, was <3 log (Figure 1).

To determine the effect of plant products on bacterial acid production, the pH of dense suspension of S. mutans was recorded during 60 min after glucose pulse. The pH of the suspension not exposed to the plant products (control) decreased rapidly from pH 7.20 to pH 4.22 and 3.57 after 5 min and 60 min, respectively (Table 2). However, the presence of both MAC and J fraction, at MIC concentrations, significantly reduced the rate of acid production by S. mutans, when compared to the control, at all the time intervals tested (5 to 60 min). MAC was actually somewhat more potent than its J fraction. The L fraction showed the weakest inhibition of acid production, significantly reducing the pH drop only at 15 and 30 min after glucose pulse.

In order to study the effect of the plant products on sucrose-dependent adherence of S. mutans, the study determined their concentration for total bacterial adherence inhibition (TBAI) to a glass surface. MAC (Figure 2) and its L fraction (data not shown) had a similar inhibitory activity (TBAI = 0.16 mg/ml). However, the J fraction was the most effective agent, showing inhibitory effect at a concentration of 0.08 mg/mL (data not shown).

Dental caries remains the most prevalent dental disease in many countries therefore being one of the greatest challenges in oral health care (Bagramian et al., 2009). Although the oral microbiota is quite diverse and complex, S. mutans has been recognized as an important etiological agent in human dental caries (Loesche, 1986; He et al., 2015).

It is widely accepted that this disease appears as a result of the breakdown of the microbial homeostasis due to a more frequent exposure of plaque to low pH following an increased frequency of sugar intake. This acidic condition provides a selective pressure that allows overgrowth of acidogenic and acid-tolerant species, such as S. mutans, whereas at the same time suppressing acid-sensitive bacteria such as S. sanguinis (Marsh, 1994).

Consequently, caries control can involve direct use of antibacterial agents to suppress bacterial overgrowth. Nevertheless, the major drawback is that antibacterial products currently in use are not selective in their action, affecting both pathogenic and beneficial bacteria (Marsh, 2010). Thus, the searches continue to find an ideal chemical agent that could control the levels of pathogenic bacteria while preserving the beneficial properties of the resident oral microbiota.

In the present study, S. mutans was more sensitive to P. aduncum extracts than S. sanguinis. Among all of the tested extracts, the J fraction yielded by gel permeation chromatography from MAC was the most effective, inhibiting the growth of S. mutans at MIC value approximately eight times lower than that of S. sanguinis (0.08 and 0.62 mg/ml, respectively). In an oral environment in which S. mutans is dominant, this differential action, in addition to well-known molecular mechanisms of interspecies competition between S. sanguinis and S. mutans (Kreth et al., 2005), would result in microbial shifts, and S. sanguinis would take the place of S. mutans in the ecosystem, restoring the health-compatible state of dental plaque.

According to Ríos and Recio (2005), a significant problem in many studies is to claim positive activity for excessively high concentrations. They consider that plant extracts that are active at concentrations lower than 0.1 mg/ml have a good potency level. Thus, based on this criterion, the study showed that J fraction has promising activity against S. mutans. It is possible to speculate that there are specific compounds in this fraction that can inhibit the growth of S. mutans at low concentrations. In addition, the study fractionation process had good result towards enhancing the antibacterial activity against S. mutans, since the J fraction showed higher activity than its corresponding crude extract (MAC) only on this bacterium.

Results of a previous study have shown that there were differences between the chromatographic profiles of the P. aduncum extracts yielded by different extraction methods. The MAC had the highest content of sesquiterpenes, which accounted for more than 97% of the total identified compounds (Santos et al., 2013). Sesquiterpenes have been extensively described in the literature for their antibacterial properties (Paduch et al., 2007; Saleem et al., 2010), including against S. mutans (Kubo et al., 1992). These latter authors reported that nerolidol, which was one of the most abundant sesquiterpenes present in MAC (Santos et al., 2013), has potent activity against S. mutans, with MIC of 0.025 mg/ml (Kubo et al., 1992). Thus, the good performance of the MAC and its J fraction can be related to their high content of sesquiterpenes. The mechanism of action of terpenes is not fully under-stood, but it is thought to involve membrane disruption associated with their lipophilic character (Cowan, 1999; Paduch et al., 2007). However, the reason for which MAC and its J fraction inhibited S. mutans and S. sanguinis to varying extent remains to be clarified.

It is important to note that the present results are different from those of Lentz et al. (1998). They reported that ethanol extract from P. aduncum collected in Honduras had no activity against S. mutans but revealed measurable antibiosis on S. sanguinis. Since in this study the agar well test was employed, MIC values were not determined. These divergent results can be explained by many factors, including differences in the geographical locations of the plants (Honduras × Brazil), extraction methods (percolation × maceration) and antibacterial assays (agar-diffusion method x dilution method). Accor-ding to Cos et al. (2006), several methods to detect the extract activity are available; however, since they are not equally sensitive or not based upon the same principle, results may be profoundly influenced by the chosen method. The diffusion method is not appropriate for testing non-polar samples or samples that do not easily diffuse into agar. Contrarily, in general, dilution methods are appropriate for assaying polar and non-polar samples. Kubo et al. (1992) showed that some active sesquiterpenes against S. mutans by broth dilution method did not have any activity by an agar-diffusion method.

Besides inhibiting bacterial growth, J fraction at a concentration of 0.08 mg/ml had also biological activity against two of the main virulence traits of S. mutans: sucrose-dependent adherence and acidogenicity. Since the concentration that inhibited bacterial growth was the same as that affected the virulence factors, it can be suggested that the anti-virulence effect of this fraction may be due to its antibacterial activity rather than due to a direct effect on specific virulence traits. However, at least for acidogenicity, the effect of J fraction occurred at early times (5 to 60 min) in which it has not bactericidal activity, according to the time kill assays.

The time kill assays were carried out for the maximum length of time of 2.5 h because, based on the present results, P. aduncum extracts could be thought to be used in oral care products. Thus, whether in this exposure time a bactericidal effect was not yet achieved in vitro, where there is fixed concentration of agent, probably under the in vivo conditions these extracts also will fail to show bactericidal activity because of their lack of persistence in the mouth. Chemical antimicrobial compounds used as a mouthwash tend to be rapidly dislodged, diluted or removed (Addy, 1994; Marsh, 2010).

In order to analyze the inhibitory effect of the extracts on adherence of S. mutans, we performed the adherence assay to glass surface. According to Limsong et al. (2004), in this assay, the adherence is mediated by glucan as well as the in vivo situation. This is an interesting issue because the establishment of S. mutans on the tooth surface is rendered irreversible only after the synthesis of sticky water-insoluble glucan from sucrose by enzymatic action of glucosyltransferase and the subsequent cell-to-cell aggregation (Nostro et al., 2004). Thus, the effect of the extracts from P. aduncum can be of value to prevent both adherence and accumulation of S. mutans on the tooth surface.

The findings that P. aduncum extracts exhibit a preferential antimicrobial activity towards S. mutans compared with S. sanguinis, in addition to their ability to inhibit sucrose-dependent adherence and reduce the level of acid production from S. mutans, suggest that this plant may have a potential for further exploitation in dentistry to prevent dental caries. Future studies should be conducted in order to find the compounds of P. aduncum responsible for its anti-S. mutans properties.

The authors have not declared any conflict of interests.

This work was supported by Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG). The authors sincerely thank Dr. Beatriz Gonçalves Brasileiro for botanical identification and Coordenadoria de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for a postgraduate fellowship to CFM. Finally, they wish to thank Dr. Luiz de Macêdo Farias (Departamento de Microbiologia, Universidade Federal de Minas Gerais, Brazil) who kindly provided the bacterial strains used in this study.