Characterization of Enterococcus faecalis in different culture conditions

The aim of this study was to investigate how carbohydrates (glucose or sucrose) affect the characteristics of Enterococcus faecalis (E. faecalis) planktonic and biofilm in vitro. For this study, E. faecalis was cultured in tryptone-yeast extract broth with 0% glucose + 0% sucrose, 0.5% glucose, 1% glucose, 0.5% sucrose, or 1% sucrose. Viability of E. faecalis was examined by colony forming unit counting assays. Biofilm formation was assessed by measuring extracellular DNA (eDNA), a component of the biofilm matrix. Quantitative real-time PCR (qRT-PCR) was performed to investigate the expression of virulence-associated genes. Field emission scanning electron microscopy analysis, confocal laser scanning microscopy analysis, and crystal violet colorimetric assay were conducted to study E. faecalis biofilms. E. faecalis showed the highest viability and eDNA levels in 1% sucrose medium in biofilms. The result of qRT-PCR showed that the virulence-associated genes expressed highest in 1% sucrose-grown biofilms and in 1% glucose-grown planktonic cultures. E. faecalis showed highly aggregated biofilms and higher bacteria and exopolysaccharide (EPS) bio-volume in sucrose than in 0% glucose + 0% sucrose or glucose. The results indicate that the production of eDNA and EPS and expression of virulence-associated genes in E. faecalis are affected by the concentration of carbohydrates in biofilm or planktonic culture.


EPS measurement and morphological observation of E. faecalis biofilms. The effect of carbo-
hydrate on E. faecalis biofilms was evaluated via bio-volume of bacteria and EPS using confocal laser scanning microscopy (CLSM) analysis (Fig. 3). E. faecalis showed the lowest level of bacteria and EPS bio-volumes in 0% glucose + 0% sucrose. The bacteria and EPS bio-volumes were higher in biofilms cultured in the presence of sucrose compared to biofilms cultured in glucose. As shown in the scanning electron microscopy (SEM) images in Fig. 4, E. faecalis biofilms aggregated to higher levels in sucrose compared to those in 0% glucose + 0% sucrose or glucose.
Biofilm formation assay. Biofilm formation of E. faecalis ATCC29212, OG1RF and clinical strains was assessed by crystal violet colorimetric assay. As shown in Fig. 5, the strains showed higher biofilm formation in sucrose than in other media. OG1RF had greater biofilm mass than ATCC29212 in 0% glucose + 0% sucrose, 0.5% glucose, 1% glucose, 0.5% sucrose, or 1% sucrose, whereas clinical isolates formed larger biofilms than ATCC29212 and OG1RF in 1% sucrose.

Discussion
E. faecalis is a pathogenic bacterium that can form biofilm, which leads to infectious disease, resistance to numerous antibiotics, and endodontic treatment failure 4,32,33 . Studies have shown that nutrients in culture media may affect bacterial biofilm formation 34,35 . Glucose was shown to enhance biofilm growth of E. faecalis by inducing Figure 1. Analysis of E. faecalis grown in biofilms and planktonic culture. Viability and eDNA levels of ATCC29212 (black) and OG1RF (white) in (a,c) planktonic culture and (b,d) biofilms grown in 0% glucose + 0% sucrose (0%G + 0%S), 0.5% glucose (0.5%G), 1% glucose (1%G), 0.5% sucrose (0.5%S), or 1% sucrose (1%S) were examined. Data represent the mean and standard deviation of 3 independent experiments. Asterisks denote a statistically significant difference compared to 0%G + 0%S (p < 0.05).  30 . However, the results of the current study showed that the expressions of the esp gene along with other biofilm-associated genes such as asa1, cylB, efaA, ahrC, ebpA, epal, epaOX, and relA were higher in biofilms cultured in the presence of sucrose compared to glucose, whereas it showed higher or similar induction in glucose than in sucrose in planktonic culture. The results of this study suggested that E. faecalis could be more virulent in sucrose than in glucose with higher expression of the virulence-associated genes and greater biofilm   www.nature.com/scientificreports/ formation in sucrose since many cases of bacterial infections are related to biofilms with 1000-fold more resistance against antibiotics than planktonic cells 7,8 . This needs further study. Interestingly, although CFUs of E. faecalis were lower in sucrose than in 0% glucose + 0% sucrose in planktonic culture, it attached to the surface of HA disks more in 0% glucose + 0% sucrose than in glucose or sucrose in biofilm conditions when the disks were not washed (data not shown). However, when the disks were gently washed, unattached E. faecalis was easily removed from the disks and the result showed that the number of CFUs in 0% glucose + 0% sucrose was lower than that in the other media, whereas there was no significant difference of CFUs in biofilms formed in glucose or sucrose with or without washing. This could be the result of a lack of biofilm components such as eDNA in 0% glucose + 0% sucrose, as shown in Fig. 1d, that could give binding sites for E. faecalis and slower growth rates of bacteria in biofilms than in planktonic culture as demonstrated by Evans et al. 17,36 . In other words, E. faecalis could grow slowly in biofilms in sucrose since the bacteria aggregated more and formed compact biofilms in sucrose than in 0% glucose + 0% sucrose or glucose (Fig. 4).
Additional studies have investigated the role of eDNA, a component of the biofilm matrix, and demonstrated that it is essential for biofilm formation 20,37,38 . Thus, eDNA levels were examined and they were higher in biofilms in sucrose than in the other media. This result is in agreement with the result of a study that suggested that the eDNA levels in biofilms were relevant to the CFU numbers in the biofilms 39 . This could be explained by the higher expression of virulence-associated genes that were upregulated in those media and also could be the result of higher EPS bio-volume that could enmesh the bacteria with biofilm components such as eDNA. Interestingly, although the planktonic growth of E. faecalis in glucose was lower than that in 0% glucose + 0% sucrose or sucrose, their eDNA levels were similar to those collected from 0% glucose + 0% sucrose or sucrose in planktonic culture. This could be the result of higher expression of genes including some virulence-associated genes in glucose than in other media, as shown in Fig. 2. This needs further study.
The effect of sucrose on biofilm formation was extensively studied with Streptococcus mutans (S. mutans), one of the major agents that are associated with dental caries, and demonstrated that S. mutans utilized sucrose  www.nature.com/scientificreports/ to produce glucans, an important component of biofilm matrix, leading to a high incidence of dental caries [40][41][42] . However, the effect of sucrose had not been studied with E. faecalis. According to the present study, E. faecalis produced higher levels of eDNA in sucrose and expressed more virulence-associated genes in biofilms than in other media. CLSM analysis showed that E. faecalis produced more EPS in sucrose, suggesting that E. faecalis could utilize sucrose and produce EPS. SEM analysis showed that E. faecalis aggregated more in the biofilm matrix in sucrose than in 0% glucose + 0% sucrose or glucose. E. faecalis biofilm formation of the strains was analyzed by crystal violet colorimetric assay and the result showed that E. faecalis formed larger biofilms in sucrose than in 0% glucose + 0% sucrose or glucose. The results of this study suggested that the biofilm formation ability of E. faecalis in sucrose could be significant clinically with greater biofilm formation and higher virulence-associated gene expression than in 0% glucose + 0% sucrose or glucose since infections of many bacteria are related to the ability of biofilm formation that induces more resistance against antibiotics than planktonic bacteria [7][8][9]34 . In summary, E. faecalis produced more components of the biofilm matrix such as eDNA and EPS and had more highly expressed virulence-associated genes when grown in biofilms in sucrose relative to 0% glucose + 0% sucrose or glucose in biofilms. In planktonic culture, glucose-induced more eDNA/CFU and expression of virulence-associated genes than in 0% glucose + 0% sucrose or sucrose.
Colony forming unit (CFU) assay and eDNA measurement. E. faecalis eDNA and CFUs were measured as previously described with some modifications 39 . Briefly, E. faecalis grown in planktonic culture was collected carefully and centrifuged at 10,000×g at 4 °C for 10 min. The pellet was resuspended in phosphate buffered saline (PBS) and sonicated using a sonifier (VCX 130 PB; Sonics and Materials Inc., Newtown, CT, USA) for 10 s at 7 W to separate eDNA from the bacteria. To collect E. faecalis in biofilms, HA disks were gently washed twice in PBS and sonicated in an ultrasonic bath (Power Sonic 410; Hwashin Technology Co., Seoul, Korea) for 10 min to break the biofilms in 2 ml PBS. The biofilms were then scraped using a spatula and sonicated with a sonifier for 10 s at 7 W. For CFU counting, an aliquot (0.1 ml) of the homogenized E. faecalis was serially diluted and plated on BHI agar plates and incubated at 37 °C. The remaining cells were centrifuged at 10,000×g at 4 °C for 10 min. The eDNA level was measured using a fluorescence microplate reader (HIDEX, Turku, Finland) at an excitation of 485 nm and emittance of 535 nm.

Quantitative real-time PCR (qRT-PCR).
After 24 h incubation in planktonic culture and in biofilms, the bacteria were collected and sonicated. Then, the homogenized bacteria were treated using Trizol Max Bacterial RNA Isolation Kit (Life Technologies, Carlsbad, CA, USA) to extract RNA according to manufacturer's instructions. Complementary DNA (cDNA) was amplified using a QuantiTech Reverse Transcription Kit (QIAGEN, Hilden, Germany). qRT-PCR was performed using a Rotor-Gene SYBR Green PCR Kit (QIAGEN, Hilden, Germany) and a Rotor-Gene Q real-time PCR machine (QIAGEN) with a 72-well rotor. The primers used in this study are shown in Table 1 43 . Relative expression was calculated by normalizing each test gene to the E. faecalis 16s RNA reference gene.

Field emission scanning electron microscopy (FE-SEM). Biofilms formed on the HA disks in 0%
glucose + 0% sucrose, 1% glucose, or 1% sucrose were fixed as previously described with some modifications 44 . In brief, HA disks were in 2.5% glutaraldehyde / 4% paraformaldehyde (Sigma-Aldrich, Saint Louis, MO, USA) in PBS at 4 °C overnight. They were then washed in PBS and a graded series of ethanol (25-100%) was used to dehydrate the samples. The samples were then air-dried, gold-coated, and observed using FE-SEM (Hitachi, Tokyo, Japan). The images were obtained on a Hitachi SU-70 using BSE detector with 10.0 kV voltage acceleration.
Confocal laser scanning microscopy (CLSM). CLSM analysis was performed to investigate the change of E. faecalis biofilms formed in media with different carbohydrates (0% glucose + 0% sucrose, 1% glucose, or 1% sucrose) as previously described 39 . E. faecalis (3 × 10 6 CFU/ml) was cultured in a medium with 1 μM Alexa Fluor 647-labeled dextran conjugate (650/668 nm; Molecular Probes, Eugene, OR, USA), which was incorporated into the EPS matrix for 24 h at 37 °C. Then, E. faecalis on the HA disks were exposed to 2.5 μM SYTO 9 green fluorescent nucleic acid stain (480/500 nm; Molecular Probes) for 20 min at room temperature. Fifteen image stacks (512 × 512 pixel tagged image file format) were obtained using an LSM 510 META microscope (Carl Zeiss, Jena, Germany). The bio-volume of the bacteria and EPS was quantified using. The laser power was 5 mW for Alexa Fluor 647-labeled dextran conjugate (red channel-HeNe laser) and 20 mW for SYTO 9 (green channel-Ar laser). Three individual experiments were performed.

Statistical analysis.
All experiments were performed at least three times. The intergroup differences were estimated by one-way analysis of variance (ANOVA). The data are presented as mean and standard deviation (SD). A p value was considered as statistically significant when it was less than 0.05.

Data availability
The data generated and analyzed in this study are available from the corresponding author upon request.