Effect of Probiotic Clostridium butyricum NCTC 7423 Supernatant on Biofilm Formation and Gene Expression of Bacteroides fragilis

Abstract

Enterotoxigenic Bacteroides fragilis (ETBF) is the main pathogen causing severe inflammatory diseases and colorectal cancer. Its biofilm plays a key role in the development of colorectal cancer. The objective of this study was to determine the antagonistic effects of cell-free supernatants (CFS) derived from Clostridium butyricum against the growth and biofilm of ETBF. Our data showed that C. butyricum CFS inhibited the growth of B. fragilis in planktonic culture. In addition, C. butyricum CFS exhibited an antibiofilm effect by inhibiting biofilm development, disassembling preformed biofilms and reducing the metabolic activity of cells in biofilms. Using confocal laser scanning microscopy, we found that C. butyricum CFS significantly suppressed the proteins and extracellular nucleic acids among the basic biofilm components. Furthermore, C. butyricum CFS significantly downregulated the expression of virulence- and efflux pump-related genes including ompA and bmeB3 in B. fragilis. Our findings suggest that C. butyricum can be used as biotherapeutic agent by inhibiting the growth and biofilm of ETBF.

Introduction

Bacteroides fragilis, an obligate anaerobe, constitutes 1% to 2% of the normal microorganisms in the intestines of a healthy human [1]. However, it is a major pathogen causing inflammatory bowel diseases (IBD) such as Crohn's disease and ulcerative colitis [2, 3]. The virulence factors of B. fragilis such as capsular polysaccharide, outer membrane proteins and enterotoxin known as fragilysin (Bft) are closely associated with biofilm formation and pathogenicity of B. fragilis [4]. Above all, enterotoxigenic B. fragilis (ETBF) secreting Bft not only causes severe inflammatory diseases, but also contributes to the development of colorectal cancer (CRC) [5]. Recent studies have reported that more than 60% of the biofilms formed by ETBF secreting Bft are frequently found in patients with severe inflammatory diseases [6]. In addition, CRC development is affected by spatial arrangement of bacterial communities in a high-dimensional structure, indicating that biofilm shows pro-carcinogenic activity and is essential for CRC initiation [7]. Therefore, it is essential to identify an effective therapeutic alternative that prevents severe inflammatory diseases or CRC by inhibiting the growth and biofilm of ETBF secreting Bft.

Bacteriotherapy using probiotics is a safe and promising approach [8]. A probiotic is defined as a live microorganism that provides health benefits when administered in appropriate amounts [9]. Probiotics have beneficial effects, such as maintaining intestinal homeostasis by regulating the host immune systems and suppressing the clustering of pathogens [10, 11]. Clostridium butyricum strains are probiotic bacteria that are used to prevent antibiotic-associated diarrhea [12]. C. butyricum, an obligate anaerobe, is a common human and animal gut commensal bacterium and accounts for 10-20% of all human stool samples by microbial culture. They secrete an abundance of short-chain fatty acids (SCFAs), mainly butyrate and acetate, which promote the proliferation of enterocytes [13] and accelerate regulatory T cell generation [14]. C. butyricum also possesses antifungal and antibacterial activities against Candida albicans, Clostridium difficile, enterotoxigenic E. coli (ETEC), Vibrio spp. and Helicobacter pylori [15-17]. Further, a recent study by Hayashi et al. revealed that C. butyricum represses acute experimental colitis in mice by stimulating intestinal interleukin-10 (IL-10)–producing macrophages [18].

Although studies have actively investigated probiotics, few studies have analyzed the inhibitory effects of C. butyricum on intestinal biofilm formation by pathogens. Especially, the effect of C. butyricum against the growth and biofilm of ETBF secreting Bft has yet to be reported. In order to investigate the effect and the possible application of metabolites formed by C. butyricum, this study evaluated the antagonistic activity of cell-free supernatant (CFS) extracted from C. butyricum against the growth, biofilm formation and gene expression of B. fragilis. Our work widens our intellectual horizon on the potential antibacterial and antibiofilm activities of CFS derived from C. butyricum in addition to its role as biotherapeutic agent against ETBF secreting Bft.

Materials and Methods

Organisms, Media and Growth Conditions

The wild-type enterotoxigenic Bacteroides fragilis 86-5443-2-2 (WT-ETBF; bft-2), a wild-type non-enterotoxigenic Bacteroides fragilis NCTC 9343 (WT-NTBF), and a recombinant strain transformed via insertion of bft-2 gene into WT-NTBF (rETBF; bft-2) and C. butyricum NCTC 7423 (ATCC 19398), a nontoxigenic strain, were used in this study [19]. B. fragilis strains and C. butyricum NCTC 7423 were cultured in brain heart infusion (BHI; Difco, Becton-Dickinson and Company, USA) supplemented with 1% glucose, 0.1% hemin, 0.5% yeast extract, and 0.05% L-cystine (BHIS) and reinforced clostridial medium (RCM; Difco, Becton-Dickinson and Company), respectively, at 37°C under anaerobic conditions [20]. All stock cultures were stored in broth with 20% glycerol at -80°C until testing.

Preparation of C. butyricum Cell-Free Supernatant

The C. butyricum cell-free supernatant (CFS) was prepared as described previously [21]. In brief, C. butyricum NCTC 7423 was incubated with the reinforced clostridial medium at 37°C for 24 h under anaerobic conditions. The culture was centrifuged at 3,000 rpm for 15 min and the CFS was filtered with 0.2 µm pore size syringe filter (Advantec, Japan). To determine whether the inhibitory activity may be affected by the bacteriocin-like inhibitory substances, the CFS was adjusted to pH 6.5 using 1N NaOH to exclude the effect of organic acid [22].

C. butyricum CFS Susceptibility Testing on B. fragilis Planktonic Cells

To determine the antibacterial activity of CFS derived from C. butyricum, the broth microdilution method was used according to the Clinical and Laboratory Standards Institute guidelines [23], with a few modifications. Briefly, overnight cultures of B. fragilis strains were grown in BHIS broth. The bacterial suspension was adjusted to the final concentration of 1 × 10⁶ CFU/ml and inoculated into a 96-well microtiter plate (BD Falcon, USA) along with the C. butyricum CFS or neutralized CFS to a total volume of 200 µl. The C. butyricum CFS or neutralized CFS was serially diluted two-fold in BHIS broth. Overnight-cultured B. fragilis strains were set as control. Following inoculation, the 96-well microtiter plate was incubated anaerobically at 37°C for 24 h. At the end of the incubation, the growth of planktonic cells was assessed at 600 nm (A₆₀₀) wavelength using a Multiskan GO plate reader (Thermo Fisher Scientific, USA).

Effect of C. butyricum CFS on Biofilm Formation and Preformed Biofilms of B. fragilis

The biofilm formation assay was performed in a 96-well microtiter plate as described previously [24, 25], with some modifications. Overnight-cultured B. fragilis suspensions were diluted in BHIS broth to a final density of 1 × 10⁶ CFU/ml and dispensed into a 96-well microtiter plate. The 2-fold serial dilutions of C. butyricum CFS in BHIS broth were added to wells of a 96-well microtiter plate and incubated at 37°C for 24 h under anaerobic conditions.

To establish the preformed biofilms, a reference to B. fragilis were prepared in BHIS broth at a density of 1 × 10⁶ CFU/ml, added to a 96-well microtiter plate and incubated anaerobically at 37°C for 24 h. At the end of the incubation, the preformed biofilms were rinsed with 0.01 M phosphate-buffered saline (PBS; pH 7.4) to remove non-adherent cells. C. butyricum CFS was diluted two-fold in fresh BHIS broth and inoculated into the preformed biofilms. A 96-well microtiter plate was incubated at 37°C for an additional 24 h under anaerobic conditions. The control was set as mentioned above. The inhibitory effect of C. butyricum CFS on biofilm formation and preformed biofilms of B. fragilis was quantitatively evaluated using the crystal violet assay as described below.

Biofilm Quantitation Via Crystal Violet Assay

The crystal violet assay was performed to evaluate the biofilm biomass quantitatively as described previously [25, 26], with minor modifications. After the biofilm formation, the medium was aspirated and the biofilm was washed with PBS to remove unattached cells from wells. The 96-well microtiter plate was dried at 60°C for 1 h. The remaining biofilms were stained with 1% crystal violet for 5 min and rinsed with sterile distilled water to remove the dye. After drying at 60°C for 1 h, the crystal violet was extracted from the stained biofilms using 33% (v/v) acetic acid. The biofilm biomass was quantitatively determined by measuring absorbance at 570 nm (A₅₇₀).

Biofilm Metabolic Activity - XTT Reduction Assay

The metabolic activity of B. fragilis biofilms was analyzed using the XTT [2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide] reduction assay as described previously [27], with slight modifications. Briefly, the 96-well microtiter plate was used to establish the biofilms of three B. fragilis strains as described above. After the preformed biofilms were washed with PBS, a fresh BHIS broth containing the 2-fold serial dilutions of C. butyricum CFS was dispensed into preformed biofilms and incubated at 37°C for 24 h in anaerobic environments. Subsequently, the preformed biofilms were washed with PBS. The metabolic activity of B. fragilis biofilms was determined using the XTT cell proliferation assay kit (ATCC, USA) in accordance with the manufacturer’s instructions. Prior to the experiment, the XTT reagent was mixed with the activation reagent at a ratio of 50:1 (v/v), and 50 µl of the XTT/activation was dispensed into the preformed biofilms and incubated at 37°C in the dark for 3 h. To analyze the metabolic activity of B. fragilis biofilms, the specific absorbances were calculated at a test wavelength of 475 nm (A₄₇₅) and a reference wavelength of 650 nm (A₆₅₀) [28]. According to the XTT cell proliferation assay protocol, the specific absorbance of the sample is expressed mathematically as follows; Specific absorbance = A₄₇₅ (Test) – A₄₇₅ (Blank) – A₆₅₀ (Test).

Confocal Laser Scanning Microscopy

Confocal laser scanning microscopy (CLSM) was conducted using biofilms of three B. fragilis strains developed on a tissue culture-treated 24-well glass bottom imaging plate (Eppendorf AG, Germany, Cat. no.: 0030741021) as reported previously [29], with some modifications. B. fragilis strains in BHIS were dispensed into the wells at a concentration of 1 × 10⁶ CFU/ml with 2-fold serial dilutions of C. butyricum CFS and incubated at 37°C for 24 h under anaerobic conditions. Overnight-cultured B. fragilis strains were employed as control. Following incubation, biofilms formed on the well were washed with PBS and fixed with 3.7%(v/v) formaldehyde for 1 h. The amino groups in the B. fragilis biofilms were visualized by staining with fluorescein isothiocyanate isomer I (FITC, 10 µg/µl; Sigma-Aldrich, Germany) for 1 h. Carbohydrates of biofilms were stained with Concanavalin A-Alexa Fluor 594 conjugate (Con A, 0.1 µg/µl; C-11253, Molecular Probes, USA) for 30 min. To monitor the extracellular nucleic acids, biofilms were stained with 4, 6-diamidino-2-phenylindoldihydrochloride (DAPI, 1 mg/l; Molecular Probes) for 45 min. After each step, the stained biofilms were washed with PBS to remove unbound staining solution. All steps were conducted in a dark room. The biofilms of B. fragilis strains were visualized at excitation wavelengths of 495, 590, and 358 nm for FITC, Con A and DAPI, respectively, using Zeiss LSM-710 confocal laser microscope (Carl Zeiss, USA) and were imaged using ZEN software (Carl Zeiss).

RNA Extraction and Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

To extract the total RNA of B. fragilis strains, 1 × 10⁶ CFU/ml of B. fragilis in BHIS was inoculated with C. butyricum CFS serially diluted 2-fold and incubated at 37°C for 24 h in anaerobic environments. At the end of incubation, the cells were collected by centrifuging each bacterial suspension at 25,000 ×g for 1 min at 4°C. The total RNA of B. fragilis strains was extracted and purified through NucleoSpin RNA mini Kit (Macherey-Nagel, Germany) in compliance with the manufacturer’s protocol. The concentration and purity of extracted RNA was assessed using BioDrop µLITE (BioDrop Ltd., UK), and 1 µg of template was reverse-transcribed into cDNA in a 20 µl reaction volume, using ReverTra Ace qPCR RT Master Mix with gDNA Remover (TOYOBO, Japan). The qRT-PCR was performed to investigate the relative gene expression of B. fragilis outer membrane protein (ompA) associated with virulence factor and RND-type efflux pump-related bmeB3. Power SYBR Green PCR Master Mix (Applied Biosystems, USA) was employed to analyze the PCR amplification products. The qRT-PCR was conducted using a StepOnePlus Real-Time PCR System (Applied Biosystems). Primer sequences used in qRT-PCR are shown in Table 1 [30-32]. 16S rRNA was used as a housekeeping gene. The thermal cycling conditions of qRT-PCR were as follows: an initial denaturation at 95°C (10 min), followed by 40 cycles of denaturation at 95°C (15 sec), annealing at 58°C (1 min), and extension at 72°C (20 sec). The annealing stage was set at 57°C (30 sec) for the ompA gene and 55°C (1 min) for the bmeB3 gene. The relative gene expression of target genes was normalized to 16S rRNA gene and evaluated using the formula of 2^-ΔΔCT.

Table 1.

Primer sequences used for qRT-PCR.

Name	Sequence (5’ → 3’)	Reference
16S rRNA	Forward AGTAGAGGTGGGCGGAATTC	[30]
	Reverse GTGTCAGTTGCAGTCCAGTG
ompA	Forward GGATATGACGGTGTTGCCAG	[31]
	Reverse TAGCAGCAGCCATGTCATTC
bmeB3	Forward GTACCGGAAGTTCAAGGTGT	[32]
	Reverse GAGCAGCCTCGATATTCTGT

Statistical Analysis

All data are representative of three independent experiments, and the data were indicated as means ± standard deviations (SD). The results were analyzed via one-way analysis of variance (ANOVA) followed by Dunnett’s test to evaluate the significant differences between the treated groups and the control group. The results of qRT-PCR were analyzed using the Student’s t-test. All statistical analyses were conducted using the GraphPad Prism version 5 (GraphPad Software, USA). Statistical significance was considered at *p < 0.05, **p < 0.01 and ***p < 0.001.

Results

Antibacterial Activity of C. butyricum CFS on B. fragilis Planktonic Cells

The antibacterial activity of C. butyricum CFS against B. fragilis strains was determined by measuring growth inhibition. A 50% C. butyricum CFS completely inhibited the growth of WT-ETBF (bft-2), rETBF (bft-2) and WT-NTBF by 99.48%, 98.38% and 100%, respectively (Figs. 1A-1C). A 25% C. butyricum CFS has a lower potential than 50%, but it significantly reduced the growth of B. fragilis strains. However, the growth of rETBF (bft-2) and WT-NTBF was not inhibited by 12.5% C. butyricum CFS. C. butyricum suppresses the growth of other microorganisms by producing a bacteriocin-like inhibitory compound [33]. In addition, neutralized probiotic CFS shows a partial similar effect to bacteriocin [21]. Therefore, to evaluate the effect of neutralized C. butyricum on the antibacterial activity, the C. butyricum CFS was adjusted to pH 6.5 to exclude organic acids in CFS. With the treatment of 50% C. butyricum CFS, the growth inhibition of three B. fragilis strains showed a statistically significant difference (Figs. 1D-1F). However, treatment with the neutralized C. butyricum CFS yielded less antibacterial activity than that of C. butyricum CFS against B. fragilis (Fig. 1), suggesting that the bacteriocin in C. butyricum CFS does not affect its growth.

Fig. 1.

Antibacterial activity of C. butyricum CFS against the growth of B. fragilis planktonic cells. WT-ETBF (bft-2) (A, D), rETBF (bft-2) (B, E) and WT-NTBF (C, F) were incubated in the presence of C. butyricum CFS (A-C) or neutralized C. butyricum CFS (D-F) at 37°C for 24 h under anaerobic conditions. The growth of B. fragilis planktonic cells was analyzed at A₆₀₀ using microplate spectrophotometers. The results are expressed as means ± standard deviations (SD). ** and *** describe significant differences at p < 0.01 and p < 0.001, respectively.

Inhibitory Effect of C. butyricum CFS on B. fragilis Biofilm Formation

To determine the inhibitory effect of C. butyricum CFS on the biofilm formation of three B. fragilis strains, the biofilm was quantified by staining with crystal violet and measuring its absorbance at 570 nm (A₅₇₀) wavelength. The biofilm formation of all strains used in this study was reduced by the C. butyricum CFS in a dose-dependent manner. More specifically, the biofilm formation of WT-ETBF (bft-2) was inhibited by 46.46%, 55.51% and 95.89% in the presence of 12.5%, 25%, and 50% C. butyricum CFS, respectively (Fig. 2A). The rETBF (bft-2) was also suppressed by 32.86%, 35.29%, and 93.48% (Fig. 2B) and WT-NTBF was inhibited by 30.16%, 48.61%, and 90.5% (Fig. 2C) following exposure to 12.5, 25, and 50% C. butyricum CFS, respectively. The results showed that the biofilm formation of all B. fragilis strains was inhibited by more than 90% by treatment with 50% C. butyricum CFS.

Fig. 2.

Inhibitory effect of C. butyricum CFS on biofilm formation of B. fragilis. WT-ETBF (bft-2) (A), rETBF (bft-2) (B) and WT-NTBF (C) were incubated with C. butyricum CFS at 37°C for 24 h under anaerobic conditions. Biofilms of B. fragilis strains were stained with 1% crystal violet and evaluated by measuring the absorbance at A₅₇₀. The data are presented as means ± standard deviations (SD). ***p < 0.001 means statistical significance compared with the control group.

Effect of C. butyricum CFS on Preformed Biofilms of B. fragilis

As shown above, the C. butyricum CFS not only inhibited the biofilm formation of B. fragilis strains, but also potentially eliminated such preformed biofilms. In the presence of 12.5 to 25% C. butyricum CFS, the preformed biofilms by WT-ETBF (bft-2), rETBF (bft-2), and WT-NTBF were eradicated by 9.38 to 26.11%, 54.22 to 72.05%, and 49.28 to 71.9%, respectively (Fig. 3). Although the preformed biofilms of WT-ETBF (bft-2) were less affected than rETBF (bft-2) and WT-NTBF upon treatment with 12.5 to 25% C. butyricum CFS, the preformed biofilms of WT-ETBF (bft-2) were significantly eliminated by 93.24% upon treatment with 50% C. butyricum CFS (Fig. 3A). The rETBF (bft-2) was also eliminated by 90.19% at the same concentration (Fig. 3B). However, only 85% of the WT-NTBF preformed biofilms were eliminated by treatment with 50% C. butyricum CFS (Fig. 3C).

Fig. 3.

Eradicative effect of C. butyricum CFS on preformed biofilms of B. fragilis. The preformed biofilms of WT-ETBF (bft-2) (A), rETBF (bft-2) (B), and WT-NTBF (C) were incubated anaerobically at 37°C for 24 h in the presence of C. butyricum CFS. Biofilms of B. fragilis strains were stained with 1% crystal violet and determined by measuring at A₅₇₀. The data are presented as means ± standard deviations (SD). *p < 0.05 and ***p < 0.001 indicate statistical significance compared with the control group.

Inhibitory Effect of C. butyricum CFS on the Metabolic Activity of Established B. fragilis Biofilms: XTT Reduction Assay

A colorimetric XTT reduction assay was performed to determine the viability of B. fragilis cells within the biofilms in the presence of C. butyricum CFS. The specific absorbance was calculated using an XTT reduction assay. Even though the viability of WT-ETBF (bft-2) biofilm was less than 78.65% compared with other strains upon treatment with 25% C. butyricum CFS, the viability of all B. fragilis strains used in this study declined by more than 90% with 50% C. butyricum CFS (Fig. 4). As shown in Fig. 4, the metabolic activities of all strains were significantly reduced by treatment with the C. butyricum CFS in a dose-dependent manner. The results suggest that treatment with C. butyricum CFS suppresses the viability of B. fragilis cells within the biofilms as well as inhibits and eliminates B. fragilis biofilms.

Fig. 4.

Inhibitory effect of C. butyricum CFS on the metabolic activity of established biofilms. The established biofilms of WT-ETBF (bft-2) (A), rETBF (bft-2) (B), and WT-NTBF (C) were incubated anaerobically at 37°C for 24 h, along with C. butyricum CFS. The metabolic activity of biofilm cells was assessed using XTT reduction assay. Specific absorbance was calculated as A₄₇₅ (Test) – A₄₇₅ (Blank) – A₆₅₀ (Test). The results are presented as means ± standard deviations (SD). Asterisks (***) signify p < 0.05 compared with the control group.

Confocal Laser Scanning Microscopy

The extracellular polymeric substances (EPS) of biofilm matrix are mainly composed of proteins, carbohydrates and extracellular DNAs [34]. To visualize the inhibitory effect of C. butyricum CSF on the biofilms of B. fragilis strains, CLSM was performed by staining biofilms with fluorescent dyes. In the absence of C. butyricum CFS, the biofilms of WT-ETBF (bft-2), rETBF (bft-2) and WT-NTBF showed a thick and compact extracellular matrix and robust bacterial growth, as shown in Figs. 5A, 5C, and 5E, respectively. In addition, bacterial cell aggregation was clearly observed in the biofilms of all strains. By contrast, biofilms of all strains treated with 25% C. butyricum CFS were dispersed and disassembled, and bacterial cells were scattered due to lack of structural components compared with the untreated group (Figs. 5B, 5D, and 5F). Proteins and extracellular nucleic acids in the biofilm were remarkably reduced after treatment with C. butyricum CFS. These results show that C. butyricum CFS alters the architecture of extracellular matrix and reduces the cell density, thickness, and biomass of biofilms.

Fig. 5.

Confocal laser scanning microscopy (CLSM) images of B. fragilis biofilms. Proteins, carbohydrates, and nucleic acids occurring in B. fragilis biofilms were stained with FITC (Green), Con A (Red), and DAPI (Blue), respectively. WT-ETBF (bft-2) control (A) and after treatment with C. butyricum CFS (B). rETBF (bft-2) control (C) and after treatment with C. butyricum CFS (D). WT-NTBF control group (E) and after treatment with C. butyricum CFS (F). Biofilms were examined at 40 × magnification. The scale bar indicates 50 μm.

Effect of C. butyricum CFS on the Expression of Virulence-and Efflux Pump-Related Genes in B. fragilis

To obtain further insight into the molecular mechanism of C. butyricum CFS underlying the inhibition of B. fragilis biofilms, qRT-PCR analysis was performed. The expression of ompA and bmeB3 genes in WT-ETBF (bft-2) and rETBF (bft-2) was considerably down-regulated by C. butyricum CFS in a concentration-dependent manner, whereas that of WT-NTBF was not affected (Fig. 6). Thus, exposure to 6.25%, 12.5%, and 25% concentrations of C. butyricum CFS downregulated the expression of ompA gene in WT-ETBF (bft-2) by 4.22-fold, 6.11-fold, and 7.57-fold, respectively (Fig. 6A), and reduced the expression of ompA gene in rETBF (bft-2) by 1.48-fold, 2.73-fold, and 3.17-fold, respectively (Fig. 6B). At similar concentrations, the expression of bmeB3 gene in WT-ETBF (bft-2) was dramatically suppressed by 101.13-fold, 131.6-fold, and 136.82-fold, respectively (Fig. 6D), and the expression of bmeB3 gene in rETBF (bft-2) was downregulated by 1.37-fold, 1.63-fold, and 1.87-fold, respectively (Fig. 6E). However, the levels of ompA and bmeB3 gene in WT-NTBF were almost unaffected by the C. butyricum CFS, and the differences were not statistically significant (Figs. 6C and 6F).

Fig. 6.

qRT-PCR analysis of B. fragilis outer membrane protein gene ompA and RND-type efflux pump-related gene bmeB3. ompA gene expression of WT-ETBF (bft-2) (A), rETBF (bft-2) (B), and WT-NTBF (C). bmeB3 gene expression of WT-ETBF (bft-2) (D), rETBF (bft-2) (E) and WT-NTBF (F). The results are presented as fold changes relative to the control group. Student’s t-test was conducted to analyze the gene expression between treated and control groups. The results are presented as means ± standard deviations (SD). *p < 0.05 and ***p < 0.001 indicate statistical significance compared with the control group.

Discussion

Biofilms of ETBF carrying the bft gene are involved in the development of severe inflammatory diseases and CRC [7, 35]. Also, biofilm can lead to side effects such as selection of antibiotic-resistant bacteria and suppression of host immune system. For this reason, probiotics represent an alternative approach to preventing growth and biofilm of B. fragilis carrying the bft gene. C. butyricum, a probiotic, can selectively kill pathogens without affecting the normal intestinal flora [24] and possess anti-diabetic [36], antibacterial, antifungal [17] and anticancer effects [37]. However, the inhibition of B. fragilis by C. butyricum has yet to be reported. Therefore, our findings show that CFS extracted from C. butyricum significantly inhibits the growth and biofilm formation by ETBF carrying the bft gene.

The Bft toxin is a major virulence factor of B. fragilis. According to a study conducted by Pierce and Bernstein [38], ETBF carrying the bft gene was formed via mutations in non-toxigenic (NTBF) strain through evolutionary changes. Therefore, this study was conducted using WT-ETBF (bft-2), rETBF (bft-2), and WT-NTBF to confirm that CFS extracted from C. butyricum was strain-specific.

Meanwhile, C. butyricum produces bacteriocin-like inhibitory substances as well as organic acids, mainly butyrate. Bacteriocins are proteinaceous toxins produced by bacteria to inhibit the growth of similar or closely related bacterial strains [39]. Bacteriocin isolated from C. butyricum exhibited antimicrobial effects on diverse bacteria, but no effects on gram-negative bacteria [33]. Consistent with previous studies, our results also show that neutralized CFS, which has the same effect as bacteriocin, do not strongly inhibit the growth of B. fragilis.

HPLC analysis of Isono’s groups revealed that C. butyricum CFS contained 13.6 mmol butyrate, 8.3 mmol acetate, and 4.3 mmol formate [40]. Also, other studies confirmed that the amount of butyric acid among SCFAs increased most significantly when C. butyricum was inoculated on the in vivo models that caused various diseases [41, 42]. Butyrate, which accounts for most of the short-chain fatty acids (SCFAs) produced by C. butyricum, is soluble in water and has amphipathic actions [43]. Since 97% of the biofilm is composed of water, it is indicated that CFS (mostly butyrate) derived from C. butyricum can effectively pass throughout the biofilm [44, 45]. In addition, many studies have determined that butyric acid has an inhibitory effect on the growth and biofilm of various microorganisms such as Trichosporon spp., Vibrio spp., Clostridium difficile [45-47]. For these reasons, we expect that the action of the butyric acid is involved in the inhibitory effect of C. butyricum CFS against B. fragilis biofilm.

A previous study demonstrated that ETBF carrying the bft gene formed a large amount of biofilm compared with the non-toxigenic (NTBF) strain [38]. Similarly, our results also show that WT-ETBF (bft-2) and rETBF (bft-2) with the bft gene form more biofilms than the WT-NTBF without the bft gene. Nevertheless, our data showed that the C. butyricum CFS more effectively inhibited and eradicated biofilms of WT-ETBF (bft-2) and rETBF (bft-2) with the bft gene than WT-NTBF without the bft gene.

Furthermore, this study analyzed the expression levels of ompA and bmeB3 genes according to the concentration of C. butyricum CFS by qRT-PCR. More specifically, the ompA, the most abundant outer membrane protein in B. fragilis, plays a structural role in capsule formation [48]. Lilian et al. have shown a positive correlation between increased adhesion and ompA gene expression in B. fragilis, suggesting that ompA gene was associated with adherence to human intestinal epithelial cells [48]. In addition, the ompA protein is involved in maintaining cell structure in the biofilms of B. fragilis [31]. Based on previous studies, it can be assumed that ompA gene plays a major role in the adhesion and biofilm formation of B. fragilis. Moreover, B. fragilis contains putative luxR orthologues that control biofilm formation, bmeB efflux pump expression, and susceptibility to antibiotics [49]. Also, the RND-type drug efflux pumps from gram-negative bacteria are related to virulence and biofilm formation [50]. This finding suggests that expression of bmeB gene related to RND-type efflux pump is associated with B. fragilis biofilm formation. Our study shows that the C. butyricum CFS inhibits biofilm formation of WT-ETBF and rETBF strains carrying the toxigenic bft-2 gene by reducing the expression of virulence- and efflux pump-related genes (ompA and bmeB3) in B. fragilis. Our results indicated that C. butyricum CFS suppressed ompA and bmeB3 genes only for B. fragilis strains with the bft gene.

Thus, our findings showed that CFS derived from C. butyricum exhibits antibacterial and antibiofilm activities against ETBF carrying the toxigenic bft-2 gene by regulating the virulence- and efflux pump-related genes (ompA and bmeB3). This study demonstrates the effectiveness of CFS obtained from C. butyricum against biofilm formation and provides scientific evidence for the development of new antibacterial and antibiofilm agents. Furthermore, these results suggest the potential of C. butyricum CFS as biotherapeutic agent to prevent and treat IBD and/or CRC caused by ETBF. Since this study does not analyze the chemical composition of C. butyricum CFS, further study needs to thoroughly investigate the effects of the one major component in C. butyricum CFS against the growth and biofilm of ETBF. Although further studies are needed to determine its clinical application, the results suggest that C. butyricum acts as an effective probiotic against ETBF secreting Bft by reducing growth and biofilm formation.