Effect of prebiotics on Bacteroides sp. adhesion and biofilm formation and synbiotic effect on Clostridioides difficile

The normal gut microbiota comprises two major phyla: Bacteroidetes and Firmicutes [1]. Bacteroidetes are anaerobic, non-spore forming Gram-negative rods, and these include Bacteroides fragilis and Bacteroides thetaiotaomicron [2,3]. Some studies have demonstrated the health benefits and probiotic potential of B. fragilis and B. thetaiotaomicron, indicating the essential role of common members of the gut microbiome [4–6]. B. fragilis is classified into non-enterotoxigenic B. fragilis (NTBF) and enterotoxigenic B. fragilis (ETBF) [2,7,8]. ETBF causes acute and chronic intestinal diseases; however, approximately 20% of the human population appears to be asymptomatic carriers of ETBF [2,9,10]. B. fragilis forms biofilms on the colonic mucosal membrane, which can help in the colonization of microorganisms and confer protection from environmental stress conditions [11–13]. Biofilm formation and production of enterotoxin (called fragilisin, encoded by the gene bft) are two main virulence factors of ETBF [13].

The consumption of antibiotics disrupts the composition of gut microbiota and reduces the number of anaerobic bacteria. The abundance of Bacteroides sp. was reported to be significantly decreased in patients with Clostridium difficile infection (CDI) [7]. C. difficile belongs to the phylum Firmicutes and is a Gram-positive anaerobe that can form biofilms [14,15]. The capacity of adherence to host tissues is essential for successful colonization of C. difficile on the gut layer [16,17]. The first-line therapy for CDI includes the use of fidaxomicin and vancomycin. Metronidazole may be used for the first episode of non-severe CDI access to vancomycin and fidaxomicin is limited [14,18]. However, high recurrence rate is the main clinical challenge, and novel therapeutic modalities need to be developed to overcome this challenge [19]. New therapies such as prebiotics, probiotics and fecal microbiota transplantation have been recently proposed to enrich gut microbiota [20]. Prebiotics are defined as a substance that are selectively utilized by host microorganisms conferring a health benefit [21]. Probiotics are live microorganisms which administration in adequate amounts confer a health benefit on the host [22]. According to the European guidelines there is insufficient evidence to recommend the use of probiotics in CDI treatment. Additional concerns relate to cases of fungemia caused by probiotic Saccharomyces boulardii in immunocompromised patients [23,24]. The advantage of prebiotics is their safety.

The prevention of adhesion using prebiotics at an early stage following the exposure of the host to pathogens could prevent the disease [25,26]. In a previous study, prebiotic fructooligosaccharides (FOS) and candidates for prebiotic mannose (MAN) exhibited anti-adhesive and anti-biofilm activity against C. difficile strains in vitro [27]. Prebiotics are commonly used in diet and medicine, but their effects on adhesion and biofilm formation by B. fragilis and B. thetaiotaomicron have not been studied. The mechanism of action of most probiotics is microbe–epithelial and immune cell interaction. Probiotics inhibit the growth of pathogens or the expression of their virulence factors. In addition, they prevent colonization by pathogenic bacteria, improve the gastrointestinal barrier and modulate mucosal and/or systemic immune response [28]. Some study demonstrated the negative correlation between C. difficile infection and the abundance of Bacteroidetes and the protective role of B. fragilis [29]. Conversely, NTBF strains have been considered as possible next-generation probiotics with the potential to attenuate pathogen-induced inflammation [30]. Evidence shows that B. fragilis can prevent CDI in a mouse model by resisting pathogen colonization and improving gut barrier integrity [31]. In vitro experiments have shown that B. fragilis directly inhibits C. difficile adherence to HT-29 cells [32].

The main aim of this study was to investigate the effects of standard and candidate prebiotics on adhesion and biofilm formation by B. fragilis and B. thetaiotaomicron to various human epithelial cell lines. We also examined the effect of different prebiotic concentrations on biofilm formation by Bacteroides sp. strains. An additional aim was to investigate the effect of NTBF or B. thetaiotaomicron in combination with FOS on the adhesion and biofilm formation by reference C. difficile 630.

Materials & methods

Bacterial strains

In this study, we used six bacterial strains: a reference C. difficile 630, B. fragilis strains (reference NTBF IPL E323, clinical NTBF no. 157/3, clinical ETBF no. 093), and two B. thetaiotaomicron strains (reference ATCC 29741 and clinical no. 1397/1) clinical strains, were isolated from the stool samples of patients suspected of Clostridioides difficile infection (CDI). The gene-encoding enterotoxin (fragilysin) of B. fragilis was detected by polymerase chain reaction [9]. All the strains were collected from Anaerobic Laboratory (Department of Medical Microbiology, Medical University of Warsaw). Tested strains were kept at -70°C using Microbank^™ system (Pro-Lab Diagnostics, UK). Before routine use, we thawed strains and cultured on Columbia agar plates with 5% sheep blood (Beckton Dickinson, USA). Strains were incubated in anaerobic conditions at 37°C for 48 h. Anaerobic atmosphere was obtained using Genbag and Genbox anaer gas generators (bioMérieux, France).

Preparation of prebiotics & candidate prebiotics

The standard prebiotics used were FOS, galacto-oligosaccharides (GOS), inulin (INU, obtained from chicory) and lactulose (LAC). Candidate prebiotics were cellobiose (CEL), MAN and raffinose (RAF). All the prebiotics except GOS (Biosynth Carbosynth, UK) were purchased from Sigma-Aldrich (USA). Their stock solutions (10% w/v) were prepared with deionized sterile water and filtered using a syringe microfilter (0.2 μm, NY, USA), and stored at 4°C.

Cell cultures

Three human epithelial cell lines were used: HT-29, phenotypically nonmucus-secreting cells (from the cell-line collection at the AL), were passaged 15–25 times, mucus-secreting cells obtained from HT-29 cultures treated with methotrexate (MTX) – HT-29 MTX, passaged 5–15 times (European Collection of Authenticated Cell Cultures [ECACC], UK) and Homo sapiens normal colon CCD 841 CoN cells, passaged 5–15-times (American Type Culture Collection [ATCC], USA). These cell lines were chosen because of their routine use in a previous study [27]. The cells were cultured in Dulbecco's modified Eagle medium (DMEM) with 4.5 g/l glucose, L-glutamine, and supplemented with 10% fetal bovine serum (FBS) and antibacterial-antifungal solution. The cells were cultured in sterile flasks and incubated at 37°C, 5% CO₂ and 95% relative humidity. The cells were harvested using trypsin-EDTA solution and seeded onto 24-well plates at a concentration of 10⁴ cells per well. The cells were observed daily and examined for growth and contamination using an inverted microscope. Medium without antibiotic/antimycotic solution was used for subculturing. The mature HT-29 and CCD 841 CoN cells obtained after 15 days of seeding and HT-29 MTX cells obtained after 21 days of seeding.

Adhesion of B. fragilis, B. thetaiotaomicron & C. difficile 630 strains to cell lines

The adhesion assay was performed according to a previous report [27]. C. difficile 630 was additionally tested with GOS and LAC because with CEL, FOS, INU, MAN and RAF were tested earlier [27]. After reaching sub-confluence (70%–80%) stage, cells of all the cell lines were washed twice with PBS and incubated for 4 h with DMEM without antibiotic/antimycotic solution with 1% of final prebiotics of desired concentration. Medium without prebiotics was employed as a negative control. Bacterial inoculum was then prepared by dissolving 24 h culture of B. fragilis or B. thetaiotaomicron (grown on Columbia agar with 5% sheep blood and adjusted to McFarland standard of 3.0) in suspension medium (bioMérieux, France). Then 100 μl inoculum was then added to each well of 24-well plate and incubated for 1 h. The medium was aspirated, and the wells were washed twice with PBS. The cells were trypsinized, afterwards contents of each well were diluted 100 times, vortexed and 20 μl was inoculated in Columbia agar with 5% sheep blood. After incubation under anaerobic conditions for 48 h at 37°C colonies were counted, and the adhesion coefficient was calculated using formula below. Every dilution was seeded in duplicate and assay was performed in triplicate.

Effect of prebiotics on biofilm formation by B. fragilis & B. thetaiotaomicron & C. difficile 630

For the biofilm formation assay, we have used a previously described method [27]. Brain heart infusion (BHI) media with 1%, 2%, 4% and 8% concentrations of prebiotics were pipetted into well of 96-well flat-bottom microplates (Nunc, Denmark) in triplicate. Afterward, the wells were inoculated with 20 μl of overnight bacterial culture diluted 1:100 in BHI. The wells with BHI broth without inoculum were considered negative controls, whereas the wells without prebiotics were considered positive controls. The plates were incubated under anaerobic conditions at 37°C for 48 h. After incubation, the liquid was draw out, walls were washed with PBS, and air-dried at 37°C for 15 min. Each well was stained with crystal violet (CV; Analab, Poland) for 10 min. CV was removed, and the wells were washed eight times with PBS. Then CV was extracted using ethanol, and the absorbance was measured at 620 nm (A₆₂₀) using a Bio-Rad 550 microplate reader (Bio-Rad, USA). All the strains were tested three times.

C. difficile 630 was additionally tested with INU, RAF, CEL, GOS and LAC because with FOS and MAN was tested earlier [27]. We extended the C. difficile study with new prebiotics to select the one with the best performance for further adhesion and biofilm studies in co-culture with Bacteroides sp.

Adhesion competition assay

The method previously described by Deng et al. was used with some modifications [31]. After reaching 70–80% confluence, the cells in 24-well plates were washed twice with PBS, and 400 μl of fresh pre-warmed (37°C) DMEM without antibiotic/antimycotic solution was added to them. We tested only two Bacteroides sp. strains, reference NTBF IPL E323 and reference B. thetaiotaomicron ATCC 29741 and C. difficile 630. The overnight cultures of bacteria grown on Columbia agar with 5% sheep blood was adjusted, C. difficile 630 to 5 × 10⁹ CFU/ml, Bacteroides strains to 5 × 10⁷ CFU/ml. Bacterial inoculums (40 μl) were added to relevant wells in the following groups: A, control (single strain); B, competition group; (C. difficile co-incubated with NTBF or B thetaiotaomicron); C, competition group with prebiotic (C. difficile co-incubated with NTBF or B. thetaiotaomicron in combination with 1% FOS for both co-cultures). FOS was chosen for this assay because of its adhesion-inhibiting properties toward C. difficile. The plates were incubated for 2 h at 37°C. After incubation, wells were washed twice with PBS and trypsinized for 10 min at 37°C. Then 500 μl of fresh media with 10% FBS was added for deactivation of trypsin. The contents of each well were diluted 100 times using PBS, vortexed, and 20 μl was inoculated into Columbia agar with 5% sheep blood. After incubation in anaerobic conditions for 48 h at 37°C dilutions were seeded in duplicate, and each assay was performed in triplicate. Colonies of C. difficile and Bacteroides sp. were visually distinguished, using UV light, and counted. The assay was performed in triplicate.

Co-culture biofilm assay

The method described by Slater et al. was used with some modifications [32]. In this experiment, only references strains were used: NTBF IPL E323, B. thetaiotaomicron ATCC 29741 and C. difficile 630. For the generation of co-culture biofilms, overnight bacterial inoculum was diluted 1:100 in BHI containing 0.1 M glucose and incubated for 48 h at 37°C under anaerobic conditions in 96-well flat-bottom microplates. The following groups were considered: A, control (single strain); B, competition group (C. difficile co-incubated with NTBF or B. thetaiotaomicron); C, competition group with prebiotic (C. difficile co-incubated with NTBF or B. thetaiotaomicron in combination with 1% FOS). After incubation, biofilm biomass was measured according to biofilm assay.

To determine the cell count of C. difficile in monoculture, co-cultured biofilms were grown in 24-well flat-bottom microplates [16]. The overnight grown bacterial cultures (1 ml) were diluted with 1:100 fresh BHI containing 0.1 M glucose in wells, and incubated for 48 h at 37°C under anaerobic conditions. After incubation, the planktonic phase was removed, and the wells were washed twice with PBS. The adherent biofilms were detached by scraping using a sterile filter tip, transferred to sterile Eppendorf tubes and vortexed to prevent cell–cell clumping. The biofilm was re-suspended in 1 ml of PBS and diluted 100-times. Dilutions were plated on Columbia agar containing 5% sheep blood, and plates were incubated anaerobically for 48 h. Colonies were differentiated based on size, colony morphology and fluorescence of C. difficile colonies under UV light at 365 nm wavelength. The assay was performed in triplicate.

Statistical analysis

We have checked normality of values distribution using the Shapiro–Wilk test. One-way analysis of variance followed by Tukey's post hoc test or Kruskal–Wallis one-way analysis of variance followed by Dunn's post hoc test was performed to determine the statistical significance. In case of two categorical variables two-way ANOVA was used. All calculations were performed using the Statistica software (version 13, StatSoft, Poland).

Results

Adhesion of tested bacteria to human epithelial cell lines without prebiotics

We compared adhesive properties of C. difficile 630, NTBF IPL E323, clinical NTBF, clinical ETBF, clinical and reference B. thetaiotaomicron ATCC 29741 to three human epithelial cell lines, without prebiotics. We observed significantly lower adhesion of C. difficile 630 and clinical NTBF compared to the other tested bacteria (p < 0.05). We found that type of cell line did not affect significantly bacterial adhesion except B. thetaiotaomicron ATCC 29741, which adhere significantly more likely to HT-29 than HT-29 MTX (p = 0.02) (Figure 1).

Effect of standard & candidate prebiotics on B. fragilis, B. thetaiotaomicron & C. difficile adhesion

The adhesion of NTBF IPL E323 strain to cell lines was significantly (p < 0.01) inhibited by standard and candidate prebiotics, except for FOS (p > 0.05) in all the cell lines and for CEL (p > 0.05) in the CCD 841 CoN cell line. The adhesion of clinical NTBF strain to HT-29 and HT-29 MTX cell lines was successfully inhibited by INU and RAF (p < 0.05). The tested prebiotics did not show significant effect on the adhesion of clinical NTBF to the CCD 841 CoN cell line. The adhesion of clinical ETBF strain to all tested cell lines was effectively inhibited by FOS and MAN (p < 0.05). INU blocked the adhesion of clinical ETBF strain to HT-29 and CCD 841 CoN cell lines (p < 0.05) (Table 1 & Figure 2).

The adhesion of B. thetaiotaomicron reference strain to the HT-29 cell line was effectively inhibited by all prebiotics (p < 0.01). GOS, CEL and MAN exhibited anti-adhesion properties when tested on the HT-29 MTX cell line (p < 0.01). INU, RAF, GOS and LAC (p < 0.001) inhibited the adhesion of the B. thetaiotaomicron reference strain to the CCD 841 CoN cell line.

FOS, INU, CEL, MAN and RAF inhibited the adhesion of clinical B. thetaiotaomicron to HT-29 and CCD 841 CoN cell lines (p < 0.001). The effect of prebiotics on the adhesion of clinical B. thetaiotaomicron to the mucus-secreting cell line was insignificant (Table 1 & Figure 3).

Interestingly, GOS did not block the adhesion of all clinical Bacteroides sp. strains (NTBF, ETBF and B. thetaiotaomicron) (p > 0.05) to any of the tested cell lines.

In this study, we tested additionally the effect of GOS and LAC on the adhesion of C. difficile 630. The effect of CEL, FOS, INU, MAN and RAF has been reported in a previous study [23]. The effect of GOS and LAC on the adhesion of C. difficile 630 to all tested cell lines was statistically insignificant (p > 0.05).

Effect of standard & candidate prebiotics on biofilm formation

Biofilm formation by NTBF IPL E323 was inhibited by 8% INU (p = 0.01), FOS (p = 0.01) and GOS (p = 0.002). We did not observe an increase in biofilm production in the presence of any tested prebiotic. Significantly high biofilm production by clinical NTBF (no. 157/03) was observed after incubation with 1%, 2% and 4% GOS (p < 0.001). None of the tested prebiotics could decrease biofilm production by this strain. We observed inhibition (p < 0.01) of biofilm formation by clinical ETBF (no. 093) in the presence of INU and FOS at the concentrations of 4% and 8%, and increase (p < 0.001) in biofilm production was observed with GOS and LAC at the concentrations of 1%, 2% and 4%. Both B. thetaiotaomicron strains produced significantly lower (p < 0.01) amounts of biofilm when incubated with FOS and INU at concentrations of 2%, 4% and 8%, and with MAN, CEL and GOS at the concentration of 8%. Biofilm formation by B. thetaiotaomicron ATCC 29741 was also inhibited by 4% and 8% RAF (p < 0.01).

In this study, we tested the effect of INU, RAF, CEL, GOS and LAC on the biofilm formation of C. difficile 630. The effect of FOS and MAN has been reported in a previous study [23]. In this experiment we observed statistically significant inhibition of biofilm formation by C. difficile 630 only by 1% and 2% CEL (p = 0.01) (data on the effect of tested prebiotics on biofilm formation are in repository 10.6084/m9.figshare.14380727).

NTBF & B. thetaiotaomicron inhibit the adherence of C. difficile to human colon epithelial cells in vitro

In this study, we tested whether two Bacteroides sp. strains, control NTBF IPL E323 and reference B. thetaiotaomicron ATCC 29741, can inhibit the adhesion of C. difficile 630 to the cell lines. In addition, we attempted to determine how FOS affects this co-incubation. Co-incubation of C. difficile with NTBF or B. thetaiotaomicron strain in combination with FOS decreased the adhesion of C. difficile to HT-29 cells from 32.67 CFU/0.02 × 10^-2 ml to 14.33 CFU/0.02 × 10^-2 ml (p = 0.0004) and 19.67 CFU/0.02 × 10^-2 ml (p = 0.0002), respectively. Similar results were observed for CCD 841 CoN and HT-29 MTX cell lines. An additional 1% FOS increased the anti-adhesion potential of Bacteroides sp. strains. We observed a decrease in adhesion of C. difficile in the presence of NTBF to HT-29 cell line, and was from 32.67 CFU/0.02 × 10^-2 ml to 5.17 CFU/0.02 × 10^-2 ml (p = 0.0002) and from 32.67 CFU/0.02 × 10^-2 ml to 13.0 CFU/0.02 × 10^-2 ml (p = 0.0004) in the presence of B. thetaiotaomicron (Figure 4). Similar results were observed in the HT-29 MTX cell line. Therefore, no statistically significant (p > 0.05) difference was observed between co-incubation of C. difficile 630 with NTBF or B. thetaiotaomicron and FOS for the CCD 841 CoN cell line (Figure 4). We observed that co-incubation of C. difficile and Bacteroides sp. not affected statistically significantly Bacteroides sp. count (p > 0.05).

NTBF & B. thetaiotaomicron affect biofilm formation by C. difficilein vitro

In our study, we tested the effect of NTBF IPL E323 or B. thetaiotaomicron ATCC 29741 on the biofilm formation by C. difficile 630. The most effective prebiotic FOS was used in biofilm formation assays in co-culture NTBF IPL E 323 with C. difficile 630 or B. thetaiotaomicton ATCC 29741 with C. difficile 630.

As mentioned previously, NTBF produced significantly more biofilms in vitro than C. difficile (p = 0.03) in monocultures. Differences between the biofilm amounts of B. thetaiotaomicron and C. difficile were not statistically significant (p = 0.06). Co-incubation of NTBF and C. difficile increased the biofilm production compared with biofilm production by C. difficile (p = 0.0002) and NTBF (p = 0.001) in monocultures. B. thetaiotaomicron co-incubated with C. difficile showed higher amount of biofilm (p = 0.0002) than that of B. thetaiotaomicron monoculture, but not C. difficile (p = 0.07). Co-incubation of Bacteroides sp. and C. difficile strains with FOS showed no statistically significant difference (p > 0.05) in biofilm formation compared with that without FOS (Figure 5A). Co-incubation of C. difficile with both Bacteroides sp. strains significantly reduced the number of C. difficile within the biofilm. The monoculture mean CFU of C. difficile was 14.0 CFU/0.02 × 10^-2 ml, and co-incubation with NTBF and B. thetaiotaomicron reduced the mean CFU of C. difficile within mixed biofilm to 1.33 CFU/0.02 × 10^-2 ml (p = 0.0001) and 3.33, respectively, CFU/0.02 × 10^-2 ml (p = 0.001). Co-incubation of Bacteroides sp. and C. difficile strains with FOS showed no significant difference (p > 0.05) CFU compared with incubation without FOS (Figure 5B). We observed that co-incubation of C. difficile and Bacteroides sp. did not statistically and significantly affect Bacteroides sp. count (p > 0.05).

Discussion

Bacterial adhesion is the initial step of infection and biofilm formation by bacteria. It has been suggested that pathogenesis and persistence of C. difficile in the gut are related to biofilm formation [15]. The Bacteroides fragilis group strains colonize the intestinal human tract as commensal bacteria. Bacteroides fragilis strains isolated from colorecal cancer (CRC) tissue showed higher biofilm-forming ability compared to isolates of normal tissue [13]. Prebiotics interact with enteropathogens which may results in the inhibition of adhesion and biofilm formation [27,33]. In our previous study [27], we examined the effects of standard prebiotics such as INU, FOS and CEL, and prebiotic candidates such as MAN and RAF on the adhesion of C. difficile to human epithelial cells in vitro. In the present study, we additionally checked the effect of two new prebiotics, GOS and LAC, and investigated their effect on the adhesion of C. difficile 630 reference strain. Because of less studies performed on the effect of prebiotics on the adhesion of commensal bacteria, we used the strains belonging to the Bacteroides genus, including reference and clinical strains with different toxigenicity profiles. In a previous study, we reported the inhibition capacity of FOS and MAN when tested on clinical C. difficile RT027 and reference, epidemic strain 630 (RT012) strains [27]. In the present study, INU was effective in blocking the adhesion of Bacteroides strains to HT-29 and CCD 841CoN (except clinical NTBF) cell lines. MAN did not inhibit the adhesion of Bacteroides to HT-29 and CCD 841 cell lines (except for B. thetaiotaomicron). FOS blocked the adhesion of Bacteroides strains to the HT-29 cell line (except NTBFs). These results indicate the high inhibition capacity of INU, MAN and FOS toward the adhesion of Bacteroides species in non-mucus-secreting cells. Other studies have reported a high ability of FOS and MAN to block bacterial adhesion. Hartman et al. reported that mannose and mannans decrease the adhesion of enteropathogenic E. coli (EPEC) [34]. Furthermore, Shoaf et al. reported 40% inhibition of adherence of enteropathogenic E. coli to Caco-2 cells in the presence of FOS [35].

Biofilms are the main forms of bacterial growth in the large intestine [36]. Carbohydrates may affect the formation of biofilm by anaerobic Bacteroides spp. and C. difficile. INU possesses the ability to decrease biofilm formation produced by all tested Bacteroides strains, except for clinical NTBF. We observed decrease in biofilm formation after incubation of this strain with 2%, 4% and 8% of INU; however, the differences were statistically insignificant. Biofilm formation by C. difficile decreased by 1% and 2% CEL. We observed a tendency to induce biofilm formation by using the low concentration (1%–4%) of GOS and LAC. Biofilm formation by clinical NTBF was reduced by GOS, whereas that of clinical ETBF was reduced by GOS and LAC. In a previous study, 8% FOS and MAN significantly decreased the biofilm formation by C. difficile 630 [27].

C. difficile is currently the leading cause of nosocomial infections. Vancomycin and fidaxomicin are the first-line therapies for CDI. However, the high recurrence rate of 20–25% was observed after treatment with these agents [37]. We found that NTBF and B. thetaiotaomicron decreased the adhesion of C. difficile to all tested cell lines in vitro. Deng et al. reported the inhibition of C. difficile adhesion to HT-29 cells and colon cell apoptosis using the B. fragilis strain. In the same study, B. fragilis exhibited protective effects in mice with CDI. All mice prophylactically treated with B. fragilis displayed lower morbidity and mortality than the untreated group [31]. We tested only NTBF strain because ETBF, which produces B. fragilis enterotoxin, has been found to be associated with the occurrence of diarrhoea, including antibiotic-associated diarrhoea in humans [7]. Results show that B. fragilis can act as a protective agent against CDI; in our study, B. thetaiotaomicron exhibited the ability to inhibit C. difficile adhesion in vitro.

Bacteroides strains may also affect biofilm formation by C. difficile in vitro. In our study, co-incubation of NTBF or B. thetaiotaomicron with C. difficile resulted in the higher amount of biofilm formation than that of strains in monoculture, and a significant reduction of C. difficile cells was observed within these biofilms. Slater et al. reported that B. fragilis produces significantly more biofilm when co-cultured with C. difficile than in monoculture with a simultaneous reduction in C. difficile cells. Slater suggested that gene expression and changes in metabolic pathways are involved in mediating B. fragilis-induced inhibition of C. difficile in co-culture. In addition, C. difficile LuxS/AI-2 may be pivotal in biofilm formation by this pathogen by mediating prophage induction and eDNA accumulation. In co-cultured communities, C. difficile AI-2 likely signals B. fragilis to induce an altered metabolic response, enabling it to outgrow C. difficile [32].

A combination of probiotics and prebiotics is known as synbiotics [38]. In the present study, we investigated the effect of NTBF or B. thetaiotaomicron with 1% FOS on adhesion by C. difficile. NTBF or B. thetaiotaomicron were able to inhibit adhesion more effectively in the presence of 1% FOS than the isolates alone. However, we did not observe the same effect in the biofilm assay. To our knowledge, this study is the first to determine the effect of Bacteroides strains in combination with FOS on the adhesion and biofilm formation by C. difficile.

The study limitation is the incubation of cell lines with bacteria in aerobic condition. Bacteroides sp. and C. difficile can survive during 6 h in aerobic conditions [39,40], but oxygen stress may slightly affect metabolic changes.

Conclusion

The human gut has abundant bacterial species that compete or coexist with each other. The composition of gut microbiota can be affected by both probiotics and prebiotics. INU, FOS and MAN may affect the adhesion of commensal Bacteroides species, which is not particularly desirable. NTBF and B. thetaiotaomicron decreased in vitro adhesion and biofilm formation by C. difficile in the presence of 1% FOS. A combination of Bacteroides sp. and FOS can enhance the anti-adhesion potential against C. difficile and might be considered for the prophylaxis of CDI, for example, in addition to antibiotic therapy. However, further studies are warranted for determining the effects in vivo.

Future perspective

Recently, there has been a growing interest in microbiome-modulating interventions such as the use of prebiotics and probiotics. Technological advance may help in the exploration of new candidate prebiotics and probiotics and investigate their impact on human health. This study contributes to existing research as the combination of Bacteroides sp., and FOS and could be considered for the prophylaxis of C. difficile infections, the prevalent cause of nosocomial infections currently, indicating the relevance to today's scenario. Prebiotics and probiotics are environmentally friendly and show no toxicity. Additionally, their effectiveness in the eradication of bacterial biofilm is confirmed. This study has strong implications in future as it showed the potential role of prebiotics and symbiotics in preventing infections, which is different from the common use of prebiotics in diet and supplements. It is especially important in times of increasing antibiotic resistance. Further research is needed on this interesting topic, especially on anaerobic bacteria.