In Vitro Antimicrobial Activity and Probiotic Potential of Bifidobacterium and Lactobacillus against Species of Clostridium

Many Clostridium species are found as commensal members of the intestinal microbiota. However, imbalances of the microbiota may lead to certain infections caused by these microorganisms, mainly Clostridium butyricum, Clostridium difficile, and Clostridium perfringens. In many cases, infection recurrence can occur after antibiotics, indicating the need for novel therapeutic options that act on the pathogens and also restore the microbiota. Herein, the in vitro antimicrobial activity and probiotic potential of clinical and reference strains of Bifidobacterium and Lactobacillus were investigated against Clostridium species. Antimicrobial activity was evaluated by the agar spot test and inhibition of gas production. Then, the probiotic potential of selected strains was assessed by analyzing their coaggregation ability, adhesive properties to host cells and mucin, tolerance to acidic pH and bile salts, and antimicrobial susceptibility profiles. Lactobacillus plantarum ATCC 8014 was the most promising strain based on its inhibitory activity against Clostridium spp. Also, this strain met criteria to be considered a probiotic based on its coaggregation ability, adhesive properties, and tolerance to harsh pH and bile acid salt conditions. The results indicate that among the studied strains, L. plantarum ATCC 8014 presents probiotic potential for controlling infections induced by the studied Clostridium species and should be further evaluated in in vivo animal models.


Introduction
In humans, the normal intestinal microbiota consists of a large number and high diversity of commensal microorganisms (about 10 13 to 10 14 ), mainly in the large intestine, where they establish a symbiotic relationship that influences the entire host organism [1][2][3]. Intestinal homeostasis is maintained through complex interactions between the host's immune system and the microbiota [4,5]. However, this mutualistic relationship can be disrupted by a variety of factors, such as changes in diet, use of antibiotics, and immunomodulatory drugs, among others, leading to changes in both bacterial function and diversity [6,7].
The agar spot test procedure was performed as described previously [41], with modifications. Briefly, in a Petri dish containing 10 mL of MRS agar, 5 µL of each probiotic culture was spotted onto one quadrant of the culture medium, followed by incubation at 37 • C for 24 h under anaerobic conditions. After incubation, 10 mL of thioglycolate agar was overlaid onto the MRS agar containing the growth of Bifidobacterium and Lactobacillus strains. After solidification of the culture medium at room temperature (25-28 • C), Clostridium spp. suspensions (McFarland standard No. 0.5, 1.5 × 10 8 colony-forming units per milliliter (CFU/mL)) were spread with the aid of a swab. The plates were incubated at 37 • C for 24 h under anaerobic conditions. The formation of a clear halo around growth of the probiotics was indicative of antimicrobial activity. The diameter of the growth inhibition halo was measured and expressed in millimeters.

Inhibition of Gas Production
The ability of probiotics to inhibit the growth of Clostridium strains was also evaluated by assessing the inhibition of gas production due to the fermentative action of the pathogens, as described previously [42], with some modifications. Briefly, the assay was performed by inoculating 1 µL (~10 7 UFC) of Clostridium culture into the upper third of the RCM agar layer (supplemented with 1.5 g/100 mL bacteriological agar), composed of 3 mL per tube. Subsequently, 3 mL of MRS containing 0.7 g% bacteriological agar was melted, cooled to 50 • C, and inoculated with 30 µL (~10 8 UFC) of each probiotic culture. The contents were homogenized by vortexing and immediately poured over the RCM agar layer in tubes inoculated with the Clostridium strains. RCM agar with Clostridium and MRS agar without inoculated probiotics were used as negative controls. The tubes were incubated under anaerobic conditions at 37 • C for 24 h. The assays were performed in triplicate, with and without buffering of the MRS medium with K 2 HPO 4 and KH 2 PO 4 (100 mM). A positive assay for antimicrobial Nutrients 2019, 11, 448 4 of 15 activity was characterized by the absence of gas production, that is, in the absence of bubbles in the culture media, or medium breakage.

Antimicrobial Activity after pH Adjustment
The strain with the highest antimicrobial activity was selected to verify if growth inhibition of Clostridium spp. was due to the acidic pH. Culture supernatants were tested after pH adjustment essentially as described by Gaspar et al. [43]. After cultivation of the microorganism in MRS broth (Difco-BD), 10 mL of the bacterial culture was heated at 70 • C for 30 min to inhibit protease activity, cooled at room temperature, and centrifuged (5,000 × g for 15 min at 4 • C). Hydrogen peroxide was eliminated by the addition of 5 mg/mL catalase from bovine liver (Sigma-Aldrich, St Louis, MO, USA) followed by filtration through a 0.2 µm pore-size cellulose acetate (Whatman ® , Clifton, NJ, USA). Antimicrobial activity of the cell-free culture supernatants (CFCN) was evaluated with and without pH adjustment to 6.5 with 10 M NaOH solution. Antimicrobial activity was evaluated by the antagonist well-diffusion method. Briefly, 20 mL of MRS soft agar (0.7% agar) was inoculated with 200 µL of each Clostridium strains. Wells with 4 mm diameter were punched in agar plates and filled with 100 µL of CFCN. Phosphate-buffered saline (PBS, pH 7.4) was used as a negative control. Plates were incubated under anaerobic conditions at 37 • C for 24 • C in an upright position. The inhibition zone diameters were measured and expressed in millimeters.

Coaggregation Test
After cultivation of the probiotic strains and Clostridium species, aliquots of 1 mL of each culture were washed twice with phosphate-buffered saline (PBS, pH 7.2), centrifuged at 5000× g for 15 min, and resuspended in PBS. The optical density of each suspension was adjusted (OD 620nm = 0.1), and 500-µL aliquots of the probiotic suspensions were mixed with 500 µL of each pathogen suspension in 24-well plates (Nunc, Roskilde, Denmark) and incubated at 37 • C for 4 h under constant stirring (100 rpm) on an orbital shaker. Plates were observed for macroscopically visible clumps and under inverted microscopy [44]. Glass slides were also prepared with 5 µL of each suspension and evaluated under the microscope for visualization of bacterial coaggregates after Gram staining. L. fermentum ATCC 23271 was used as a positive control in this assay, since high coaggregation scores were previously demonstrated [45]. Control assays were performed with individual bacterial samples to assess their ability to autoaggregate.

Mucin Binding Assay
The ability of selected probiotic strains to bind to mucin was evaluated essentially as described by Tallon et al. [46]. A volume of 100 µL of a 10 mg/mL mucin solution in PBS (pH 7.2) was added to the wells of polystyrene microtiter plates (Nunc) and incubated overnight at 4 • C. The wells were washed twice with 200 µL PBS and saturated with a 2% (w/v) bovine serum albumin (BSA) solution (Sigma-Aldrich, St. Louis, MO, USA) for 4 h at 4 • C. Finally, the wells were washed twice with 200 µL PBS. At least four replicates were used to estimate the adhesion of a given strain. Probiotic cultures in MRS broth were washed three times in PBS, and the final suspension was standardized by spectrophotometry (OD 600nm = 0.1). Aliquots of 100 µL of the bacterial suspension were added to each well, and the microplates were incubated at 37 • C for 1 h. After this, the wells were washed 12 times with 1 mL PBS to remove non-adherent bacteria. The wells were treated with 200 µL of 0.5% Triton X-100 (Sigma-Aldrich), and the plates were then incubated for 2 h at room temperature under orbital shaking to release the adhered bacteria. Then, the wells were scraped with a sterile tip and the number of bacteria with binding ability to mucin was estimated by serial decimal dilutions in PBS and plating on MRS agar, followed by incubation at 37 • C for 24 h under anaerobic conditions. L. fermentum ATCC 23271 was used as a positive control [45]. Mucin-containing wells without bacteria were used as negative controls.

Adhesion to Eukaryotic Cells
Adhesion to HeLa was evaluated according to the method of Carmo et al. [45]. A 300-µL aliquot of each potential probiotic cultured in MRS broth was washed three times with PBS (pH 7.4, Sigma-Aldrich) and the bacterial pellet was resuspended in 300 µL Dulbecco's modified Eagle's medium (DMEM, Sigma-Aldrich). Monolayers of HeLa cells grown in 24-well microplates (Nunc) containing DMEM supplemented with 10% fetal bovine serum (Gibco, Gaithersburg, MD, USA), with or without glass coverslips, were inoculated with 50 µL (~2.3 × 10 7 CFU) of bacterial suspension and incubated at 37 • C under 5% CO 2 for 3 h. Then, each well was washed three times with PBS to remove non-adherent bacteria. For quantification of the adherent bacteria, the HeLa cell monolayers in the wells without coverslips were treated with 1 mL of 0.1% Triton X-100 (Sigma-Aldrich) for 5 min and scraped with the aid of a tip. Thereafter, serial decimal dilutions were spread on MRS agar plates and incubated at 37 • C for 24 h. The total number of bacteria adhered to the cells was expressed as CFU per milliliter. Wells with HeLa cells incubated in the absence of bacteria were used as negative controls. Visualization of bacterial adherence to eukaryotic cells was performed after fixation with methanol (Amresco, Gymea, Australia) and staining with May-Grunwald and Giemsa (Amresco). Gram staining was also used to better visualize gram-positive Lactobacillus adhered to the cells. The coverslips were then mounted on glass slides and visualized by light microscopy under a 100× oil immersion objective.

Tolerance to Acidic pH and Bile Salts
Tolerance of selected bacteria to acidic pH (2 and 4) and bile salts (0.5% and 1%, Oxgall, Sigma-Aldrich) was evaluated as previously described, with minor modifications [47]. Briefly, 900 µL MRS, adjusted to pH 2 or 4, or non-adjusted (control), or supplemented with 0%, 0.5%, or 1.0% (w/v) Oxgall (Sigma-Aldrich), was inoculated with 100 µL of a 24-h culture, which had been previously washed three times with PBS and resuspended in the same volume of MRS broth. After incubation at 37 • C for 3 h under anaerobic conditions, the percentage of viable bacteria relative to that in the control was determined by plate counting on MRS agar.

Statistical Analyses
Statistical analyses were performed using NCSS 11 Statistical Software (2016; NCSS, Kaysville, UT, USA). Adherence to eukaryotic cells and to mucin was expressed as log CFU/mL (±SD). Tolerance to acidic pH and to bile salts was compared to growth in standard MRS medium. The Shapiro-Wilk test was carried out and confirmed that all variables were normally distributed. Thus, all comparisons were carried out by the Student t-test. Statistical significance was established at p < 0.05. All assays were performed in three independent experiments conducted on three different days.

Selection of Strains with Antimicrobial Activity against Clostridium spp.
The antimicrobial effects of potential probiotics, as determined by the agar spot test, are shown in Table 1. C. butyricum was inhibited by 11 strains (91.7%), whereas C. difficile and C. perfringens were inhibited by 9 (75%) strains each. The diameter of the inhibition zone varied among the clinical and reference probiotic strains. Of the 12 species tested, 8 (66.7%) exhibited antimicrobial activity against all three Clostridium species. The largest inhibition zones were produced by L. plantarum ATCC 8014, mainly against C. butyricum (17 mm). Only B. animalis 56.1 showed no inhibitory activity against any Clostridium species in this assay. To assess whether or not the inhibitory activity on Clostridium spp. was due to acid produced by the probiotics, a preliminary attempt was made by supplementing MRS with phosphate buffer to evaluate the inhibition of gas production by C. butyricum ATCC 860 in culture medium. This test was also carried out in MRS medium without phosphate buffer. C. butyricum ATCC 860 was selected because this strain usually produces a large amount of gas resulting from its fermentative activity in thioglycolate medium or RCM. In the presence of buffer, the strains L. brevis ATCC 367, L. delbrueckii ATCC 9649, L. paracasei ATCC 335, L. plantarum ATCC 8014, L. rhamnosus ATCC 9595, B. animalis 56.1, and B. longum ATCC 15707 inhibited gas production by Clostridium spp., whereas the other strains did not ( Figure 1). Among the strains that inhibited gas production in the absence of the phosphate buffer, only B. bifidum 14.2 gave a negative result when the assay was carried out in the of presence phosphate buffer in the MRS medium (Table 2).
L. plantarum ATCC 8014, the strain with the highest antimicrobial activity in the agar spot test for the three Clostridium strains, was selected for investigation of the antimicrobial activity in cell-free culture supernatants (CFCN) with and without pH adjustment to 6.5 with NaOH 10N, after growth in MRS broth. CFCN at pH 4.3, the final pH after cultivation of L. plantarum ATCC 8014, showed growth inhibition zones ranging from 14.2 (±0.8) mm to 16.5 (±0.8) mm, whereas CFCN at pH 6.5 presented inhibition zones from 13.2 mm (±0.4) mm to 14.7 (±0.5) mm. However, the inhibition zones of the CFSN at pH 4.3 in comparison to CFSN at pH 6.5 only presented statistical significance against C. butyricum (p < 0.05, Table 3). thioglycolate medium or RCM. In the presence of buffer, the strains L. brevis ATCC 367, L. delbrueckii ATCC 9649, L. paracasei ATCC 335, L. plantarum ATCC 8014, L. rhamnosus ATCC 9595, B. animalis 56.1, and B. longum ATCC 15707 inhibited gas production by Clostridium spp., whereas the other strains did not (Figure 1). Among the strains that inhibited gas production in the absence of the phosphate buffer, only B. bifidum 14.2 gave a negative result when the assay was carried out in the of presence phosphate buffer in the MRS medium (Table 2).

L. plantarum ATCC 8014 Presents Probiotic Potential
L. plantarum ATCC 8014 was selected for further analysis and characterization as a potential probiotic. In terms of its coaggregation capacity, we observed that L. plantarum ATCC 8014 interacted clearly with the pathogens, forming bacterial aggregates with the three species of Clostridium ( Figure 2).   Investigation of the adhesion properties of L. plantarum ATCC 8014 revealed its ability to interact with eukaryotic cells, as observed in the assay with HeLa cells (Figure 3). In fact, adherence values for L. plantarum ATCC 8014 were higher than those observed for L. fermentum ATCC 23271 (Table 4, p = 0.0037). In contrast, L. fermentum ATCC 23271 exhibited a greater ability to bind mucin than L. plantarum ATCC 8014 (Table 4, p < 0.0001).    L. plantarum ATCC 8014 exhibited growth at pH 2 and 4 after 180 min of exposure, with growth values of 70.3% and 97.8%, respectively, relative to that observed in standard MRS medium (Table 4). Assays performed at bile salt concentrations of 0.5% and 1.0% resulted in growth values of 110.8% and 90.1%, respectively, relative to that observed in standard medium. No differences were observed relative to the growth of L. fermentum strain ATCC 23271 (Table 5). The antibiotic susceptibility profile of L. plantarum ATCC 8014 was evaluated using the overlay diffusion method. L. plantarum ATCC 8014 showed resistance to ciprofloxacin and vancomycin and sensitivity to all other antibiotics tested (Table 6). Table 6. Antimicrobial susceptibility of L. plantarum ATCC 8014 by the agar overlay diffusion method.

Antibiotics
Inhibition

Discussion
Some probiotic strains have been successfully used in clinical studies for the treatment or prevention of AAD [34,35] and NEC [36,37]. Their mechanisms of action still remain unclear, perhaps because it seems that this is not important since some probiotics interventions have been effective in certain patients with some reliability [40,50]. On the other hand, controversial findings regarding AAD and NEC interventions have been reported in the literature, as a positive effect does not always occur when used in cases of dysbiosis [51]. A better understanding of the relevant "central probiotic properties" that contribute to their inhibitory effects on major etiological agents of these clinical syndromes could therefore aid in the design of a more rational strategy for selecting and producing more effective probiotics [31].
In this study, we showed that different clinical and reference strains of Bifidobacterium and Lactobacillus present different levels of antimicrobial efficacy against C. butyricum, C. difficile, and C. perfringens. Evaluation of the antimicrobial activity of these bacteria (12 potential probiotic bacteria) indicated that L. plantarum strain ATCC 8014 exhibited the greatest capacity to inhibit the growth of the three reference strains of Clostridium, based on the activity detected using screening methods. In addition to L. plantarum ATCC 8014, other species showed inhibitory activity against one or all Clostridium strains. However, their antimicrobial activities were evidenced by lower zones of inhibition or were variable in comparison to the inhibition test of gas production. This variability in performance has previously been reported and suggests that more than one method should be used to assess the antimicrobial activity of probiotics, given that the conditions of each methodology may interfere with the results [52].
Various species of Lactobacillus are able to produce compounds with antimicrobial activities, including organic (acetic and lactic) acids, low-molecular-weight compounds, antifungal peptides, and antibacterial peptides (bacteriocins) [53,54]. It appears, however, that the inhibition of Clostridium growth exhibited by the probiotic strains was not the result of the overproduction of acids, since the addition of buffer to the MRS media did not affect the inhibitory activity of the majority of these species. Furthermore, some species of Clostridium exhibit intense fermentative activity, resulting in the production of large quantities of organic acids, including acetic, lactic, formic, butyric, and propionic acid, among other substances; thus, they would likely already be habituated to these and would survive acidic pH conditions [54,55]. In addition, as Lactobacillus and Bifidobacterium can produce a large amount of organic acids [55] and the concentration of the phosphate buffer used in the assay might not be sufficient to neutralize them, we tested the culture supernatants of L. plantarum ATCC 8014 at pH 4.3 and 6.5, but the inhibitory effect continued to be evidenced even after pH adjustment, although there was a higher activity against L. butyricum at pH 4.3.
This evidence allows us to suggest that other molecules produced by the probiotic strain are expected to be involved in their inhibitory actions on Clostridium strains. Indeed, a recent study demonstrated that Lactobacillus metabolites isolated from vaginal smears presented with in vitro bacteriostatic effects against C. perfringens [56]. These have been associated with the production of bacteriocins, which are produced by several species of probiotics and have bactericidal or bacteriostatic actions. Bacteriocins may increase the permeability of the inner membrane of bacteria, thus contributing to their rupture and interfering with bacterial cell wall synthesis, resulting in pore formation by binding to the peptidoglycan precursor lipid [57]. Lacticin 3147, for example, produced by Lactococcus lactis, forms selective pores in the cell walls of some pathogenic gram-positive bacteria, including C. difficile, resulting in its death [58].
The nature of the compound(s) produced by L. plantarum ATCC 8014 and their potential bactericidal or bacteriostatic actions are not yet known. However, even if such compounds are bacteriostatic or capable of inducing only sporulation, such actions could be relevant to controlling the clinical manifestation of infections caused by Clostridium, neutralizing the metabolic activity of the pathogen and, consequently, the production of toxins and other virulence factors involved. In addition, it is worth mentioning that Clostridium spp. are gram-positive bacteria and bacteriocins have a more targeted inhibitory action against this type of microorganism rather than against gram-negative bacteria [52,54].
Several criteria are used to define microorganisms as probiotics, including the ability to: (1) coaggregate with microbial pathogens; (2) adhere to eukaryotic cells and mucus; and (3) tolerate conditions of acidic pH and bile salts, among other properties [59,60]. L. plantarum ATCC 8014 fulfilled all these criteria, as it tolerated acidic pH and bile salts under the conditions tested, coaggregated with Clostridium spp., and exhibited adhesive properties suggesting its capacity for in vivo colonization. Although this bacterium demonstrated lower mucin binding than the L. fermentum strain ATCC 23271, the results obtained were of great relevance.
A worrying issue in the selection of probiotics to be used in foods or supplements is their potential to transmit genes involved in antibiotic resistance, especially if the microorganism in question carries plasmids [61,62]. Herein, L. plantarum ATCC 8014 showed resistance to ciprofloxacin and vancomycin. However, despite having plasmids [63], resistance to these antibiotics is considered intrinsic and non-transmissible in this species [64,65]. Of importance, vancomycin is among the antibiotic options to treat Clostridium infections [66,67]. Thus, we can assume that the development of a therapeutic strategy for intestinal disorders consisting of vancomycin administration in association with L. plantarum ATCC 8014 could be more effective than the single use of the antibiotic, if a synergistic effect is to be proven.
Currently, there is a large panel of probiotic strains in use. However, in most cases, probiotic action has been shown to be species-specific or even strain-specific [31,[68][69][70]. Thus, the possibility of a potential probiotic with proven antimicrobial action against multiple Clostridium species, which are commonly associated with pathologies resulting from an imbalance in the microbiota, represents a significant advance for the reduction of morbidity and mortality rates arising from these clinical syndromes. In addition to being a more rational therapeutic approach, use of a probiotic would not have a negative impact on the intestinal microbiota and would not exert pressure for the selection of resistant bacteria, as with conventional antibiotic therapy [71].
Although the results presented herein are promising in terms of novel therapeutic strategies to treating intestinal infections caused by Clostridium spp., especially as L. plantarum ATCC 8014 is readily available to the scientific community as a culture collection strain, certain limitations of the study should be taken into consideration. Firstly, although there was an apparent cell monolayer integrity under microscopic examination of HeLa cells, its viability following incubation with the tested probiotic strain was not evaluated; thus, it is uncertain whether the adherence of L. plantarum ATCC 8014 was facilitated due to HeLa cell death. Also, it would be important to confirm the probiotic potential of L. plantarum ATCC 8014 in assays with clinical isolates of Clostridium, as not all reference strains tested in the present study are pathogenic, and to assure whether there is any variability in the spectrum of the antimicrobial activity against a larger panel of pathogenic isolates of C. butyricum, C. difficile, and C. perfringens. It is also important to highlight the need for addressing the safety of L. plantarum ATCC 8014 for oral administration in humans, especially in patients who have predisposing factors, such as preterm newborns or critically ill children [40]. Therefore, further studies in animal models and clinical trials would be essential to fill these gaps of knowledge and determine the in vivo benefits of L. plantarum ATCC 8014 against Clostridium spp.

Conclusions
Our findings allow us to conclude that L. plantarum strain ATCC 8014 has probiotic potential, with antimicrobial activity against C. butyricum ATCC 860, C. difficile ATCC 9689, and C. perfringens ATCC 12924. Additionally, this microorganism fulfills essential criteria to survive the harsh conditions of the gastrointestinal tract, as well as to colonize it.