Lactobacillus rhamnosus GG Protects Cells from Clostridium difficile Toxins

Aims: To determine the anti-cytotoxic effects of Lactobacillus rhamnosus GG (LGG) against extracellular and intracellular Clostridium difficile toxins. Study Design: Co-culture system. Place and Duration of Study: Division of Infectious Diseases, Department of Medicine, School of Medicine and Public Health and Department of Pathobiological Sciences, School of Veterinary Medicine, Madison, Wisconsin, between April 2010 and August 2011. Methodology: In this study, we investigated the effects of a probiotic LGG (Culturelle®) against a toxigenic C. difficile strain (ATCC 9689) and a non-toxigenic C. difficile strain (ATCC 700057) in a co-culture system. Co-cultures were prepared with 3 ml of 1:10, 1:100 or 1:1000 dilution of an overnight culture of LGG and 2 ml of 1:100 dilution of either the toxigenic or the non-toxigenic strain. Cytotoxic effects of cell-free culture supernatants (CFS) and cell lysates of the toxigenic strain on Vero cells were evaluated after coculturing. The relative abundance of toxin A (TcdA) and Toxin B (TcdB) genes in 72 h cocultures were determined using real time PCR. Results: In co-cultures with 1:10 or 1:100 dilution of LGG, counts of the toxigenic C. difficile strain were about one log unit lower than control pure cultures after incubation for Research Article British Microbiology Research Journal, 3(2): 165-175, 2013 166 48 h. In all co-cultures, counts of the non-toxigenic strain were two log units lower than those of controls. Accordingly, LGG resulted in a significant decrease (p < 0.05) in the relative abundance of TcdA and TcdB in target DNA prepared from co-cultures containing the 1:10 or the 1:100 dilution of the probiotic. Co-culturing the toxigenic strain with the probiotic (1:10 and 1:100) decreased (P < 0.05) the cytotoxic effect of both extracellular and intracellular clostridial toxins resulting in up to 30% increase in cell viability. Conclusion: LGG inhibits the growth of C. difficile in a dose-dependent manner and protects cells from C. difficile induced cytotoxicity.


INTRODUCTION
Clostridium difficile is a pathogen of tremendous public health importance. According to a recent report by the Centers for Disease Control and Prevention (CDC), C. difficile infections (CDI) are at an all-time high [29]. CDI are linked to 14,000 deaths in the US each year. Between 2000 and 2007, deaths related to C. difficile increased by 400%. Almost half of the infections occur in people younger than 65 while more than 90% of deaths occur in older people. The risk factors associated with the increasing rate of CDI include exposure to antibiotics such as cephalosporins, clindamycin, fluoroquinolones and ampicillin [3].
Toxin A (TcdA) and toxin B (TcdB) are the causative agents of C. difficile-associated diarrhea and colitis [18]. In cultured cells, the two exotoxins are known to induce actin reorganization, cytopathic effects and cell death [20]. The binding and internalization of the toxins leads to a serious of inflammatory cascades that involve the production of Interleukin (IL)-8 and increased permeability of the epithelial barrier [4]. Activation of TcdR gene triggers the induction of TcdA and TcdB gene transcription while TcdC is a negative regulator of C. difficile toxin synthesis [9,26].
The emergence of antibiotic resistant virulent C. difficile strains has warranted the assessment of non-antibiotic based novel treatment options. TcdA and TcdB are targets of CDI therapy. Antibodies specific for TcdA and TcdB have been shown to effectively treat CDI and prevent disease relapse in animal models and in humans [12]. Recently, Lowy et al. [19] reported reduction in CDI recurrence when human monoclonal antibodies developed against TcdA and TcdB were administered with antibiotics. However, antibody treatment is expensive and not commercially available at present.
There is emerging evidence that probiotics may be a promising approach to reduce the burden of CDI [21]. Probiotics are live microorganisms that are available over the counter and represent a low-cost, well tolerated, safe, non-antibiotic based strategy that may have efficacy as adjunctive treatment of infections without the attendant risks of promoting antimicrobial resistance. Saccharomyces boulardii is effective in preventing antibioticassociated diarrhea and recurrent CDI in adults [16]. However, due to the potential risk offungemia associated with this probiotic, it may be preferable to use other probiotics such as lactic acid bacteria to prevent CDI [13,28]. Some studies show the potential of Lactobacillus rhamnosus GG (LGG) to decrease the incidence of antibiotic-associated diarrhea caused by some antibiotics [16,23]. However, there is little evidence on the efficacy and mode of action of either probiotics to treat CDI [25].
The aim of this study was to investigate the effect of LGG (Culturelle®) against C. difficile and cytotoxicity induced by both its extra-and intra-cellular toxins using a co-culture system. Cytotoxic effects of the cell-free culture supernatants (CFS) and the cell lysates of the toxigenic strain were evaluated on Vero cells. The relative quantities of TcdA and TcdB in 72 h co-cultures were also determined using real time PCR.

Bacteria and Cell Culture Conditions
A toxigenic C. difficile strain (ATCC 9689) and a non-toxigenic C. difficile strain (ATCC 700057) were obtained from the American Type Culture Collection (ATCC) (Manassas, VA, USA). Both were cultured in PRAS-Brain Heart Infusion (BHI) broth (Anaerobe Systems) and plate counts were performed by plating serial dilutions of cultures on Cycloserine Cefoxitin Fructose Agar (CCFA: Anaerobe Systems). LGG capsules obtained from i-Health, Inc. /Culturelle (Cromwell, CT) were cultured in MRS broth and plated on MRS agar. Both the C. difficle strains and the LGG cultures were incubated at 37ºC in anaerobic chamber.
For the cytotoxicity studies, Vero cells were cultured in Minimum Essential Medium (MEM) with 10% heat-deactivated fetal bovine serum (FBS), 1% v/v antibiotic antimycotic solution, 0.6% v/v tylosin and 1% glutamine. All incubations involving cell cultures were undertaken at 37ºC in a humidified 5% CO 2 incubator.

Challenge Experiments
Overnight cultures of C. difficile and LGG at ca. 1 x 10 9 CFU were used for co-cultures. Three milliliters of 1:10 or 1:100 or 1:1000 dilutions of LGG were mixed with 2 ml of the 1:100 dilution of the toxigenic or the non-toxigenic C. difficile strain and incubated in anaerobic chamber at 37ºC for 48 h. Aliquots of the co-cultures were used to determine pH and counts of C. difficile. The rest of the co-cultures were used for determination of cytotoxic effects on Vero cells. The relative amounts of TcdA and TcdB in 72 h co-cultures were measured using real-time PCR.

Preparation of Cell-Free Culture Supernatants (CFS) and Cell Lysates
TcdA and TcdB, which have been implicated to be responsible for the cytotoxic effects of toxigenic C. difficile strains, are present both intra-and extra-cellularly [8,24]. Thus, CFS, for the extra-cellular toxins and cell lysates, for the intra-cellular toxins, were prepared as described below. After culturing C. difficile with the three different dilutions of LGG for 48 h, culture supernatants were collected and centrifuged at 13, 600 x g for 15 min. The supernatants were filtered through a 0.22 µm pore size filter membrane and used in the subsequent experiments. The residual bacterial cells were used to prepare the cell lysates. Cells we washed three times with PBS (1x) and were suspended in 3 ml of PBS. The suspension was subjected to a serious of freeze-thaws (3x) and brought to 5 ml with PBS for sonication. The sonication was performed on ice at 30% duty cycle, output set between 3 and 4, for 10 min with 1 min interval. The sonicates were centrifuged (at 14, 000 rpm for 10 min) and filter-sterilized before used in the following cytotoxicity assays.

Cytotoxicity of CFS and Cell Lysates on Vero Cells
Monolayers of Vero cells (5 x 10 4 cells) were prepared in 96-well culture plates. Cells were washed with 1% PBS and incubated with the 1:2 dilution of the CFS or cell lysates of the cocultures in MEM for 24 h. Positive control wells were incubated with the CFS or the cell lysates of the toxigenic C. difficile strain. Negative control wells were incubated with the CFS or the cell lysates of the non-toxigenic C. difficile cultures or that of LGG at the three different dilutions. In addition, some control wells were incubated in MEM without any CFS or cell lysate (untreated). After incubation, MTT assay, which serves as an index of living cells by measuring the mitochondrial function, was performed. The assay was carried out as previously described [10]. About 250 mg MTT was suspended in 50 ml of sterile 1% PBS to make 5 mg/ml MTT solution. Ten micro-litters of the MTT solution were added to each well and plates were further incubated at 37ºC for 4 h. The media were replaced with 100 μl of DMSO solution (94.6 ml DMSO, 0.6 ml acetic acid and 10 g SDS). Plates were then immediately read in a standard ELISA plate reader at 550 nm. The percent viability of Vero cells was calculated using the expression: (OD 550nm of each treatment/OD 550nm of control untreated cells) x 100

DNA Preparation
After incubation for 72 h, DNA samples were prepared from pellets of the co-cultures using QIAamp DNA mini kit (Qiagen) for Gram-positive bacteria. Cultures were centrifuged at 2,800 g for 25 min. Pellets were resuspended in distilled water and re-centrifuged. Following Qiagen´s protocol, the bacteria were lyzed and DNA samples were prepared from pure cultures of the toxigenic C. difficle strain and co-cultures of the toxigenic C. difficle strain with the three different dilutions of LGG. The final DNA extracts were suspended in ddH 2 O and 2 μl of each suspension was used as a target DNA for PCR analyses.

PCR Detection of Toxin A and Toxin B
Primer sequences specific for TcdA and TcdB were obtained from the literature [2]. Primers were synthesized (IDT; Carlsbad, CA) and diluted to 100 μM with ddH 2 O. The other components of the PCR reactions were obtained from Promega (Madison, WI). A 25 μl PCR reaction mixture contained 11.13 μl ddH 2 O, 6.67 Buffer (5x), 1.25 μl of each of the forward and the reverse primers, 2.5 μl of dNTPs (2 mM), 2 μl DNA template and 0.2 μl Taq (5 U/μl). PCR was carried out according to the program: 95ºC for 4 min, followed by 35 cycles of 95ºC for 30 sec, 48ºC (for TcdA) or 45ºC (for TcdB) for 30 sec, 72ºC for 1 min and a final extension at 72ºC for 4 min. PCR products were ran on a pre-cast 1.2% Agarose FlashGel (Lonza, Rockland, ME) and images were visualized and documented using a Gel Doc2000 (BioRad; Hercules, CA).

Quantitative real-time PCR
To compare the relative abundance of TcdA and TcdB genes in the co-cultures and the pure toxigenic C. difficle strain, 25 μl reaction mixtures containing 12.5 μl of Sigma SYBR® Green JumpStart™ TaqReadyMix™ (Sigma-Aldrich Corp, St. Louis, MO), 0.3 μl of 100 μM of each of the forward and the reverse primer and 2.5 μl of DNA were prepared. Quantitative realtime PCR was run with a Strata gene MX3005P TM quantitative PCR system (Agilent Technologies, LaJolla, CA) following the protocol: 94ºC melt step for 4 min and 44 cycles of 94ºC for 20 sec, 45ºC (for TcdA) or 48ºC (for TcdB) for 20 sec and 72ºC for 30 sec. This was followed by a melting curve analysis, which ran for 1 cycle at 95ºC for 1 min, 45ºC for 30 min and 95ºC for 30 sec. The relative abundance of TcdA and TcdB were related to the cycle threshold (Ct). DNA templates of pure cultures of LGG and the non-toxigenic C. difficile strain were used as negative controls.

Data Analyses
All experiments were repeated at least three times allowing the mean and the standard deviations to be calculated. Statistical significances were determined by one-way repeated measures analysis of variance (ANOVA) (SPSS 12.0). A probability (P) value equal to or less than 0.05 was considered significant.

Counts of C. difficile in Co-cultures
After 48 h, counts of the non-toxigenic C. difficile strain (CNT cont) in all co-cultures were about 2 logs lower than cultures without the probiotic (Fig. 1A). Relative to control pure cultures, counts of the toxigenic strain (CT cont) were lower by one log unit when co-cultured with LGG at 1:10 and 1:100. There was no significant difference in the counts of the toxigenic strain co-cultured with LGG at 1:1000 dilution (Fig. 1A). The pH of the co-cultures did not vary significantly from both the control pure toxigenic and the non-toxigenic C. difficile cultures and were higher than the pure LGG cultures at the three dilutions (Fig. 1B). Interestingly, C. difficile counts in co-cultures containing the 1:10 and 1:100 dilutions of LGG remained unchanged from the initial counts used to start the co-cultures (ca. 1 x 10 7 CFU/ml). This suggests that LGG suppressed the growth of C. difficile, which increased in control pure cultures by more than 1 (CT cont) or 2 logs (CNT cont) (Fig. 1A). Trejo et al. [27] reported a significant decrease in C. difficile toxins in co-cultures containing strains of bifidobacteria while the growth of clostridial strains in co-cultures was similar to pure clostridial control cultures. Low pH conditions caused by organic acid secretion were shown to have inhibitory effects on the growth and cytotoxicity of C. difficile [30]. However, in our study the pH of co-cultures containing LGG was not significantly different from that of the pure clostridial cultures. This suggests the presence of a different mechanism by which LGG suppresses the growth of C. difficile and protects cells from its cytotoxic effects.

Cytotoxic Effect of Extracellular Toxin
CFS of 48 h old co-cultures and that of the pure C. difficile cultures were used to determine the effect of LGG on the cytotoxic effect of extracellular C. difficile toxins. Vero cells were incubated with either of the CFS for 24 h. The CFS of all cultures containing the toxigenic C. difficile strain caused cytopathic effects (cell rounding and detachment). Treatment of Vero cells with the CFS of the toxigenic C. difficile strain resulted in the most significant decrease in the number of viable cells (Fig. 2). There was significant (p < 0.05) protection by LGG resulting in 15-30% increase in the percent of viability when the toxigenic strain was coincubated with LGG at the different dilutions (Fig. 2).
Co-cultures have been used to demonstrate the potential of various probiotic bacteria to decrease the number and virulence of C. difficle [11,27]. In our co-culture system, although the decrease in the number of the toxigenic C. difficile strain relative to the controls was only a log, LGG resulted in a significant decrease in the cytotoxic effects of clostridial toxins as compared to pure clostridial cultures. Similarly, Banerjee et al. [1] reported that the CFS of the probiotic L. bulgaricus LDB B-30892 reduced C. difficile-induced cytotoxicity. They hypothesized possible mechanisms of detoxification such as proteolytic cleavage of toxin or toxin receptors. In addition, S. boulardii has been shown to have proteolytic activity on C. difficile toxins [7]. In contrast, Carasi et al. [5] showed that proteolytic activity is not involved in the protective effect of kefir lactobacilli against clostridial toxins.

Cytotoxic Effects of Intracellular Toxin
The cell lysates of the co-cultures were used to study the cytotoxic effects of intracellular C. difficile toxins after co-culturing with LGG. As compared with the CFS, the cell lysates of the toxigenic C. difficile strain cultures had lesser cytopathic effects on Vero cells. When the cell lysates were added on monolayers of Vero cells, the viability of the cells decreased significantly after 24 h (Fig. 3). Co-incubating the toxigenic strain with culturelle had significant protective effect resulting in a 20%, 30% and 15% increase in cell viability with the 1:10, 1:100 and 1:1000 dilution of the probiotic, respectively (Fig. 3). Probiotics have many potential mechanisms of actions; including competitive exclusion, increasing host epithelial barrier function and modulation of host immune responses [14,17]; although in general the mechanisms of actions have not been adequately studied. Recently, different authors have reported protective effects of various probiotics against cytotoxic effects induced by cell-free culture supernatants (CFS) containing clostridial toxins [27,30].
In this study, we showed that the growth of C. difficile in the presence of LGG leads to CFS and cell lysates with significantly lower cytotoxic effect as compared with those from pure clostridial cultures. Most previous studies investigated the effect of probiotics only on the cytotoxicity of secreted C. difficile toxins in CFS. However, it is very important to study the protective effects on the intracellular clostridial toxins. Release of cytotoxins may be delayed by various factors such as the incubation period [24]. Furthermore, it has long been reported that the minimum dose of intracellular clostridial toxin causing cytotoxicity is lower than extracellular toxin [22]. In this regard, we were able to show that LGG decreased the cytopathic effect of not only the released toxins but also the cell-associated C. difficile toxins.

Relative Quantities of Toxin A and Toxin B Genes
After incubation for 72 h, TcdA (Fig. 4A) and TcdB (Fig. 4B) were detected in both the pure cultures of the toxigenic C. difficile strain (CT) and the co-cultures of the strain with the probiotic at the three different dilutions. The non-toxigenic C. difficile strain (CNT) was confirmed to be negative for both toxins (Fig. 4).
The relative abundance of TcdA and TcdB genes in DNA templates from the pure cultures of the toxigenic C. difficile strain (CT) and the co-cultures after 72 h was related to the Ct values (Fig. 5). The Ct values are inversely proportional with the number of each gene and the counts of the toxigenic strain. Co-cultures with LGG at 1:10 and 1:100 dilutions provided significantly higher Ct values (P < 0.05) than the pure C. difficile cultures (Fig. 5); indicating inhibition of the growth of C. difficile by the probiotic.
LGG dilution at 1:1000 did not affect the relative amount of TcdA (Fig. 5A). Our results showed LGG could inhibit the growth of C. difficile in co-cultures. This was supported by the lower quantities of TcdA and TcdB genes in co-cultures containing LGG at 1:10 and 1:100 dilutions (Fig. 5). In co-cultures with the 1:1000 dilution of LGG, the amount of Toxin A was almost the same as that in pure C. difficile cultures. This was in line with the counts of the toxigenic strain ( Fig 1A) that were not affected by the probiotic at the lowest dilution. This suggests the ability of the probiotic to suppress the growth of C. difficile in a dose-dependent manner.
The mode of action of probiotics by which they antagonize gastrointestinal pathogens by decreasing the expression levels of their virulent genes has not been well investigated. Carey et al. [6] reported the down-regulation of Shiga toxin 2 subunit A in a Shiga toxin producing Escherichia coli (STEC) strain after co-culturing with different probiotics. Various possible mechanisms of antagonism of C. difficile by lactic acid bacteria have been suggested [15,27]. Our study encourages further investigation on the effects of metabolic products of LGG on the regulation of clostridial toxins. In this study, the mRNA expression levels of Toxin A and Toxin B were not determined. This would have given more valuable information on the direct effect of LGG on the expression of the toxin genes. Furthermore, additional studies are warranted to determine the role of the probiotic in reducing levels of secreted and cell-associated clostridial toxins.

CONCLUSION
There is an increasing interest towards the use of probiotics in prevention and treatment of CDI [23]. In this study, we showed that Culturelle (LGG) has antagonistic effects against C. difficile by protecting eukaryotic cells from cytotoxicity induced both by intracellular and extracellular clostridial toxins. This was shown by the growth inhibition of C. difficile in cocultures, which was confirmed by the decreased relative abundance of Toxin A and Toxin B genes in cultures containing the probiotic. Our study suggests further investigation on the role of the probiotic in regulating these genes and other genes in the pathogenicity locus of C. difficle that are involved in the synthesis of clostridilal toxins.