Lactobacillus plantarum CS24.2 prevents Escherichia coli adhesion to HT-29 cells and also down-regulates enteropathogen-induced tumor necrosis factor-α and interleukin-8 expression

Authors

  • Akhilesh S. Dhanani,

    1. Department of Microbiology and Biotechnology Centre, Faculty of Science, Maharaja Sayajirao University of Baroda, Vadodara, Gujarat, India
    Search for more papers by this author
  • Tamishraha Bagchi

    Corresponding author
    • Department of Microbiology and Biotechnology Centre, Faculty of Science, Maharaja Sayajirao University of Baroda, Vadodara, Gujarat, India
    Search for more papers by this author

Correspondence

Tamishraha Bagchi, Department of Microbiology and Biotechnology Centre, Faculty of Science, Maharaja Sayajirao University of Baroda, Vadodara, Gujarat 390 002, India.

Tel: +91 265 2794396; fax: +91 265 2792508; email: mailforbagchi@yahoo.com

ABSTRACT

The aim of the present study was to evaluate the potential of Lactobacillus plantarum CS24.2 to antagonize Escherichia coli adhesion and modulate expression of the responses by HT-29 cells of inflammatory molecules to E. coli adhesion. Experiments were performed under different adhesion conditions and findings compared with the responses of Lactobacillus rhamnosus GG. Tests of competitive adhesion, adhesion inhibition and displacement assays were performed for lactobacilli (L. rhamnosus GG and L. plantarum CS24.2) and E. coli O26:H11 to HT-29 cells. Both the lactobacilli significantly reduced E. coli adhesion to HT-29 cells (P < 0.05). The ability of lactobacilli to modulate tumor necrosis factor-α and interleukin-8 expression was analyzed in HT-29 cells stimulated with E. coli using qRT-PCR. L. plantarum CS24.2 significantly down regulated expression of both the genes induced by E. coli in HT-29 cells at 6 hr as well as 24 hr, which was more significant than the corresponding findings for L. rhamnosus GG. The present findings suggest that L. plantarum CS24.2 inhibits pathogen adhesion to a similar extent as does the established probiotic strain L. rhamnosus GG. It may also attenuate tumor necrosis factor-α and interleukin-8 expression in HT-29 cells stimulated with E. coli.

List of Abbreviations
CFS

cell free supernatant

cfu

colony forming units

DMEM

Dulbecco's modified Eagle's minimal essential medium

E. coli

Escherichia coli

EPEC

enteropathogenic E. coli

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

IBD

inflammatory bowel diseases

IL

interleukin

L.

Lactobacillus

MRS

de Man–Rogosa–Sharpe

NFκB

nuclear factor of κ light polypeptide gene enhancer in B cells

qRT-PCR

quantitative reverse transcriptase polymerase chain reaction

SD

standard deviation

TNF

tumor necrosis factor

The human gastrointestinal tract is colonized by a large number of both aerobic and anaerobic bacterial species. The gut epithelial layer is covered with a thick mucus layer that protects the underlying epithelial cells from direct contact with gut microbiota and food antigens [1, 2]. The local adaptive immune system also protects the gut epithelial layer, which includes gut-associated lymphoid tissue [3]. Together, the innate and adaptive systems maintain gut homeostasis under normal conditions. However, this balance is disturbed under various circumstances, including secretion of toxins and degradation of the mucin layer upon adhesion of gut pathogens and the consequent deleterious effect of these pathogens on tight junction integrity of the epithelial layer. Pathogenic strains of E. coli cause various gastrointestinal infections [4]. Adhesion of pathogenic bacteria to the epithelial layer is essential to manifest their virulence attributes and cause infection. Recent studies that analyzed gut microbiota in IBD patients showed increased numbers of adhesive and enterohemorrhagic E. coli on the surfaces of inflamed tissues [5]. Enteropathogenic E. coli also reportedly adhere to enterocytes and have a role in ulcerative colitis [6, 7]. Various mechanisms can protect the gut epithelium from colonization by pathogens and their resultant adverse effects. Several strains of Lactobacillus reportedly compete with enteropathogens and antagonize adhesion sites on gut epithelium. Lactobacilli produce antimicrobial substances like lactic acid, superoxide radicals and/or antimicrobial peptides such as bacteriocins that also contribute to their probiotic effects [8]. HT-29 cell line, a human colonic adenocarcinoma cell type which, upon differentiation, has the structural properties of mature enterocytes, is considered as an ideal model for studies related to gut epithelium [9].

The gut epithelial cells play an important role in maintaining gut homeostasis. The enterocytes of the epithelial layer act as immunocompetent cells and secrete various signaling molecules, such as cytokines and chemokines, upon adhesion and invasion by gut pathogens [10]. Upon stimulation, specialized immune cells such as neutrophils and macrophages migrate to the site of infection and prevent pathogen entry. Under certain conditions, inflammation becomes self-sustaining even after elimination of the pathogens which led to the inflammatory disorder because of ineffective down regulation of pro-inflammatory molecules [11]. TNF-α and IL-8 are reportedly strongly expressed in Crohn's disease and acute colitis patients [12, 13]. Probiotic lactobacilli have the potential to interact with the mucosal layer and can thereby remarkably reduce expression of these pro-inflammatory molecules [14, 15].

The aims of the present study were to evaluate the well characterized putative probiotic isolate L. plantarum CS24.2 (originally isolated from a child's fecal sample) for its ability to inhibit adhesion of enteropathogenic E. coli O26:H11 to the intestinal epithelial cell line HT-29 and to compare it with the well established probiotic strain L. rhamnosus GG. We also analyzed the ability of L. plantarum CS24.2 to normalize the degree of expression of TNF-α and IL-8 in HT-29 cells induced by adherent E. coli O26:H11 in various in vitro assays for possible application in the treatment of IBD.

MATERIALS AND METHODS

Bacterial strains and culture conditions

The L. plantarum CS24.2 isolate used in this study was originally obtained from a fecal sample from a healthy human child. The isolate was analyzed for various probiotic properties, including acid and bile tolerance, adhesion to Caco-2 cells and antimicrobial activity against various pathogens [16]. This isolate was found to be the best of the isolates in the laboratory collection [17]. The established probiotic strain, L. rhamnosus GG, was obtained as a kind gift from Dr. Shira Doron, MD, Department of Medicine, Tufts–New England Medical Center, USA [18]. The adhesive EPEC serotype O26:H11 was isolated from an acute colitis patient and maintained in the culture collection of our department. Lactobacilli were grown in MRS (Hi-media, Mumbai, India) broth at 37°C for 18–20 hrs before the actual experiments. E. coli O26:H11 was grown aerobically at 37°C in Luria broth (Hi-media) for 16–18 hrs.

Intestinal epithelial cell line

The human colonic adenocarcinoma cell line, HT-29 was obtained from the National Centre for Cell Science, Pune, India and routinely cultured in DMEM (Sigma–Aldrich, St. Louis, MO, USA) at 37°C temperature in an incubator with 5% CO2 and 95% air atmosphere. The medium was supplemented with 10% (v/v) FBS (Sigma–Aldrich), 10 mM non essential amino acids, 1 mM Na pyruvate and 50 µg/mL gentamicin. The media lacked gentamicin whenever antibiotic free medium was used.

Inhibition of enteropathogen adhesion to HT-29 cells with lactobacilli and cell free supernatants

For adhesion assays, the HT-29 cells were seeded at a density of 1 × 105 cells/well in 24-well standard tissue culture plates (Corning, Corning, NY, USA) and maintained for at least 7 days to achieve confluent growth. The culture medium was replaced every alternate day. Before the experiment, the culture medium was replaced with antibiotic free medium of pH 6.5 adjusted with 1 N HCl and the cells maintained in this medium for 4 hrs [17]. Lactobacilli and E. coli were harvested by centrifugation (10,000 g, 2 mins and 4°C) and washed twice with Dulbecco's PBS, pH 7.0 (Sigma–Aldrich). The cell density was adjusted by measuring absorbance at 600 nm to get 1 × 108 cfu in 50 µL cell suspension for each strain. The exact number of viable lactobacilli and E. coli used in the assays was determined for each experiment by plate counting on MRS agar and Luria agar, respectively.

To mimic different in vivo conditions, various adhesion assays were designed for competitive inhibition, adhesion inhibition and displacement of enteropathogenic E. coli O26:H11 by lactobacilli. For the competitive inhibition assay, lactobacilli and pathogen were provided with an equal chance for binding at the same ratio. The adhesion inhibition assay was performed to investigate the role of lactobacilli growth and its ability to protect intestinal cells from being colonized by pathogen. In the displacement assay, the ability of lactobacilli to displace colonized pathogen from intestinal epithelium was evaluated. The pathogen was allowed to adhere first to HT-29 cells before lactobacilli adhesion. For competitive inhibition, the same number (1 × 108 cfu/well) of lactobacilli and E. coli were added at the same time to the HT-29 cells, and co-incubated for 90 mins. For both the adhesion inhibition and displacement assays, lactobacilli and E. coli were allowed to adhere to HT-29 cells for 90 mins. Non-adherent bacterial cells were then removed by washing the wells thrice with 1 mL PBS each time. E. coli and lactobacilli were then added to different wells and incubated for an additional 90 mins. At the end of each assay, HT-29 cells were lysed by treatment with 0.5 mL of 0.05% (v/v) Triton X-100 in PBS for 20 mins at 37°C. The HT-29 lysates, including bound bacterial cells, were plated after appropriate dilution on Luria and MRS agar plate for E. coli and lactobacilli, respectively. Enumeration was performed after 18–24 hrs incubation at 37°C. HT-29 cells co incubated with E. coli alone were considered controls and the number of bacteria adhering to HT-29 cells was considered as 100%. Treatment with 0.05% (v/v) Triton X-100 in PBS for 30 mins at 37°C did not affect the viability of E. coli and lactobacilli (data not shown).

To check the role of any secretary molecules in adhesion inhibition of E. coli, the CFS of both lactobacilli were analyzed for antimicrobial activity. CFS was obtained from a culture of lactobacilli grown overnight by separating cells by centrifugation (10,000 g, 15 mins and 4°C). One microliter of filter sterilized CFS was mixed with equal volume of 1 × 108 cfu in PBS and incubated at 37°C for 1 hr. After incubation, the cells were washed twice with PBS before being resupended in 1 mL of DMEM and co incubated with HT-29 cells for 90 mins. Enumeration was performed as described above. Incubation of E. coli with MRS did not affect the viability of bacteria in 90 mins (data not shown). All assays were performed with between 40 and 60 passages of HT-29 cells and repeated three times in duplicate.

Stimulation of HT-29 cells with lactobacilli and Escherichia coli

To analyze the degree of transcript expression of TNF-α and IL-8 in HT-29 cells upon stimulation with lactobacilli and/or E. coli O26:H11 in co culture conditions, the various adhesion assays described above were performed. HT-29 cells were stimulated with 1 × 108 cfu/mL of lactobacilli and E. coli O26:H11 and co-incubated for 90 mins at 37°C in a CO2 incubator to assess competitive adhesion. For the adhesion inhibition and displacement assays, HT-29 cells were pre-incubated with lactobacilli and E. coli,respectively, for 90 mins. Thereafter, the unbound bacterial cells were removed by washing as described above and the HT-29 cells further incubated for 90 mins with E. coli and lactobacilli for the adhesion inhibition and displacement assays. To analyze the ability of lactobacilli and E. coli to modulate expression of pro inflammatory molecules, HT-29 cells were co incubated with either lactobacilli or E. coli alone. After this initial incubation, the culture medium was replaced with gentamicin-containing media (50 µg/mL) to prevent bacterial growth and further incubated under the same conditions for another 4 hrs and 30 mins and 22 hrs and 30 mins for analysis of degree of transcription at 6 and 24 hr, respectively. At the end of incubation, culture supernatants were collected and stored at −80°C. HT-29 cells were lysed in the presence of guanidine thiocyanate, which is included in the total RNA extraction kit (Bangalore Genei, Bangalore, India) employed. Further steps were performed as per the manufacturer's instructions.

cDNA synthesis and quantitative reverse transcriptase polymerase chain reaction

Total RNA was isolated from control HT-29 cells (to which no bacteria had been added) and those that had been co incubated with lactobacilli and/or E. coli O26:H11. The quality of the RNA samples was analyzed by inspecting the integrity of 28S and 18S bands on agarose gel electrophoresis. The total RNA in each sample was quantified by NanoPhotometer Pearl (Implen GmbH, München, Germany). For cDNA synthesis, 1 µg of each total RNA sample was mixed with anchored oligo-dT in a 20 µL system using verso cDNA synthesis kit based on Moloney murine leukaemia virus reverse transcriptase (Thermo Fisher Scientific, Surrey, UK) according to the manufacturer's instructions. Briefly, the RNA was mixed with oligo dT, RT enhancer, which contains DNAse I, dNTP mix and enzyme mix, followed by incubation at 50°C for 30 mins and then 95°C for 2 mins to inactivate the enzyme in a thermal cycler (Eppendorf, Hamburg, Germany). Quantitative PCR amplifications were then performed in a CFX96 real-time thermal cycler (Bio-Rad Laboratories, Hercules, CA, USA) with specific primers for TNF-α and IL-8 with 1 µL of first strand cDNA, 1.25 pmol each of the forward and reverse primers and 5 µL of 2× SsoFast EvaGreen supermix (Bio-Rad Laboratories) in a 10 µL system. The amplification conditions were as follows: initial denaturation at 94°C for 3 mins, followed by 45 cycles of denaturation at 94°C for 10 s, annealing and extension for 30 s at 60°C. The fluorescence was recorded after each cycle. Each sample was run in triplicate and the cycle threshold was used for gene expression analysis. Transcript expression of TNF-α and IL-8 in each sample was normalized to a GAPDH transcript expression of the same sample using CFX manager software (Bio-Rad Laboratories). Data were analyzed using the math formula method. The product specificity was confirmed by single peak in melt curve analysis (from 65 to 95°C in 0.5°C/5 s increments). The negative controls were set with the total RNA without reverse transcription (data not provided).

Interleukin-8 quantification in culture supernatants

Interleukin-8 in culture supernatants was measured by ELISA using a Mini ELISA Development Kit (PeproTech Asia, Rehovot, Israel). The steps were performed as per the manufacturer's instructions. A standard plot was generated for quantifying IL-8 in samples using the known concentration of recombinant human IL-8.

Statistics

Values are given as mean and standard deviation (SD) from three experimental replicates. Significant ANOVAs were followed by Dunnett's test in the case of the various adhesion assays to HT-29 cells and compared with the respective controls (P < 0.05). Statistical analysis was conducted using SigmaStat 3.5 software.

RESULTS

Ability of lactobacilli to antagonize adhesion of Escherichia coli O26:H11 to HT-29 cells

For each combination of Lactobacillus with E. coli, three competitive conditions were tested: competitive inhibition, adhesion inhibition and displacement (Fig. 1). For competitive inhibition, lactobacilli and E. coli were given the opportunity to adhere to HT-29 cells for the same time and in equal ratio. L. rhamnosus GG and L. plantarum CS24.2 significantly reduced the adhesion of E. coli by 47.5% and 41.2%, respectively (P < 0.05). To study the ability of adhered lactobacilli to inhibit adhesion of E. coli, an adhesion inhibition assay was carried out. Pre-colonization of HT-29 cells by L. rhamnosus GG and L. plantarum CS24.2 inhibited E. coli adhesion by 36.5% and 33.9%, respectively. To analyze the ability of lactobacilli to displace adhered enteropathogens from intestinal epithelial cells, a displacement assay was carried out by allowing adhesion of E. coli prior to addition of lactobacilli to HT-29 cells. L. rhamnosus GG and L. plantarum CS24.2 significantly displaced E. coli by 58.4% and 62.9%, respectively (P < 0.05). The CFS of both the lactobacilli showed antimicrobial activity. Adhesion of E. coli was reduced by 48.7% and 59.1% in the presence of culture supernatant of L. rhamnosus GG and L. plantarum CS24.2, respectively.

Figure 1.

Adhesion of Escherichia coli O26:H11 to HT-29 cells in the presence of CFS and lactobacilli under different competitive conditions. Adhesion of E. coli in the absence of lactobacilli served as control. Each bar represents mean value and standard deviation as error bar of three independent experiments. The square box above each bar shows the percentage reduction in adhesion of E. coli as compared to control. Significant ANOVAs were followed by Dunnett's test for multiple comparisons versus control group. Black bars, L. rhamnosus GG or L. plantarum CS24.2; white bars, E. coli. AI, adhesion inhibition; C, competition inhibition; CFS, cell free supernatants; D, displacement. *Mean value of adhesion significantly lower than that of control (P < 0.05). †Within L. rhamnosus GG or L. plantarum CS24.2, mean value of adhesion of E. coli statistically similar in the same adhesion assay (P < 0.05).

It should be noted that both the lactobacilli were able to adhere to HT-29 cells under all the three competitive conditions. Overall, the ability of L. plantarum CS24.2 to inhibit adhesion of E. coli O26:H11 to an intestinal epithelial cell line under different adhesion conditions was very similar to that of the established probiotic strain L. rhamnosus GG.

Immunomodulatory effects of lactobacilli on HT-29 cells stimulated with Escherichia coli O26:H11

The ability of lactobacilli to modulate the degree of transcript expression of pro-inflammatory molecules in HT-29 cells stimulated with and without E. coli O26:H11 was analyzed under different adhesion assays at 6 hrs (Fig. 2a, b) and 24 hrs (Fig. 2c, d). Compared to controls to which no bacteria had been added, there were no significant changes in expression of TNF-α and IL-8 in HT-29 cells co cultured with lactobacilli at either 6 hrs or 24 hrs (P < 0.05). The EPEC strongly induced TNF-α and IL-8 expression at 6 hrs with increases in fold expression of 1.8 and 2.4, respectively. Fold expression continued to increase with time, reaching 6.1 and 9.0 at 24 hrs for TNF-α and IL-8, respectively.

Figure 2.

Comparison of effects of L. rhamnosus GG and L. plantarum CS24.2 on mRNA expression of tumor necrosis factor (TNF)-α and interleukin (IL)-8 in HT-29 cells upon stimulation with enteropathogenic E. coli O26:H11 under different adhesion assays. The degree of transcript expression by HT-29 cells was analyzed (a, b) at 6 hrs and (c, d) 24 hrs. Expression of TNF-α and IL-8 was normalized to the internal reference (GAPDH expression). The relative degree of expression in the control (to which no bacteria added) was set at one for fold expression analysis of the other experimental groups. All data are presented as the mean ± SD of three independent experiments. Significant ANOVAs were followed by Dunnett's test for multiple comparisons versus control group. AI, adhesion inhibition; C, competition inhibition; D, displacement. *The mean value of relative expression in all experimental groups is significantly higher than that of the control (P < 0.05). †The mean value of relative expression in different adhesion assays is significantly lower than that of E. coli (P < 0.05).

In the competitive adhesion assay, L. rhamnosus GG was unable to prevent up-regulation of TNF-α and IL-8 in HT-29 cells co-stimulated with E. coli; the degree of expression was the same as observed with stimulation by E. coli alone. Under similar assay conditions, L. plantarum CS24.2 significantly normalized expression of both the genes till 6 hrs. Expression was increased with further incubation up to 24 hrs, the fold increase in expression being 2.4 and 4.7 for TNF-α and IL-8, respectively. Interestingly, expression was significantly less than that of HT-29 cells stimulated with E. coli alone (P < 0.05). In the adhesion inhibition assay, both lactobacilli maintained expression of TNF-α and IL-8 to the basal level up to 6 hrs upon stimulation with E. coli after lactobacilli adhesion. The fold expression of both genes was increased at 24 hrs but it was significantly less than that of HT-29 cells stimulated by E. coli alone; the exception was expression of TNF-α in the case of L. rhamnosus GG. The fold expression of TNF-α and IL-8 was 4.8 and 6.1, respectively, for L. rhamnosus GG, whereas it was 3.3 and 4.6, respectively, for L. plantarum CS24.2. In the displacement assay, both lactobacilli prevented up-regulation of TNF-α and IL-8 up to 6 hrs in HT-29 cells pre-incubated with E. coli. The fold expression of both genes was significantly increased at 24 hrs in HT-29 cells, the exception being IL-8 in the case of L. plantarum CS24.2 (P < 0.05). When compared with HT-29 cells stimulated with E. coli alone, there was no significant decrease in fold expression of either gene at 24 hrs in the case of L. rhamnosus GG. L. plantarum CS24.2 significantly reduced the fold expression of TNF-α and IL-8 to 2.9 and 1.6, respectively, when compared to E. coli induced expression at 24 hrs. Adhesion of L. plantarum CS24.2 to HT-29 cells pre-colonized with E. coli protected the cells from up-regulation of IL-8 even after 24 hrs incubation.

The ability of lactobacilli to inhibit IL-8 secretion from HT-29 cells stimulated with E. coli O26:H11 was also confirmed at 24 hrs using ELISA (Fig. 3). There was no significant difference in secretion of IL-8 on stimulation with L. rhamnosus GG (180 ± 12 pg/mL) and L. plantarum CS24.2 (185 ± 10 pg/mL) compared to controls (148 ± 10 pg/mL). E. coli significantly increased IL-8 secretion from HT-29 (786 ± 102 pg/mL). L. rhamnosus GG was unable to down-regulate IL-8 synthesis by HT-29 cells stimulated with E. coli under different competitive conditions, the exception being the adhesion inhibition assay (662 ± 17 pg/mL). Compared to the same HT-29 cells stimulated with E. coli., L. plantarum CS24.2 significantly reduced secretion of IL-8 in all adhesion assays The degree of IL-8 secretion by HT-29 cells was statistically similar to the control values in the displacement assay with L. plantarum CS24.2 (187 ± 36 pg/mL). Overall, transcript expression and secretion of IL-8 was comparable in all adhesion assays.

Figure 3.

Comparison of effect of L. rhamnosus GG and L. plantarum CS24.2 on interleukin (IL)-8 secretion by HT-29 cells upon stimulation with enteropathogenic E. coli O26:H11 under different adhesion assays at 24 hrs. All data are presented as the mean ± SD of three independent experiments. Significant ANOVAs were followed by Dunnett's test for multiple comparisons versus the control group. AI, adhesion inhibition; C, competition inhibition; D, displacement. *The mean value of secreted IL-8 in all experimental group is significantly higher than that of the control (P < 0.05). †The mean value of secreted IL-8 in different adhesion assays is significantly lower than that of E. coli (P < 0.05).

DISCUSSION

Normal host–microbial interactions in the gastrointestinal tract are important for maintaining human health. Interruption of these interactions and the resultant impaired microbial balance leads to gastrointestinal disorders such as IBD and infectious gastroenteritis [19, 20]. Probiotics act therapeutically by replacing pathogenic bacteria with potentially beneficial microbes. The effects of probiotics are reportedly strain-specific and vary among different strains of the same species [21]. Thus, careful selection of the appropriate probiotic lactobacilli is important when targeting specific diseases. L. rhamnosus GG has been widely studied and is considered one of the best probiotic strains. It is an effective therapeutic agent in several gut-associated infections and disorders [18]. In the present study, we evaluated the ability of two Lactobacillus strains (L. plantarum CS24.2 and L. rhamnosus GG) to antagonize adhesion by EPEC to intestinal epithelial cells (HT-29 cells). In all adhesion assays, both Lactobacillus strains significantly inhibited adhesion of E. coli to HT-29 cells. Competitive inhibition reflects the affinity of surface molecules on competing bacteria for the available adhesion sites on the gut epithelium [22]. Both lactobacilli reduced adhesion of E. coli by between 40% and 50%, which is similar to our previous report using Caco-2 cells [16]. The similar large numbers of adherent lactobacilli and E. coli suggest that they compete for the same attachment sites. It is interesting to note that pre-colonization of L. plantarum CS24.2 to Caco-2 cells did not prevent E. coli adhesion according to a previous study [16]. However, in the present study, the same strain did prevent pathogen adhesion to HT-29 cells. This could be due to differences in the surface profiles of these cell lines, both of which reportedly produce some mucin, which also acts as an attachment site [23]. Both the lactobacilli assessed displaced adherent E. coli from HT-29 cells. Their displacement ability was stronger than their competition and adhesion inhibition abilities. Production of antimicrobial compounds or anti adhesion factors by probiotic bacteria, rather than mere competition for common attachment sites, reportedly mediates such displacement [24]. Our study also supports this finding. The ability of lactobacilli to adhere was poor when HT-29 cells had been pre-colonized with E. coli and the CFS of both lactobacilli were able to inhibit adhesion of E. coli to HT-29 cells. Thus, the optimal use of this ability would be for treatment of gastrointestinal disorders involving pre-colonized adhesive pathogenic bacteria. Overall, the ability of L. plantarum CS24.2 to oppose pathogen adhesion to HT-29 cells was comparable with that of the well-established probiotic strain L. rhamnosus GG. In all adhesion assays, L. plantarum CS24.2 did adhere to HT-29 cells and inhibit adhesion of E. coli;these findings can be related to its strong adhesion and effacing abilities.

Interactions between microbes and epithelial cells modulate cytokine expression by enterocytes [25]. The lipopolysaccharides of gram-negative bacteria increase expression of TNF-α in gut mucosa [26]. IL-8 is a chemokine that acts as a chemo attractant, drawing neutrophils to sites of infection [27]. One of the key regulators of chronic inflammation in IBD is reportedly enhanced expression of pro-inflammatory molecules, including TNF-α and IL-8 [12, 13]. In the present study, both Lactobacillus strains were unable to induce expression of TNF-α and IL-8 in HT-29. However, adhesive E. coli strongly induced expression of both these pro-inflammatory molecules within 6 hrs and their expression continued to increase with the time. Our gene expression studies involving co-incubation of lactobacilli and pathogens with HT-29 cells showed different ability to regulate expression of these molecules in different adhesion assays. When competing with E. coli, L. rhamnosus GG did not maintain the basal degree of expression of TNF-α and IL-8 by HT-29 cells. For the initial 6 hrs of the adhesion inhibition and displacement assays, both lactobacilli maintained expression of these inflammatory molecules to the degree found with unstimulated HT-29 cells. However, at 24 hrs, expression of these molecules increased in the case of L. rhamnosus GG and reached the same degree as E. coli stimulated HT-29 cells, the exception being IL-8 in the adhesion inhibition assay. In a similar assay, Lopez et al. showed that pre-colonized L. rhamnosus GG down-regulates flagellin-induced IL-8 production by Caco-2 cells [28]. When competing with E. coli in the various assays, L. plantarum CS24.2 was unable to protect HT-29 cells from induction of TNF-α and IL-8 at 24 hrs; however, the fold expression of both molecules was significantly less than that of E. coli stimulated HT-29 cells. Surprisingly, IL-8 expression was similar to that of the control after we added L. plantarum CS24.2 to HT-29 cells pre-stimulated with E. coli. We also observed the same effect in secretion of IL-8 from HT-29 cells.

The exact mechanisms of immunomodulation by probiotic bacteria are still under investigation; however, several reports suggest that lactobacilli mediate anti-inflammatory effects by modulating NF-κB signaling pathways in the gut epithelium [29, 30]. Of note, in this study, the ability of L. plantarum CS24.2 to attenuate IL-8 secretion in the displacement assay was markedly stronger than its abilities in the other two assay conditions. Suppression of activation of NF-κB and subsequent IL-8 expression by some secretory molecules of L. plantarum CS24.2 may explain this difference. The fact that, in the displacement assay, the unbound E. coli were removed before adding 1 × 108 cfu of L. plantarum CS24.2 to HT-29 cells may have led to increased accumulation of such anti-inflammatory molecules. In the competitive inhibition assay, equal numbers of lactobacilli and E. coli simultaneously stimulated the HT-29 cells, the balance being in favor of E. coli because it is a strong inducer of IL-8. Similarly, in the adhesion inhibition assay, the number of bound lactobacilli and molecules secreted by them would have been fewer when 1 × 108 cfu of E. coli was added after removal of unbound lactobacilli. These considerations explain why the activities in these two assays were not as strong as in the displacement assay. Additionally, in the displacement assay the number of lactobacilli added was large compared to the number of E. coli that remained adherent to the HT-29 cells after the initial wash step. The number of bacteria has been shown to play an important role in immunomodulation [31]. However, for L. plantarum CS24.2, both the purported secretory molecules and the mechanisms involved have yet to be identified. Furthermore, both lactobacilli showed similar ability to inhibit adhesion of E. coli but different abilities to attenuate IL-8 expression in the displacement assay, suggesting that pathogen inhibition and immunomodulation are independent and strain specific properties. Although, L. rhamnosus GG is reportedly effective in various infectious diseases, recent clinical data suggests that its efficacy in IBD is mixed [32].

The ability of L. plantarum CS24.2 to inhibit EPEC O26:H11 adhesion to an intestinal epithelial cell line suggests the suitability of this strain for replenishing protective microbes in pathogen-associated gut diseases. Along with its antagonistic ability, the inability of L. plantarum CS24.2 to elicit expression of pro-inflammatory molecules and down-regulate expression of TNF-α and IL-8 induced by E. coli O26:H11 in HT-29 cells shows the suitability of this strain for maintaining protective immunity in both healthy subjects and inflammatory disease patients. This primary study strongly suggests the efficacy of L. plantarum CS24.2 for treatment of inflammatory diseases. The findings need further confirmation by an in vivo study.

ACKNOWLEDGMENTS

The authors thank the Department of Biotechnology, New Delhi, India for financial support to carry out research work (Grant number BT/PR-7496/PID/20/292/2006). ASD was supported by a fellowship from the University Grant Commission, New Delhi, India.

DISCLOSURE

A. S. Dhanani and T. Bagchi have no conflicts of interest to disclose.

Ancillary