Comparison of the immunomodulatory properties of three probiotic strains of Lactobacilli using complex culture systems: prediction for in vivo efficacy

Abstract

Background: While the use of probiotics to treat or prevent inflammatory bowel disease (IBD) has been proposed, to this point the clinical benefits have been limited. In this report we analyzed the immunological activity of three strains of Lactobacillus to predict their in vivo efficacy in protecting against experimental colitis.

Methodology/principal findings: We compared the immunological properties of Lactobacillus plantarum NCIMB8826, L. rhamnosus GG (LGG), L. paracasei B21060 and pathogenic Salmonella typhimurium (SL1344). We studied the stimulatory effects of these different strains upon dendritic cells (DCs) either directly by co-culture or indirectly via conditioning of an epithelial intermediary. Furthermore, we characterized the effects of these strains in vivo using a Dextran sulphate sodium (DSS) model of colitis. We found that the three strains exhibited different abilities to induce inflammatory cytokine production by DCs with L. plantarum being the most effective followed by LGG and L. paracasei. L. paracasei minimally induced the release of cytokines, while it also inhibited the potential of DCs to both produce inflammatory cytokines (IL-12 and TNF-alpha) and to drive Th1 T cells in response to Salmonella. This effect on DCs was found under both direct and indirect stimulatory conditions - i.e. mediated by epithelial cells - and was dependent upon an as yet unidentified soluble mediator. When tested in vivo, L. plantarum and LGG exacerbated the development of DSS-induced colitis and caused the death of treated mice, while, conversely L. paracasei was protective.

Conclusions: We describe a new property of probiotics to either directly or indirectly inhibit DC activation by inflammatory bacteria. Moreover, some immunostimulatory probiotics not only failed to protect against colitis, they actually amplified the disease progression. In conclusion, caution must be exercised when choosing a probiotic strain to treat IBD.

Figure 1. DCs are phenotypically similarly activated by Lactobacilli.

DCs were incubated or not with the reported live bacterial strains for 1 h in medium without antibiotics, washed and incubated for 23 h in medium with antibiotics. Cells were stained for HLA-DR and CD80 expression and analyzed by FACS. A. % of cells highly positive for the marker is reported. B. Mean fluorescence intensity (MFI) expression of markers is reported. C. Viability of the cells after 24 h incubation with bacteria. Cells were stained with propidium iodide (PI) and annexin V and analyzed by FACS. Cells double negative for both markers are considered viable cells. Error bars: standard deviations on values obtained on 4 different donors.

Figure 2. DCs incubated with different bacterial strains produce a distinct cytokine profile.

A. DCs were incubated or not with the reported live bacterial strains for 1 h in medium without antibiotics, washed and incubated for 23 h in medium with antibiotics. Culture supernatants were collected and tested for cytokine contents by ELISA. Each symbol represents a different DC donor. Red lines represent mean values. *, p<0.05; **, p<0.01. S. typhim.: S. typhimurium; L. plant.: L. plantarum; L. parac.: L. paracasei. B. To analyze the kinetic of cytokine production, DCs were incubated or not with the reported live bacterial strains for 1 h in medium without antibiotics, washed and incubated for 3–5 h in medium with antibiotics. Culture supernatants were collected and tested for cytokine release by ELISA. Error bars: standard deviations on values obtained on 4 different donors.

Figure 3. Lactobacilli-treated DCs have different ability to induce T cell proliferation and cytokine production.

A. T cell proliferation: DCs were incubated or not with the reported live bacterial strains for 1 h in medium without antibiotics, washed and incubated for 23 h in medium with antibiotics. Bacteria-treated DCs were washed and incubated with naïve CD4⁺CD45RA⁺ cells for 3 days, followed by a 16-hours pulse with 1 µCi [³H] thymidine (Amersham, Milan). ³H-thymidine incorporation is shown. Each symbol represents a different DC donor. Red lines represent mean values. *, p<0.05. S. typh: S. typhimurium; L. plan: L. plantarum; L. par: L. paracasei. B. Cytokine release: Bacteria-treated DCs were incubated with naïve CD4⁺CD45RA⁺ cells for 5 days (Ratio 1∶10 DC∶T cells). Cell culture supernatants were collected and cytokines measured by ELISA or CBA Flex set. Error bars: standard deviation on values obtained on 3 different donors. *, p<0.05; **, p<0.01.

Figure 4. L. paracasei induces increased levels of TGF-β and TSLP.

Caco-2 cells were grown as monolayers in the upper chamber of a transwell filter and incubated with live bacteria (5×10⁷ CFU/TW) upon the apical surface (top chamber). One hour after incubation, bacteria were washed out and medium was changed with one containing antibiotics. Culture supernatants were collected 3, 5, 23 and 47 hours later from the bottom chamber and tested for cytokine release. Error bars: standard deviations on values obtained in 2 different experiments. *, p<0.05; **, p<0.01.

Figure 5. L. paracasei inhibits the release of inflammatory cytokines both directly and indirectly on DCs.

Each treatment is schematically reported below the graphs. Three situations were analyzed (a, b, c). a. DCs were incubated or not with the reported live bacterial strains either separately (SL, Salmonella; LP, L. paracasei) or together (LP+SL) for 1 h in medium without antibiotics, washed and incubated for 23 h in medium with antibiotics. b. Caco-2 cells were grown as monolayers in the upper chamber of a transwell filter. 24 h from achievement of a TER of 300 Ohm•cm² supernatants (sn Caco) were collected from the bottom chamber and used to pre-treat DCs for 24 h before bacterial incubation as in a. c. Caco-2 cells were grown as monolayers in the upper chamber of a transwell filter and incubated with L. paracasei (5×10⁷ CFU/TW) upon the apical surface (top chamber). One hour after incubation, bacteria were washed out and medium was changed with one containing antibiotics. Culture supernatants (sn caco LP) were collected 24 hours later from the bottom chamber, filtered and used to pre-treat DCs for 24 h before bacterial incubation as in a. 24 h after bacterial treatment of DCs cell culture supernatants were collected and cytokines analyzed by ELISA. Error bars: standard deviations on values obtained on 3 different donors. *, p<0.05; **, p<0.01

Figure 6. L. paracasei culture supernatant is responsible for the anti-inflamamtory activity of the bacterium.

DCs were incubated or not with the reported live bacterial strains either separately (SL, Salmonella; LP, L. paracasei) or together (LP+SL) or in the presence of culture supernatants of L. paracasei corresponding to the exponential growth of the same amount of CFU of bacteria used to treat the DCs. The culture supernatant (sn LP) was used either undiluted or diluted 1/5, 1/10, 1/100 that correspond to nearly 7%, 1,4%, 0,7%, and 0,07% volume/volume of tissue culture medium, respectively. Cells were incubated with the different treatments for 1 h in medium without antibiotics, washed and incubated for 23 h in medium with antibiotics. Cytokine release was analyzed by ELISA. Error bars: standard deviations on values obtained on 3 different donors. *, p<0.05; **, p<0.01

Figure 7. L. paracasei inhibits the ability of DCs to activate T cells.

Three situations were analyzed (a, b, c) as in fig. 6 DCs were incubated or not with the reported live bacterial strains either separately (SL, Salmonella; LP, L. paracasei) or together (LP+SL) for 1 h in medium without antibiotics, washed and incubated for 23 h in medium with antibiotics. a. Caco-2 cells were grown as monolayers in the upper chamber of a transwell filter. 24 h from achievement of a TER of 300 Ohm•cm² supernatants (sn Caco) were collected from the bottom chamber and used to pre-treat DCs for 24 h before bacterial incubation as in a. b. Caco-2 cells were grown as monolayers in the upper chamber of a transwell filter and incubated with L. paracasei (5×10⁷ CFU/TW) upon the apical surface (top chamber). One hour after incubation, bacteria were washed out and medium was changed with one containing antibiotics. Culture supernatants (sn caco LP) were collected 24 hours later from the bottom chamber, filtered and used to pre-treat DCs for 24 h before bacterial incubation as in a. Cells were then washed and incubated with naïve CD4+CD45RA+ cells for 5 days (Ratio 1∶10 DC∶T cells). Cell culture supernatants were collected and cytokines measured by ELISA or CBA Flex set. Error bars: standard deviations on values obtained on 3 different donors. *, p<0.05; **, p<0.01.

Figure 8. L. paracasei protects against DSS colitis.

Mice (n = 6) were administered intra gastrically (i.g.) once a day for 7 days with 200 µl PBS containing 10¹⁰ CFUs of bacteria or plain PBS as a control. Mice were then fed with 2% DSS dissolved in the drinking water for 5 days without bacteria, followed by 7 days of plain water and assessed over time for colitis development. A. Body weight was measured at baseline and every day for the duration of the experiment. Weight change was calculated as percentage change in weight compared with baseline. L. plant.: L. plantarum; L. parac.: L. paracasei. Asterisks refer to statistical analysis of the groups LGG or L. plantarum versus DSS PBS positive control group. *, p<0.05; **, p<0.01;  = , dead animals. B. Disease activity index (DAI) was measured as reported in Materials and Methods. DAI at 4, 5, 7 days is shown per each group. The scatterplot shows a line at the mean of each group with error bars. Dashed lines identify one standard deviation above and below the group means. *, p<0.05; **, p<0.01.