Pre-treatment with Lactobacillus plantarum prevents severe pathogenesis in mice infected with Leptospira interrogans and may be associated with recruitment of myeloid cells

Abstract

Recent estimates on global morbidity and mortality caused by Leptospirosis point to one million cases and almost 60,000 deaths a year worldwide, especially in resource poor countries. We analyzed how a commensal probiotic immunomodulator, Lactobacillus plantarum, affects Leptospira interrogans pathogenesis in a murine model of sub-lethal leptospirosis. We found that repeated oral pre-treatment of mice with live L. plantarum restored body weight to normal levels in mice infected with L. interrogans. Pre-treatment did not prevent L. interrogans access to the kidney but it affected the inflammatory response and it reduced histopathological signs of disease. Analysis of the immune cell profiles in lymphoid tissues of mice pre-treated with L. plantarum showed increased numbers of B cells as well as naïve and memory CD4+ helper T cell populations in uninfected mice that shifted towards increased numbers of effector CD4+ helper T in infected mice. CD8+ cytotoxic T cell profiles in pre-treated uninfected and infected mice mirrored the switch observed for CD4+ except that CD8+ memory T cells were not affected. In addition, pre-treatment led to increased populations of monocytes in lymphoid tissues of uninfected mice and to increased populations of macrophages in the same tissues of infected mice. Immunohistochemistry of kidney sections of pre-treated infected mice showed an enrichment of neutrophils and macrophages and a reduction of total leucocytes and T cells. Our results suggest that complex myeloid and T cell responses orchestrate the deployment of monocytes and other cells from lymphoid tissue and the recruitment of neutrophils and macrophages to the kidney, and that, the presence of these cells in the target organ may be associated with reductions in pathogenesis observed in infected mice treated with L. plantarum.

Fig 1. Body weight and Leptospira load in body fluids after infection of C3H-HeJ mice pre-treated with L. plantarum.

A. Treatment/infection schedule: groups of 5 week old mice received repeated oral treatments of L. plantarum (o, n = 30) or PBS and were infected intraperitoneally with L. interrogans serovar Copenhageni on week 11; uninfected groups of mice were kept as controls. B. Body weight measurements (g) were recorded for 2 weeks post-infection and normalized to 100% on the day of infection (d0). C. Blood and D. urine were collected for determination of the number of Leptospira 16s rRNA per ul of sample by qPCR. Statistics: body weight differences between all groups, Two-way ANOVA, p<0.0001; and paired t-test between infected groups * p<0.0001. Number of mice: Lp and Lp/Lepto n = 4 per group, PBS and PBS/Lepto n = 5 per group. Data represents one of two experiments.

Fig 2. Histopathology, Leptospira load and measurement of inflammatory markers in kidney.

A. H&E staining (20X) and silver stain (40X) of kidney sections of treated uninfected mice and of mice infected with Leptospira after treatment. B. Histopathology was empirically quantified by scoring interstitial nephritis blindly, and glomeruli were scored by measuring size (um) in 5 fields per sample and averaging per group. C. Numbers of Leptospira in kidney tissue were quantified by qPCR of 16s rRNA; D. qPCR of pro-inflammatory transcripts (CxCL1, CxCL2, CCL5, TNFα and IFNγ) in kidney. Statistics by unpaired Mann-Whitney U test and by unpaired t test with Welch’s correction between infected groups, B, Interstitial nephritis p = 0.0079, Glomeruli size p<0.0001; D, CxCL1 p = 0.0308, CxCL2 p = 0.0545, CCL5 p = 0.1132, TNFα p = 0.7592, IFNγ p = 0.5788; ** and * p<0.05. Number of mice: Lp and Lp/Lepto n = 4 per group, PBS and PBS/Lepto n = 5 per group. Data represents one of two experiments.

Fig 3. Analysis of renal fibrosis.

A. Interstitial collagen deposition (fibrosis, black arrows, 400x) was evaluated by Masson’s trichrome staining of kidney sections from infected mice pre-treated with PBS and with L. plantarum. B. Digital quantification of fibrosis as determined by the % area (pixels) where blue staining exceeds a threshold. The slides were processed and stained in parallel and images were taken using the same illumination. C. qPCR quantification of fibrosis markers (iNOS and ColA1) in kidney tissue. Statistics between groups by Two-Way ANOVA % fibrosis p = 0.0067 and by unpaired t test with Welch’s correction, iNOS p = 0.1131, ColA1 p = 0.0342; * p<0.05. Number of mice, Lp/Lepto n = 4, PBS/Lepto n = 5. Data represents one of two experiments.

Fig 4. Antibody response and chemokine immunomediators in serum.

A. Total concentration of IgM, IgG, IgG1, IgG2a and IgG3 antibodies, and B. CxCL1, CxCL2, CCL5, TNFα and IFNγ, which were quantified by ELISA two weeks post-infection. Statistics: unpaired t test with Welch’s correction between infected groups, Total IgM, p = 0.0457; Total IgG p = 0.0133; Total IgG1 p<0.0001; Total IgG2a p = 0.0083; Total IgG3 p = 0.0085; CxCL1 p = 0.0349; CxCL2 p = 0.0749; CCL5 p = 0.0141; TNFα p = 0.2929; IFNγ p = 0.7142; * p<0.05. Number of mice, Lp and Lp/Lepto n = 4 per group, PBS and PBS/Lepto n = 5 per group. Data shown in A is representative of one of two experiments and data shown in B represents one experiment.

Fig 5. Percentage of B cells, T cells, CD4+ helper T cells, CD8+ cytotoxic T cell and double negative T cell (DN) populations in spleen and lymph nodes.

Flow cytometric analysis of immune cells isolated from spleen and peripheral lymph nodes from uninfected and infected mice previously treated with L. plantarum. Cells were labelled with anti-CD19, anti-CD3, anti-CD4 and anti-CD8 lineage surface markers. Statistics: Multiple t tests: A, Spleen CD19+B cell p = 0.0142, CD4+T cell p = 0.0156; B, Spleen CD4+T cell p = 0.0029, CD8+ T cell p = 0.0002, DN T cell p = 0.0205; C, Lymph Nodes CD19+B cell p = 0.0061; D, Lymph Nodes CD4+ T cell p = 0.0003. A, B, C, D * p<0.05. N = 3 to 5 mice per group x 4 groups (PBS, Lp, PBS/Lepto, Lp/Lepto). Data represents one of two experiments.

Fig 6. Percentage of naive, effector and memory CD4+ and CD8+ T cell subsets.

Flow cytometric analysis of immune cells isolated from spleen and peripheral lymph nodes from uninfected and infected mice previously treated with L. plantarum. Cells were labelled with anti-CD3, anti-CD4, anti-CD8, anti-CD44 and anti-CD62L lineage surface markers. Statistics by Multiple t tests: A, CD4+ Naive p = 0.0168, Early Effector p = 0.0001, Effector p = 0.0004, Memory p<0.0001; CD8+ Naive p<0.0001, Early Effector p = 0.0106, Effector p<0.0001, Memory p = 0.8073; B, CD4+ Naive p<0.0001, Early Effector p<0.0001, Effector p = 0.0494, Memory p = 0.0002; CD8+ Naive p<0.0001, Early Effector p = 0.0008, Effector p<0.0001, Memory p = 0.2038; C, CD4+ Naive p = 0.0003, Early Effector p = 0.0001, Effector p<0.0001, Memory p<0.0001; CD8+ Naive p<0.0001, Early Effector p = 0.0002, Effector p<0.0001, Memory p = 0.1248; D, CD4+ Naive p<0.0001, Early Effector p<0.0001, Effector p = 0.1018, Memory p<0.0001; CD8+ Naive p<0.0001 Early Effector p = 0.0017, Effector p<0.0001, Memory p = 0.0997; A, B, C, D * p<0.05. N = 3 to 5 mice per group x 4 groups (PBS, Lp, PBS/Lepto, Lp/Lepto). Data represents one of two experiments.

Fig 7. Percentage of neutrophils, monocytes, macrophages and dendritic cell populations in spleen and lymph nodes.

Flow cytometric analysis of myeloid cells isolated from spleen and peripheral lymph nodes from uninfected and infected mice previously treated with L. plantarum. Cells were labeled with anti-MHC II, anti-CD11b, anti-CD11c and anti-Ly6C lineage surface markers. Statistics by Multiple t tests: A, neutrophils p = 0.0015, monocytes p = 0.0900, macrophages p<0.0001, dendritic p = 0.0219; B, neutrophils p = 0.6008, monocytes p = 0.0122, macrophages p = 0.0007, dendritic p = 0.4653; C, neutrophils p = 0.0003, monocytes p = 0.0077, macrophages p = 0.3968, dendritic p = 0.3892; D, neutrophils p = 0.0002, monocytes p = 0.0114, macrophages p = 0.0012, dendritic p = 0.0232; A, B, C, D * p<0.05. N = 3 to 5 mice per group x 4 groups (PBS, Lp, PBS/Lepto, Lp/Lepto). Data represents one of two experiments.

Fig 8. Oral treatment with L plantarum leads to recruitment of myeloid cells in kidney.

Immunostaining of kidney sections from groups of treated mice, in the presence or absence of Leptospira infection using various leukocyte markers. We determined the number of CD45+, NIMP-R14+, F4/80+ and CD3+ T cells per millimeter squared of kidney. Data are mean SEM of 4–5 mice per group. Scale bar represents 200X for (A) and 400x for (B), (C) and (D). Statistics by two-tailed paired t-test *p < 0.05, ** p < 0.01, *** p < 0.001. Data is representative of one of two experiments.