Mast cell interleukin-10 drives localized tolerance in chronic bladder infection

Abstract

The lower urinary tract's virtually inevitable exposure to external microbial pathogens warrants efficient tissue-specialized defenses to maintain sterility. The observation that the bladder can become chronically infected in combination with clinical observations that antibody responses after bladder infections are not detectable suggest defects in the formation of adaptive immunity and immunological memory. We have identified a broadly immunosuppressive transcriptional program specific to the bladder, but not the kidney, during infection of the urinary tract that is dependent on tissue-resident mast cells (MCs). This involves localized production of interleukin-10 and results in suppressed humoral and cell-mediated responses and bacterial persistence. Therefore, in addition to the previously described role of MCs orchestrating the early innate immunity during bladder infection, they subsequently play a tissue-specific immunosuppressive role. These findings may explain the prevalent recurrence of bladder infections and suggest the bladder as a site exhibiting an intrinsic degree of MC-maintained immune privilege.

Figure 1. see also Figure S1. Bladders fail to eradicate persistent E. coli

(A) Mice were infected either in the bladder or with concurrent bladder and kidney infections and the organs were homogenized and plated on bacterial agar for growth, to quantitate the CFUs of bacteria per organ. While the kidneys are able to completely clear infection within 3d, the UPEC in the bladder persist for over 4 weeks. All data are representative of three independent experiments, n≥3 per time point; error bars represent ±SEM. (B) E. coli (red) remain within the intermediate (green) and superficial bladder epithelium (blue) 2 weeks p.i. (basement membrane denoted by white line). Scale bar=50μm.

Figure 2. see also Figure S2. A mouse model distinguishing between cystitis and pyelonephritis

(A) Similar bacterial CFUs are obtained from bladders during both methods of infection; bacteria are only cultured from kidneys of pyelonephritis-infected mice. All data are representative of 2 independent experiments, n=3-4 per time point; error bars represent ±SEM. (B) Only mice with acute pyelonephritis show significant serum anti-UPEC IgG geometric mean titers (GMT) following infection. Dotted line denotes threshold of detection. *p<0.05; **p<0.01. Error bars represent the 95% confidence level; n=4-6.

Figure 3. Bladder-contained infections fail to evoke significant antibody responses unless primary immunity was already established

(A) Mice were infected transurethrally with 1×10⁸ CFU to induce cystitis-only or pyelonephritis (first arrow) and serum GMT anti-UPEC antibodies were measured (x-axis shows days p.i.). On day 21 (second arrow) cystitis-only was induced in both groups. *p<0.05; **p<0.01; ***p<0.001, compared to naïve. Error bars represent the 95% confidence level with n=4-7. Dotted line denotes threshold of detection. (B) Naïve mice that were adoptively transferred splenocytes from mice that previously had acute pyelonephritis show enhanced bacterial clearance 5d post-cystitis challenge, compared to control naïve mice or mice adoptively transferred splenocytes from mice that previously had cystitis only. *p<0.05. Error bars represent ±SEM with n=4-6.

Figure 4. see also Figure S3. Enhanced Il10 transcription typifies the anti-inflammatory program upon infection, limits bacterial clearance and serum antibodies

(A-D) Cytokine transcriptional profile in the bladder and kidney at various times p.i. determined by real time PCR for Il6, KC, TNF-α and Il10. (E-F) In addition to Il10 upregulation, significantly increased levels of other major anti-inflammatory mediators – TGF-β and SOCS3 – were also seen only in the bladder and not detected in the kidney. * p < 0.05, ** p ≤ 0.01; error bars represent ±SEM. All data are representative of two or more independent experiments, n ≥ 3. Il10^-/- mice have better bacterial clearance in the bladder as compared to WT, (G) 5d and (H) 6 weeks after infection. * p<0.05, *** p<0.001; n=10-12 mice per group (5d) and n=5-6 mice per group (6 weeks) (I) Significant serum GMT anti-UPEC IgM antibodies are seen in both cystitis-only and mice with pyelonephritis in Il10^-/- mice at comparable titers to mice with concurrent cystitis and pylenophritis infection at 7d p.i. There are no significant differences in antibody titers between WT mice with pyelonephritis and Il10^-/- infected mice. UPEC-specfic IgM was not detected in WT mice with cystitis alone. Dotted line denotes threshold of detection. ND=not detected, *** p<0.001. Error bars represent the 95% confidence level with n=5-7.

Figure 5. see also Figure S4. Suppression of DC activation in ILN during cystitis

(A) Numbers of DCs draining to the ILN and RLN upon cystitis or pyelonephritis were determined by flow cytometry by staining for MHC-II and CD11c, 24h after infection with 1×10⁸ UPEC. Data does not differ significantly. (B) Increased numbers of activated DCs (MHCII⁺ CD11c⁺ CD86⁺) were present in the RLNs of WT mice and in ILN of Il10^-/- mice, similarly infected. WT mice have fewer (C) T_H cells (CD4⁺) and (D) germinal center B cells (CD19⁺GL7⁺) in LNs 7d p.i. * p<0.05, ** p<0.01; error bars represent ±SEM. All data are representative of 2 or more experiments with n=3-7. Representative flow cytometry plots are shown in Figure S4.

Figure 6. see also Figure S5. MCs in the bladder promote Il10 expression

(A) A whole mount picture of the luminal side of a murine bladder, showing the abundance of MCs (red, avidin-TRITC stained) under the epithelium. (B) Wsh mice, even at basal levels, have lower expression of Il10 in the bladder as compared to WT bladders, determined by real time PCR. (C) With infection, at 6h after bacterial instillation, Il10 expression increases in bladders of WT but not Wsh mice. (D) Il10 is expressed at higher basal levels in the kidneys of Wsh compared to WT mice, as determined by real time PCR. (E) However, upon infection, the high Il10 levels of expression in Wsh kidneys drop dramatically to be not significantly different from those of WT kidneys at 6h. (F) Image shows a bladder section stained for MCs (MMCP6, green), and for IL-10 (red) 24h after UPEC infection. (G) Intracellular staining for IL-10 reveals IL-10⁺ cells in the bladder 24h p.i. Though Treg cells (foxp3⁺CD4⁺) and macrophages (MHC-II⁺CD11b⁺) contribute to the IL-10 in the bladder, the majority of cells that are IL-10⁺ are MCs (ckit⁺). (H) Rescue of increased Il10 expression is seen in infected bladders of Wsh mice reconstituted with WT BMMCs but not Il10^-/- BMMCs at 24h p.i. (I) Wsh mice reconstituted with WT BMMCs have significantly greater activated DCs draining to the RLNs, compared to ILNs 24h p.i. Reconstitution with BMMC from Il10^-/- mice results in comparable numbers of activated DCs in both LNs. *p<0.05, **p ≤0.01, ***p≤0.001; data are representative of ≥3 independent experiments with n≥3. Error bars represent ±SEM. (J) Mcpt5-Cre Il10Fl mice showed the same phenotype as Il10^-/- mice and Il10^-/- BMMC-reconstituted Wsh mice: significantly greater numbers of mature DCs present in both ILN and RLN as compared to WT ILN 24h p.i. (K) Serum UPEC-specific IgG was detected in Mcpt5-Cre Il10Fl mice 7d p.i. Error bars represent the 95% confidence level with n=5.

Figure 7. see also Figure S6. MC-derived IL-10 promotes bacterial persistence in bladder

(A) Wsh mice reconstituted with Il10^-/- BMMCs had significantly lower persistent bacteria in bladders compared to both WT mice and Wsh mice reconstituted with WT BMMCs. Bladders were harvested 2 weeks p.i. (1×10⁸ UPEC) for quantification of bacterial persistence by CFU. (B) This phenotype was also observed in Il10^-/- and Mcpt5-Cre Il10Fl mice. The bladders of WT mice retained significantly greater numbers of intracellular bacteria 6 weeks p.i. (C) Increased numbers of MCs are present in the bladder at 24h, as determined by quantification from tissue sections. (D) This increase of MCs persisted in the bladder 6 weeks p.i. *p<0.05, **p≤0.01; n=3. Error bars represent ±SEM.