Immunoregulatory actions of epithelial cell PPAR gamma at the colonic mucosa of mice with experimental inflammatory bowel disease

Abstract

Background: Peroxisome proliferator-activated receptors are nuclear receptors highly expressed in intestinal epithelial cells (IEC) and immune cells within the gut mucosa and are implicated in modulating inflammation and immune responses. The objective of this study was to investigate the effect of targeted deletion of PPAR gamma in IEC on progression of experimental inflammatory bowel disease (IBD).

Methodology/principal findings: In the first phase, PPAR gamma flfl; Villin Cre- (VC-) and PPAR gamma flfl; Villin Cre+ (VC+) mice in a mixed FVB/C57BL/6 background were challenged with 2.5% dextran sodium sulfate (DSS) in drinking water for 0, 2, or 7 days. VC+ mice express a transgenic recombinase under the control of the Villin-Cre promoter that causes an IEC-specific deletion of PPAR gamma. In the second phase, we generated VC- and VC+ mice in a C57BL/6 background that were challenged with 2.5% DSS. Mice were scored on disease severity both clinically and histopathologically. Flow cytometry was used to phenotypically characterize lymphocyte and macrophage populations in blood, spleen and mesenteric lymph nodes. Global gene expression analysis was profiled using Affymetrix microarrays. The IEC-specific deficiency of PPAR gamma in mice with a mixed background worsened colonic inflammatory lesions, but had no effect on disease activity (DAI) or weight loss. In contrast, the IEC-specific PPAR gamma null mice in C57BL/6 background exhibited more severe inflammatory lesions, DAI and weight loss in comparison to their littermates expressing PPAR gamma in IEC. Global gene expression profiling revealed significantly down-regulated expression of lysosomal pathway genes and flow cytometry results demonstrated suppressed production of IL-10 by CD4+ T cells in mesenteric lymph nodes (MLN) of IEC-specific PPAR gamma null mice.

Conclusions/significance: Our results demonstrate that adequate expression of PPAR gamma in IEC is required for the regulation of mucosal immune responses and prevention of experimental IBD, possibly by modulation of lysosomal and antigen presentation pathways.

Figure 1. Genotyping of PPAR γ flfl; Villin Cre+ (VC+) and Villin Cre- (VC-) control mice.

Conditional deletion of the PPAR γ gene via Villin Cre-mediated recombination was examined in mouse intestine by PCR analysis. The floxed (fl) allele at 275 bp and the null allele at 400 bp. (A) Left to right: depicts fl/fl in homogenized whole duodenum, jejunum, ileum, cecum and colon without recombination (VC-) (lanes 1, 3, 5, 7 and 9) or with recombination (VC+) (lanes 2, 4, 6, 8 and 10). (B) Left to right: depicts fl/fl in epithelial cells isolated from duodenum, jejunum, ileum, cecum and colon without recombination (VC-) (lanes 1, 3, 5, 7 and 9) or with recombination (VC+) (lanes 2, 4, 6, 8 and 10).

Figure 2. Effect of epithelial cell-specific PPAR γ deletion on disease severity.

PPAR γ flfl; Villin Cre+ (VC+) or PPAR γ flfl; Villin Cre- (VC-) mice in a mixed FVB/C57BL/6J (FVB) or C57BL/6J (B6) background were treated with 2.5% dextran sodium sulfate (DSS) or water (no DSS) for 7 days. The disease activity index (DAI), a composite score reflecting clinical signs of the disease (i.e. perianal soiling, rectal bleeding, diarrhea, and piloerection) was assessed daily (A and C) and the average daily loss in body weights (B and D) throughout the 7-day DSS challenge was calculated. Data are represented as mean ± standard error. Points with an asterisk are significantly different (P<0.05).

Figure 3. Effect of intestinal epithelial cell-specific PPAR γ deletion on colon histopathology and inflammation.

PPAR γ flfl; Villin Cre+ (VC+) or PPAR γ flfl; Villin Cre- (VC-) mice with a mixed FVB/C57BL6/J (FVB) or C57BL6/J (B6) background were treated with 2.5% dextran sodium sulfate (DSS) or water (no DSS) for 7 days. Representative photomicrographs of colonic samples from VC- (A) and VC+ (B) FVB mice with DSS colitis (Original magnification, 40×). Colonic specimens from FVB mice underwent blinded histological examination and were scored (1–4) on leukocyte infiltration (C), and mucosal wall thickening (D), and epithelial erosion (E) on day 7 of the challenge. In B6 mice spleen (F), colon (G), and mesenteric lymph nodes MLN (H) were scored based on macroscopic signs of inflammation on days 2 and 7. Data are represented as mean ± standard error. Points with an asterisk are significantly different (P<0.05).

Figure 4. Effect of epithelial cell-specific PPAR γ deletion on immune cell subsets in blood and spleen in FVB/C57BL/6J mice.

Blood (A–D) and spleen (E–H) from PPAR γ flfl; Villin Cre+ (VC+) or PPAR γ flfl; Villin Cre- (VC-) mice with a mixed FVB/C57BL/6J background (FVB) were immunophenotyped. Data were collected on days 0, 2, and 7 of DSS challenge and were analyzed with FACS Diva software. Data are represented as mean ± standard error. There were no statically significant differences between groups (P<0.05).

Figure 5. Effect of epithelial cell-specific PPAR γ deletion on immune cell subsets in mesenteric lymph nodes in FVB/C57BL/6J mice.

Mesenteric lymph nodes (MLN) from PPAR γ flfl; Villin Cre+ (VC+) or PPAR γ flfl; Villin Cre- (VC-) mice with a mixed FVB/C57BL/6J background (FVB) were immunophenotyped to identify immune cell subsets by flow cytometry. Data were collected on days 0, 2, and 7 of DSS challenge and were analyzed with FACS Diva software. Data are represented as mean ± standard error. Points with an asterisk are significantly different (P<0.05).

Figure 6. Effect of epithelial cell-specific PPAR γ deletion on immune cell subsets in spleen and mesenteric lymph nodes in C57BL/6J mice.

Spleen (A–F) and mesenteric lymph nodes (MLN) (G–H) from PPAR γ flfl; Villin Cre+ (VC+) or PPAR γ flfl; Villin Cre- (VC-) mice with a C57BL/6J background (B6) were immunophenotyped to identify immune cell subsets through flow cytometry. Data were collected on days 0, 2, and 7 of DSS challenge and were analyzed with FACS Diva software. Data are represented as mean ± standard error. Points with an asterisk are significantly different (P<0.05).

Figure 7. Effect of epithelial cell-specific PPAR γ deletion on target gene expression in colonic mucosa.

Expression of PPAR γ targets Plin2 (ADRP), Fabp2 (FABP); and pro-inflammatory genes TNF-α, IL-6 and IL-1β in the colonic mucosa in VC- and VC+ mice at 0, 2 and 7 days after dextran sodium sulfate (DSS) challenge. Compared to VC- mice, VC+ mice show reduced expression of Plin2 (ADRP) at day 7 of DSS challenge and of Fabp2 (FABP) at earlier time points (days 0 and 2). Expression of pro-inflammatory genes TNF-α, IL-6 and IL-1β are increased in VC+ mice compared to VC- mice after DSS challenge. Statistical significance is indicated by asterisks (P<0.05).

Figure 8. Effect of epithelial cell-specific PPAR γ deletion on target gene expression in colonic mucosa in C57BL/6J mice.

Expression of IL-6 (A), IL-1β (B) and IRAK-1 (C) was assessed in the colonic mucosa in PPAR γ flfl; Villin Cre+ (VC+) or PPAR γ flfl; Villin Cre- (VC-) in mice with a C57BL6/J (B6) background. Expression levels were assessed at days 0, 2 and 7 of DSS challenge and normalized relative to the housekeeping gene β-actin. Data are represented as mean ± standard error. There were no statistically significant differences between groups (P<0.05).