Vitamin D deficiency in mice impairs colonic antibacterial activity and predisposes to colitis

Abstract

Vitamin D insufficiency is a global health issue. Although classically associated with rickets, low vitamin D levels have also been linked to aberrant immune function and associated health problems such as inflammatory bowel disease (IBD). To test the hypothesis that impaired vitamin D status predisposes to IBD, 8-wk-old C57BL/6 mice were raised from weaning on vitamin D-deficient or vitamin D-sufficient diets and then treated with dextran sodium sulphate (DSS) to induce colitis. Vitamin D-deficient mice showed decreased serum levels of precursor 25-hydroxyvitamin D(3) (2.5 +/- 0.1 vs. 24.4 +/- 1.8 ng/ml) and active 1,25-dihydroxyvitamin D(3) (28.8 +/- 3.1 vs. 45.6 +/- 4.2 pg/ml), greater DSS-induced weight loss (9 vs. 5%), increased colitis (4.71 +/- 0.85 vs. 1.57 +/- 0.18), and splenomegaly relative to mice on vitamin D-sufficient chow. DNA array analysis of colon tissue (n = 4 mice) identified 27 genes consistently (P < 0.05) up-regulated or down-regulated more than 2-fold in vitamin D-deficient vs. vitamin D-sufficient mice, in the absence of DSS-induced colitis. This included angiogenin-4, an antimicrobial protein involved in host containment of enteric bacteria. Immunohistochemistry confirmed that colonic angiogenin-4 protein was significantly decreased in vitamin D-deficient mice even in the absence of colitis. Moreover, the same animals showed elevated levels (50-fold) of bacteria in colonic tissue. These data show for the first time that simple vitamin D deficiency predisposes mice to colitis via dysregulated colonic antimicrobial activity and impaired homeostasis of enteric bacteria. This may be a pivotal mechanism linking vitamin D status with IBD in humans.

Figure 1

Serum vitamin D metabolites and calcium in mice raised on vitamin D-sufficient and vitamin D-deficient diets. A, Serum concentrations of 25OHD₃ (25D₃) (ng/ml) in C57BL/6 mice raised on vitamin D-sufficient (normal) or vitamin D-deficient (D-deficient) diets from weaning (wk 4) until wk 9. At wk 8, mice were exposed to either regular tap water (control, C) or water with 2.5% dextran sodium sulfate (DSS) for 1 wk. B, Serum concentrations of 1,25(OH)₂D₃ (1,25D₃) (pg/ml). Panel C, Serum concentrations of calcium (mg/dl). Values are means ± sem (n = 8). ***, Statistically different from corresponding vitamin D-sufficient mice (normal diet), P < 0.001; **, P < 0.01.

Figure 2

DSS-induced weight loss in mice raised on vitamin D-sufficient and vitamin D-deficient diets. Eight-week-old mice raised on vitamin D-sufficient (normal) or vitamin D-deficient (D-deficient) diets were exposed to either regular tap water (C) or water with 2.5% DSS. Changes in body weight for DSS-treated mice relative to control (tap water) equivalents for mice raised on vitamin D-sufficient (normal,filled circles) and vitamin D-deficient (D-deficient, open circles) diets (% change).

Figure 3

DSS-induced colitis in mice raised on vitamin D-sufficient and vitamin D-deficient diets. Eight-week-old mice raised on vitamin D-sufficient (normal) or vitamin D-deficient (D-deficient) diets were exposed to either regular tap water (C) or water + DSS. A, Animal health as determined by clinical scores. B, Histological scoring of colitis using colonic tissue sections. C–F, Example H&E staining for vitamin D-sufficient mice receiving tap water (control) (C) and vitamin D-sufficient mice receiving water with 2.5% DSS (control + DSS) (D). Arrow indicates immune infiltrate, double-lined arrow indicates loss of colonic epithelial morphology. E, Vitamin D-deficient mice receiving tap water (D-deficient). F, Vitamin D-deficient mice receiving water with 2.5% DSS (D-deficient + DSS). Values are means ± sem (n = 8). ***, Statistically different from corresponding vitamin D-sufficient mice (normal diet), P < 0.001; **, P < 0.01.

Figure 4

Dysregulation of splenic cytokines in mice raised on vitamin D-sufficient and vitamin D-deficient diets. Wild-type C57BL/6 mice were raised on either a normal diet or vitamin D-deficient chow from weaning until 9 wk of age when they were treated either with regular water or water + dextran sodium sulfate (DSS). Data show changes in: 1) spleen weight, 2) mRNA expression for IL-1α, 3) IL-17, 4) Cyp27b1, 5) IL-10, and 6) IFNγ. Values are means ± sem (n = 8). ***, Statistically different from corresponding vitamin D-sufficient mice (normal diet), ***, P < 0.001; **, P < 0.01; *, P < 0.05.

Figure 5

Severe IBD in vitamin D deficient mice is associated with significantly altered T-cell and B-cell proportions. Splenocytes from mice raised on either a normal or vitamin D-deficient diet treated with or without dextran sodium sulfate (DSS) were analyzed by flow cytometry for expression of T-cell (CD3), B-cell (CD19), and T-cell subset (CD4 and CD8) makers. Data are shown as the percentage of cells positive for each marker.

Figure 6

Validation of differentially expressed genes in kidney, colon, and spleen tissue from vitamin D-sufficient vs. vitamin D-deficient mice. Real-time RT-PCR analysis of mRNA expression for genes showing differential patterns of colonic expression in vitamin D-deficient mice (open bars) relative to mice on a normal diet (closed bars). Genes were selected from those originally identified by DNA array analysis as showing more than 2-fold change in expression (see genes highlighted in bold type in Table 1). PLAGL-1, Pleiomorphic adenoma gene-like 1; ATF3, activating transcription factor 3; NT5E, 5′ nucleotidase, ecto; FK506BP, FK506 binding protein 11. Values are represented as mean ± sem (n = 8) and significantly different as follows: ***, statistically different from corresponding vitamin D-sufficient mice (normal diet), P < 0.001; *, P < 0.05.

Figure 7

The effects of vitamin D deficiency and DSS treatment on colonic expression of Ang4 protein and colonic bacterial load. A–D, Immunohistochemical analysis of colonic Ang4 protein expression in vitamin D-sufficient or vitamin D-deficient mice. A, Vitamin D-sufficient mice receiving tap water (control). B, Vitamin D-sufficient mice receiving water with 2.5% DSS (+DSS). C, Vitamin D-deficient mice receiving tap water (d-def). D, Vitamin D-deficient mice receiving water with 2.5% DSS (D-def + DSS). E, quantification of Ang4 expression (% expression relative to control samples). F, Bacterial load in vitamin D-sufficient (control) and vitamin D-deficient (D-def) colon tissue quantified by real-time PCR for 16S rDNA. All quantification using n = 6 separate colons. Values are represented as mean ± sem. ***, Statistically different from vitamin D-sufficient control mice, P < 0.001; **, P < 0.01; *, P < 0.05.