Vitamin D receptor negatively regulates bacterial-stimulated NF-kappaB activity in intestine

Abstract

Vitamin D receptor (VDR) plays an essential role in gastrointestinal inflammation. Most investigations have focused on the immune response; however, how bacteria regulate VDR and how VDR modulates the nuclear factor (NF)-kappaB pathway in intestinal epithelial cells remain unexplored. This study investigated the effects of VDR ablation on NF-kappaB activation in intestinal epithelia and the role of enteric bacteria on VDR expression. We found that VDR(-/-) mice exhibited a pro-inflammatory bias. After Salmonella infection, VDR(-/-) mice had increased bacterial burden and mortality. Serum interleukin-6 in noninfected VDR(+/+) mice was undetectable, but was easily detectable in VDR(-/-) mice. NF-kappaB p65 formed a complex with VDR in noninfected wild-type mouse intestine. In contrast, deletion of VDR abolished VDR/P65 binding. P65 nuclear translocation occurred in colonic epithelial cells of untreated VDR(-/-) mice. VDR deletion also elevated NF-kappaB activity in intestinal epithelia. VDR was localized to the surface epithelia of germ-free mice, but to crypt epithelial cells in conventionalized mice. VDR expression, distribution, transcriptional activity, and target genes were regulated by Salmonella stimulation, independent of 1,25-dihydroxyvitamin D3. Our study demonstrates that commensal and pathogenic bacteria directly regulate colonic epithelial VDR expression and location in vivo. VDR negatively regulates bacterial-induced intestinal NF-kappaB activation and attenuates response to infection. Therefore, VDR is an important contributor to intestinal homeostasis and host protection from bacterial invasion and infection.

Figure 1

VDR-null mutant mice have worse outcomes with Salmonella-induced infection. A: Serum IL-6 increased in VDR deficient mice. Normal mice (VDR^+/+) and VDR-deficient mice were infected with wild-type S. typhimurium for six hours. Mouse serum was collected for IL-6 enzyme-linked immunosorbent assay. Data are presented as the mean ± SD from a single experiment assayed in triplicate. *P < 0.05. B: Cecum was dissected out from the indicated mouse and photographed. Cecum shortening was found in the mouse intestine with Salmonella infection. C: H&E staining and scores of the mouse intestine with or without Salmonella infection. D: Relative body weight change in mice with or without Salmonella infection. E: Survival percentage of the VDR^+/+ and VDR^−/− mice administrated S. typhimurium. wild-type group n = 10, VDR^−/− group n = 12. *P < 0.05, **P < 0.01.

Figure 2

VDR expression reduces Salmonella invasion in cells. A: Salmonella location (green) in the VDR^+/+ and VDR^−/− mouse cecum. Normal mice and VDR deficient mice were infected with wild-type Salmonella typhimurium for five days. B: Salmonella concentrations in the VDR^+/+ and VDR mouse ceca. C: Salmonella invasion in infected VDR^+/− and VDR^−/− MEFs. MEFs were stimulated with wild-type Salmonella for 30 minutes, washed, and incubated in fresh Dulbecco’s Modified Eagle Medium for 30 minutes. The mean ± SD is from three replicate experiments. **P < 0.01.

Figure 3

The expression of NFκB p65 in intestinal epithelial cells in VDR knockout mice. A: NF-κB p65 expression in normal mice and VDR-deficient mice. Colonic epithelial cells were collected and lysates were immunoblotted with antibody against NF-κB p65. B: Relative NF-κB p65 band intensity in normal and VDR^−/− mice. Data are presented as the mean ± SD from three repeated experiments. C: VDR physically interacts with NF-κB p65 in intestinal epithelial cells in vivo. VDR deletion abolishes this interaction. Mouse intestinal epithelial cells were collected. Cell lysates were immunoprecipitated with anti-VDR antibody, then the precipitated complex was probed with anti-p65 antibody by Western blot (W.B. left). Note that the background level of p65 was detectable in VDR^−/− cells. An aliquot of the cell lysates was probed with antibodies against VDR, villin, and β-actin to confirm equal input (right). D: VDR/p65 binding in mouse colon using cell fraction. Mouse colon epithelial mucosa was lysed and the nuclear proteins were extracted. Cell fraction was used for co-immunoprecipitation assay using anti-VDR antibody. N: nuclei; C: cytoplasm. **P < 0.01.

Figure 4

The location and activity of NFκB p65 in intestinal epithelial cells in VDR knockout mice. A: NF-κB p65 location in colon in normal mice (VDR^+/+) and VDR^−/− mice without any treatment. Arrows indicate the p65 nuclear location (green). B: Western blot analysis of p65 levels in cytosolic (C) and nuclear (N) extracts isolated from VDR^+/+ and VDR^−/− mouse colons. C: The p65 level in cytosolic and nuclear extracts isolated from VDR^+/− and VDR^−/− MEFs. The nuclear protein Sp1, which is absent in the cytosolic fraction, serves as a nuclear protein loading control. D: Higher phospho-p65 level in the VDR^−/− MEFs without any treatment. *P < 0.05. E: Increased phospho-p65 level in the VDR^−/− mouse colonic epithelial cells without any treatment. Relative phospho-p65 band intensity in normal and VDR^−/− mice is shown. Villin is used as a marker for epithelial cells and an internal control. β-actin is also used as an internal control. The mean ± SD is from three repeats. **P < 0.01.

Figure 5

The distribution of VDR in normal mouse colon. A: VDR distribution in the “Swiss roll” made with the SPF mouse colon. Arrow indicates the direction of the proximal and distal colon. B: Higher magnification of VDR distribution in the normal mouse proximal colon. White arrow indicates the nuclear staining of VDR at the top of colonic crypts. C: VDR protein expression in the proximal and distal colon determined by Western blot. DC: distal colon; PC: proximal colon. D: Relative VDR band intensity in normal mice. Data are presented as the mean ± SD; n = four mice per group. *P < 0.05.

Figure 6

The expression and location of VDR protein in mouse intestinal epithelial cells infected with or without bacteria. A: Total VDR protein in mouse colonic epithelial cells increased after infection with pathogenic Salmonella typhimurium for six hours. B: Normal mice were gavaged without (−) or with either wild-type Salmonella typhimurium or E. coli F18 for six hours. Colonic epithelial cell lysates were immunoblotted with antibodies against VDR or β-actin. Data are from a single experiment and are representative of three separate mouse cells. *P < 0.05. C: Relocation of VDR after Salmonella for six hours. D: VDR distribution in germ-free and conventionalized mouse intestine seven days after colonization with SPF slurries. E: Total VDR protein in mouse colonic epithelial cells increased after GF mice mono-associated with E.coli F18 for six days. **P < 0.01.

Figure 7

Salmonella directly increases VDR expression and transcriptional activity in human intestinal epithelial cells. A: Salmonella directly increases VDR protein expression in human intestinal epithelial CaCo2 BBE cells in the absence of 1,25-dihydroxyvitamin D3 (−). 1,25-Dihydroxyvitamin D3 (20 nmol/L) stimulated VDR protein was measured by Western blot. B: Salmonella colonization increased the VDR transcription activity in human epithelial cells. The cells were transfected with Cignal Vitamin D Reporter (luc) Kit. After transfection for 24 hours, cells were colonized with Salmonella for indicated time and lysed, and luciferase activity was determined. Firefly luciferase activity was normalized to Renilla luciferase activity, and the activity was expressed as relative luminescence units. C: Salmonella colonization increased the mRNA level of the VDR target genes in human epithelial cells. *P < 0.05; **P < 0.001. Data are from a single experiment and are representative of two to three separate experiments.

Figure 8

Working model of VDR in regulation of bacterial invasion, NF-κB activity, host protection, and intestinal homeostasis.