Tight junction CLDN2 gene is a direct target of the vitamin D receptor

Abstract

The breakdown of the intestinal barrier is a common manifestation of many diseases. Recent evidence suggests that vitamin D and its receptor VDR may regulate intestinal barrier function. Claudin-2 is a tight junction protein that mediates paracellular water transport in intestinal epithelia, rendering them "leaky". Using whole body VDR(-/-) mice, intestinal epithelial VDR conditional knockout (VDR(ΔIEC)) mice, and cultured human intestinal epithelial cells, we demonstrate here that the CLDN2 gene is a direct target of the transcription factor VDR. The Caudal-Related Homeobox (Cdx) protein family is a group of the transcription factor proteins which bind to DNA to regulate the expression of genes. Our data showed that VDR-enhances Claudin-2 promoter activity in a Cdx1 binding site-dependent manner. We further identify a functional vitamin D response element (VDRE) 5΄-AGATAACAAAGGTCA-3΄ in the Cdx1 site of the Claudin-2 promoter. It is a VDRE required for the regulation of Claudin-2 by vitamin D. Absence of VDR decreased Claudin-2 expression by abolishing VDR/promoter binding. In vivo, VDR deletion in intestinal epithelial cells led to significant decreased Claudin-2 in VDR(-/-) and VDR(ΔIEC) mice. The current study reveals an important and novel mechanism for VDR by regulation of epithelial barriers.

Figure 1

VDR status in intestinal epithelial cells leads to the change of Claudin-2 at both mRNA and protein levels in vivo. (A) Claudin-2 mRNA level and (B) Claudin-2 protein level in the intestinal epithelial cells of VDR^+/+, VDR^+/-, or VDR^-/- mice. (C) Claudin-2/VDR protein relative in the intestinal epithelial cells of VDR^+/+, VDR^+/-, or VDR^-/- mice. (D) Claudin-2 mRNA level and (E) Claudin-2 protein level in the intestinal epithelial cells of VDR KO (VDR^ΔIEC) mice. Data are expressed as mean ± SD. *P < 0.05. n = 3 mice/group. (F) Location and quantification of Claudin-2 protein in colons of mice in vivo. Images for each protein shown represent three separate experiments. n = 3 mice/group.

Figure 2

High levels VDR lead to increased Claudin-2 in human colonic epithelial SKCO15 cells in vitro. (A) Claudin-2 mRNA level increased post vitamin D₃ treatment. SKCO15 cells were treated with vitamin D₃ (20 nM) for 24 hours. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments. (B) Claudin-2 protein level and vitamin D₃ dose-dependent curve. SKCO15 cells were treated with indicated vitamin D₃ concentrations for 24 hours. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments. (C) Protein synthesis of Claudin-2 is high in vitamin D₃-treated SKCO15 cells. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments. (D) Claudin-2 expression after SKCO15 cell transfection with human VDR in a pCMV-hVDR plasmid. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments.

Figure 3

VDR binds to the Claudin-2 promoter. (A) CHIP-PCR amplification demonstrated binding of VDR to the promoter regions of human Claudin-2 in SKCO15 cells. PCR were performed including input-positive controls and IgG/villin-negative controls. n = 3 separate experiments. (B) ChIP-PCR assays of VDR binding to the Claudin-2 promoter in mouse colon. n = 3 separate experiments. (C) PCR analysis of RXR in VDR^+/+ or VDR^-/- mice. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments.

Figure 4

Lacking VDR decreases Claudin-2 protein and mRNA expression. (A) Claudin-2 protein and (B) mRNA were reduced by using siRNA against VDR. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments. (C) Claudin-2 protein and (D) mRNA were decreased in MEF VDR^+/-/VDR^-/- cells. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments. (E) VDR protein and (F) mRNA did not change after Claudin-2 was knocked down with Claudin-2 siRNA for 72 hours in SKCO15 cells. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments.

Figure 5

Identification of a functional VDRE sequence in the Claudin-2 promoter. (A) A schematic representation of transcriptional binding sites in the WT Claudin-2 promoter and deletion mutants. Plasmids include wild-type (WT), binding site deletions of NFκB (ΔNFκB), STAT (ΔSTAT), or Cdx1 (ΔCdx) in the Claudin-2 promoter. (B) WT or mutant Claudin-2 reporter plasmids were transfected in HCT116 and (C) CaCO2 cells. Luciferase activity was measured in the cell monolayers incubated in the absence or presence of vitamin D₃ (20 nM) for 24 hours. Dual luciferase assays were performed and firefly luciferase activity was normalized to renilla luciferase activity. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments. (D) A schematic representation of VDRE deletion construct plasmids. Putative VDR-binding sites (containing AGATAACAAAGGTCA sequence) are designated as VDRE. Deletions of all VDRE binding sites (ΔVDRE), deletion of VDRE binding sites and adjacent bases (ΔD2), and non-VDRE deletion controls (ΔD3/ΔD4). (E) WT Claudin-2 reporter gene plasmids and the deletion mutant plasmids were transfected to HCT116 and (F) CaCO2 cells. Luciferase activity was measured in the cell monolayers incubated in the absence or presence of vitamin D₃ (20 nM) for 24 hours. Data are expressed as mean ± SD. *P < 0.05. n = 3 separate experiments.