Others and we previously showed that the vitamin D receptor (VDR) is subject to degradation by the 26S proteasome and that treatment with 1,25-dihydroxyvitamin D(3) (1,25D(3)) inhibited this degradation. In the present study, we found that in osteoblasts, but not in intestinal epithelial cells, the VDR was susceptible to degradation by the 26S proteasome. The subcellular site for degradation of the VDR in osteoblasts is the cytoplasm and the site for ligand-dependent protection of the VDR from the 26S proteasome is the chromatin. These direct relationships between nuclear localization and protection of the VDR from 26S proteasome degradation led us to hypothesize that the unoccupied cytoplasmic VDR is a substrate for polyubiquitination, which targets VDR for degradation by the 26S proteasome, and that nuclear localization has the ability to protect the VDR from polyubiquitination and degradation. To test these hypotheses, we used Cos-1 cells transfected with human VDR and histidine-tagged ubiquitin expression vectors. We found that unoccupied VDR was polyubiquitinated and that 1,25D(3) inhibited this modification. Mutations in the nuclear localization signal of VDR (R49W/R50G and K53Q/R54G/K55E) or in the dimerization interface of VDR with retinoid X receptor (M383G/Q385A) abolished the ability of 1,25D(3) to protect the VDR from polyubiquitination, although these mutations had no effect on the ligand-binding activity of VDR. Therefore, we concluded that in some cellular environments unoccupied cytoplasmic VDR is susceptible to polyubiquitination and proteasome degradation and that ligand-dependent heterodimerization and nuclear localization protect the VDR from these modifications.
J. Cell. Biochem. 110: 926-934, 2010. (c) 2010 Wiley-Liss, Inc.