Stimulation of Sirt1-regulated FoxO protein function by the ligand-bound vitamin D receptor

Abstract

Hormonal vitamin D, 1,25-dihydroxyvitamin D (1,25D), signals through the nuclear vitamin D receptor (VDR). 1,25D regulates cell proliferation and differentiation and has been identified as a cancer chemopreventive agent. FoxO proteins are transcription factors that control cell proliferation and survival. They function as tumor suppressors and are associated with longevity in several organisms. Accumulating data have revealed that 1,25D and FoxO proteins regulate similarly common target genes. We show here that the ligand-bound VDR regulates the posttranslational modification and function of FoxO proteins. 1,25D treatment enhances binding of FoxO3a and FoxO4 within 4 h to promoters of FoxO target genes and blocks mitogen-induced FoxO protein nuclear export. The VDR associates directly with FoxO proteins and regulators, the sirtuin 1 (Sirt1) class III histone deacetylase (HDAC), and protein phosphatase 1. In addition, phosphatase activity and trichostatin A-resistant HDAC activity coimmunoprecipitate with the VDR. 1,25D treatment rapidly (in <4 h) induces FoxO deacetylation and dephosphorylation, consistent with activation. In contrast, ablation of VDR expression enhances FoxO3a phosphorylation, as does knockdown of Sirt1, consistent with the coupling of FoxO acetylation and phosphorylation. 1,25D regulation of common VDR/FoxO target genes is attenuated by blockade of phosphatase activity or by small interfering RNA (siRNA)-mediated knockdown of Sirt1 or FoxO protein expression. Finally, 1,25D-dependent cell cycle arrest is blocked in FoxO3a-deficient cells, indicating that FoxO proteins are key downstream mediators of the antiproliferative actions of 1,25D. These studies link 1,25D signaling through the VDR directly to Sirt1 and FoxO function and provide a molecular basis for the cancer chemopreventive actions of 1,25D.

FIG. 1.

The VDR and FoxO proteins regulate common target genes. (A) RT-PCR analysis of regulation by 1,25D (100 nM) of CCND2, CCNG2, BNIP3, and MMP3 in SCC25 cells. CYP24 and GAPDH are included as positive and negative controls, respectively, for 1,25D-regulated gene expression. (B) Treatment of SCC25 cells with 1,25D reduces cyclin D2 levels. SCC25 cells were incubated with 1,25D (100 nM) as indicated prior to isolation of protein for Western blotting. (C) Western blotting of the expression of FoxO1, FoxO3a, and FoxO4 in SCC25 cells treated with 100 nM 1,25D over a 72-h period. (D) SCC25 cells were incubated with vehicle (−), insulin (Ins), or insulin and 1,25D as indicated for 1 h. Subcellular localization of FoxO3a was determined by immunocytochemistry. Nuclei were stained with DAPI. See Materials and Methods for details. (E) SCC25 cells were treated with 1,25D as indicated prior to harvesting and isolation of nuclear and cytoplasmic fractions. Western blots of FoxO3a and actin controls are presented.

FIG. 2.

Ablation of VDR expression blocks regulation by 1,25D of FoxO target genes. (A) siRNA-mediated knockdown of VDR expression. SCC25 cells were transfected with siRNAs recognizing the VDR gene or scrambled siRNA controls, and expression of the VDR was assessed by Western blotting. (B) Effect of VDR knockdown on regulation by 1,25D of VDR target gene CYP24. (C) Effects of VDR knockdown on regulation by 1,25D of FoxO target genes CCND2, BNIP3, CCNG2, and MMP3. Error bars represent SD. *, P < 0.05 for control (siCont) versus VDR-specific (siVDR) siRNAs at 24 h.

FIG. 3.

Short-term 1,25D treatment induces binding of FoxO3a to target gene promoters. (A) FHREs in the promoter-proximal regions of the CCND2 and BNIP3 genes. (B and C) ChIP analysis of 1,25D-dependent binding of FoxO3a to the CCND2 (B) and BNIP3 (C) promoters in SCC25 cells as assessed by qRT-PCR. (D) ChIP analysis of 1,25D-dependent association of the VDR with the FoxO binding regions of the CCND2 and BNIP3 promoters. Error bars represent SD. *, P < 0.05.

FIG. 4.

Interactions of the VDR with FoxO proteins and their regulators Sirt1 and PP1. (A and B) 1,25D induces dephosphorylation of FoxO3a (A) and FoxO4 (B). FoxO3a and FoxO4 phosphorylation on specific residues were assessed by Western blotting using specific antibodies recognizing FoxO3a phosphorylated on Thr32 or FoxO4 phosphorylated on Ser193. (C) FoxO3a and FoxO4 associate with the VDR, as determined by immunoprecipitation of the VDR from SCC25 cells treated with 1,25D as indicated, followed by Western blotting for FoxO3a, FoxO4, or the VDR. Coimmunoprecipitation of p160 coactivator AIB1 with the VDR was used as a control for 1,25D-dependent association of the VDR with a cofactor. (D) Association of the VDR with PPI and Sirt1. The VDR was immunoprecipitated from SCC25 cells, and immunoprecipitates were probed for the presence of the catalytic subunits of PP1, PP2a, Sirt1, or the VDR. (E) 1,25D augments the association of Sirt1 with FoxO3a. FoxO3a was immunoprecipitated from SCC25 cells, and immunoprecipitates were probed for the presence of Sirt1, the VDR, or FoxO3a. (F) Sirt1 was immunoprecipitated from SCC25 cells treated with vehicle or 1,25D, followed by Western blotting for associated FoxO3a. (G) Analysis by a GST pulldown assay of the binding of in vitro-translated FoxO3a to the VDR ligand binding domain in the absence and presence of 1,25D. (H) Analysis of 1,25D-dependent binding of in vitro-translated SIRT1 to the VDR ligand binding domain by a GST pulldown assay. (I) Effect of siRNA-mediated VDR ablation on expression levels of Sirt1, total FoxO3a, and RXRα and an analysis of the effect of VDR ablation on the level of Th32 FoxO3a phosphorylation.

FIG. 5.

Phosphatase inhibition blocks 1,25D-induced FoxO3a dephosphorylation and activation. (A) SCC25 cells were treated with 1,25D for 8 h alone or in the presence of the phosphatase inhibitor okadaic acid (Ok; 1 nM). Expression of total and phospho-FoxO3a was assessed by Western blotting. (B) 1,25D treatment enhances total phosphatase activity in SCC25 cells. Extracts of SCC25 cells were incubated with 1,25D, as indicated, and total phosphatase activity was determined. (C) Association of phosphatase activity with the VDR. SCC25 cells were treated for 4 h with vehicle or 1,25D, and extracts were immunoprecipitated with an αVDR antibody. Immunoprecipitates were assayed for phosphatase activity. (D) Inhibition of phosphatase activity blocks or attenuates regulation by 1,25D of FoxO target genes. SCC25 cells were treated with 1,25D or vehicle for 24 h in the absence of inhibitors (Cont.) or in the presence of the phosphatase inhibitor okadaic acid (Ok; 1 nM) prior to isolation of RNA for qRT-PCR analysis. (E) Okadaic acid (1 nM) blocks suppression by 1,25D of cyclin D2 expression, as analyzed by Western blotting. Error bars represent SD. *, P < 0.05.

FIG. 6.

Association of the VDR with TSA-resistant lysine deacetylase activity and the effects of Sirt1 on FoxO3a phosphorylation and target gene regulation. (A) Association of the VDR with lysine deacetylase activity. SCC25 cells were treated with vehicle or 1,25D for 4 h, and extracts were immunoprecipitated with an αVDR antibody or IgG control and assayed for lysine deacetylase activity in the presence of NAD and the class I and II HDAC inhibitor TSA. (B) 1,25D treatment induces deacetylation of FoxO3a. SCC25 cells were treated with 1,25D for 0, 1, or 4 h, and extracts were immunoprecipitated with an anti-FoxO3a antibody, followed by Western blotting for acetyllysine. As a control for specificity of 1,25D-induced deacetylation, SCC25 cells were treated with 1,25D for 0, 1, or 4 h and extracts analyzed by Western blotting with an anti-acetylated-p53 antibody. Note that the treatment at the zero time point was performed in duplicate in this experiment. (C) Ablation of Sirt1 expression enhances phosphorylation of FoxO3a. SCC25 cells were transfected with siRNAs recognizing Sirt1 or scrambled siRNA controls, and expression of Sirt1 was assessed by Western blotting. Expression of total and Th32-phosphorylated FoxO3a in extracts of cells transfected with control or SIRT1-specific siRNA was analyzed. (D) Western blotting of the effects of siRNA-mediated ablation of Sirt1 or FoxO3a expression on levels of the VDR and RXRα. (E) Effect of Sirt1 ablation on regulation of 1,25D target gene CYP24. SCC25 cells transfected with control or SIRT1-specific siRNAs were incubated with 1,25D as indicated, and regulation of CYP24 was analyzed by qRT-PCR. (F) Effect of Sirt1 ablation on regulation by1,25D of FoxO target genes. SCC25 cells transfected with control or SIRT1-specific siRNA were incubated with 1,25D as indicated, and regulation of FoxO target genes was assessed by qRT-PCR. (G) ChIP analysis of association of transiently expressed Flag-Sirt1 with the FoxO binding regions of the CCND2 and BNIP3 promoters. (H) Analysis of the effects of Sirt1 ablation on 1,25D-dependent binding of the VDR to the FoxO binding region of the BNIP3 promoter. Error bars represent SD. *, P < 0.05.

FIG. 7.

Ablation of FoxO expression attenuates or blocks inhibition of SCC25 cell proliferation by 1,25D. (A) siRNA-mediated knockdown of FoxO3a or FoxO4. SCC25 cells were transfected with siRNAs recognizing FOXO3A, FOXO4, or control siRNA, and expression of FoxO proteins was assessed by Western blotting. (B) Ablation of FoxO expression does not affect regulation by 1,25D of a VDRE-containing promoter. SCC25 cells were transfected with the VDRE3/tk-luc expression vector along with control siRNAs or siRNAs directed against FOXO3A and/or FOXO4, as indicated. Cells were treated with vehicle or 1,25D, and extracts were assayed for luciferase activity. (C) Effects of ablation of FoxO expression on regulation by 1,25D of target gene CYP24. SCC25 cells were transfected with control or FoxO-specific siRNAs, and transfected cells were treated with 1,25D as indicated. Expression of CYP24 was analyzed by qRT-PCR. (D) Effects of ablation of FoxO expression on regulation by 1,25D of FoxO target genes. SCC25 cells were transfected with control or FoxO-specific siRNA, and transfected cells were treated with 1,25D as indicated. Expression of FoxO target genes was analyzed by qRT-PCR. (E) Effects of FoxO knockdown on regulation by 1,25D of cyclin D2 expression. SCC25 cells were transfected with control or FoxO-specific siRNAs, and transfected cells were treated with 1,25D for 8 h. Cyclin D2 expression and actin control were assessed by Western blotting. Error bars represent SD. *, P < 0.05.

FIG. 8.

Knockdown of FoxO3a expression blocks the antiproliferative activity of 1,25D. Dose (A)- and time (B)-dependent effect of 1,25D on SCC25 cell proliferation as assessed by an EdU incorporation assay. (C) SCC25 cells were transfected with control or FOXO-specific siRNAs and then treated with vehicle or 1,25D for 24 h. Cell proliferation was analyzed by determining EdU incorporation. Quantification of results and analysis by the t test are presented on the right. Error bars represent SD. *, P < 0.05.