DNA methylation-related vitamin D receptor insensitivity in breast cancer

Abstract

Calcitriol (1α, 25(OH)(2)-Vitamin D3) binds to the vitamin D receptor (VDR) and regulates differentiation of the normal mammary gland, and may therefore be useful in breast cancer treatment or prevention. Many breast cancer cells are, however, resistant to Calcitriol. In this study, we investigated the resistance mechanism and the role of epigenetic silencing of VDR by promoter hypermethylation. Bisulfite sequencing of the VDR promoter region revealed methylated CpG islands at -700 base pairs (bp) upstream and near the transcription start site. VDR CpG islands were demethylated by 5'deoxy-azacytidine treatment, and this was accompanied by a parallel increase in VDR mRNA levels in breast cancer cell lines. Quantitative methylation-specific PCR analyses confirmed hypermethylation of these CpG islands in primary tumors, and its absence in normal breast tissue. VDR transcripts detected in breast cancers were predominantly 5'-truncated, while normal breast tissue expressed full-length transcripts. Consistent with this observation, genes containing the VDR-responsive element (VDRE), such as cytochrome p450 hydroxylases, p21 or C/EBP were underexpressed in breast cancers compared to normal breast samples. Expression of the active longer transcripts of VDR was restored with 5'deoxy-Azacytidine (AZA) treatment, with a concurrent increase in expression of VDRE-containing genes. Thus, promoter methylation-mediated silencing of expression of the functional variants of VDR may contribute to reduced expression of downstream effectors of the VDR pathway and subsequent Calcitriol insensitivity in breast cancer. These data suggest that pharmacological reversal of VDR methylation may re-establish breast cancer cell susceptibility to differentiation therapy using Calcitriol.

Figure 1

Structure of the 5′ VDR gene exons and splice variant-selective primer design. The V primers (VF & VR) detect all 5′ VDR variants (VDR total, VT), and span exons 3 and 4, producing a 223 bp amplicon. Table 1 lists the oligonucleotide sequences of all primers used in this report. The V1 primers (V1F & V1R) detect three VDR variant products (V1, V2, V3) spanning the 5′ flanking region of exon 1a to the end of exon 2. Variant V1 (167 bp product) lacks exons 1d and 1b. Variant V2 (290 bp product) is devoid of exon 1d and Variant V3 (85 bp) is devoid of exons 1d, 1b and 1c. V1 and V2 both encode the same active 427 amino acid VDRA protein. The V1d primers (V1dF & V1dR) detect three VDR variant products (V1d, V1d′ and V1d″) extending from the 5′ region of exon 1d to exon 2. V1d′ (279 bp product) contains both exon 1b and exon 1c, V1d (156 bp product) is lacking exon 1b, and V1d″ (87 bp product) contains neither exon 1b nor exon 1c. Variants V1d and V1d″ encode active VDR proteins of 477 amino acids (VDRB1) and 450 amino acids (VDRB2), respectively. The second column depicts the variants that we identified in breast cancer tissue by cloning and sequencing the gel purified RT-PCR products; the third column lists the predicted VDR peptides. Alternative translational start sites are indicated with a * at the top of exon 1d (coding for VDRB proteins) and exon 2 (coding for VDRA proteins).

Figure 2

Demethylation potentiates Calcitriol-induced growth inhibition and VDR Expression. The immortalized normal breast cell line HBL100 and the breast cancer cell lines HS578T, 21PT, MCF-7 and T47D were treated with AZA (5-azacyitidine; 5 µM), Calcitriol (D; 1.5 µM), alone or in combination and compared to vehicle control (CON). After 96 hours, growth inhibition (A) and VDR expression (B) was assessed. Significance is expressed as (*), p < 0.05; (**), p < 0.01. (A) MMT assay: For each cell line, the percentage of viable cells recovered after each drug treatment is indicated. The means and standard deviations of triplicate experiments are shown. (B) Quantitative RT-PCR assay of total VDR expression: Expression levels of VDR are expressed as ratio of inverse Ct values of VDR to β-actin transcripts. The V and BA primer sets were used to assess total VDR and beta actin, respectively.

Figure 3

Bisulfite sequencing of the VDR promoter in breast cell lines and tissues. The diagram shows individual CpG dinucleotides as beads on a string representing the VDR promoter region from −789 to +380 bp relative to the exon 1a transcriptional start site (TS). Methylation levels are shown as average of three independent reads, where white circles indicate 0–25% methylation, white-black circles indicate 25–50% methylation, and black circles indicate 50–100% methylation. For each breast cell line sequenced, lines 1, 2 and 3 represent untreated control cells, Calcitriol-treated and AZA-treated cells, respectively. The sequences in the lower half of the diagram were obtained from 8 breast cancer samples, 7 adjacent normal breast tissue samples, and 3 organoid preparations from normal breast tissue. The arrowheads indicate the locations of the primers used for the MSP assays (see Fig. 4). Grey, black, striped and white vertical bars at the top represent DNA binding sites of Sp1, NFγB, AP-1 and AP-2 transcription factors, respectively.

Figure 4

Quantification of VDR methylation by Methylation-Specific PCR. The fraction of methylated VDR promoter is expressed as percent methylation level in 15 breast cancers and 7 normal breast tissues. Round symbols indicate DNA obtained from fresh frozen tissue, squares indicate FFPE-derived DNA. The box plots show significantly higher VDR methylation in breast cancers than normal tissue controls (p < 0.0002 by Wilcoxon rank sum test).

Figure 5

Different VDR variants are present in breast cancer tissue and normal breast tissue organoids. (A) Agarose gel electrophoresis: The products of reverse-transcriptase PCR amplifications using V1 (top gel) and V1d primers (bottom gel) in breast cancer tissues (Cancer) and in normal breast organoids (Normal) are shown. Standard qRT-PCR conditions (40 cycles) were used. See Figure 1 for a description of the variants detected. β-actin transcript levels are shown below. (B and C) Quantitation of VDR transcripts: An agarose gel electrophoresis showing the products of RT-PCR amplifications using V1 and V1d primers is shown in (B). Semi-quantitative PCR reaction conditions (28 cycles) were used in order to remain in the linear product range of the amplifications. The bars in the matching bar graph (C) represent ratios of VDR variants to beta actin, in cancer tissue (grey) and in normal breast organoids (white). The right-most bars (VT) reflect the results obtained using the V primers that detect exon 3 & 4 transcripts. Means and standard deviations of triplicate experiments are shown. Significant differences between normal and cancer are indicated (*p < 0.05; **p < 0.01). (D) Quantitation of VDR variant expression after Calcitriol and AZA: Cell lines were treated as in Figure 2A. RT-PCR reactions were performed with V1 and V1d primer sets, and GAPDH was used to normalize VDR expression levels. The bar graphs show means and standard deviations of the five cell lines. Significant differences compared to vehicle-only controls (fold induction = 1) are indicated (*p < 0.05; **p < 0.01).

Figure 6

Expression of VDR Responsive Element (VDRE)-containing genes in breast cancer in vivo and in vitro. (A) Assessment of VDRE-containing gene expression in vivo: Agarose gel electrophoresis of RT-PCR of CYP3A4 (24/25-hydroxylase), CYP27B1 (1-hydroxylase), CYP24A1 (24-hydroxylase) and p21. Assays were performed in triplicate on a total of 8 breast cancers and 3 normal organoid preparations. (B) Quantitation of VDRE-containing gene expression in vivo: The matching bar graph shows ratios of individual gene transcripts to beta actin, in cancer tissue (grey) and in normal breast organoids (white). Means and standard deviations are shown. Significant differences between normal and cancer are indicated (*p < 0.05; **p < 0.01). (C) Induction of VDRE-containing gene expression after Calcitriol and AZA treatment in vitro: The bars represent the fold induction of VDR downstream gene targets (CYP3A4 [24/25 hydroxylase]-white, CYP27B1 [1 hydroxylase]-grey, and C/EBP-hatched) over untreated control (horizontal line = 1.0). Means and standard deviations of triplicate experiments are shown (*p < 0.05; **p < 0.01).