Functional enhancers at the gene-poor 8q24 cancer-linked locus

Abstract

Multiple discrete regions at 8q24 were recently shown to contain alleles that predispose to many cancers including prostate, breast, and colon. These regions are far from any annotated gene and their biological activities have been unknown. Here we profiled a 5-megabase chromatin segment encompassing all the risk regions for RNA expression, histone modifications, and locations occupied by RNA polymerase II and androgen receptor (AR). This led to the identification of several transcriptional enhancers, which were verified using reporter assays. Two enhancers in one risk region were occupied by AR and responded to androgen treatment; one contained a single nucleotide polymorphism (rs11986220) that resides within a FoxA1 binding site, with the prostate cancer risk allele facilitating both stronger FoxA1 binding and stronger androgen responsiveness. The study reported here exemplifies an approach that may be applied to any risk-associated allele in non-protein coding regions as it emerges from genome-wide association studies to better understand the genetic predisposition of complex diseases.

Figure 1. Transcript landscape in the 8q24 region.

Shown are results from high resolution transcriptional profiling in 20 prostate tissues (upper bars) and 4 cancer cell lines (lower tracts). Color intensity represents RNA abundance, while known genes are plotted at the bottom and the cancer-linked regions are indicated at the top on coordinate axis. Note that the tissues' data set is coherent and also strongly correlated with LNCaP cells. Region 1 does not show any traces of significant transcriptional activity, while region 3 may include a weak transcript involving the POU5F1 gene fragment (Figure S2), and upstream transcriptional activity is strongly indicated in Region 2.

Figure 2. Epigenetic map of 5-Mb around the 8q24 risk loci.

(A) Color coding for 6 combinatorial patterns of epigenomic marks and factors (the means of the tracks in each cluster are shown and demarcated into domains I-VI). (B) Epigenomic profiles derived from LNCaP and PC3 cells, as well as histone acetylation data form these two as well as MCF7 and HCT116 cells. Yellow bars indicate enriched marks, and blue bars indicate negative enrichment. RNA data from Figure 1 (in red) is provided for reference. The top panel presents the results of a spatial clustering analysis, showing a color-coded partitioning of the 5 Mb region into the combinatorial patterns of epigenomic marks and factors as defined in (A). Note a genomic block that includes the MYC gene (domain I, red) and the cancer-linked region is flanked by two H3K27me3-dominated clusters (domain IV, blue). Note also that the risk regions are associated with specific clusters (domain VI, green) that are defined by presence of active chromatin marks in LNCaP and weak or no transcriptional activity.

Figure 3. Enhancer-chromatin patterns on the risk regions.

Epigenomic patterns at the risk intervals. Shown are the epigenomic profiles we derived for the 8q24 risk intervals. Specific regions enriched with AcH3 and AROR occupancy were isolated for further analysis (black marks labeled AcP1 to AcP15). Risk regions 1 and 3 are shown in higher resolution in the inset.

Figure 4. Constitutive enhancer activity of AcH3 peak sequences at 8q24.

The DNA sequence containing each of the 15 identified AcH3 sites or a control sequence from the neighboring unacetylated region was inserted upstream of TK-luciferase reporter vector. The constructs were transfected into 5 different cell lines (LNCaP, PC3, HCT 115, COLO 205, and MCF7) along with pRL-TK Renilla luciferase plasmid for 24 h. Dual luciferase assays were conducted. The results were normalized against the internal Renilla control for each transfection. The luciferase activity of the control region was defined as 1. Relative luciferase activity values are presented as mean±SD of triplicate transfections.

Figure 5. Region 1 ARORs contain DHT-mediated enhancer activity and influence of SNP rs11986220.

(A) LNCaP cells were cultured in hormone-depleted medium for 3 days and then treated with 10 nM DHT (+) or ethanol vehicle (−) for 4 h. Conventional site-specific ChIP assays were performed with antibody against AR or normal IgG. Three ARORs identified by ChIP-chip and 1 negative control (CT) were examined by qPCR. The PSA enhancer served as a positive control. The values are presented as percentage of input. (B) LNCaP cells were transiently transfected with AROR containing TK-firefly-luciferase reporter plasmids, PSA-luc , or negative control (CT) plasmid and then incubated with 10 nM DHT or ethanol vehicle for 24 h. Luciferase assays were conducted. The results are presented as mean±SD of triplicate transfections, and because DHT affected Renilla luciferase expression, the firefly luciferase activities were normalized to the protein content of the extracts. (C) Linkage disequilibrium patterns of ARORs14 and -15. Resequencing of AROR14 (128,588,859–128,589,357) and AROR15 (128,600,697–128,601,159) was conducted in 172 chromosomes from prostate cancer cases of European ancestry from the Multiethnic Cohort. One SNP was identified in AROR14 and 6 at AROR15. The risk variant, rs10090154, is shown to the far right of the LD plot and is located 250-bp outside of AROR15. Two SNPs (at AROR15) were highly correlated with rs10090541 (rs11986220, r² = 1.0; rs11988857, r² = 0.923). (D) As indicated, four allele-specific AROR15-containing TK-luciferase reporter plasmids were transfected into LNCaP cells. DHT-mediated luciferase activity was determined as described in (B) and presented as DHT-mediated fold activities. Values are means±SD of six independent clones of each allele (n = 6). For each clone average values of three independent transfections were used. The experiment was replicated three times and a representative experiment is shown. Two-sided p-value was calculated using the student t-test. (E) FoxA1 site specific ChIP-qPCR was conducted in LNCaP cells treated as in (A). (F) EMSA was performed with a ³²P-labeled oligonucleotide probe containing a FoxA1 consensus sequence and a LNCaP cell extract. Complex formation was challenged, as indicated, by anti-FoxA1 antibody (Ab), by the same Ab that had been denaturated by boiling (b), or by increasing concentrations of unlabeled oligonucleotide competitors containing either a FoxA1 consensus sequence, or a sequence centered around rs11986220 with the T SNP, or the same sequence with the A SNP. Results are representative of three experiments.