KDM2A represses transcription of centromeric satellite repeats and maintains the heterochromatic state

Abstract

Heterochromatin plays an essential role in the preservation of epigenetic information, the transcriptional repression of repetitive DNA elements and inactive genes, and the proper segregation of chromosomes during mitosis. Here we identify KDM2A, a JmjC-domain containing histone demethylase, as a heterochromatin-associated and HP1-interacting protein that promotes HP1 localization to chromatin. We show that KDM2A is required to maintain the heterochromatic state, as determined using a candidate-based approach coupled to an in vivo epigenetic reporter system. Remarkably, a parallel and independent siRNA screen also detected a role for KDM2A in epigenetic silencing. Moreover, we demonstrate that KDM2A associates with centromeres and represses transcription of small non-coding RNAs that are encoded by the clusters of satellite repeats at the centromere. Dissecting the relationship between heterochromatin and centromeric RNA transcription is the basis of ongoing studies. We demonstrate that forced expression of these satellite RNA transcripts compromise the heterochromatic state and HP1 localization to chromatin. Finally, we show that KDM2A is required to sustain centromeric integrity and genomic stability, particularly during mitosis. Since the disruption of epigenetic control mechanisms contributes to cellular transformation, these results, together with the low levels of KDM2A found in prostate carcinomas, suggest a role for KDM2A in cancer development.

Figure 1. KDM2A localizes to heterochromatin

(A) Indirect immunofluorescence analysis of HeLa cells transfected with constructs encoding FLAG-tagged KDM2A or FLAG-tagged KDM2B, as indicated. Cells were stained with an anti-FLAG antibody and DAPI. (B) Indirect immunofluorescence analysis of U2OS cells transfected with a construct encoding FLAG-tagged KDM2A. Cells were stained with an anti-FLAG antibody, antibodies to heterochromatin protein 1 variants (HP1-α, HP1-β and HP1-γ), as indicated, and DAPI. Scatter-plots used to determine Pearson (r_p) and Spearman correlation coefficients (r_s) are shown. (C) Graph shows the average colocalization coefficients between KDM2A and HP1 proteins (n=10, +/- SD), as determined by quantification of Pearson (r_p) (open columns) and Spearman correlation coefficients (r_s) (filled columns). The value given for 100% was set as 1. (D) Endogenous KDM2A, but not endogenous KDM2B, associates with HP1 variants. The indicated HA-tagged HP1 proteins were expressed in 293T cells. Whole cell extracts (WCE) were subjected to immunoprecipitation with anti-HA resin followed by immunoblotting with antibodies to the indicated proteins. The asterisk indicates the IgG light chain.

Figure 2. KDM2A maintains the heterochromatic state

(A) Representative example showing knockdown efficiency for KDM2A siRNA oligos by Western blot. Immunoblot of cell extracts from NIH-3T3 cells transfected with control oligos or KDM2A siRNA oligos. Efficacy of siRNA was assessed by immunoblotting to detect KDM2A, and α-Tubulin levels were used as loading controls. (B) Immunoblot of cell extracts from HCT-116 cells transfected with control oligos or KDM2A siRNA oligos. Efficacy of siRNA was assessed by immunoblotting as in A. (C) Indirect immunofluorescence of HP1-γ in NIH-3T3 cells transfected with control oligos or KDM2A siRNA oligos. Cells were pre-extracted with TritonX-100 detergent prior to fixation and permeabilization with 100% methanol to visualize loss of HP1-γ. (D) Indirect immunofluorescence of HP1-γ in mitotic HeLa cells transfected with control oligos or KDM2A siRNA oligos. (E) Reactivation of a GFP reporter gene in GFP(-) HeLa cells transfected with control or KDM2A-siRNA oligos as measured by FACS. (F) Reactivation of a GFP reporter gene in GFP(-) HeLa cells transfected with control, HDAC1, KDM2A or KDM2B siRNA oligos as measured by FACS.

Figure 3. KDM2A represses transcription of pericentric satellite repeats in a JmjC-dependent manner

(A) KDM2A binds to the pericentric region of a human chromosome. Enrichment at the pericentric region of human chromosome 4 was obtained with anti-FLAG antibody as determined by ChIP analysis using chromatin prepared from 293T cells transfected with FLAG-EV (empty vector), FLAG-KDM2A, FLAG-KDM2A(CXXC) and FLAG-KDM2B (open columns). Enrichment at the pericentric region of human chromosome 4 was additionally obtained after DNA complexes were treated with the methylation-sensitive HpaII (5'-CCGG-3') enzyme that only cleaves unmethylated CpG sequences (closed columns). DNA-binding was quantified by real-time PCR. The value given for the amount of PCR product in EV-transfected cells was set at 1 (n=3, +/- SD). (B) Representative example showing knockdown efficiency for KDM2A siRNA oligos by Western blot. Immunoblot of cell extracts from HCT-116 cells transfected with control oligos or KDM2A siRNA oligos. Efficacy of siRNA was assessed by immunoblotting to detect KDM2A, and α-Tubulin levels were used as loading controls. (C) RT-PCR analysis with oligo-dT-primed, reverse-transcribed (RT) cDNA derived from total RNA that was prepared from HeLa cells transfected with control or KDM2A siRNA oligos. PCR amplification using 25 cycles was conducted using locus-specific primers for the indicated chromosome satellites and ARPP P0 (as control). Panel on the right shows the quantification of α-satellite RNA expression in C averaged with one additional, independent experiment. Levels of ARPP P0 were used for normalization. The value given for the amount of α-satellite RNAs in control siRNA cells was set as 1 (n=2, +/- SD). (D) Increase of H3K36me2 at the pericentric region of human chromosome 4 after downregulation of KDM2A. Analysis of H3K36me2 at the pericentric region of human chromosome 4 (open columns) and the GAPDH promoter (closed columns). ChIP was performed as in A with samples from cells treated with control (open columns) or KDM2A (filled columns) siRNA. The value given for the amount of PCR product present in IgG control samples was set at 1 (n=3, +/- SD). (E) RT-PCR was conducted as in C using NIH-3T3 cells transfected with control or KDM2A siRNA oligos specific for mouse cells. Locus-specific primers for major and minor satellite repeats and levels of GAPDH (as control) were used. (F) RT-PCR analysis of major satellite repeats and GAPDH in KDM2A siRNA-treated NIH-3T3 cells, as performed as in E, following transfection with vectors encoding FLAG-tagged KDM2A or FLAG-tagged KDM2A(JmjC). KDM2A expression levels were analyzed by immunoblotting using anti-FLAG antibody. Panel on the right shows the quantification of Major satellite RNA expression in E averaged with one additional, independent experiment. Levels of GAPDH were used for normalization. The value given for the amount of Major satellite RNAs in control cells (EV) was set as 1 (n=2, +/- SD). (G) Forced expression of an α-satellite repeat results in disruption of heterochromatin. RT-PCR was conducted as in C using HeLa cells transfected with an empty expression vector (EV) or an expression vector driving transcription of one α-satellite repeat (171-bp) from a human chromosome (chromosome 4). Locus-specific primers for RNAs encoded by a4 and GAPDH (as control) were used. Panel on the right shows quantification of the expression of a GFP reporter gene in GFP(-) HeLa cells transfected with an empty expression vector (EV) or an expression vector driving transcription of one α-satellite repeat (171-bp) from human chromosome 4 (n=300, +/- SD).

Figure 4. Silencing of KDM2A compromises mitotic fidelity and low levels of KDM2A are found in prostate cancer

(A) Indirect immunofluorescence analysis of mitotic HCT-116 cells transfected with control or KDM2A siRNA oligos. Cells were stained with antibodies to CENP-A (centromere protein A) and α-Tubulin, as indicated, and DAPI. Panel on the right shows the percentage of mitotic cells with misaligned centromeres in HCT-116 cells transfected with control or KDM2A siRNA oligos. (B) Representative chromosome bridges observed in HCT-116 cells transfected with control or KDM2A siRNA oligos. Chromosome bridges were visualized by phosphorylated histone H3 immunostaining. Panel on the right shows the percentage of mitotic cells with chromosome bridges in HCT-116 cells transfected with control or KDM2A siRNA oligos. (C) KDM2A expression data from Dhanasekaran et al., 2001. Normal indicates normal prostate from adult tissues, and cancer indicates prostate carcinoma.