Coordinated methyl and RNA binding is required for heterochromatin localization of mammalian HP1alpha

Abstract

In mammalian cells, as in Schizosaccharomyces pombe and Drosophila, HP1 proteins bind histone H3 tails methylated on lysine 9 (K9). However, whereas K9-methylated H3 histones are distributed throughout the nucleus, HP1 proteins are enriched in pericentromeric heterochromatin. This observation suggests that the methyl-binding property of HP1 may not be sufficient for its heterochromatin targeting. We show that the association of HP1alpha with pericentromeric heterochromatin depends not only on its methyl-binding chromo domain but also on an RNA-binding activity present in the hinge region of the protein that connects the conserved chromo and chromoshadow domains. Our data suggest the existence of complex heterochromatin binding sites composed of methylated histone H3 tails and RNA, with each being recognized by a separate domain of HP1alpha.

Figure 1

HP1α specifically binds methylated histone H3 tails in overlay assays. (A) Schematic representation of the far-western-type overlay assays. Recombinant GST–HP1 fusion proteins produced in E. coli were incubated with western-blotted membranes or paraformaldehyde-fixed cells. After washing, retained GST–HP1 was detected using anti-GST antibodies produced in mouse and then labelled anti-mouse antibodies. (B) Nuclear extracts were resolved by SDS–PAGE and either stained with Coomassie Blue (lane 1) or transferred to nitrocellulose. The membrane was cut to separate the lanes, and each lane was incubated with either GST–HP1α (lane 2) or GST (lane 3). Size markers are indicated (in kDa). (C) Purified histones were resolved by SDS–PAGE and treated as in (A). Each individual lane was incubated with GST–HP1α either in the absence of competing peptide (lane 2) or in the presence of either unmethylated (lane 3), tri-methylated H3 peptide (lane 4) or RNase A (lane 5). Lane 1 shows Coomassie Blue staining of the gel. (D, E and I) Fixed NIH 3T3 cells were incubated with anti-HP1α antibodies and then washed and incubated with recombinant GST–HP1α fusion protein as described in (A). (F, G, J and K) Competition experiments were performed by incubating cells with GST–HP1α and the same peptides used in (C). (H and L) Cells were incubated simultaneously with GST–HP1α and RNase A. Scale bar: 6 μm.

Figure 2

Binding of HP1α to pericentromeric heterochromatin requires both the chromo domain and the hinge. (A) Schematic representation of the HP1α deletion constructs. (B) Purified histones were resolved by SDS–PAGE as in Figure 1C and then incubated with the indicated GST–HP1α constructs as described in Figure 1A. (C–N) Fixed NIH 3T3 cells were incubated with indicated GST–HP1α constructs as described in Figure 1A. Dots observed upon staining with HP1α(67–119) correspond to the nucleoli. DNA was stained with DAPI (I–N). Scale bar: 6 μm.

Figure 3

A conserved region of the hinge is required for pericentromeric binding. (A) Amino acid alignment of the hinge region present in human HP1α, HP1β and HP1γ, part of the suggested alignment with human centromere protein C (CENP-C; Sugimoto et al., 1996), and a schematic representation of the HP1α deletion constructs. (B–M) Fixed NIH 3T3 cells were incubated with indicated GST–HP1α constructs as described in Figure 1A. DNA was stained with DAPI (H–M). Scale bar: 6 μm.

Figure 4

HP1α is an RNA-binding protein. (A) Approximately 1 μg of GST–HP1α (lanes 2–6) was incubated in the presence of a radioactively labelled RNA probe. Incubation was performed either in the absence (lane 2) or in the presence of either unlabelled probe RNA (lanes 3 and 4), single-stranded DNA with the same sequence as the RNA probe (lanes 5 and 6) or the anti parallel sequence (lanes 7 and 8), or double-stranded DNA with the same sequence (lanes 9 and 10). (B) As in (A), 1 μg of GST–HP1α (lanes 2–12 and 14–20) was incubated in the presence of a radioactively labelled RNA probe. Incubation was performed either in the absence (lanes 2 and 14) or in the presence of tRNA (lanes 3 and 4), rRNA (lanes 5 and 6), nuclear RNA (lanes 7 and 8 and 18–20), AU₁₀ (lanes 9 and 10) and GC₁₀ (lanes 11 and 12) oligoribonucleotides, or unlabelled probe RNA (lanes 15–17) in the amounts indicated. (C) EMSAs were performed as in (A), using the indicated GST–HP1α fusion proteins.

Figure 5

Divergent RNA-binding properties of HP1γ. (A–C) Paraformaldehyde-fixed NIH 3T3 cells were indirectly stained with anti-HP1α, anti-HP1γ or GST–HP1γ as indicated. (D–G) NIH 3T3 cells were permeabilized with Triton X-100 and then incubated either in the absence (D and E) or in the presence of RNase A (F and G). Cells were then fixed and indirectly stained with anti-HP1α and anti-HP1γ antibodies as indicated. DNA was stained with DAPI (H–K). Scale bar: 6 μm. (L) Approximately 1 μg of the indicated HP1α constructs (lanes 2 and 3) or 1 and 3 μg of the indicated HP1γ constructs were incubated in the presence of a randomly chosen radioactively labelled RNA probe. (M) North-western assays. Either GST–HP1α (lane 1), GST–HP1α(3 × K→A) (lane 2), GST (lanes 3–5), 6×HIS-HP1γ (lanes 6–8) or 6×HIS-HP1α (lanes 9–11) were resolved by SDS–PAGE, transferred to nitrocellulose, re-natured and then incubated with the radioactively labelled RNA probe.