Downregulation of histone H3 lysine 9 methyltransferase G9a induces centrosome disruption and chromosome instability in cancer cells

Background

Modifications of the histone amino-terminal tails affect access of regulatory factors and complexes to chromatin and thereby influence biological processes. Cancer cells are characterized by prominent epigenetic dysregulation, including histone modifications. However, the functional roles of the histone methyltransferases (HMT) in cancer remain unclear.

Methodology/Principal Findings

We studied RNAi-based inhibition (knockdown, KD) of 2 different H3K9 HMTs, SUV39H1 and G9a. Knockdown of the 2 HMTs in PC3 cancer cell line markedly inhibited cell growth and caused profound morphological changes with loss of telomerase activity and shortened telomeres. SUV39H1 KD cells showed substantial increase in G2/M fraction. G9a KD cells showed increased DNA content (1.7-fold in 2 independent clones) compared with FACS analyses to control. Karyotype analyses showed that this was due to an increased number of chromosomes (from 61 to 102) in G9a KD cells compared to parental PC3. Intriguingly, we found abnormal centrosome morphology and number in about 25% of the G9a KD cells, while centrosomes were morphologically normal in control cells. Microarray analyses after KD of SUV39H1 or G9a showed very few genes up-regulated among the 39,000 genes. The silenced tumor-suppressor genes p16 and RASSF1A were not activated in KD cells.

Conclusions/Significance

These data suggest that the 2 HMTs, SUV39H1 and G9a are required to perpetuate the malignant phenotype. Furthermore, G9a plays a critical role in regulating centrosome duplication presumably through chromatin structure rather than through affecting gene expression in cancer cells. Targeting these histone methyltransferases may be of therapeutic benefit in cancers.     

Reversal of H3K9me2 by a small-molecule inhibitor for the G9a histone methyltransferase

Histone lysine methylation has important roles in the organization of chromatin domains and the regulation of gene expression. To analyze its function and modulate its activity, we screened for specific inhibitors against histone lysine methyltransferases (HMTases) using recombinant G9a as the target enzyme. From a chemical library comprising 125,000 preselected compounds, seven hits were identified. Of those, one inhibitor, BIX-01294 (diazepin-quinazolin-amine derivative), does not compete with the cofactor S-adenosyl-methionine, and selectively impairs the G9a HMTase and the generation of H3K9me2 in vitro. In cellular assays, transient incubation of several cell lines with BIX-01294 lowers bulk H3K9me2 levels that are restored upon removal of the inhibitor. Importantly, chromatin immunoprecipitation at several G9a target genes demonstrates reversible reduction of promoter-proximal H3K9me2 in inhibitor-treated mouse ES cells and fibroblasts. Our data identify a biologically active HMTase inhibitor that allows for the transient modulation of H3K9me2 marks in mammalian chromatin.     

Histone H3.3 G34 Mutations Alter Histone H3K36 and H3K27 Methylation In Cis

Histone H3 encoding genes, particularly H3F3A and H3F3B, the genes encoding the variant histone H3.3, are mutated at high frequency in pediatric brain and bone malignancies. Compared to the extensive studies on K27M and K36M mutations, little is known about the mechanism of G34 mutations found in pediatric glioblastoma or giant cell tumors of the bone. Here we report that unlike the K27M or K36M that affect global histone methylation, the giant cell tumors of the bone G34 mutations (G34L/W) only affect histone H3K36 and H3K27 methylation on the same mutated histone tails (in cis), a mechanism distinct from known histone mutations.     

Histone H3R2 symmetric dimethylation and histone H3K4 trimethylation are tightly correlated in eukaryotic genomes

The preferential in vitro interaction of the PHD finger of RAG2, a subunit of the V(D)J recombinase, with histone H3 tails simultaneously trimethylated at lysine 4 and symmetrically dimethylated at arginine 2 (H3R2me2sK4me3) predicted the existence of the previously unknown histone modification H3R2me2s. Here, we report the in vivo identification of H3R2me2s . Consistent with the binding specificity of the RAG2 PHD finger, high levels of H3R2me2sK4me3 are found at antigen receptor gene segments ready for rearrangement. However, this double modification is much more general; it is conserved throughout eukaryotic evolution. In mouse, H3R2me2s is tightly correlated with H3K4me3 at active promoters throughout the genome. Mutational analysis in S. cerevisiae reveals that deposition of H3R2me2s requires the same Set1 complex that deposits H3K4me3. Our work suggests that H3R2me2sK4me3, not simply H3K4me3 alone, is the mark of active promoters and that factors that recognize H3K4me3 will have their binding modulated by their preference for H3R2me2s.     

Histone H3K27 dimethylation landscapes contribute to genome stability and genetic recombination during wheat polyploidization

Bread wheat (Triticum aestivum) is an allohexaploid that was formed via two allopolyploidization events. Growing evidence suggests histone modifications are involved in the response to ‘genomic shock’ and environmental adaptation during polyploid formation and evolution. However, the role of histone modifications, especially histone H3 lysine-27 dimethylation (H3K27me2), in genome evolution remains elusive. Here we analyzed H3K27me2 and H3K27me3 profiles in hexaploid wheat and its tetraploid and diploid relatives. Although H3K27me3 levels were relatively stable among wheat species with different ploidy levels, H3K27me2 intensities increased concurrent with increased ploidy levels, and H3K27me2 peaks were colocalized with massively amplified DTC transposons (CACTA family) in euchromatin, which may silence euchromatic transposons to maintain genome stability during polyploid wheat evolution. Consistently, the distribution of H3K27me2 is mutually exclusive with another repressive histone mark, H3K9me2, that mainly silences transposons in heterochromatic regions. Remarkably, the regions with low H3K27me2 levels (named H3K27me2 valleys) were associated with the formation of DNA double-strand breaks in genomes of wheat, maize (Zea mays) and Arabidopsis. Our results provide a comprehensive view of H3K27me2 and H3K27me3 distributions during wheat evolution, which support roles for H3K27me2 in silencing euchromatic transposons to maintain genome stability and in modifying genetic recombination landscapes. These genomic insights may empower breeding improvement of crops.     

Substrate specificity and kinetic mechanism of mammalian G9a histone H3 methyltransferase

Lysine-specific murine histone H3 methyltransferase, G9a, was expressed and purified in a baculovirus expression system. The primary structure of the recombinant enzyme is identical to the native enzyme. Enzymatic activity was favorable at alkaline conditions (>pH 8) and low salt concentration and virtually unchanged between 25 and 42 degrees C. Purified G9a was used for substrate specificity and steady-state kinetic analysis with peptides representing un- or dimethylated lysine 9 histone H3 tails with native lysine 4 or with lysine 4 changed to alanine (K4AK9). In vitro methylation of the H3 tail peptide resulted in trimethylation of Lys-9 and the reaction is processive. The turnover number (k(cat)) for methylation was 88 and 32 h(-1) on the wild type and K4AK9 histone H3 tail, respectively. The Michaelis constants for wild type and K4AK9 ((K(m)(pep))) were 0.9 and 1.0 microM and for S-adenosyl-L-methionine (K(m)(AdoMet)) were 1.8 and 0.6 microM, respectively. Comparable kinetic constants were obtained for recombinant histone H3. The conversion of K4AK9 di- to trimethyl-lysine was 7-fold slower than methyl group addition to unmethylated peptide. Preincubation studies showed that G9a-AdoMet and G9a-peptide complexes are catalytically active. Initial velocity data with peptide and S-adenosyl-L-methionine (AdoMet) and product inhibition studies with S-adenosyl-L-homocysteine were performed to assess the kinetic mechanism of the reaction. Double reciprocal plots and preincubation studies revealed S-adenosyl-L-homocysteine as a competitive inhibitor to AdoMet and mixed inhibitor to peptide. Trimethylated peptides acted as a competitive inhibitor to substrate peptide and mixed inhibitor to AdoMet suggesting a random mechanism in a Bi Bi reaction for recombinant G9a where either substrate can bind first to the enzyme, and either product can release first.     

Identification of nuclear localization signals of Drosophila G9a histone H3 methyltransferase

G9a is one of the well-characterized histone methyltransferases. G9a regulates H3K9 mono- and dimethylation at euchromatic region and consequently plays important roles in euchromatic gene regulation. Mammalian G9a contains several distinct domains, such as GHD (G9a homology domain), ANK, preSET, SET and PostSET. These domains are highly conserved between mammals and Drosophila. Although mammalian G9a has nuclear localization signal (NLS) in its N-terminal region, the amino acid sequences of this region are not conserved in Drosophila. Here we have examined the subcellular localization of a series of truncated forms of Drosophila G9a (dG9a). The identified region (aa337-aa470) responsible for nuclear localization of dG9a contains four short stretches of positively charged basic amino acids (NLS1, aa334-aa345; NLS2, aa366-aa378; NLS3, aa407-aa419; NLS4, aa461-aa472). Each of NLS1, NLS3 and NLS4 is sufficient for the nuclear localization when they are fused with the enhanced green fluorescent protein. These NLSs of dG9a are distinct from those of mammalian G9a in their positions and amino acid sequences.     

Heterochromatic Genome Stability Requires Regulators of Histone H3 K9 Methylation

Heterochromatin contains many repetitive DNA elements and few protein-encoding genes, yet it is essential for chromosome organization and inheritance. Here, we show that Drosophila that lack the Su(var)3-9 H3K9 methyltransferase display significantly elevated frequencies of spontaneous DNA damage in heterochromatin, in both somatic and germ-line cells. Accumulated DNA damage in these mutants correlates with chromosomal defects, such as translocations and loss of heterozygosity. DNA repair and mitotic checkpoints are also activated in mutant animals and are required for their viability. Similar effects of lower magnitude were observed in animals that lack the RNA interference pathway component Dcr2. These results suggest that the H3K9 methylation and RNAi pathways ensure heterochromatin stability.