EZH2

Function

EZH2 is the catalytic subunit of the Polycomb Repressive Complex 2 (PRC2). EZH2's catalytic activity relies on its formation of a complex with at least two other PRC2 components, SUZ12 and EED. As a histone methyltransferase (HMTase), EZH2's primary function is to methylate Lys-27 on histone 3 (H3K27me) by transferring a methyl group from the cofactor S-adenosyl-L-methionine (SAM), although recent studies have indicated that it is also capable of methylating non-histone proteins. EZH2 is capable of mono-, di-, and tri-methylation of H3K27 and has been associated with a variety of biological functions, including transcriptional repression and activation, hematopoiesis, development, and cell differentiation.

Transcription repression

EZH2, as a part of PRC2, catalyzes trimethylation of H3K27 (H3K27me3), which is a histone modification that has been characterized as part of the histone code. The histone code is the theory that chemical modifications, such as methylation, acetylation, and ubiquitination, of histone proteins play distinctive roles in epigenetic regulation of gene transcription. EZH2-mediated catalysis of H3K27me3 is associated with long term transcription repression.

EZH2, as well as other Polycomb group proteins, are involved in establishing and maintaining gene repression through cell division. This transcriptionally repressive state is thought to be due to PRC2/EZH2-EED-mediated H3K27 methylation and subsequent recruitment of PRC1 which facilitates condensation of chromatin and formation of heterochromatin. Heterochromatin is tightly packed chromatin which limits the accessibility of transcription machinery to the underlying DNA, thereby suppressing transcription.

During cell division, heterochromatin formation is required for proper chromosome segregation. PRC2/EED-EZH2 complex may also be involved in the recruitment of DNA methyltransferases (DNMTs), which results in increased DNA methylation, another epigenetic layer of transcription repression. Specific genes that have been identified as targets of EZH2-mediated transcriptional repression include HOXA9, HOXC8, MYT1, CDKN2A and retinoic acid target genes.

Transcription activation

In cancer, EZH2 may play a role in activation of transcription, independently of PRC2. In breast cancer cells, EZH2 has been demonstrated to activate NF-κB target genes, which are involved in responses to stimuli. The functional role of this activity and its mechanism are still unknown.

Development and cell differentiation

EZH2 plays an essential role in development. In particular, it helps control transcriptional repression of genes that regulate cell differentiation. In embryonic stem cells, EZH2-mediated trimethylation of H3K27me3 in regions containing developmental genes appears to be important for maintenance of normal cell differentiation. H3K27me3 is also important in driving X-inactivation, the silencing of one X-chromosome in females during development. During X-inactivation, it is thought that EZH2 is involved in initiating heterochromatin formation by trimethylating H3K27 and that other histone methyltransferases and histone marks may be involved in maintaining the silenced state.

Further, EZH2 has been identified as an essential protein involved in development and differentiation of B-cells and T-cells. H3K27me3 is involved in suppressing genes that promote differentiation, thus maintaining an undifferentiated state of B- and T-cells and playing an important role in regulating hematopoiesis.

Regulation of EZH2 activity

The activity of EZH2 is regulated by the post-translational phosphorylation of threonine and serine residues on EZH2. Specifically, phosphorylation of T350 has been linked to an increase in EZH2 activity while phosphorylation of T492 and S21 have been linked to a decrease in EZH2 activity. Phosphorylation of T492 has been suggested to disrupt contacts between human EZH2 and its binding partners in the PRC2 complex, thus hindering its catalytic activity.

In addition to phosphorylation, it has also been shown that PRC2/EZH2-EED activity is antagonized by transcription-activating histone marks, such as acetylation of H3K27 (H3K27ac) and methylation of H3K36 (H3K36me).

Enzymatic activity

EZH2 function is highly dependent upon its recruitment by the PRC2 complex. In particular, WD40-repeat protein embryonic ectoderm development (EED) and zinc finger protein suppressor of zeste 12 (SUZ12) are needed to stabilize the interaction of EZH2 with its histone substrate Recently, two isoforms of EZH2 generated from alternative splicing have been identified in humans: EZH2α and EZH2β. Both isoforms contain elements that have been identified as important for EZH2 function including the nuclear localization signal, the EED and SUZ12 binding sites as well as the conserved SET domain. Most studies have thus far focused on the longer isoform EZH2α, but EZH2β, which lacks exons 4 and 8, has been shown to be active. Furthermore, PRC2/EZH2β complexes act on distinct genes from that of its PRC2/EZH2α counterpart suggesting that each isoform may act to regulate a specific subset of genes. Additional evidence suggests that EZH2 may also be capable of lysine methylation independent of association with PRC2, when EZH2 is highly upregulated.

Lysine methylation

Methylation is the addition of a -CH₃, or methyl group, to another molecule. In biology, methylation is typically catalyzed by enzymes, and methyl groups are commonly added to either proteins or nucleic acids. In EZH2-catalyzed methylation, the amino acid lysine in the histone h3 is methylated. This amino acid residue can be methylated up to three times on its terminal ammonium group. These methylated lysines are important in the control of mammalian gene expression and have a functional role in heterochromatin formation, X-chromosome inactivation and transcriptional regulation. In mammalian chromosomes, histone lysine methylation can either activate or repress genes depending the site of methylation. Recent work has shown that at least part of the silencing function of the EZH2 complex is the methylation of histone H3 on lysine 27. Methylation, and other modifications, take place on the histones. Methyl modifications can affect the binding of proteins to these histones and either activate or inhibit transcription.

Mechanism of catalysis

EZH2 is a member of the SET domain family of lysine methyltransferases which function to add methyl groups to lysine side chains of substrate proteins. SET methyltransferases depend on a S-Adenosyl methionine (SAM) cofactor to act as a methyl donor for their catalytic activity. SET domain proteins differ from other SAM-dependent methyltransferases in that they bind their substrate and SAM cofactor on opposite sides of the active site of the enzyme. This orientation of substrate and cofactor allows SAM to dissociate without disrupting substrate binding and can lead to multiple rounds of lysine methylation without substrate dissociation.

Although neither a substrate-bound or SAM-bound crystal structure for EZH2 has been determined, STAMP structure alignment with the human SET7/9 methyltransferase shows conserved tyrosine residues in almost identical positions within the putative active site of EZH2.

It had been previously suggested that tyrosine 726 in the EZH2 active site was acting as a general base to de-protonate the substrate lysine but kinetic isotope effects have indicated that active site residues are not directly involved in the chemistry of the methyltransferase reaction. Instead these experiments support a mechanism in which the residues lower the pKa of the substrate lysine residue while simultaneously providing a channel for water to access the lysine side chain within the interior of the active site. Bulk solvent water can then easily deprotonate the lysine side chain, activating it for nucleophilic attack of the SAM cofactor in an S_N2-like reaction resulting in transfer of the methyl group from SAM to the lysine side chain.

EZH2 primarily catalyzes mono- and di-methylation of H3K27 but a clinically relevant mutation of residue tyrosine 641 to phenylalanine (Y641F) results in higher H3K27 tri-methylation activity. It is proposed that the removal of the hydroxyl group on Y641 abrogates steric hindrance and allows for accommodation of a third methyl group on the substrate lysine. This EZH2 Y641F mutant is associated with many cancer phenotypes and implies that Y641 may be involved in regulating the number of methyl groups added to a single lysine residue.

Cancer

EZH2 is an attractive target for anti-cancer therapy because it helps cancerous cells divide and proliferate. It is found in larger amounts than in healthy cells in a wide range of cancers including breast, prostate, bladder, uterine, and renal cancers, as well as melanoma and lymphoma. EZH2 is a gene suppressor, so when it becomes overexpressed, many tumor suppressor genes that are normally turned on, are turned off. Inhibition of EZH2 function shrinks malignant tumors in some reported cases because those tumor suppressor genes are not silenced by EZH2. EZH2 typically is not expressed in healthy adults; it is only found in actively dividing cells, like those active during fetal development. Because of this characteristic, overexpression of EZH2 can be used as a diagnostic marker of cancer and some neurodegenerative disorders. However, there are cases where it is difficult to tell whether overexpression of EZH2 is the cause of a disease, or simply a consequence. If it is only a consequence, targeting EZH2 for inhibition may not cure the disease. One example of a cancer pathway in which EZH2 plays a role is the pRB-E2F pathway. It is downstream from the pRB-E2F pathway, and signals from this pathway lead to EZH2 overexpression. Another important characteristic of EZH2 is that when EZH2 is overexpressed, it can activate genes without forming PRC2. This is an issue because it means the methylation activity of the enzyme is not mediated by complex formation. In breast cancer cells, EZH2 activates genes that promote cell proliferation and survival. It can also activate regulatory genes like c-myc and cyclin D1 by interacting with Wnt signaling factors. Importantly, the mutation of tyrosine 641 to phenylalanine in the active SET domain of EZH2 results in preference for H3K27 tri-methylation and has been linked to lymphoma.

Inhibitors

Developing an inhibitor of EZH2 and preventing unwanted histone methylation of tumor suppressor genes is a viable area of cancer research. EZH2 inhibitor development has focused on targeting the SET domain active site of the protein. Several inhibitors of EZH2 have been developed as of 2015, including 3-deazaneplanocin A (DZNep), EPZ005687, EI1, GSK126, and UNC1999. DZNep has potential antiviral and anti-cancer properties because it lowers EZH2 levels and induces apoptosis in breast and colon cancer cells. DZNep inhibits the demethylation of S-adenosyl-L-methionine, the cofactor of EZH2, to form S-adenosyl-L-homocysteine, therefore blocking the transfer of the methyl group to a histone. However, DZNep is not specific to EZH2 and also inhibits other DNA methyltransferases.

In 2012, a company called Epizyme revealed EPZ005687, an S-adenosylmethionine (SAM) inhibitor that is more selective than DZNep; it has a 50-fold increase in selectivity for EZH2 compared to EZH1. The drug blocks EZH2 activity by binding to the SET domain active site of the enzyme. EPZ005687 can also inhibit the Y641 and A677 mutants of EZH2, which may be applicable for treating non-Hodgkin's lymphoma. In 2013, Epizyme began Phase I clinical trials with another EZH2 inhibitor, tazemetostat (EPZ-6438), for patients with B-cell lymphoma.

Sinefungin is another SAM-competitive inhibitor similar to DZNep, however, like DZNep, it is not specific to EZH2. It works by binding in the cofactor binding pocket of DNA methyltransferases to block methyl transfer. EI1 is another inhibitor, developed by Novartis, that showed EZH2 inhibitory activity in lymphoma tumor cells, including cells with the Y641 mutation. The mechanism of this inhibitor also involves competing with the SAM cofactor for binding to EZH2. GSK126 is a potent, SAM-competitive EZH2 inhibitor developed by GlaxoSmithKline, that has 150-fold selectivity over EZH1 and a K_i of 0.5-3 nM. UNC1999 was developed as an analogue of GSK126, and was the first orally bioavailable EZH2 inhibitor to show activity. However, it is less selective than its counterpart GSK126, and it binds to EZH1 as well, increasing the potential for off-target effects.

Combination therapies are being studied as possible treatments when primary treatments begin to fail. Etoposide, a topoisomerase inhibitor, when combined with an EZH2 inhibitor, becomes more effective for non-small cell lung cancers with BRG1 and EGFR mutations. However, EZH2 and lysine methylation can have tumor suppressing activity, for example in myelodysplastic syndrome, indicating that EZH2 inhibition may not be beneficial in all cases.

Weaver Syndrome

Mutations in the EZH2 gene have been linked with Weaver syndrome, a rare disorder characterized by advanced bone age, macrocephaly, and hypertelorism. The histidine residue in the active site of the wild-type EZH2 was mutated to tyrosine in patients with diagnosed with Weaver syndrome. The mutation likely interferes with cofactor binding and causes disruption of the natural function of the protein.

Taxonomic distribution

Enhancer of zeste (E(z)) was originally identified in Drosophila melanogaster, and its mammalian homologs were subsequently identified and named EZH1 (enhancer of zeste homolog 1) and EZH2 (enhancer of zeste homolog 2). EZH2 is highly conserved through evolution. It and its homologs play essential roles in development, cell differentiation, and cell division in plants, insects, fish, and mammals. The following taxonomic tree is a depiction of EZH2's distribution throughout a wide variety of species.