The PR-Set7 binding domain of Riz1 is required for the H4K20me1-H3K9me1 trans-tail ‘histone code’ and Riz1 tumor suppressor function

Abstract

PR-Set7/Set8/KMT5a is the sole histone H4 lysine 20 monomethyltransferase (H4K20me1) in metazoans and is essential for proper cell division and genomic stability. We unexpectedly discovered that normal cellular levels of monomethylated histone H3 lysine 9 (H3K9me1) were also dependent on PR-Set7, but independent of its catalytic activity. This observation suggested that PR-Set7 interacts with an H3K9 monomethyltransferase to establish the previously reported H4K20me1-H3K9me1 trans-tail ‘histone code’. Here we show that PR-Set7 specifically and directly binds the C-terminus of the Riz1/PRDM2/KMT8 tumor suppressor and demonstrate that the N-terminal PR/SET domain of Riz1 preferentially monomethylates H3K9. The PR-Set7 binding domain was required for Riz1 nuclear localization and maintenance of the H4K20me1-H3K9me1 trans-tail ‘histone code’. Although Riz1 can function as a repressor, Riz1/H3K9me1 was dispensable for the repression of genes regulated by PR-Set7/H4K20me1. Frameshift mutations resulting in a truncated Riz1 incapable of binding PR-Set7 occur frequently in various aggressive cancers. In these cancer cells, expression of wild-type Riz1 restored tumor suppression by decreasing proliferation and increasing apoptosis. These phenotypes were not observed in cells expressing either the Riz1 PR/SET domain or PR-Set7 binding domain indicating that Riz1 methyltransferase activity and PR-Set7 binding domain are both essential for Riz1 tumor suppressor function.

Somatic PRDM2 c.4467delA mutations in colorectal cancers control histone methylation and tumor growth

“…Notably, the same C-terminal region was found to interact with both the PR domain of PRDM2 and the SET domain of PR-Set7, indicating that these C-terminal truncations may have broader effects besides modulating PRDM2 activity. While the H3K9 methyltransferase activity of PRDM2 has been shown both in vitro and in vivo , it remains unclear whether PRDM2 has monomethyltransferase, dimethyltransferase or dual activity [8, 26, 27]. Moreover, some reports provide evidence for PRDM2 acting as a global enzyme [26], whereas others show its activity to be localized to specific loci [27].…”

“…While the H3K9 methyltransferase activity of PRDM2 has been shown both in vitro and in vivo , it remains unclear whether PRDM2 has monomethyltransferase, dimethyltransferase or dual activity [8, 26, 27]. Moreover, some reports provide evidence for PRDM2 acting as a global enzyme [26], whereas others show its activity to be localized to specific loci [27]. First, while G9a, a closely related H3K9 methyltransferase, targets the chromatin to place both mono and di methylation of H3K9 ubiquitously, we find that PRDM2 only places H3K9 dimethylation.…”

“…In vitro characterization of truncation mutants has demonstrated the importance of the PR-binding motif in regulation of PRDM2 activity through PR binding and intra- and intermolecular interactions [9]. Direct evidence for its role in regulating PRDM2 function was provided recently by the demonstration that the C-terminus is necessary for H3K9 methyltransferase activity involved in establishing the H4K20me1-H3K9me1 trans -tail histone code at specific loci [15]. Despite their high prevalence the functional significance of PRDM2 C-terminal truncating mutations in CRC and cancer in general has not been clarified.…”

Tumor Suppressor RIZ1 in Carcinogenesis

“…A very recent study by Congdon et al showed that RIZ1 was recruited to chromatin by PR-Set7 via direct binding of their C-terminal domains [80]. The RIZ1-PR-Set7 complex was able to establish an H4K30 me1 -H3K9 me1 trans-tail 'histone code' at an ectopic locus to repress gene transcriptions [80].…”

“…The RIZ1-PR-Set7 complex was able to establish an H4K30 me1 -H3K9 me1 trans-tail 'histone code' at an ectopic locus to repress gene transcriptions [80]. Regardless of which route/routes RIZ1 may use to carry out its tumor suppressing functions, it is still unknown how the different functional domains of RIZ proteins coordinate one another during tumor-suppressing by RIZ1 or carcinogenesis by RIZ2.…”