Specific Recognition of Arginine Methylated Histone Tails by JMJD5 and JMJD7

Abstract

We have reported that JMJD5 and JMJD7 (JMJD5/7) are responsible for the clipping of arginine methylated histone tails to generate “tailless nucleosomes”, which could release the pausing RNA polymerase II (Pol II) into productive transcription elongation. JMJD5/7 function as endopeptidases that cleave histone tails specifically adjacent to methylated arginine residues and continue to degrade N-terminal residues of histones via their aminopeptidase activity. Here, we report structural and biochemical studies on JMJD5/7 to understand the basis of substrate recognition and catalysis mechanism by this JmjC subfamily. Recognition between these enzymes and histone substrates is specific, which is reflected by the binding data between enzymes and substrates. High structural similarity between JMJD5 and JMJD7 is reflected by the shared common substrates and high binding affinity. However, JMJD5 does not bind to arginine methylated histone tails with additional lysine acetylation while JMJD7 does not bind to arginine methylated histone tails with additional lysine methylation. Furthermore, the complex structures of JMJD5 and arginine derivatives revealed a Tudor domain-like binding pocket to accommodate the methylated sidechain of arginine, but not lysine. There also exists a glutamine close to the catalytic center, which may suggest a unique imidic acid mediated catalytic mechanism for proteolysis by JMJD5/7.

The potential underlying mechanism of the leukemia caused by MLL‐fusion and potential treatments

“…Interestingly, the protease activities of JMJD5 on lysine residues was independently reported by another group, 80 which was also observed by our group and was deemed as nonspecific 10 . The deep and narrowed substrate binding pocket may suggest that JMJD5 can only accommodate substrates that contain 2 to 3 residues at the N‐termini within histone tails, 11 thus preventing proper engagement of any potential site within long peptides. Interestingly, another later report, which shows hydroxylase activities of JMJD5 on an arginine within a cytosol ribosomal protein S6, 81 was in accordance with our discovery of specific recognition on arginine by the Tudor‐like binding pocket, 11 though disputed enzymatic activities.…”

“…Most interestingly, H3, H4, and their arginine methylated isoforms are greatly increased in cell lines with deficiency of either JMJD5 or JMJD7 in vivo 10 . Structural characterization reveals that there is a unique structural feature within the catalytic core of JMJD5 to accommodate the methylated sidechains of arginine, which is highly similar to that of methyl‐arginine recognizing Tudor domain 11 . Unique structural features within the Tudor domain‐like pocket differentiates methylated arginines from methylated lysine 11 .…”

“…Furthermore, the structure show that there exists a unique glutamine residue (not existing in other JmjC containing members) near the catalytic center within JMJD5 and JMJD7 to be essential and possibly acts as a proton acceptor, 11 which reflects a novel catalytic enzymatic function through an imidic acid intermediate, as observed in some enzymes 82‐84 . Based on these novel discoveries, we propose that both JMJD5 and JMJD7 could cleave arginine methylated histone tails with different combinations of modifications, so as to generate tailless nucleosomes at the TSSs, mostly at +1 positions, 10,11 which have been characterized as a major barrier to paused pol II 85‐90 . Furthermore, we propose that Pol II could overcome the tailless nucleosomes at +1 regions without the assistance of additional components, 10,11 akin to how RNA polymerase works within Archaea, in which there only exists tailless nucleosomes and no chromatin modification or remodeling components 91‐93 .…”

“…Thus, it is essential to understand how paused Pol II is regulated in higher eukaryotes, which remains elusive to this day. In recent years, we fortuitously found that a group of orphan family members of Jumonji domain containing protein family, including JMJD5, JMJD6, and JMJD7, may work together with the super elongation complex to play critical roles in the release of paused Pol II 10‐13 (Liu et al, unpublished). The pausing of RNA Pol II at promoter regions is a unique transcription regulation mechanism seen in higher eukaryotes 48‐54 .…”

“…The deep and narrowed substrate binding pocket may suggest that JMJD5 can only accommodate substrates that contain 2 to 3 residues at the N‐termini within histone tails, 11 thus preventing proper engagement of any potential site within long peptides. Interestingly, another later report, which shows hydroxylase activities of JMJD5 on an arginine within a cytosol ribosomal protein S6, 81 was in accordance with our discovery of specific recognition on arginine by the Tudor‐like binding pocket, 11 though disputed enzymatic activities. Furthermore, the structure show that there exists a unique glutamine residue (not existing in other JmjC containing members) near the catalytic center within JMJD5 and JMJD7 to be essential and possibly acts as a proton acceptor, 11 which reflects a novel catalytic enzymatic function through an imidic acid intermediate, as observed in some enzymes 82‐84 .…”