The functional diversity of protein lysine methylation

Abstract

Large-scale characterization of post-translational modifications (PTMs), such as phosphorylation, acetylation and ubiquitination, has highlighted their importance in the regulation of a myriad of signaling events. While high-throughput technologies have tremendously helped cataloguing the proteins modified by these PTMs, the identification of lysine-methylated proteins, a PTM involving the transfer of one, two or three methyl groups to the ε-amine of a lysine side chain, has lagged behind. While the initial findings were focused on the methylation of histone proteins, several studies have recently identified novel non-histone lysine-methylated proteins. This review provides a compilation of all lysine methylation sites reported to date. We also present key examples showing the impact of lysine methylation and discuss the circuitries wired by this important PTM.

Figure 1. PKMT–substrate association maps suggest that lysine methylation is found in complex regulatory networks

Each PKMT or substrate node of the methylation networks is color‐coded according to its functional classification (see Supplementary Table S1). In Eukarya, 34 PKMTs methylate > 65 substrates other than histones. SET7/9 is by far the most promiscuous PKMT targeting close to half of eukaryotic substrates reported to this day. In contrast to eukaryotes, only 8 unique PKMTs have been identified in prokaryotes and 2 in Archaea. Those interactions, together with the 1,018 methylation sites listed in Supplementary Table S1, demonstrate the complexity of this modification and its regulatory potential for the proteome.

Figure 2. Detection of lysine methylation

(A) Most common experimental approaches in target‐specific detection of lysine methylation. Edman degradation and direct detection either by mass spectrometry or by immunoblotting allows for the analysis of in vivo samples. In vitro radiolabeling is commonly used to confirm the PKMT associated to a given site. (B) Recent high‐throughput approaches enabled large‐scale identification of methyl‐lysine proteins. Methylated peptides or proteins can be enriched, either by pan‐methyllysine antibodies or methyl‐binding protein domains. Alternately, proteins can be specifically labeled (isotopically, radioactively) to allow an easier identification of methylated peptides.

Figure 3. Methyllysine residues on canonical histone H2A, H2B, H3 and H4

Bold numbers indicate the methylated residue, italics indicate the organisms in which these modifications are found: At, Arabidopsis thaliana; Bt, Bos taurus; Ce, Caenorhabditis elegans; Dm, Drosophila melanogaster; Dr, Danio rerio; Gg, Gallus gallus; Hs, Homo sapiens; Mm, Mus musculus; Nc, Neurospora crassa; Pb, Paramecium bursaria chlorella virus; Sc, Saccharomyces cerevisiae; Sp, Schizosaccharomyces pombe; Tt, Tetrahymena thermophila. Known methylation states are indicated in parenthesis. A * indicates methyllysine residues modified by an unidentified enzyme.

Figure 4. Molecular mechanisms of lysine methylation

(A) Lysine methylation impacts protein ubiquitination. Numerous instances of methylated lysine residues regulate protein turnover in preventing ubiquitination (see Molecular functions of lysine methylation). (B) Lysine methylation indirectly controls, in cis, deposition of other PTMs. Methyl “switches” are known to positively or negatively regulate the installation of other PTMs on neighboring residues by recruiting other protein‐modifying enzymes or preventing their association with their substrates. (C) Lysine methylation controls protein‐protein interactions (Examples shown in D). (D) Methyllysine residues recruit specific effector proteins. “Readers” such as the chromo, PHD finger and MBT domains can specifically bind methylated lysine residues. In addition to numerous effector proteins able to bind methylated lysine residues located on histone tails (reviewed in Musselman et al 2012), a few examples have been reported for non‐histone substrates. In addition to HP1, the chromodomains of MPP8 and Cbx3 recognize (above) methyllysine residues (green) of non‐histone proteins through residues forming an aromatic cage (blue) (PDB ID 3SVM and 3DM1). In addition, mono‐methylated K382 and di‐methylated K370 of p53 are bound, respectively, by the second MBT repeat of L3MBTL1 and the second Tudor domain of PHF20 (PDB ID 3OQ5 and 2LDM), respectively.