National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health, Bethesda, Maryland, USA.
Prog Mol Biol Transl Sci. 2011;101:105-76. doi: 10.1016/B978-0-12-387685-0.00004-4.
In eukaryotes, methylation of nucleosomal histones and other nuclear proteins is a central aspect of chromatin structure and dynamics. The past 15 years have seen an enormous advance in our understanding of the biochemistry of these modifications, and of their role in establishing the epigenetic code. We provide a synthetic overview, from an evolutionary perspective, of the main players in the eukaryotic chromatin protein methylation system, with an emphasis on catalytic domains. Several components of the eukaryotic protein methylation system had their origins in bacteria. In particular, the Rossmann fold protein methylases (PRMTs and DOT1), and the LSD1 and jumonji-related demethylases and oxidases, appear to have emerged in the context of bacterial peptide methylation and hydroxylation systems. These systems were originally involved in synthesis of peptide secondary metabolites, such as antibiotics, toxins, and siderophores. The peptidylarginine deiminases appear to have been acquired by animals from bacterial enzymes that modify cell-surface proteins. SET domain methylases, which display the β-clip fold, apparently first emerged in prokaryotes from the SAF superfamily of carbohydrate-binding domains. However, even in bacteria, a subset of the SET domains might have evolved a chromatin-related role in conjunction with a BAF60a/b-like SWIB domain protein and topoisomerases. By the time of the last eukaryotic common ancestor, multiple SET and PRMT methylases were already in place and are likely to have mediated methylation at the H3K4, H3K9, H3K36, and H4K20 positions, and carried out both asymmetric and symmetric arginine dimethylation. Inference of H3K27 methylation in the ancestral eukaryote appears uncertain, though it was certainly in place a little later in eukaryotic evolution. Current data suggest that unlike SET methylases, which are universally present in eukaryotes, demethylases are not. They appear to be absent in the earliest-branching eukaryotic lineages, and emerged later along with several other chromatin proteins, such as the Dot1-methylase, prior to divergence of the kinetoplastid-heterolobosean lineage from the remaining eukaryotes. This period also corresponds to the point of origin of DNA cytosine methylation by DNMT1. Origin of major lineages of SET domains such as the Trithorax, Su(var)3-9, Ash1, SMYD, and TTLL12 and E(Z) might have played the initial role in the establishment of multiple distinct heterochromatic and euchromatic states that are likely to have been present, in some form, through much of eukaryotic evolution. Elaboration of these chromatin states might have gone hand-in-hand with acquisition of multiple jumonji-related and LSD1-like demethylases, and functional linkages with the DNA methylation and RNAi systems. Throughout eukaryotic evolution, there were several lineage-specific expansions of SET domain proteins, which might be related to a special transcription regulation process in trypanosomes, acquisition of new meiotic recombination hotspots in animals, and methylation and associated modifications of the diatom silaffin proteins involved in silica biomineralization. The use of specific domains to "read" the methylation marks appears to have been present in the ancestral eukaryote itself. Of these the chromo-like domains appear to have been acquired from bacterial secreted proteins that might have a role in binding cell-surface peptides or peptidoglycan. Domain architectures of the primary enzymes involved in the eukaryotic protein methylation system indicate key features relating to interactions with each other and other modifications in chromatin, such as acetylation. They also emphasize the profound functional distinction between the role of demethylation and deacetylation in regulation of chromatin dynamics.
在真核生物中,核小体组蛋白和其他核蛋白的甲基化是染色质结构和动态的核心方面。在过去的 15 年中,我们对这些修饰的生物化学及其在建立表观遗传密码中的作用有了巨大的认识。我们从进化的角度对真核染色质蛋白甲基化系统的主要参与者进行了综合概述,重点介绍了催化结构域。真核蛋白甲基化系统的几个组成部分起源于细菌。特别是,罗斯曼折叠蛋白甲基转移酶(PRMT 和 DOT1)以及 LSD1 和与 jumonji 相关的去甲基酶和氧化酶,似乎是在细菌肽甲基化和羟化系统的背景下出现的。这些系统最初涉及肽类次生代谢物的合成,如抗生素、毒素和铁载体。肽酰精氨酸脱亚氨酶似乎是动物从修饰细胞表面蛋白的细菌酶中获得的。SET 结构域甲基转移酶,显示出β-夹折叠,显然首先在原核生物中从碳水化合物结合结构域的 SAF 超家族中出现。然而,即使在细菌中,SET 结构域的一部分也可能与 BAF60a/b 样 SWIB 结构域蛋白和拓扑异构酶一起进化出与染色质相关的作用。在上一个真核生物共同祖先出现的时候,已经有了多种 SET 和 PRMT 甲基转移酶,并且很可能在 H3K4、H3K9、H3K36 和 H4K20 位置进行甲基化,并进行不对称和对称的精氨酸二甲基化。在祖先真核生物中推断 H3K27 甲基化似乎不太确定,尽管在真核生物进化的后期它肯定已经存在。目前的数据表明,与普遍存在于真核生物中的 SET 甲基转移酶不同,去甲基酶并非如此。它们似乎在最早分支的真核生物谱系中不存在,并且在与其余真核生物分离的动基体-异变形生物谱系之后,与其他几个染色质蛋白一起出现,如 Dot1-甲基转移酶。这一时期也对应于 DNMT1 对 DNA 胞嘧啶甲基化的起源点。Trithorax、Su(var)3-9、Ash1、SMYD 和 TTLL12 和 E(Z)等主要 SET 结构域家族的起源可能在建立多个不同的异染色质和常染色质状态方面发挥了初步作用,这些状态可能以某种形式存在于真核生物进化的大部分过程中。这些染色质状态的细化可能与多个 jumonji 相关和 LSD1 样去甲基酶的获得以及与 DNA 甲基化和 RNAi 系统的功能联系同时进行。在真核生物进化过程中,SET 结构域蛋白发生了几次谱系特异性扩张,这可能与锥虫中特殊的转录调节过程有关,在动物中获得了新的减数分裂重组热点,以及与涉及硅生物矿化的硅藻丝氨酸蛋白的甲基化和相关修饰有关。在祖先真核生物中,似乎已经存在使用特定结构域“读取”甲基化标记的情况。其中,类染色质结构域似乎是从细菌分泌蛋白中获得的,这些蛋白可能在结合细胞表面肽或肽聚糖方面发挥作用。真核生物蛋白甲基化系统中主要酶的结构域架构表明了与彼此以及与染色质中其他修饰(如乙酰化)相互作用的关键特征。它们还强调了去甲基化和去乙酰化在调节染色质动力学方面的功能区别。