This week's seminar at Birkbeck was given by another alumnus of the School of Crystallography, Wyatt Yue, who studied for his PhD here from 1999-2003 working with Susan Buchanan. Wyatt is now at the Structural Genomics Consortium at Oxford University, but the work he presented was all done when he was a post-doc with Lawrence Pearl at the Institute of Cancer Research in London. He described some elegant work on the structure and function of one member of a family of enzymes that catalyse the addition of methyl groups to arginine residues in histones.
Histones are proteins that are involved in the coiling and compacting of DNA, so the long DNA molecules fit into the cell structures. In eukaryotes, DNA is first wound round an assembly of eight histone molecules to form a structure called a nucleosome. The DNA then resembles beads on a string; simplistically, this structure is further compacted by coiling into a chromatin fibre, from which the chromosome structures are formed. (Links here are to Wikipedia.) The histone molecules have N-terminal "tails" which protrude from the folded structure and are typically not seen in crystal structures. Residues on these tails can be chemically modified by e.g. phosphorylation, acetylation and methylation. These modifications are dynamic and form the so-called "histone code" which is one of the processes that control gene expression.
One of the most important of these changes is the methylation of nitrogen atoms on the side chain of arginine residues; this process is catalysed by a family of enzymes called protein arginine methyltransferases (PRMTs). Six members of this family are found in mammals; the protein that Wyatt has been working on is PRMA4, otherwise known as coactivator-associated methyltransferase 1 (CARM1). All these proteins share a common, catalytically active core domain; CARM1 is the only member with an additional C-terminal domain. The structure of the core is known (PDB 2oqb); it has two domains, an N-terminal Rossman fold, with two extra helices, and a C-terminal beta barrel with an insertion of a helix-turn-helix motif known as the arm. Core structures of two other PRMT proteins have also been known; compared to the others, the CARM1 core has a piece of ordered structure at the end of its C-terminal beta strand (the beginning of the C-terminal domain) and the helices of the arm are longer. Therefore, the cavity between the monomers of the crystallographic dimer is larger, leaving space for the C-terminal domains.
Lawrence Pearl's group, including Wyatt, has now solved the structure of the core domain bound to its co-factor, S-adenosyl methionine (PDB 2V74). This has revealed that it is only the co-factor binding that creates the cavity into which the arginine substrate is bound, indicating that the co-factor must bind first. This binding order has also been demonstrated in kinetic studies. The arginine pocket is close to the methyl group that is transferred during catalysis and is lined by negatively charged residues to attract the positively charged arginine.
CARM1 methylates three arginine residues in the N-terminal tail of the histone H3: R2, R17 anad R26. However, the mechanism of each methylation is not exactly the same. Studies with chimeras have shown that the pre-core region is necessary for methylation of R26 but not R17; Arg 17 methylation is faster if the neighbouring residue Lys 18 is acetylated.
You can read more about this work in the publication: Yue et al. (2007), EMBO J. 26(20): 4402-12. (Link to PubMed). The online version of EMBO Journal is available in the Birkbeck e-libarary.