Beyond A, C, T, and G lies an epigenetic level of heritable control that has reached the forefront of biomedical research. Histone modification patterns, like multicolored highlighting in the genome's book of life, tell eukaryotic cells when to turn genes on and off, thus making sense of its DNA instructions. In 1993, researchers first pondered whether these alterations on N-terminal tails of histone proteins created a readable code1 (See also 5-Prime). Eight years later, groups from Austria and the United Kingdom wrote what many consider as the first complete chapter on the histone code hypothesis.2,3
Using mammalian cells and fission yeast, these two groups built on findings that histone methyltransferase SU(VAR)3-9 and its yeast homolog, Clr4, specifically methylate the ninth lysine (K9) on histone H3's tail. This methylated lysine, and none other, acts as a binding site for heterochromatin protein HP1, which was long associated with silent heterochromatic regions of the genome. "To put it to you as simply as I [can]," says University of Virginia's C. David Allis, who first coined the term histone code in 2000,1 "something is writing this code and something is reading this code."
Myriad details of SU(VAR)3-9 the writer and HP1 the reader, each well known in the Drosophila community, clued scientists into the mechanisms of heterochromatin formation, maintenance, and inheritance. Yet, plot twists continue to surface, guaranteeing additional chapters.
Courtesy of Dmitry Pruss
Bryan M. Turner, University of Birmingham Medical School, one of the first to hypothesize about histone-tail modifications regulating expression1 says, "To my mind, the story that's going to come out is that K9 methylation is involved in HP1 binding, but it's part of a binding platform that HP1 recognizes, which will involve other things as well." Indeed, stacks of articles on histone methylation implicate other methyltransferases and binding partners, a connection with DNA methylation, and even a role for RNA interference (RNAi) in silencing some genomic areas. Still, little is known about what prompts these chromatin signals and how, if at all, they are removed and modified.
RAISING A FEW EYEBROWS The scopes came out in 2001, when scientists observed methyltransferase activity of the SET-domain protein SU(VAR)3-9 in Drosophila and its mammalian and yeast counterparts SUV39H1 and Clr4.4 Both SU(VAR)3-9 and HP1 were associated with silent hetero-chromatin. Thomas Jenuwein's group at the Research Institute of Molecular Pathology in Vienna examined HP1 binding both biochemically and in mutant murine fibroblasts. HP1 consistently bound methylated lysine 9 but not, for example, methylated lysine 4--a nearby mark associated with active euchromatin. Tony Kouzarides' group at the University of Cambridge also did studies in the fission yeast Schizosaccharomyces pombe, testing the HP1 yeast homolog Swi6. Marker genes imbedded in usually silent centromeric heterochromatin switched on without Clr4.
Both groups proposed the same mechanism by which lysine 9 methylation would propagate itself. A second active domain in HP1 appears to recruit SU(VAR)3-9, meaning that when one HP1 docks on a histone tail, the methyltransferase is already there to mark the next tail, or perhaps the histones around which newly-replicated DNA wraps. "That suggests a mechanism for epigenetic events to be passed on to the next generation," says Kouzarides.
This, says Allis, raised many an eyebrow. "The theme of an enzyme ... carrying in its body, literally, its own Velcro® patch for its modification was very exciting in the field."
But, the long histories of the proteins involved have made some wary of easy answers. Though he agrees with the findings, Turner says these results were "pushing against an open door." Relying on antibody specificity can lead to problems, he says, and no clear demonstration has shown the extent to which lysine-9 methylation is a marker for heterochromatin. The story was ripe for the picking in the literature. "They've got a set of results which I guess are exactly what everybody wanted," says Turner.
As evidence that researchers don't yet have the full picture, Turner points to more recent findings that RNAi may set up this silencing pathway5 and that the inactive X chromosome in mammalian females is methylated at H3 lysine 9, but independently of SU(VAR)3-9 or HP1.6 Good structural studies show the binding can happen, Turner says.7,8 "[But] if you go with the simplistic approach, you lose a lot of interesting stuff."
Reprinted with permission from Elsevier Science
Allis agrees that RNAi's role is poorly understood. Judging from the data in his lab and elsewhere, it appears that small double-stranded RNAs mark the genome in a way that "establishes that this is the chromatin that needs to be silenced." As for the differences in inactive-X silencing, Allis suggests the process might require an X-chromosome-specific protein similar to HP1, or a different histone methyltransferase, or both. "I think only time will tell on that one."
METHYLS, METHYLS EVERYWHERE Another level of epigenetic control further complicates results. A group in Oregon published observations on fungi that histone methylation is required for DNA methylation, which is also associated with gene silencing9 Kouzarides' group recently observed just the opposite--that DNA methylation leads to lysine-9 methylation.10 Kouzarides suggests that this is indicative of a positive feedback loop. "The DNA methylase ... brings in, indirectly through methyl-binding proteins, the lysine-9 methylase. Which brings the DNA methylase in. So, the loop is closed and that gene is completely off. It can't be switched on again." Such a feedback loop, if necessary, might suggest a more transient methyl mark than previously envisioned, hence obviating the need for the elusive histone demethylase. "I think that is the hottest area at the moment.... Is there a demethylase?" says Kouzarides. Says Allis: "We've struck out on that completely."
As mysteries continue to envelop SU(VAR)3-9, other histone methyltransferases are attracting attention. Enhancer of Zeste, E(Z), is part of a Drosophila polycomb group complex associated with gene silencing during development. E(Z), like SU(VAR)3-9, contains a SET domain, and appears to methylate H3 at lysine 27 and possibly at lysine 9.11,12 Interest also lies around H3 lysine-4 methylation and its possible link to MLL leukemia. MLL is a human homolog of the Drosophila gene Trithorax, another SET-domain containing methyltransferase.13 Says Allis, whose group worked on the MLL research, "I think it's now remarkably hot stuff because we've got almost a 30-year grip on these Drosophila gene products that are conserved from fly to man [and] that are probably potent epigenetic regulators."
These are only three of at least 13 known modifications that can be made to histone H3. Account for the other three types of histones that form the core of a single nucleosome, and that number is near 30. Even amid this overwhelming complexity, scientists are unraveling the story of the histone code and how it plays out in biological systems and human disease. Says Jenuwein in an E-mail: "Based on these mechanistic insights, the field of 'Epigenetics' is on an unprecedented high."
Brendan A. Maher can be contacted at firstname.lastname@example.org.
1. B.A. Maher, "Researchers focus on histone code," The Scientist, 15:15, 2001.
2. M. Lachner et al., "Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins," Nature, 410:116-20, 2001. (Cited in 202 papers)
3. A.J. Bannister et al., "Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain," Nature, 410:120-4, 2001. (Cited in 224 papers)
4. S. Rea et al., "Regulation of chromatin structure by site-specific histone H3 methyltransferases," Nature, 406:593-9, 2001.
5. T.A. Volpe et al., "Regulation of heterochromatic silencing and histone H3 lysine-9 methylation by RNAi," Science, 297:1833-7, Sept. 13, 2002.
6. A.H. Peters et al., Histone H3 lysine 9 methylation is an epigenetic imprint of facultative heterochromatin," Nat Genet, 30:77-80, 2002.
7. S.A. Jacobs et al., "Structure of HP1 chromodomain bound to a lysine 9-methylated histone H3 tail," Science, 295:2080-3, 2002.
8. P.R. Neilsen et al., "Structure of the HP1 chromo-domain bound to histone H3 methylated at lysine 9," Nature, 416:103-7, 2002.
9. H. Tamaru et al., "A histone H3 methyltransferase controls DNA methylation in Neurospora crassa," Nature, 414:277-83, 2001.
10. F. Fuks et al., "The methyl-CpG-binding protein MeCP2 links DNA methylation to histone methylation," J Biol Chem, Nov. 9, 2002, advance online publication, DOI 10.1074/jbc.M210256200.
11. B. Czermin et al., "Drosophila enhancer of Zeste/ESC complexes have a histone H3 methyltransferase activity that marks chromosomal polycomb sites," Cell, 111:185-96, Oct. 18, 2002.
12. J. Muller et al., "Histone methyltransferase activity of a Drosophila polycomb group repressor complex," Cell, 111:197-208, Oct. 18, 2002.
13. T.A. Milne et al., "MLL targets SET domain methyltransferase activity to hox gene promoters," Molec Cell, 10:1107-17, November 2002.