ReviewHistone H2A variants H2AX and H2AZ
Introduction
Eucaroytic cells contain genomes ranging in size from 12×106 base pairs in Saccharomyces cerevisiae to many times larger than the human of 6×109 base pairs. These sizes correspond to combined DNA fiber lengths of 4 mm in the former to >2 meters in the latter. This DNA is integrated into cells ∼0.01 mm in diameter by histone binding and condensation into nucleosomes. The nucleosome comprises 145 bp DNA and eight histones, two from each of the four core histone families—H4, H3, H2B and H2A. A minimum of another 20 bp of DNA stretches between nucleosomes complexed with the linker histone H1. In mammals, the result is that the 4 meters of DNA in a G2 human cell exists as 180 mm of 30 nm diameter fibers and is further condensed to 120 μm of 700 nm diameter arms of mitotic chromosomes.
Although researchers recognized that the complex modification patterns and sequence variations of the histone proteins gave multitudinous opportunities for the regulation of chromosome functions, only recently have the tools and techniques become available to characterize these. Antibodies specific to site-specific modifications coupled with chromatin immunoprecipitation (ChIP) and PCR techniques have made it possible to study the effects of a particular histone modification on a particular gene region in chromatin. Most histone modifications take place on their amino- and carboxy-terminal tails, presumably because these are on the chromatin surface accessible to the prerequisite enzymes. Regions buried in the nucleosome may not be accessible to modifying enzymes and variations in those regions originate at the gene rather than protein level, leading to sequence variants. In this review, we concentrate on the variants of the histone H2A family, in particular H2AX and H2AZ. The term variant can confuse because some sequence differences may not confer a discernible differential function on a protein. Originally the term variant was an operational one, referring to related protein species that were resolvable by a method such as gel electrophoresis in the presence of Triton X-100™ [1]. However, the question was whether these sequence differences were the result of allowable evolutionary diversity for the same protein function, perhaps with a different pattern or timing of expression, or whether the differences conferred a unique function on the protein 2., 3..
While the major interest at the protein level is in the conferring of novel function, the issue is complicated by the coexistence of both types of sequence diversity. With the sequencing of the human genome, the sequences of all the most abundant H2A species are known as well as those of some (and possibly all) of the rare species (Fig. 1). The human genome contains 10 genes that encode for H2A peptides classified as H2A1 variants, six identical in sequence and four that vary in up to three of four positions. These peptides are not resolvable by gel electrophoresis in the presence of Triton X-100™. H2A gene family member O (Oh!) encodes a peptide in which leucine residue 51 has been replaced by methionine; this change confers an altered electrophoretic mobility to the protein which is named H2A2 [2]. The H2A1 and H2A2 variants comprise the bulk of the mammalian H2A, all migrate as a single band on SDS gels and do not have any known differential functions. These proteins have also been known as the major variants because of their abundance in mammals.
There are five other human H2A genes that encode peptides the sequences of which differ considerably from the bulk H2A sequences (Fig. 1). The proteins they encode are present in smaller amounts and have been known as minor variants because of their rarity. However, it is now becoming apparent that the minor variants may have major roles in chromatin metabolism. Two of the minor variants, H2AX and H2AZ [4], were identified in the 1980s, two, macroH2A1 [5] and macroH2A2 6., 7., in the 1990s and one, H2A-Bbd [8], just recently. Several of the variants differ from the consensus sequence at many positions throughout the sequence, whereas one, H2AX, differs primarily in the carboxyl terminus. Two variants, H2AX and H2AZ, are highly conserved as unique H2A species from S. cerevisiae to human. The macroH2A species may have arisen more recently in evolution and appear to be involved in X-chromosome silencing. (The reader is referred to the review by Cohen and Lee in this issue [pp 219–224] for coverage of that topic.) H2A-Bbd has just been reported [8]. We limit this review to describing recent developments in understanding the roles of H2AX and H2AZ.
Section snippets
H2AX has a conserved SQ(E,D)(I,L,F,Y)* motif
It had been surmised on the basis of size and peptide mapping that H2AX contained a carboxy-terminal region longer than those of the bulk H2A species, but it was the isolation of the human H2AX cDNA that revealed a protein with a carboxy-terminal region highly homologous with the H2A species of S. cerevisiae and Schizosaccharomyces pombe [9]. This result suggested that the conserved motif centering on serine four residues from the carboxyl terminus may have a conserved function. As histone
H2AX SQ motif is massively and rapidly phosphorylated in response to DNA DSBs
In mammals, Xenopus laevis, D. melanogaster, and S. cerevisiae, the serine four residues from the carboxyl terminus of an H2A species becomes phosphorylated in response to ionizing radiation and other agents that introduce DNA double-strand breaks (DSBs) 12., 13.. In mammals, the second (S,T)Q motif also becomes phosphorylated to a lesser extent. For simplicity and clarity, because the H2A species with the SQ motif have different names and lengths in different organisms (Fig. 2), this
γ-H2AX forms after accidental DNA DSBs
Different kinases may respond to different stimuli. γ-H2AX is formed in response to replicational stress, induced by hydroxyurea or ultraviolet light [19••]. Only S-phase cells in a culture form γ-H2AX foci under these conditions, whereas all cells in the culture form γ-H2AX foci in response to ionizing radiation. ATM−/− cells show an undiminished γ-H2AX response to hydroxyurea and ultraviolet light. In contrast, ATR ‘kinase dead’ cells show greatly diminished responses to these two agents, but
γ-H2AX forms after programmed DNA DSB
In addition to the environmentally and metabolically caused breaks, programmed DNA DSBs are essential steps in several important eukaryotic processes, such as DNA rearrangement during immune system development in mammals, meiotic recombination during germ cell formation, and programmed cell death by apoptosis. The first process occurs by NHEJ, the second by homologous recombination, and the third by a caspase-activated DNase. Nevertheless, studies have demonstrated γ-H2AX formation in all three.
H2AX in vitro studies
Frog oocytes and eggs contain a large store of histones sufficient for the fertilized egg to develop into a blastula without histone synthesis. The major form of stored H2A is H2AX [24], which can become phosphorylated during nucleosome assembly and replication of DNA added to a cell-free extract prepared from eggs [25]. γ-H2AX formation has been demonstrated on NotI-digested chromatin added to these extracts whereas added undigested nuclei do not induce γ-H2AX formation. However, if the
Possible models
The commonality of γ-H2AX formation in these processes is its tight correlation with the formation and presence of DNA DSBs irrespective of origin. As γ-H2AX forms at the sites of the breaks, a physical signal, such as a DNA end or a chromatin deformation near the site of the break, could initially present a high-affinity site for a PI3 kinase such as ATM or ATR. PI3 kinases recruited thus could then phosphorylate H2AX molecules starting near the break and progressing away from it. Comparing
H2AZ is a necessary but not sufficient H2A
The H2AZ protein was identified in mammalian cells as an H2A variant in 1980 by gel and tryptic peptide analysis [4] and shown to be present in nucleosomes [27]. The cDNA for a chicken variant H2AF was isolated in 1985 [28]; when the mammalian H2AZ cDNA sequence was reported [29], it was found that these two proteins were homologs— hence H2AF/Z. In those organisms where they have been determined, H2AZ protein levels are ∼10% of the total H2A complement. H2AZ provides an essential function to
H2AZ essential motif
Chimeric constructs injected into D. melanogaster H2AZ null embryos showed that the essential portion of H2AvD is the αC helix and H3/H4-binding domains (Fig. 1) [33]. Similar chimeric constructs restore the wild-type phenotype to S. cerevisiae [31•]. However, other studies show that H2AZ may serve functions other than just providing an H2A–H2B dimer with an altered interface to the H3–H4 tetramer. In T. thermophila, changing the six lysine residues in the H2AZ amino-terminal to arginines is
H2AZ structural and physical studies
The crystal structure of H2AZ-containing nucleosomes has been determined at 2.6Å [39••]. It shows a core particle similar to that with H2A but with some intriguing differences. The L1 loop domain, which helps mediate interactions between H2A molecules in the same nucleosome, is altered in such a way that may favor binding to another H2AZ, resulting in nucleosomes with two H2AZ molecules. Secondly the tetramer docking domain, where the H2AZ–H2B dimer interfaces with the H3–H4 tetramer, has a
H2AZ functional studies
Several studies have focussed on H2AZ functions in S. cerevisiae. In silencing studies, strains were generated in which a mutant Sir1 protein can be recruited to HMR silencers but once there cannot silence [41•]. A screen for overexpressed proteins that could overcome this defect yielded H2AZ. Deletion of H2AZ by itself resulted in only a partial derepression of silencing at HMR (ADE2 reporter), but a profound loss of telomeric silencing of a URA3 gene. Such variability in silencing effects
Conclusions
It is now accepted that histones not only package DNA but also monitor whether the package is intact, and when and where it is to be opened and closed. Although histones are not necessary to study the basic chemical pathways in transcription and replication in vitro, understanding histones is essential to understanding how those processes are harnessed in a eucaryotic cell.
Insight into the role of histone H2AX in eucaryotes is likely to come from studies on organisms in which the H2AX has been
Update
Mice have been produced that lack H2AX; the mice survive but are impaired in several processes of which impairments in class switch recombination has been documented [44••]. Class switch recombination is a region-specific recombination that replaces one immunoglobulin heavy-chain constant region with another. B cells from H2AX−/− mice have 50–86% reduction in surface IgG1 levels compared to their normal littermates. The results with CSR do not imply that H2AX−/− mice are not deficient in V(D)J
References and recommended reading
Papers of particular interest, published within the annual period of review,have been highlighted as:
• of special interest
•• of outstanding interest
References (45)
Resolution of histones by polyacrylamide gel electrophoresis in presence of nonionic detergents
Methods Cell Biol
(1978)- et al.
MACROH2A2, a new member of the MARCOH2A core histone family
J Biol Chem
(2001) - et al.
DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139
J Biol Chem
(1998) - et al.
A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage
Curr Biol
(2000) - et al.
ATM phosphorylates histone H2AX in response to DNA double-strand breaks
J Biol Chem
(2001) - et al.
Nuclear retention of Atm at sites of DNA double strand breaks
J Biol Chem
(2001) - et al.
Histone H2AX is phosphorylated in an ATR- dependent manner in response to replicational stress
J Biol Chem
(2001) - et al.
Initiation of DNA fragmentation during apoptosis induces phosphorylation of H2AX histone at serine 139
J Biol Chem
(2000) - et al.
Two complexes that contain histones are required for nucleosome assembly in vitro: role of nucleoplasmin and N1 in Xenopus egg extracts
Cell
(1987) - et al.
Mre11 protein complex prevents double-strand break accumulation during chromosomal DNA replication
Mol Cell
(2001)