Human Molecular Genetics, 2001, Vol. 10, No. 10 1101-1113
© 2001 Oxford University Press
Histone H2A variants and the inactive X chromosome: identification of a second macroH2A variant
Department of Genetics, Case Western Reserve University School of Medicine and Center for Human Genetics and Research Institute, University Hospitals of Cleveland, 10900 Euclid Avenue, Cleveland, OH 44106-4955, USA
Received 13 February 2001; Revised and Accepted 15 March 2001.
DDBJ/EMBL/GenBank accession nos AF336304 and AF336305.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
MacroH2A1 is an unusual variant of the core histone H2A which is enriched in chromatin on the inactive X chromosome of female mammals. The N-terminal third of the protein shares 65% amino acid identity with core histone H2A, while the remaining two-thirds of the protein are novel, with a small stretch of basic amino acids and a putative leucine zipper motif. We have now cloned a second macroH2A gene, encoding macroH2A2 which shares 80% amino acid identity with macroH2A1. Despite mapping to different chromosomes, the genomic organization of the macroH2A2 and macroH2A1 genes are nearly identical. The leucine zipper motif of macroH2A1 is not conserved in macroH2A2. Like macroH2A1, macroH2A2 forms a Macro Chromatin Body in the nuclei of female cells which is coincident with an X chromosome and co-localizes with macroH2A1. To address the distribution of other histone H2A variants in relation to macroH2A and the inactive X chromosome, we constructed a series of epitope-tagged versions of other histone H2A variants. Like the recently described H2A-Bbd (Barr body-deficient) variant, the histone variant H2A.Z was found to be deficient in chromatin on the inactive X chromosome in a significant proportion of female nuclei. This study provides further information about the nucleosomal composition of chromatin on the inactive X chromosome and indicates that a number of H2A variants are non-randomly distributed on the active and inactive X chromosomes.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Nucleosomes are composed of two molecules each of core histones H2A, H2B, H3 and H4 (1) and are dynamic structures that directly influence the transcriptional status of local chromatin environments (2,3). All histones contain a core structural motif referred to as the histone fold domain (4), which is flanked by N- and C-terminal tails that were originally identified by protease treatment (5). Histone tails are the targets of several forms of covalent modification including methylation, ubiquitination, phosphorylation, acetylation and ADP-ribosylation. Each tail contains multiple modification sites, and each form of modification can confer different physical characteristics on a nucleosome, reflecting a major role in regulating nucleosome conformation and function (6).
In addition to covalent modification, a number of variants of the core histones have been identified that can likely substitute for their core histone counterparts and further influence nucleosomal conformations in specific biological contexts (7). To date, more variants have been identified for the core histone H2A than for any other histone type. The human core histone H2A gene family consists of over a dozen intronless H2A genes, most of which are organized into a gene cluster with other histone genes at 6p21.3–p22 (8). Core H2A genes are replication-dependent with expression restricted to the S phase of the cell cycle. In addition, unlike a majority of transcripts, histones are not polyadenylated but the 3'-end of the transcript is processed via an alternative method (9,10).
The first H2A variants to be identified were H2A.X (11), H2A.Z (12) and hv1 (13). H2A.X shares 95% amino acid identity with core H2A and also shares several characteristic features with replication-dependent H2A genes. Although located at chromosome 11q23, outside of the major histone gene cluster (14), the H2A.X gene is intronless with signals for both polyadenylation and histone 3'-end processing (11). H2A.X has recently been implicated in recruiting DNA repair factors to sites of DNA damage (15–17), a feature that involves the phosphorylation of a novel motif at the C-terminus (18,19). H2A.Z shares 63% amino acid identity with core H2A, and its gene is organized into five exons at chromosome 4q24 (20,21). In Saccharomyces cerevisiae, the H2A.Z homolog has been implicated in gene silencing (22) and in the regulation of transcription in a redundant fashion with nucleosome remodeling complexes (23). The crystal structure of a nucleosome containing H2A.Z (24) is almost completely identical to the previously described crystal structure of a nucleosome containing core H2A (25). Nonetheless, some changes were identified that apparently destabilize the H2A.Z–H2B heterodimer interaction with the H3–H4 core and, in addition, a metal ion is found to associate with the nucleosome. These subtle alterations are likely to be sufficient for alteration of nucleosome conformation to facilitate or inhibit protein factor access to DNA. The H2A variant hv1 was identified through its enrichment in macronuclei of Tetrahymena thermophilia (13). This protein more closely resembles H2A.Z (84% identity) than core H2A (60% identity). Although this variant associates with regions of active transcription, a direct human homolog, other than H2A.Z, is not obvious.
Probably the most intriguing of the H2A variants reported to date is macroH2A1. MacroH2A1 was originally identified as a protein tightly associated with nucleosomes (26). The N-terminal third of the protein shares 65% amino acid identity with conventional H2A with a large C-terminal tail of unknown function that makes up two-thirds of the protein size. MacroH2A1 has a generalized nuclear distribution in male nuclei, while in female nuclei there is, in addition, a densely staining accumulation of macroH2A1, which is referred to as a Macro Chromatin Body (MCB). This sexual dimorphism led to the finding that macroH2A1 is enriched on the inactive X chromosome (27). X-inactivation is the mammalian dosage compensation mechanism, which achieves comparable levels of X-linked gene expression between males and females (28,29). The inactive X chromosome is heterochromatic and can often be observed as a densely staining mass at the periphery of human nuclei; a structure referred to as the Barr body (30). The association of macroH2A1 with the inactive X chromosome is dependent upon the continued presence of the X-inactive specific transcript (XIST) (31), the large untranslated RNA that associates in cis along the length of the inactive X (32–34). Although reported to be physically associated with XIST RNA (35), a specific role for macroH2A1 in X inactivation has not yet been determined.
We recently identified a novel variant of histone H2A, called H2A-Bbd (Barr body-deficient) (36). H2A-Bbd is slightly smaller than conventional H2A and is the most diverged of the variants reported to date, with only 47% amino acid identity to core H2A. Immunolocalization studies demonstrated that the inactive X chromosome was largely deficient for H2A-Bbd in an almost mutually exclusive distribution with macroH2A1. As part of studies to define the composition of X chromosome chromatin, here we describe the identification, organization and nuclear distribution of a second macroH2A gene in man and mouse called macroH2A2 and describe the results of a study of H2A variant distribution in relation to the inactive X chromosome.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Cloning of human and mouse macroH2A2
Using the nucleotide sequence of human macroH2A1, we searched the public databases and identified a group of expressed sequence tags (ESTs) that were highly similar, but not identical to macroH2A1. A representative sample of cDNA clones was obtained and completely sequenced (accession no. AF336304). An open reading frame was identified extending from nucleotides 188 to 1306. As with macroH2A1, the cDNA clone was polyadenylated. The cDNA encodes a predicted 372 amino acid protein, with a molecular weight of 40.1 kDa and a pI of 9.7. An alignment of the predicted protein sequence with macroH2A1 revealed 80% amino acid identity, and we have therefore named the protein macroH2A2. A comparison of the full cDNA sequence of macroH2A2 with entries in the human EST database revealed identical ESTs from a wide range of tissues, indicating that macroH2A2 is likely to be ubiquitously expressed. In order to identify a mouse homolog of macroH2A2, the human sequence was compared to entries in the mouse EST databases. Two representative clones were obtained and completely sequenced (accession no. AF336305). An open reading frame was identified from nucleotides 142 to 1260 that encodes a protein of 372 amino acids with a predicted molecular weight of 40.1 kDa and a pI of 9.7, sharing 98% amino acid identity with human macroH2A2.
Figure 1 shows an alignment of human and mouse macroH2A2 with human and mouse macroH2A1. The core histone domain is the most conserved region between the macroH2A proteins at 84% identity (Fig. 1). This region of histone H2A is under the greatest evolutionary constraint in amino acid composition, as demonstrated by the high level of sequence identity seen between different organisms for the individual histone subtypes.
|
MacroH2A1 contains a basic domain directly after the core histone region and a putative leucine zipper motif (Fig. 1). Although the basic domains of both macroH2A1 and macroH2A2 contain nearly 50% basic amino acids, the location of each residue is not conserved. If the region simply serves as a patch of basic character, the site of each residue may not be crucial for function. The C-terminal portion of the putative leucine zipper motif is conserved between two alternatively spliced forms of macroH2A1, designated macroH2A1.1 and macroH2A1.2 (26), and between the rat and chicken macroH2A1.1 and macroH2A1.2 isoforms (37). In the case of human and mouse macroH2A2 proteins, only three out of the four residues necessary for a functional leucine zipper motif are hydrophobic, suggesting that, at least in the case of macroH2A2, this region of the protein is not involved in protein–protein interactions via a leucine zipper motif.
Outside of the core histone region, macroH2A2 shares highest amino acid identity with the macroH2A1.2 splice isoform (68%). The highest region of conservation within the macroH2A tail is within the C-terminal third of the protein.
The human macroH2A1 and macroH2A2 genes map to different chromosomes
The human macroH2A1 gene (official gene symbol H2AFY) was mapped to chromosome 5q21–q31 with a LOD score of >21 by PCR typing of the Genebridge 4 radiation Hybrid Mapping Panel (Materials and Methods). The location of macroH2A1 on human chromosome 5 was confirmed by PCR analysis of monochromosomal mapping panels. A BAC clone containing the 3'-end of macroH2A1 hybridized to chromosome 5q31 by fluorescence in situ hybridization (FISH) (data not shown).
A comparison of the human macroH2A2 cDNA sequence with entries in the working draft of the human genome identified two overlapping BAC entries (RP11-367H5 and RP11-379O18) that contained the complete macroH2A2 gene (official gene symbol H2AFY2). Both BAC clones mapped to chromosome 10, and contain 10 sequence tag sites (STSs) that map to chromosome 10q22.3, close to genetic marker D10S537. This conclusively places the macroH2A2 gene on a chromosome separate from macroH2A1, ruling out the possibility of a local gene duplication event.
The genomic organization of human macroH2A1 and macroH2A2 suggests recent evolution from a common ancestor
Alignment of the cDNA sequences of macroH2A1 and macroH2A2 with their corresponding genomic sequences allowed elucidation of the genomic structure of both genes. The macroH2A1 gene consists of eight coding exons, one non-coding upstream exon and two alternative exon 6’s resulting in either macroH2A1.1 or macroH2A1.2. All splice donor and acceptor sites conform to the GT/AG rule (38). The macroH2A1 gene covers 
65.5 kb in 5q31. The genomic organization of the human macroH2A1 gene is almost identical to the previously characterized mouse macroH2A1 locus in the syntenic region at mouse chromosome 13 (39). The exception is the boundary of exon 2, which is 5 bp further 3' in mouse, and consequently the acceptor boundary of exon 3 is 5 bp further downstream. Human isoform macroH2A1.2 exists as two different lengths in the EST database, resulting in either a 371 or 372 amino acid protein with a lysine at residue 160. Analysis of the exon acceptor site of exon 5 shows the sequence aagaagAAG, where the capital letters represent the lysine codon shared by both protein forms. The presence of three aag codons at an exon boundary allows for choice of either the first or the second ag acceptor sequence (indicated in bold), resulting in an alternative splice acceptor site. Analysis of mouse ESTs in the public databases also reveals two groups of transcript that display an alternative splice acceptor site at exon 5.
The genomic organization of the human macroH2A2 gene was determined by comparison of the cDNA sequence to the genomic BAC sequence. MacroH2A2 is composed of eight coding exons and one non-coding upstream exon. With the exception of the alternative splice acceptor site at the boundary of exon 5 of human macroH2A1, all exon–intron boundaries are conserved for macroH2A2. However, the alternative splice acceptor site at the edge of exon 5 is not conserved. As with macroH2A1, all splice donor and acceptor sites conform to the GT/AG rule (38). Figure 2 shows a schematic representation of the genomic organization of macroH2A1.1 and macroH2A 1.2 compared to macroH2A2. The conservation of exon–intron boundaries is intriguing and suggests that both genes originate from a common ancestor. However, the intron sizes between conserved exons vary significantly. Exon 6 is alternatively spliced in humans to obtain either the 1.1 or 1.2 isoforms (Fig. 2) (37). Exon 6 of macroH2A2 most resembles macroH2A1.2 in length (100 bp for 1.2 and 2, compared to 91 bp for 1.1) and at the coding level (48% between 1.2 and 2, versus 27% between 1.1 and 2). The BAC sequence for macroH2A2 was used with the GENSCAN exon prediction server to predict any additional exons in the exon 5–7 interval. No additional exons were predicted for macroH2A2, suggesting that no macroH2A1.1-like splice variants exist. In addition, no macroH2A2 EST entries contain alternative sequences corresponding to the exchange of exon 6.
|
MacroH2A2 forms a MCB in female nuclei and is coincident with macroH2A1
To investigate the subcellular localization of macroH2A2, a green fluorescent protein (GFP) fusion to the C-terminus of macroH2A2 was generated and transfected into female cells. As with macroH2A1-GFP, macroH2A2-GFP was distributed throughout the nucleus with a single region of enrichment, or MCB, that is coincident with the Barr body evident by 4,6-diaminidino-2-phenylin-dole (DAPI) staining (Fig. 3). To confirm the coincidence of macroH2A2 with an X chromosome, a C-terminal Myc epitope-tagged version of macroH2A2 was generated and transfected into female cells. An X chromosome-specific
-satellite probe was used with FISH to identify the two X chromosomes in the transfected cells. Figure 3C and D shows a clear coincidence of an X -satellite signal for both macroH2A1-Myc- and macroH2A2-Myc-transfected cells. To investigate the spatial relationship of macroH2A1 with macroH2A2, the GFP-tagged macroH2A2 was transfected into a macroH2A1-Myc stably transfected 293 cell line, which contains a single active X chromosome and a variable number of inactive X chromosomes (1–4 copies in different cells). Figure 4 clearly indicates a complete overlap of macroH2A1 and macroH2A2 signals, indicating that the two proteins are coincident with the inactive X chromosome.
|
|
Antisera generated to the non-histone region of macroH2A1 detect a protein of expected size in total cell extracts by western analysis (Fig. 5A). The same signal is detected in a preparation of micrococcal nuclease-digested chromatin that was separated by sucrose gradient ultracentrifugation, confirming specificity for a 40 kDa nucleosomal protein. To determine whether macroH2A2, like macroH2A1, is a nucleosomal protein, nucleosomes were isolated from a cell line stably transfected with macroH2A2-CT-Myc. Figure 5C demonstrates the cofractionation of macroH2A2 with nucleosomes.
|
According to the Lyon hypothesis (40) and the known behavior of Barr bodies, we expected to observe one less MCB than the total number of X chromosomes in a nucleus (the ‘n–1’ rule). We tested this prediction in 46,XY, 46,XX, 47,XXX, 49,XXXXY and 49,XXXXX cell lines. As predicted, we demonstrated a direct correlation between the number of X chromosomes and ‘n–1’ MCBs (Fig. 6). Each MCB coincides with a Barr body. Table 1A shows the results of 50 nuclei for each cell line that were scored for the number of MCBs detected using the macroH2A1 antisera. Table 1B and C show that transiently transfected macroH2A1-GFP and macroH2A2-GFP behave in an almost identical fashion to the endogenous protein and that both proteins are equally efficient at associating with the inactive X chromosome.
|
|
Other histone H2A variants and the inactive X chromosome
As shown above, the macroH2A histone variants are enriched in chromatin of the inactive X chromosome; in contrast, the histone variant H2A-Bbd is almost completely excluded from the inactive X and from the Barr body (36). To investigate whether this non-random distribution was a feature of other H2A variants, or was a feature specific to these particular H2A variants, we cloned and Myc epitope-tagged H2A.X and H2A.Z (Fig. 7). As controls, we also cloned a core histone H2B gene and a replication-independent polyadenylated H2A gene from chromosome 12p which shares 96% amino acid identity with the family of core H2A genes (a level of divergence that is representative of the sequence variation observed between the core H2A family: 96–100%). No H2B variants have yet been identified and, as expected, we observed a uniform nuclear distribution (data not shown). Each H2A variant was transfected into female 46,XX cells, and the H2A-Myc, H2A-Bbd-Myc, H2A.X-Myc and H2A.Z-Myc constructs were counterstained for macroH2A by indirect immunofluorescence to identify the position of the inactive X chromosome. Table 2 shows the results of this survey. Only macroH2A1-Myc localized to MCBs, at a level comparable to the endogenous and GFP-tagged proteins (Table 1). Confirming previous results (36), the distribution of H2A-Bbd-Myc was consistently markedly deficient for the inactive X chromosome (Table 2), in a mutually exclusive distribution with macroH2A (Fig. 8B). To extend the previous observations, H2A-Bbd-Myc was transfected into the 46,XY, 46,XX, 47,XXX, 49,XXXXY and 49,XXXXX cell lines and scored for the presence of ‘n–1’ exclusions that coincide with an X-
-satellite FISH signal. Table 3 indicates a clear correlation of ‘n–1’ exclusions for H2A-Bbd.
|
|
|
|
Like H2B-Myc, H2A-Myc and H2A.X-Myc were neither enriched nor deficient for chromatin of the inactive X chromosome (Table 2) and showed a uniform nuclear distribution (Fig. 8A and C). The histone variant H2A.Z-Myc had a uniform nuclear distribution in most cells (Fig. 8D); however, in a significant proportion of nuclei (Table 2), H2A.Z-Myc was deficient over the Barr body and the inactive X chromosome (Fig. 8E). The degree of exclusion was not as marked as for H2A-Bbd-Myc, but was a consistently reproducible observation.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
X inactivation normalizes the level of X-linked gene expression between male and female mammals by silencing gene expression from much of one X chromosome in female cells (28,29). The process of X inactivation is random and occurs early in development, but the mechanism(s) by which the number of X chromosomes are counted and the choice of chromosome to inactivate is made is not clearly understood (41).
Several specific features of nucleosomes on the inactive X chromosome have been identified. These include hypoacetylation of the N-terminal tails of histones H3 and H4 (42–44), enrichment for the histone variants macroH2A1 (27) and, as shown here, macroH2A2, and deficiency of the histone variant H2A-Bbd (36). The acetylation of histone tails is a general marker of the transcriptional status of chromatin, with active chromatin associated with hyperacetylated histones (6,45), and therefore one would expect all transcriptionally silent regions to be hypoacetylated. A specific role for macroH2A1 or macroH2A2 in X inactivation is unclear, although the non-histone region of macroH2A1 can reduce the transcription level of a reporter gene in yeast (46). The association of macroH2A1 with the inactive X chromosome appears to occur after the initiation and propagation of X inactivation (47,48) and is not required for the maintenance of gene silencing on the inactive X (31). The discovery of macroH2A2, however, suggests the possibility of redundant functions that might complicate analysis of macroH2A function by standard genetic approaches. Further, of likely significance is the apparently mutually exclusive distribution of the macroH2As with the histone variant H2A-Bbd (36).
We have described the cloning of a second macroH2A gene encoding macroH2A2, which shares 80% amino acid identity with macroH2A1 and cofractionates with nucleosomes (Fig. 5C). Despite mapping to different chromosomes, the two genes share almost identical genomic organization, with highly conserved exon–intron boundaries (Fig. 2), suggesting that both loci have originated from a common ancestor. With the exception of an apparent splice acceptor variant site for macroH2A1, both proteins are co-linear with no insertions or deletions. This may reflect spatial constraints of domains within the tail region of the protein that need to be positioned a certain distance from the nucleosome for interactions with other proteins. Outside of the histone domain, macroH2A1 and macroH2A2 are most similar at the very C-terminus. The high amino acid identity could indicate residues necessary for a common function of both proteins.
The histone domain of macroH2A1 and macroH2A2 has the highest amino acid identity between the proteins (Fig. 1). If macroH2A1 and macroH2A2 share similar functional roles, it is likely that an identical nucleosome conformation is required, which would account for the degree of evolutionarily conservation. The histone variant H2A.Z shares >90% amino acid identity from man to Drosophila and, like macroH2A, is significantly diverged from core H2A (63% identity). This conservation most likely represents the need to confer the same subtle conformational changes to the nucleosome (24), to remodel the local chromatin environment (23).
Polyclonal antisera raised to the non-histone region of macroH2A1 cross react at a reduced efficiency with shared epitopes within the non-histone region of macroH2A2 (data not shown) due to the high degree of amino acid identity (Fig. 1). For this reason we chose to investigate the specific subcellular localization of macroH2A2 using GFP and Myc epitope-tagged expression constructs. Female cell lines transfected with either a C-terminal GFP or Myc epitope-tagged macroH2A2 show formation of an MCB that is identical in appearance to that observed with macroH2A antisera (Fig. 6) or with epitope-tagged macroH2A1 constructs (Fig. 3). The macroH2A2 MCB was coincident with an X -satellite FISH signal (Fig. 3D) and co-localizes with macroH2A1 (Fig. 4), indicating association with the inactive X chromosome. Transfection of macroH2A2-GFP into cell lines containing 1–5 X chromosomes demonstrated adherence to the ‘n–1’ rule (Table 1C), also observed for macroH2A1-GFP and macroH2A antisera (Table 1), further supporting association of macroH2A2 with the inactive X chromosome (40).
The maintenance of the inactive state is a combination of likely redundant mechanisms, including DNA methylation, histone hypoacetylation, late replication and the differential use of H2A variants. The removal of a single component does not appear to result in reactivation of the entire chromosome (31,49–53). In addition, the association of macroH2A1 with the inactive X chromosome in development appears to occur after initiation and propagation of X inactivation (47,48). Antisera used in those studies was specific for the macroH2A1 protein with macroH2A2 reactive antisera removed. Therefore the timing of accumulation of macroH2A2 with the inactive X chromosome in development is not known and requires investigation. What role macroH2A1 or macroH2A2 plays at any stage of the X inactivation process is in question, but the clear enrichment of the proteins on the inactive X seems unlikely to be a chance occurrence. This has important implications for gene targeting studies of macroH2A in mouse. The identification of a second macroH2A gene that is highly similar to macroH2A1 implies functional redundancy between the two proteins. Therefore, targeting a single gene may not be fully informative.
H2A variants and the inactive X chromosome
The obviously non-random distribution of three H2A variants, macroH2A1, macroH2A2 and H2A-Bbd, led us to survey other known human H2A variants for their distribution in relation to the inactive X chromosome. As expected, macroH2A1-Myc was enriched on the inactive X, H2A-Bbd-Myc was consistently deficient on the inactive X (Fig. 8B), while H2A-Myc (Fig. 8A) and H2B-Myc were neither enriched nor deficient on the inactive X chromosome (Table 2). Female cells transfected with H2A.X-Myc showed a uniform distribution of the protein throughout the nucleus (Fig. 8C). H2A.X is likely to require access to all chromatin including the inactive X, as it is involved in DNA damage repair (15–17).
In contrast to these variants, H2A.Z-Myc in a significant proportion of nuclei (Table 2) resembled H2A-Bbd-Myc and was deficient on the inactive X chromosome (Fig. 8E). In S.cerevisiae, H2A.Z is implicated in the regulation of gene expression. Deletion of the yeast H2A.Z gene, HTZ1, resulted in dependence on chromatin remodeling complexes for regulation of transcription (23). In a separate study, Htzp1 was found to restore gene silencing in Sir1p mutants (22). H2A.Z was found to associate with both euchromatin and heterochromatin in a non-random distribution in Drosophila melanogaster (54). These results implicate H2A.Z in aspects of both gene activation and silencing. The inactive X chromosome is for the most part transcriptionally silent (55), but maintains islands of actively transcribed genes on a background of silenced chromatin. The partial deficiency of H2A.Z on chromatin of the inactive X chromosome raises a number of plausible models implicating it in the creation and/or maintenance of different chromatin states on the inactive X.
We have demonstrated that chromatin of the inactive X chromosome is either enriched or deficient for certain histone H2A variants while no restriction is observed for other H2A variants or for core histones H2A and H2B. This is in stark contrast to the data presented by Perche et al. (46) in which the Barr body was reported to be enriched for the histones H2A, H2B, H3 and macroH2A as a direct consequence of the inactive X chromosome having a higher nucleosome density. It seems unlikely that higher nucleosome density alone accounts for the enrichment of macroH2A with the inactive X chromosome, however, since the inactive X chromosome at metaphase is clearly enriched for macroH2A1 (27,47). The contrasting observations between the two studies may reflect differences in the experimental approaches taken, the levels of expression of the transfected histone constructs or the different constructs used.
Other than CENP-A, a histone H3 variant implicated in centromere function (56,57), all other identified human histone variants are related to H2A. Surprisingly, no H2B or H4 variants have been identified to date. This may reflect a lower structural requirement for the H2A position in the nucleosome that has allowed H2A to diverge and acquire unique functional roles that are exerted through subtle conformational changes, as reported for H2A.Z (24).
Table 4 summarizes our observations for known histone H2A variant relationships with the inactive X chromosome and the Barr body. Enrichment of H2A variants in chromatin of the inactive X chromosome is not a feature of all histone variants, but is specific for macroH2A1 and macroH2A2, while deficiency is primarily a characteristic of H2A-Bbd and less so for H2A.Z. This places substantial focus on the relative distribution of H2A-Bbd and H2A.Z with macroH2A1 and macroH2A2, and what influence these variants may have on structuring chromatin on the active and inactive X chromosomes. Specific nucleosomal conformations could be adopted that present the chromatin in such a way that accessibility of the transcriptional machinery to the DNA is altered. How chromatin is assembled on the inactive X chromosome to include some histone variants and exclude others is an open question, both during the onset of X inactivation and throughout subsequent cell divisions. How proteins that share such high levels of sequence identity can be ordered in such non-random distributions may be an important aspect of understanding the process of X inactivation. Specific questions concerning the precise distribution of macroH2A1, macroH2A2, H2A-Bbd and H2A.Z can be addressed by chromatin immunoprecipitation experiments (58,59). The ability of the macroH2A proteins to be targeted to the inactive X chromatin can be addressed by analysis of targeted disruptions of the MCB competent expression constructs.
|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Isolation of human and mouse cDNA clones for macroH2A2
The nucleotide sequence of human macroH2A1 (AF054174) was used in TBLASTX searches against entries in the human EST database at GenBank using the NIH BLAST server (www.ncbi.nlm.nih.gov/cgi-bin/BLAST/nph-newblast). Four cDNA clones that were related, but not identical to macroH2A1 (IMAGE numbers 50058, 309275, 2518298 and 2519153) were obtained from Research Genetics. DNA was prepared with the Wizard-plus miniprep DNA purification system (Promega), and the cDNAs were sequenced with a fluorescence-labeled dye-terminator cycle sequencing kit according to the manufacturer’s instructions (PRISM Ready DyeDeoxy Terminator Premix, Applied Biosystems) and electrophoresed on an ABI 373 (Perkin-Elmer). The complete human macroH2A2 sequence (accession no. AF336304) was used in BLAST searches against entries in the mouse EST database at GenBank using the NIH BLAST server (www.ncbi.nlm.nih.gov/cgi-bin/BLAST/nph-newblast). Two cDNA clones were obtained from Research Genetics. (IMAGE numbers 1381320 and 1361199), and completely sequenced as described for human macroH2A2 above (accession no. AF336305).
Chromosomal assignment of human macroH2A1
The human macroH2A1 gene was mapped in the human genome by PCR screening of the GeneBridge-4 radiation hybrid mapping panel (60), obtained from Research Genetics. PCR-positive radiation hybrid clones were organized into the official GeneBridge-4 order and mapping data was obtained from the Whitehead Institute server (http://carbon.wi.mit.edu:8000/cgi-bin/contig/rhmapper.pl). MacroH2A1 mapped 2.63 cR from AFMA057VG5 (typing data: 00000 10100 11100 00020 00011 10011 00000 01000 00000 11000 10100 00000 11100 10000 01000 10000 00010 00000 100). The chromosomal location of macroH2A1 was confirmed by PCR screening of the monochromosomal hybrid panels (61), obtained from Research Genetics. PCR primers were designed for the 3'-untranslated region of macroH2A1 and titrated for a unique human-specific PCR product, as follows: primer 1, 5'-TGA GCA ATG ACA GAA CCA GC-3'; primer 2, 5'-AGA ACG CCA CTG GAG GAT G-3'; 94°C for 2 min; 40 cycles of 94°C for 30 s, 60°C for 30 s, 72°C for 30 s; cooled to 4°C; product size, 393 bp.
Using the PCR conditions outlined above, the Research Genetics PCR-screenable BAC library IV (CITB human BAC DNA pools B&C libraries) was screened for macroH2A1 positive clones. A single clone, CITB176A1, was isolated and confirmed to contain macroH2A1 by PCR and sequencing. Clone CITB176A1 was labeled with biotin using nick-translation (Boehringer-Mannheim) with the addition of 1 µl DNaseI/µg probe DNA. FISH was performed as described by Miller et al. (62). Denatured probe DNA in 10 µl of hybridization solution VII (50% formamide, 2x SSC; Oncor) was applied to denatured slides and hybridized for 16 h at 37°C. Post-hybridization washes were performed in 50% formamide, 2x SSC (43°C, 15 min) and 2x SSC (37°C, 8 min). Signal detection and counterstaining were performed using reagents supplied by Oncor, as recommended by the manufacturer.
Genomic organization of human macroH2A1 and macroH2A2
Database searches with the cDNA sequence of macroH2A1, using the working draft of the human genome (http://www.ncbi.nlm.nih.gov/genome/seq/HsBlast.html), identified a BAC clone (CTC-203F4) from chromosome 5 that contained the full macroH2A1 sequence. Two chromosome 10 BAC clones were identified that contained the full macroH2A2 cDNA sequence (RP11-367H5 and RP11-379O18). Sequence alignments between the cDNA and genomic sequences allowed determination of exon–intron boundaries.
To identify any alternative exon 6 sequences for macroH2A2, the genomic sequence was used with the GENSCAN exon prediction server (http://genes.mit.edu/GENSCAN.html). As a control, the genomic sequence of macroH2A1 was used to predict exon 6 of macroH2A1.1 and macroH2A1.2; both exons were predicted with probabilities of 0.321 and 0.641, respectively. Exon 6 of macroH2A2 was predicted from the corresponding genomic sequence with a probability of 0.486.
Mammalian expression constructs
A full-length C-terminal Myc epitope-tagged human macroH2A1, H2A-Bbd and H2B were constructed as described previously (36). All data for epitope-tagged macroH2A1 refers to the macroH2A1.2 splice isoform. A C-terminal GFP-tagged macroH2A1 was generated by subcloning the Myc epitope-tagged insert into pcDNA3.1-CT-GFP (Invitrogen). The full coding sequence from clone 50058 was PCR-amplified with primers incorporating KpnI and EcoRI restriction enzyme recognition sites (primer 1, 5'-GGG GTA CCA TGT CAG GCC GGA GCG GG-3'; primer 2, 5'-GGA ATT CCT TGG TGT CCA GTT TGG CC-3') and subcloned using standard techniques (63). Both a C-terminal Myc epitope-tagged and GFP fusion were generated by cloning into pcDNA3.1-CT-Myc-His and pcDNA3.1-CT-GFP (Invitrogen), and subclones were verified for sequence integrity as above. The nucleotide sequence of human H2A.X (NM_002105) was used in BLASTN searches against entries in the human EST database at GenBank using the NIH BLAST server. A representative cDNA clone (IMAGE 1676605) was obtained from Research Genetics. The full coding sequence from clone 1676605 was PCR-amplified with primers incorporating BamHI and EcoRI restriction enzyme recognition sites (primer 1, 5'-CGG GAT CCA TGT CGG GCC GCG GCA AG-3'; primer 2, 5'-GGA ATT CGT ACT CCT GGG AGG CCT GG-3') and subcloned into pcDNA3.1-CT-Myc-His and pcDNA3.1-CT-GFP as described above. H2A.Z cDNA was PCR amplified from reverse transcribed poly(A)+ RNA prepared from 1 x 108 HEK 293 cells (Fast-Track 2.0 system, Invitrogen) using standard techniques (63). H2A.Z was amplified using primers covering the full coding sequence of H2A.Z and incorporating BamHI and EcoRI restriction enzyme recognition sites (primer 1, 5'-CGG GAT CCA TGG CTG GCG GTA AGG CTG-3'; primer 2, 5'-GGA ATT CGA CAG TCT TCT GTT GTC CTT TC-3'). Database searches with the coding sequence of core histone H2A identified an H2A entry for a polyadenylated H2A gene (AK001765) that was contained in a BAC clone from chromosome 12p (RP11-174G6; AC010168). The H2A gene was amplified using primers covering the full coding sequence of H2A and incorporating BamHI and EcoRI restriction enzyme recognition sites (primer 1, 5'-GGA TCC ATG TCC GGT CGC GGG AAA C-3'; primer 2, 5'-GGA ATT CCT CCA CTT TTT GAA AAA TAT GGC-3'). Both H2A.Z and H2A PCR products were subcloned into pcDNA3.1-CT-Myc-His as described above.
Generation of macroH2A antisera
The macroH2A1-CT-GFP clone was used to PCR amplify the non-histone region of macroH2A1 (amino acids 126–371), using primers incorporating BamHI and EcoRI restriction enzyme recognition sites and a 3'-TAG stop codon (primer 1, 5'-CGG GAT CCA TCA TCA CAC CAC CC-3'; primer 2, 5'-GGA ATT CCT AGT TGG CGT CCA GCT TG-3'). The PCR product was subcloned into pGEX-2T (Amersham Pharmacia) using standard techniques (63).
pGEX-2T-macroH2A1 was transformed into competent BL21(DE3)pLysS Escherichia coli bacteria strain (Invitrogen) using standard techniques (63). GST-macroH2A1 (amino acids 127–371) fusion protein was overexpressed in BL21(DE3)pLysS bacteria grown in SOB media (20 g/l Bacto tryptone, 5 g/l Bacto yeast extract, 0.5 g/l NaCl, 186 mg/l KCl, 10 mM MgCl2, pH 7.0) at 37°C, and induced with 1 mM IPTG at OD600 = 0.6 for 2 h. Recombinant protein was solubilized in PBS containing 1 mM DTT, 1 M NaCl, 1 mM EDTA, 1% Triton X-100 and supplemented with a protease inhibitor cocktail (complete, EDTA-free, Boehringer Mannheim), and purified using glutathione–Sepharose according to the manufacturer (Amersham Pharmacia).
For immunizations, purified protein (0.5 mg in Freund’s complete adjuvant) was injected subcutaneously into New Zealand white rabbits (Covance), with subsequent intramuscular injections (0.25 mg in Freund’s incomplete adjuvant) on days 21, 42, 63, 84, 105 and 147. Immune sera was obtained on days 31, 52, 73, 94, 116 and 157.
Cell culture and transfection
Human cell lines used in this study were HEK 293, a female embryonic kidney tumor cell line; RPE1, a 46,XX telomerase-immortalized cell line derived from a retinal pigment epithelial cell line RPE-340 (64) (Clontech); GM04626 and GM00254, both 47,XXX primary fibroblast strains; GM1202, a 49,XXXXY primary fibroblast cell line; GM6061B, a 49,XXXXX primary fibroblast cell line (National Institute of General Medical Sciences Cell Repository, Camden, NJ); and Y87, a 46,XY primary fibroblast strain.
All cell lines, with the exception of RPE1, were maintained in culture as monolayers in -MEM supplemented with 20% FBS, 0.1 mg/ml antibiotics (penicillin and streptomycin) and 2 mM L-glutamine (Gibco BRL) at 37°C in a 5% CO2 atmosphere. RPE1 was maintained as a monolayer in DMEM:F-12 supplemented with 10% FBS, 0.1 mg/ml antibiotics (penicillin and streptomycin) and 2 mM L-glutamine, and 0.348% sodium bicarbonate (Gibco BRL) at 37°C in a 5% CO2 atmosphere.
Transfections were performed in serum-free media using Superfect reagent according to the manufacturer’s recommendations (Qiagen).
To establish stably transformed 293 cell lines, 5 µg of macroH2A1-CT-Myc or macroH2A2-CT-Myc expression construct (see section entitled ‘Mammalian expression constructs’) were transfected into HEK 293 cells with Superfect, and grown for 48 h before selection with 300 µg/ml neomycin (G418) (Gibco BRL).
Immunofluorescence and FISH
Transfected and non-transfected cells were grown directly on microscope slides and were permeablilized with 0.1% Triton X-100 in PBS containing 3.7% formaldehyde for 10 min at room temperature. Slides were blocked for 1 h at room temperature in PBS-Tween supplemented with 3% BSA and washed once for 2 min at room temperature in PBS before proceeding with immunostaining. Slides were incubated with a 1:100 dilution of anti-Myc mAb (Invitrogen) and/or a 1:200 dilution of anti-macroH2A1 polyclonal sera in PBS-Tween and 1% BSA for 3 h at room temperature. Detection was achieved using a 1:200 dilution of goat anti-mouse IgG and/or goat anti-rabbit IgG conjugated with either FITC or Rhodamine in PBS-Tween and 1% BSA at room temperature for 1 h (Jackson ImmunoResearch). Slides were fixed in 3.7% formaldehyde in PBS for 10 min at room temperature and washed twice in water before application of 200 ng/ml DAPI in diazabicyclol-2-2-2-octane (DABCO) (Sigma).
Sequential FISH was performed after fixation in 3.7% formaldehyde in PBS for 10 min. Slides were denatured in 70% formamide, 2x SSC for 14 min at 72°C before dehydration through a 70–80–100% ethanol series and then air-dried. A direct-labeled human X chromosome -satellite probe was obtained from Vysis (Spectrum Orange). The probe was denatured for 5 min at 72°C before placing at 42°C, and hybridization was carried out overnight in a humid chamber at 37°C. Slides were washed twice in 50% formamide, 2x SSC for 8 min at 42°C and once in 2x SSC at 42°C before DAPI staining as above. Images were collected with a Vysis imaging system equipped with a cooled CCD camera (Photometrics) controlled via the Quips M-FISH software (Vysis).
Cells transfected with GFP-tagged proteins were detected with a FITC excitation filter (Chromatech) using a Vysis Quips imaging system. Slides were fixed with 3.7% formaldehyde, 0.1% Triton X-100 for 10 min, before washing with PBS and counterstaining nuclei with DAPI.
Total protein extracts, isolation of nucleosomes and immunoblotting
Nuclei were isolated from HEK 293 cells and a stably transfected 293-mH2A2-CT-Myc line. 1 x 108 cells were homogenized in resuspension buffer at 4°C (80 mM NaCl, 20 mM EDTA, 1% Triton X-100), centrifuged at 1500 g and washed twice more in the same buffer. Nuclei were resuspended in digestion buffer (10 mM NaCl, 10 mM NaButyrate, 10 mM Tris pH 7.2, 2 mM MgCl2 and 1 mM CaCl2), digested with 15 U of micrococcal nuclease (Worthington) and incubated at 37°C for 8 min. Digestion was stopped by addition of EDTA to 10 mM final concentration and chilling of the suspension on ice. Nuclei were centrifuged at 10 000 g for 4 min and supernatant S1 retained. The pellet was resuspended in lysis buffer (10 mM Tris pH 7.2, 10 mM NaButyrate and 250 mM EDTA) and incubated on ice for 30 min before centrifugation at 10 000 g for 4 min. The supernatant S2 was retained and pooled with S1. Nucleosome oligomers were separated by ultracentrifugation through 5–30% sucrose gradient containing 10 mM Tris pH 7.2, 10 mM NaButyrate and 250 mM EDTA at 37 000 r.p.m. for 16 h at 4°C in a Beckman SW41 rotor.
Proteins were separated by gel electrophoresis on an 18% SDS–PAGE gel before transfer to PVDF membrane (BioRad) as previously described (65). Blots was probed with either a 1:400 dilution of anti-macroH2A1 polyclonal antisera or a 1:3000 dilution of anti-Myc monoclonal antisera (Invitrogen) in 5% non-fat milk, PBS, 0.1% Tween-20, followed by incubation with a horseradish peroxidase conjugated goat anti-rabbit or anti-rabbit secondary antibody. Detection was performed using the ECL chemiluminescence substrate (Amersham Pharmacia).
|
ACKNOWLEDGEMENTS |
|---|
We are grateful to Karen Gustashaw for technical advice and for FISH mapping of the human macroH2A1 gene, to Cory Valley for identification of the mouse macroH2A2 cDNA clone, and to members of the Willard lab for helpful discussions and advice. This work was supported by research grant GM 45441 to H.F.W. from the National Institutes of Health and by an award from the Board of Regents of the State of Ohio. B.P.C. is supported by a postdoctoral fellowship from the Rett Syndrome Research Foundation.
|
FOOTNOTES |
|---|
+ To whom correspondence should be addressed at: Department of Genetics, Case Western Reserve University School of Medicine, BRB 731, 2109 Adelbert Road, Cleveland, OH 44106–4955, USA. Tel: +1 216 368 1617; Fax: +1 216 368 3030; Email: willard@uhri.org
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Ramakrishnan, V. (1997) Histone structure and the organization of the nucleosome. Annu. Rev. Biophys. Biomol. Struct., 26, 83–112.[ISI][Medline]
2 Kornberg, R.D. and Lorch, Y. (1999) Twenty-five years of the nucleosome, fundamental particle of the eukaryote chromosome. Cell, 98, 285–294.[ISI][Medline]
3 Wu, J. and Grunstein, M. (2000) 25 years after the nucleosome model: chromatin modifications. Trends Biochem. Sci., 25, 619–623.[ISI][Medline]
4 Argents, G. and Moudrianakis, N.N. (1995) The histone fold: a ubiquitous architectural motif utilized in DNA compaction and protein dimerization. Proc. Natl Acad. Sci. USA, 92, 11170–11174.
5 Bohm, L. and Crane-Robinson, C. (1984) Proteases as structural probes for chromatin: the domain structure of histones. Biosci. Rep., 4, 365–386.[ISI][Medline]
6 Strahl, B.D. and Allis, C.D. (2000) The language of covalent histone modifications. Nature, 403, 41–45.[Medline]
7 Wolffe, A.P. and Pruss, D. (1996) Deviant nucleosomes: the functional specialization of chromatin. Trends Genet., 12, 58–62.[ISI][Medline]
8 Albig, W., Trappe, R., Kardalinou, E., Eick, S. and Doenecke, D. (1999) The human H2A and H2B histone gene complement. Biol. Chem., 380, 7–18.[ISI][Medline]
9 Williams, A.S. and Marzluff, W.F. (1995) The sequence of the stem and flanking sequences at the 3'-end of histone mRNA are critical determinants for the binding of the stem–loop binding protein. Nucleic Acids Res., 23, 654–662.
10 Dominski, Z. and Marzluff, W.F. (1999) Formation of the 3' end of histone mRNA. Gene, 239, 1–14.[ISI][Medline]
11 Mannironi, C., Bonner, W.M. and Hatch, C.L. (1989) H2A.X. a histone isoprotein with a conserved C-terminal sequence, is encoded by a novel mRNA with both DNA replication type and polyA 3'-processing signals. Nucleic Acids Res., 17, 9113–9126.
12 Hatch, C.L. and Bonner, W.M. (1988) Sequence of cDNAs for mammalian H2A.Z, an evolutionarily diverged but highly conserved basal histone H2A isoprotein species. Nucleic Acids Res., 16, 1113–1124.
13 Allis, C.D., Glover, C.V., Bowen, J.K. and Gorovsky, M.A. (1980) Histone variants specific to the transcriptionally active, amitotically dividing macronucleus of the unicellular eucaryote, Tetrahymena thermophila. Cell, 20, 609–617.[ISI][Medline]
14 Porcher, C. and Grandchamp, B. (1995) Structure of the mouse H2A.X gene and physical linkage to the UPS locus on chromosome 9: assignment of the human H2A.X gene to 11q23 by sequence analysis. Genomics, 25, 312–313.[ISI][Medline]
15 Paull, T.T., Rogakou, E.P., Yamazaki, V., Kirchgessner, C.U., Gellert, M. and Bonner, W.M. (2000) A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr. Biol., 10, 886–895.[ISI][Medline]
16 Chen, H.T., Bhandoola, A., Difilippantonio, M.J., Zhu, J., Brown, M.J., Tai, X., Rogakou, E.P., Brotz, T.M., Bonner, W.M., Ried, T. et al. (2000) Response to RAG-mediated V(D)J cleavage by NBS1 and 
-H2AX. Science, 290, 1962–1965.
17 Downs, J.A., Lowndes, N.F. and Jackson, S.P. (2000) A role for Saccharomyces cerevisiae histone H2A in DNA repair. Nature, 408, 1001–1004.[Medline]
18 Rogakou, E.P., Pilch, D.R., Orr, A.H., Ivanova, V.S. and Bonner, W.M. (1998) DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J. Biol. Chem., 273, 5858–5868.
19 Rogakou, E.P., Nieves-Neira, W., Boon, C., Pommier, Y. and Bonner, W.M. (2000) Initiation of DNA fragmentation during apoptosis induces phosphorylation of H2AX histone at serine 139. J. Biol. Chem., 275, 9390–9395.
20 Hatch, C.L. and Bonner, W.M. (1990) The human histone H2A.Z gene. Sequence and regulation. J. Biol. Chem., 265, 15211–15218.
21 Popescu, N., Zimonjic, D., Hatch, C. and Bonner, W. (1994) Chromosomal mapping of the human histone gene H2AZ to 4q24 by fluorescence in situ hybridization. Genomics, 20, 333–335.[ISI][Medline]
22 Dhillon, N. and Kamakaka, T.R. (2000) A histone variant, Htz1p, and a Sir1p-like protein, Esc2p, mediate silencing at HMR. Mol. Cell, 6, 769–780.[ISI][Medline]
23 Santisteban, M.S., Kalashnikova, T. and Smith, M.M. (2000) Histone H2A.Z regulates transcription and is partially redundant with nucleosome remodeling complexes. Cell, 103, 411–422.[ISI][Medline]
24 Suto, R.K., Clarkson, M.J., Tremethick, D.J. and Luger, K. (2000) Crystal structure of a nucleosome core particle containing the variant histone H2A.Z. Nat. Struct. Biol., 7, 1121–1124.[ISI][Medline]
25 Luger, K., Mader, A.W., Richmond, R.K., Sargent, D.F. and Richmond, T.J. (1997) Crystal structure of the nucleosome core particle at 2.8 Å resolution. Nature, 389, 251–260.[Medline]
26 Pehrson, J.R. and Fried, V.A. (1992) MacroH2A, a core histone containing a large nonhistone region. Science, 257, 1398–1400.
27 Costanzi, C. and Pehrson, J.R. (1998) Histone macroH2A1 is concentrated in the inactive X chromosome of female mammals. Nature, 393, 599–601.[Medline]
28 Heard, E., Clerc, P. and Avner, P. (1997) X-chromosome inactivation in mammals. Annu. Rev. Genet., 31, 571–610.[ISI][Medline]
29 Willard, H.F. (2000) Sex chromosomes and X chromosome inactivation. In Scriver, C.R., Beaudet, A.L., Sly, W.S., and Valle, D. (eds), The Molecular and Metabolic Bases of Inherited Disease, 8th Ed. McGraw-Hill Publishing Co., New York, NY.
30 Barr, M.L. and Bertram, E.G. (1949) A morphological distinction between neurones of the male and female, and the behaviour of the nucleolar satellite during accelerated nucleoprotein synthesis. Nature, 163, 676–677.
31 Csankovszki, G., Panning, B., Bates, B., Pehrson, J.R. and Jaenisch, R. (1999) Conditional deletion of Xist disrupts histone macroH2A localization but not maintenance of X inactivation. Nature Genet., 22, 323–324.[ISI][Medline]
32 Brown, C.J., Ballabio, A., Rupert, J.L., Lafreniere, R.G., Grompe, M., Tonlorenzi, R. and Willard, H.F. (1991) A gene from the region of the human X inactivation centre is expressed exclusively from the inactive X chromosome. Nature, 349, 38–44.[Medline]
33 Brockdorff, N., Ashworth, A., Kay, G.F., McCabe, V.M., Norris, D.P., Cooper, P.J., Swift, S. and Rastan, S. (1992) The product of the mouse Xist gene is a 15 kb inactive X-specific transcript containing no conserved ORF and located in the nucleus. Cell, 71, 515–526.[ISI][Medline]
34 Brown, C.J., Hendrich, B.D., Rupert, J.L., Lafreniere, R.G., Xing, Y., Lawrence, J. and Willard, H.F. (1992) The human XIST gene: analysis of a 17 kb inactive X-specific RNA that contains conserved repeats and is highly localized within the nucleus. Cell, 71, 527–542.[ISI][Medline]
35 Gilbert, S.L., Pehrson, J.R. and Sharp, P.A. (2000) XIST RNA associates with specific regions of the inactive X chromatin. J. Biol. Chem., 275, 36491–36494.
36 Chadwick, B.P. and Willard, H.F. (2001) A novel chromatin protein, distantly related to histone H2A, is largely excluded from the inactive X chromosome. J. Cell Biol., 152, 375–384.
37 Pehrson, J.R. and Fuji, R.N. (1998) Evolutionary conservation of histone macroH2A subtypes and domains. Nucleic Acids Res., 26, 2837–2842.
38 Breathnach, R. and Chambon, P. (1981) Organization and expression of eucaryotic split genes coding for proteins. Annu. Rev. Biochem., 50, 349–383.[ISI][Medline]
39 Rasmussen, T.P., Huang, T., Mastrangelo, M.A., Loring, J., Panning, B. and Jaenisch, R. (1999) Messenger RNAs encoding mouse histone macroH2A1 isoforms are expressed at similar levels in male and female cells and result from alternative splicing. Nucleic Acids Res., 27, 3685–3689.
40 Lyon, M.F. (1961) Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature, 190, 372–373.[Medline]
41 Avner, P. and Heard, E. (2001) X-Chromosome Inactivation: counting, choice and initiation. Nat. Rev., 2, 59–67.
42 Jeppesen, P. and Turner, B.M. (1993) The inactive X chromosome in female mammals is distinguished by a lack of histone H4 acetylation, a cytogenetic marker for gene expression. Cell, 74, 281–289.[ISI][Medline]
43 Boggs, B.A., Connors, B., Sobel, R.E., Chinault, A.C. and Allis, C.D. (1996) Reduced levels of histone H3 acetylation on the inactive X chromosome in human females. Chromosoma, 105, 303–309.[ISI][Medline]
44 Gilbert, S.L. and Sharp, P.A. (1999) Promoter-specific hypoacetylation of X-inactivated genes. Proc. Natl Acad. Sci. USA, 96, 13825–13830.
45 Workman, J.L. and Kingston, R.E. (1998) Alteration of nucleosome structure as a mechanism of transcriptional regulation. Annu. Rev. Biochem., 67, 545–579.[ISI][Medline]
46 Perche, P., Vourc’h, C., Konecny, L., Souchier, C., Robert-Nicoud, M., Dimitrov, S. and Khochbin, S. (2000) Higher concentrations of histone macroH2A in the Barr body are correlated with higher nucleosome density. Curr. Biol., 10, 1531–1534.[ISI][Medline]
47 Mermoud, J.E., Costanzi, C., Pehrson, J.R. and Brockdorff, N. (1999) Histone macroH2A1.2 relocates to the inactive X chromosome after initiation and propagation of X-inactivation. J. Cell Biol., 147, 1399–1408.
48 Rasmussen, T.P., Mastrangelo, M.A., Eden, A., Pehrson, J.R. and Jaenisch, R. (2000) Dynamic relocalization of histone MacroH2A1 from centrosomes to inactive X chromosomes during X inactivation. J. Cell Biol., 150, 1189–1198.
49 Mohandas, T., Sparkes, R.S. and Shapiro, L.J. (1981) Reactivation of an inactive human X chromosome: evidence for X inactivation by DNA methylation. Science, 211, 393–396.
50 Driscoll, D. and Migeon, B. (1990) Sex differences in methylation of single-copy genes in human meiotic germ cells: implications for X chromosome inactivation, parental imprinting, and origin of CpG mutations. Somat. Cell Mol. Genet., 16, 267–282.[ISI][Medline]
51 Brown, C.J. and Willard, H.F. (1994) The human X-inactivation centre is not required for maintenance of X-chromosome inactivation. Nature, 368, 154–156.[Medline]
52 Gartler, S.M. and Goldman, M.A. (1994) Reactivation of inactive X-linked genes. Dev. Genet., 15, 504–514.[ISI][Medline]
53 Rack, K.A., Chelly, J., Gibbons, R.J., Rider, S., Benjamin, D., Lafreniere, R.G., Oscier, D., Hendriks, R.W., Craig, I.W., Willard, H.F. et al. (1994) Absence of the XIST gene from late-replicating isodicentric X chromosomes in leukaemia. Hum. Mol. Genet., 3, 1053–1059.
54 Leach, T.J., Mazzeo, M., Chotkowski, H.L., Madigan, J.P., Wotring,