Role of late replication timing in the silencing of X-linked genes
Role of late replication timing in the silencing of X-linked genesR. Scott Hansen1,*, Theresa K. Canfield1, Alan D. Fjeld1 and Stanley M. Gartler1,2
Departments of 1Medicine and 2Genetics, University of Washington, Seattle, WA 98195, USA
Received May 10, 1996;Revised and Accepted June 20, 1996
Cytosine methylation at promoter regions and late replication timing have both been implicated in the regulation of genes subject to X chromosome inactivation. Reported here are studies of X-linked gene replication in normal male and female cells as well as in cell hybrids that contain either a normal active X, a normal inactive X, or an inactive X chromosome that has been treated with the demethylating agent, 5-azacytidine (5aC). The relationship between replication timing and transcriptional activity was examined for XIST, XPCT, PGK1, HPRT, F9, FMR1, IDS, and G6PD, and earlier replication was generally found to be associated with increased transcriptional activity. The HPRT and G6PD genes in an untreated inactive X hybrid were among the few exceptions to this correlation in that they remain inactive, yet replicate earlier than their inactive X alleles present in normal human diploid cells. This condition of earlier replication timing may contribute to the high rates of 5aC-induced reactivation for HPRT and G6PD in this hybrid relative to other inactive X hybrids. Other anomalous cases include 5aC-induced advances in replication time for genes such as XIST and F9 whose transcription was unaltered by treatment. These and other data support a model for regulation of X-inactivated genes that involves at least two levels of control:(i) large chromosomal domains are placed into a transcriptionally nonpermissive state by late replication and (ii) transcription is blocked at the local level by promoter methylation. In addition, our observations of continued XIST expression in 5aC-treated hybrids with reactivated genes indicates that such expression is not sufficient for the maintenance of X inactivation.
Attempts to reactivate genes that have been silenced by X chromosome inactivation (1 ) are nearly as old as the Lyon hypothesis (2 ). However, consistent reactivation of inactive X-linked genes only became possible when the demethylation agent 5-azacytidine (5aC) was used in somatic cell hybrids by Mohandas et al. (3 ). The rationale for using 5aC was based on the suggestion that DNA methylation plays a role in X inactivation and differentiation (4 -6 ) and the fact that Jones and colleagues had found an induction of cellular differentiation with 5aC treatment of embryonic mouse cells (7 -9 ). The initial reactivation finding of Mohandas et al. (3 ) has been confirmed and extended by a number of studies examining reactivation of X-inactivated genes in either somatic cell hybrids (10 -12 ) or transformed cells (13 ,14 ). Such 5aC-induced reactivation occurs with very low frequency in normal rodent cells (14 ), however, and is apparently undetectable in normal human cells (15 ).
A critical requirement for the reactivation of inactive X-linked housekeeping genes by 5aC is the demethylation of their heavily methylated promoters (16 -20 ). The differential reactivation response of normal and transformed cells to 5aC does not appear to involve promoter structure, however, because both cell types have similar patterns of promoter methylation (18 ,19 ,21 ). It follows that a mechanism other than promoter methylation may also be involved in the developmental process resulting in X inactivation. Two-step models of X inactivation have previously been proposed to explain differential reactivation rates in transformed versus normal cells (22 ), and in undifferentiated versus differentiated murine embryonal carcinoma cells (23 ). In our current replication timing studies of active, inactive, and 5aC-reactivated X-linked genes, we find support for replication timing being a control element involved in X inactivation that can be disturbed in somatic cell hybrids independently of promoter methylation.
Earlier replication timing in expressing tissues versus nonexpressing tissues has been documented for most genes that have been examined, including alpha fetoprotein, serum albumin, T-cell receptor beta chain, beta cluster globin genes, immunoglobulin kappa genes, and immunoglobulin heavy chain genes (24 -26 ). In the case of X-linked genes, there is a great deal of cytogenetic evidence that the active X chromosome replicates earlier than the inactive X (27 ), and some molecular studies of individual gene replication have been reported (28 -33 ). Advanced replication time following 5aC treatment has also been demonstrated cytologically for segments of the inactive X chromosome (34 ,35 ), and one report of this phenomenon has been described at the gene level (31 ). Consistent with these data, we find that reactivation of genes in somatic cell hybrids at six different loci is associated with an advancement in replication time, often to that of the normal, active X gene.
Exceptions to this general correlation were also found, however. We observed that XIST, the gene that appears to play a major role in the establishment of X chromosome inactivation (36 ), can also be induced to replicate earlier by treatment of the inactive X with 5aC even though this is the expressed allele (37 ,38 ) whose promoter region is unmethylated (39 -42 ). Although the factor IX gene (F9) lacks a 5' CpG island (43 ) and is not expressed in fibroblasts, an advance in replication time can also occur for this gene following 5aC exposure. In addition, a few cases were found of partially-advanced replication time without associated reactivation for genes capable of reactivation, including one example where the promoter is known to be fully methylated. In the latter case, the rate of 5aC-induced reactivation of HPRT is much higher than in cells with normal, later replication. These data suggest that reactivation of X-inactivated genes that are widely-expressed involves a demethylation-induced alteration of at least two control elements: the promoter and a replication time control locus, both of which must be altered to bring about reactivation.
To better understand the relationship of transcription to replication timing on the active and inactive X chromosomes, we have studied the replication of eight X-linked genes (XIST, XPCT, PGK1, HPRT, F9, FMR1, IDS, and G6PD) in normal cells (fibroblasts and lymphocytes), lymphoblastoid cells, and human-hamster somatic cell hybrids containing either an active or an inactive human X chromosome. For each of these genes, the active X allele replicates earlier in the cell cycle than the inactive X allele in all cell types examined. Except for F9, which is not expressed in these cell types (see Materials and Methods section), and XIST, which is expressed on the inactive X and not the active X, this means that the expressed alleles replicate before the repressed ones.
To study replication timing for reactivated inactive X-linked genes, a human-hamster somatic cell hybrid with an inactive human X chromosome (X8-6T2S1) was treated with 5aC, transferred to medium selective for cells with reactivated HPRT, and the resultant clones were isolated and expanded for further study (18 ,44 ). We have shown previously that the promoter region of PGK1 in X8-6T2S1 is heavily methylated and that 5aC-derived clones with reactivated PGK1, including those described below, are all extensively demethylated in this region (18 ,19 ). The 5' CpG island of HPRT is also heavily methylated in X8-6T2S1 (17 ) and several of the reactivant clones were found to be unmethylated at HpaII and HhaI restriction sites in the promoter region (R. S. Hansen, unpublished observations).
To determine if and to what extent coreactivation of HPRT and other unselected genes had occurred, RNA from such clones was analyzed by semiquantitative reverse transcriptase-PCR (RT-PCR). Transcript levels in the 5aC-treated hybrids are compared to those of an active X hybrid (GM06318). A summary of the reactivation status of six genes with 5' CpG islands (XPCT, PGK1, HPRT, FMR1, IDS, and G6PD) is shown in Table 1 for seven selected hybrids. Reactivation of HPRT was found in all treated hybrids, as expected from their growth in HPRT-selective media. Including these instances of HPRT reactivation, 25 cases of reactivation were identified among this set of 5aC-treated hybrids. Replication timing was examined for 22 of these cases and, in all except three cases of HPRT reactivation, a significant advance in replication time relative to the parental inactive X allele was observed.
Representative replication profiles for reactivated genes are shown in Figure 1 ; replication profiles for active X and the parental inactive X alleles are presented for comparison. Replication of those genes that are strongly reactivated usually occurs at a time in S that is similar to that of the corresponding active X allele. The replication profiles of HPRT in III-5, III-14S7, and V-6CS1 were exceptional in that, despite reactivation of the gene, replication was not advanced relative to X8-6T2S1, the untreated inactive X hybrid (data not shown). It appears, however, that replication of HPRT in X8-6T2S1 is advanced relative to a normal inactive X allele to a time that is compatible with expression (see below).
XIST differs from the other genes studied in that it is expressed exclusively on the inactive X chromosome and it is expressed in X8-6T2S1 cells before 5aC treatment at levels similar to those in normal female fibroblasts (30 ). In fibroblasts and in X8-6T2S1, we found that the inactive X allele of XIST replicates in the latter portion of S and the active X allele replicates in early S (30 ). Because the promoter of the expressed, inactive X allele is unmethylated in normal human cells (40 ) as well as X8-6T2S1 (data not shown), we wondered if the late replication of XIST in this cell line could be altered by demethylation at some other locus. Examination of our 5aC-treated hybrid clones of X8-6T2S1 revealed a clear advance in the time of XIST replication in the 19AS2 clone (Fig. 2 A). Although the replication time of XIST in 19AS2 is similar to that of active X alleles, no reduction in XIST expression is apparent in this cell line as determined by semiquantitative RT-PCR analyses (data not shown). These data indicate that the time of XIST replication is controlled by methylation at a site other than the promoter and is independent of transcription. In addition, the continued expression of XIST in 19AS2 and other 5aC-treated hybrids (R. S. Hansen, unpublished observations) indicates that XIST expression is not sufficient to repress those genes reactivated in these cells (Table 1 ).
F9 also differs from the other genes studied in that it is a tissue-specific gene not expressed in fibroblasts and its 5' region, which is deficient in CpG dinucleotides (43 ), has a similar pattern of methylation in male and female fibroblasts (45 ). We and others, however, have previously observed that F9 replicates earlier on the active X chromosome than on the inactive X in fibroblasts (30 ) or fibroblast-derived mouse-human hybrids (31 ). It was of interest, therefore, to determine if methylation has a transcription-independent role in the replication timing of F9 on the inactive X chromosome by examining our 5aC-treated inactive X hybrids for alterations in replication timing. We found that the time of F9 replication in one 5aC-treated clone, 19AS2, was advanced to that of an active X allele (Fig. 2 B). F9 reactivation was not observed in this hybrid (described in Materials and Methods section), as expected from the probable absence of appropriate tissue-specific transcription factors and from the observation that F9 is not expressed in active X hybrids. Thus, the 5aC-induced advance in F9 replication suggests that the control of replication time on the inactive X is methylation-dependent and transcription-independent with respect to this gene.
The observation that F9 replication could be advanced by 5aC treatment suggested that advanced replication timing might also be observed for other genes in weak or nonreactivated 5aC-treated hybrids. PGK1 reactivation in one clone, VI-1C1, was only about 5% the activity of active X alleles as determined by RT-PCR analysis [although PGK1 enzymatic activity was not detected (18 )]. The replication profile for PGK1 in VI-1C1 was not as advanced as in more fully reactivated clones, exhibiting a replication peak around S3 as compared to S2 for the more reactivated clones (Fig. 2 C). Similar to these examples of advanced PGK1 replication, partial advances in replication time were observed for XPCT in some 5aC-treated clones (compared to active X alleles; Fig. 1 F: III-14S7; Fig. 2 D: 15A and 19AS2) and these advances are associated with either partial (III-14S7) or no reactivation (15A and 19AS2). These cases of partially advanced replication do not appear to represent an average replication time for mosaic populations of shifted and unshifted cells within the hybrid clones because the replication peaks are as sharp as those of normal active X alleles.
Reactivation of inactive X alleles of HPRT or G6PD has not been observed in normal human fibroblasts treated with 5aC (15 ,46 ), and the frequency of reactivation in somatic cell hybrids varies considerably (10 -12 ,47 ,48 ). If replication timing has a significant influence on the rate of reactivation of these genes, the variable rate of reactivation among the different cell lines or strains could be derived from differences in replication time. In the fibroblast-derived inactive X hybrid, X8-6T2S1, both HPRT and G6PD are inactive and their replication is retarded relative to the corresponding genes in active X hybrids (Fig. 1 A and B). However, in contrast to the replication patterns of other tested genes in X8-6T2S1 that appear to be like those of the normal inactive X alleles (Fig. 1 , Fig. 3 , and data not shown), HPRT and G6PD replication in X8-6T2S1 is advanced compared to their counterparts in normal female fibroblasts (Fig. 3 ). Advanced replication is also seen for the inactive HPRT gene in 8121-TGRD, another somatic cell hybrid with an inactive X chromosome (in which G6PD is deleted). In this case, the replication pattern of the inactive HPRT gene is intermediate between the inactive gene in normal fibroblasts and that of the HPRT gene in X8-6T2S1 (Fig. 3 ). Although the HPRT promoter is extensively methylated in both of these hybrids (17 ), HPRT reactivation frequencies following 5aC treatment differed markedly between them: 3.4% HAT resistance for X8-6T2S1 clones and 0.46% for 8121-TGRD clones. The more advanced replication timing in X8-6T2S1 may, therefore, determine a more permissive chromosomal state with respect to reactivation. The high frequency of G6PD reactivation in X8-6T2S1 (48 ) suggests that advanced replication time also provides a more permissive state for reactivation of this gene.
Presented here are replication and transcription data for eight X-linked loci (XIST, XPCT, PGK1, HPRT, F9, FMR1, IDS, and G6PD). Strong reactivation of expression was associated with advanced time of replication in the cell cycle, as expected from the general finding that a gene replicates earlier in an expressing cell than in a nonexpressing cell. Among the more striking of our observations is that a reactivation-associated replication time advance is not always to the time of the active X gene and that a significant advance in time of replication can occur without reactivation. The latter observation is especially significant because it reveals the existence of previously unsuspected features regarding replication timing control and expression of X-linked genes.
The 5aC-induced advance in XIST replication time (Fig. 2 A) indicates that the control of replication time for a gene can be independent of transcription and promoter methylation as these properties were not affected by treatment. This conclusion is supported by the observation of advanced replication time for F9 following 5aC treatment (Fig. 2 B) because (i) F9 is not expressed in fibroblasts, (ii) its promoter is deficient in CpG dinucleotides, and (iii) no obvious sex-specific differences in methylation have been found in the vicinity of the promoter either in fibroblasts or in liver, a F9-expressing tissue (45 ).
Furthermore, replication timing for HPRT is not primarily controlled by promoter methylation or transcription factor binding because advanced replication of the gene in the nonexpressing X8-6T2S1 hybrid relative to a normal inactive X allele (Fig. 3 A) is not associated with any change in these parameters (17 ,49 ). Consistent with this view is our previous finding that inactive X `alleles' of HPRT and G6PD with advanced replication time are reactivated at high frequency in X8-6T2S1 (48 ). That such an advanced state of replication is more permissive to reactivation is supported by our report here that replication of HPRT in three 5aC-treated reactivant clones (III-5, III-14S7, and V-6CS1) was not advanced relative to replication in X8-6T2S1 (data not shown).
XIST expression is required for the establishment of X chromosome inactivation (36 ), but the function of the XIST gene product, a processed RNA, has not been established in somatic cells. Although XIST is ubiquitously expressed in female somatic cells, this expression is apparently not required for the maintenance of X inactivation because such repression is retained in cells with XIST deletions (50 ,51 ). These data do not exclude a role for XIST in the maintenance of X inactivation, however, because repression in cells lacking XIST is probably mediated by promoter methylation and/or other X inactivation mechanisms such as late replication. Our study of XIST argues against a role in the maintenance of X inactivation because 5aC-induced reactivation of several genes can occur in cells with normal levels of XIST; XIST expression, therefore, is notsufficient to maintain X inactivation in somatic cells.
In addition to our finding that the expressed allele of XIST and the repressed allele of F9 became earlier replicating after 5aC treatment, observations of advanced replication time following 5aC exposure have been reported for several nontranscribed loci, such as those containing satellite sequences (52 ). These data, and our other observations of advanced replication time following 5aC exposure, suggest that replication control loci are influenced by cytosine methylation at nonpromoter sites. The earlier replication of HPRT and G6PD in the X8-6T2S1 hybrid compared to normal inactive X alleles is expected to derive from demethylation events that occurred during cell fusion or subsequent cell culture.
We propose that the effect of 5aC on the timing of replication initiation involves demethylation of a defined control region at the affected locus, similar to the requirement of promoter demethylation for transcriptional activation. Alternatively, trans acting factors that influence replication timing might be activated by 5aC. Such factors would be expected to have pleiotropic effects, however, and none were observed in our 5aC-treated hybrid clones because they differed from each other in the number and identity of loci with advanced replication time.
One candidate locus in which methylation could directly affect the timing of replication is the region of replication initiation itself. Although mammalian replication can initiate in large zones, the preferred sites of initiation are more limited (53 ,54 ). The time in S phase when a replicon begins replication may depend on methylation of sites in the region of preferred initiation or methylation of a greater number of sites in the larger zone of initiation. Cytosine methylation has recently been reported for two mammalian replication origins (55 ), but its relevance to the stable transmission of replication timing information to daughter cells is unclear because it is not restricted to CpG sites and is only present in a limited stage of the cell cycle.
Locus control regions (LCRs) are also candidate regions for methylation control of replication time. X-linked loci of this type may have a primary role in the establishment and maintenance of X chromosome inactivation (56 ,57 ). The globin LCR, whose activity is altered by tissue-specific modifications, is known to affect the timing of origin activity (58 ,59 ) and is apparently required for normal origin function (60 ) even though it is located 50 kb away from that locus. Although X-linked LCRs have yet to be identified, methylation of such regions could, therefore, have a regulatory influence on replication initiation at distant origins.
Thus, we propose that a high level of methylation at replication origins, LCRs, or other yet-unidentified replication control regions on the inactive X chromosome results in late replicating, heterochromatic regions and that the corresponding loci on the active X chromosome are unmethylated. The unmethylated state of the active X loci leads to an advanced replication state, but not necessarily to a transcriptionally active one. Transcription within the replicon would require additional conditions; the primary condition for housekeeping genes with a 5' CpG island, such as those studied here, is an unmethylated promoter. Even under conditions of induced demethylation, therefore, reactivation of an inactive X-linked gene in normal cells would be unlikely because demethylation of several sites would be required in both the replication control region and the promoter.
Partial or full demethylation of the replication control region would bring about an advance in replication time and a permissive state in which reactivation could occur at a much higher frequency because only promoter demethylation is required. A compelling argument for the permissive effect of earlier replication can be invoked from the reactivation rates for the HPRT gene in two different hybrid cell lines containing an inactive human X chromosome (X8-6T2S1 and 8121-TGRD). The replication time shift of HPRT in X8-6T2S1 is more advanced than that of HPRT in 8121-TGRD (Fig. 3 ), and, as expected from the above hypothesis, the reactivation rate for X8-6T2S1 is an order of magnitude greater than that for 8121-TGRD. The inactive X in 8121-TGRD is not uniformly later than in X8-6T2S1 because PGK1 replicates earlier in 8121-TGRD than in X8-6T2S1 (Fig. 3 ). The lower frequency of PGK1 reactivation in X8-6T2S1 relative to HPRT (R. S. Hansen, unpublished RT-PCR data) may reflect the more normal replication timing of PGK1 in these cells (Fig. 3 ). These observations allow us to suggest that the wide variability in reactivation frequency reported for HPRT, G6PD, PGK1, and other X-inactivated genes in different hybrid cell lines (10 -12 ,47 ,48 ) could be explained in large part by differences between hybrids in the extent to which replication time has advanced for each gene.
A recent study of 5-methyl dCTP- and 5aC-induced alterations in HPRT expression provides an example of 5aC-induced reactivation in normal male fibroblasts (61 ). In this study, a high frequency of gene inactivation was observed following the introduction of 5-methyl dCTP into cells by electroporation, presumably because of increased methylation. Consistent with this view is their observation that reactivation of HPRT in such cultures could be induced by treatment with 5aC. Although no molecular or clonal analysis was carried out, this study appears to be the first example of 5aC-induced reactivation in a normal human fibroblast. As this is not a case of reversing X inactivation, the relatively high frequency of gene reactivation could be explained by our model of multiple, methylation-sensitive transcriptional control loci. It is likely that the 5-methyl dCTP-induced inactivation of the active X allele of HPRT occurred by methylation of only one methylation-sensitive control locus (the promoter or the replication time control locus) and the gene would, therefore, be in a more permissive state for reactivation by 5aC than would a normal inactive X allele.
Increased reactivation in somatic cell hybrids could also result from perturbation of additional elements that control the expression of an X-inactivated gene. Such an element might be a histone acetylation control mechanism because, in addition to promoter methylation and replication timing, histone acetylation also distinguishes the active and inactive X chromosomes (62 ). Differences in the extent to which general and locus-specific X inactivation control elements are perturbed in inactive X hybrids may explain more fully the observation (noted above) that such hybrids have different reactivation rates for the same gene.
Whether a methylation-sensitive control locus for replication timing is at the replication origin or at a more distant site, comparative restriction mapping with methyl-sensitive endonucleases between hybrids with shifted and unshifted alleles may reveal promising candidate regions for replication control. Candidate regions could then be tested for function as replication initiation sites and for properties of an LCR.
Standard growth conditions for cultured cells were as previously described (44 ). CHO-YH21 is a Chinese hamster ovary cell line (63 ). X8-6T2S1 is a 6-thioguanine-resistant human-hamster hybrid cell subclone that contains a single, human inactive X chromosome derived from a human fibroblast; Y162-11CS3 is a similarly derived hybrid that contains an active X chromosome (18 ,21 ,44 ,63 ). Hypoxanthine, aminopterin, thymidine (HAT)-resistant hybrid clones derived from X8-6T2 cells were described previously or similarly derived by 5aC treatment (18 ,44 ). GM06318 was obtained from the NIGMS Human Genetic Mutant Cell Repository and is a human fibroblast-hamster hybrid cell line that contains a single, active human X chromosome. 8121-TGRD is a human-hamster hybrid containing an inactive human X chromosome that is terminally-deleted distal to Xq26 (17 ,64 ). X8-6T2S1 and 8121-TGRD reactivation rates for HPRT were determined simultaneously by treating cells in log-phase growth with 5aC (44 ) and then selecting HPRT-positive clones with HAT medium after only 24 hours. Reactivation frequency was determined by counting the number of colonies that arose in plates with HAT medium in comparison to those with normal medium as described previously (44 ). HeLa cells were cultured in RPMI 1640 with 15% fetal bovine serum. This medium was also used to culture mitogen-stimulated peripheral lymphocytes from a normal female (194 female). Normal human fibroblasts (such as the female strain 78-18) were obtained as frozen stocks of early-passage strains from G. Martin (Department of Pathology, University of Washington) and grown in Chang's Medium (Irvine Scientific).
. Sequence tagged sites used for replication timing.
Gene
Primers
Sequence
Size (bp)
Reference1
F9
ixu5
ggcctcactcttgctagttcct
463
(29)
ixu5r
tggtgtttgggatgcctctccat
FMR1
fmr6904
tctgaccacagagacgaactcag
710
(29)
fmr8616
aagggatccatctgttgttcttcc
FMR1
fmr7U.1
gcagaaatgggcgttctggccct
320
(29)
fmr8L.1
cggccctccaccggaagtgaaac
FMR1
fmr9U.1
ggctgaagagaagatggaggagc
450
(29)
fmr10L.2
ggatcccgctgggagatgatgtttag
G6PD
g6pdd
gcagcagtggggtgaaaatacg
860
(29)
pd9
tgcaggccaacaatgtggtcct
HPRT
hpi
aggactgcgtgtgggaagagaa
380
(30)
hpt2312L
cctaatgccttttccctagttcc
IDS
idsE
ggttggcttcaatcctgatg
190
(68)
idsD
gctggaagggagcacatcac
MIC2
mic2
gcctgtcacagccacgccct
240
(30)
rmic2
ctccaccgccgcagatggaca
PGK1
pgkC2
gggttggggttgcgccttttccaa
220
(29)
pgkD2
acgccgcgaaccgcaaggaacct
PGK1
pgkix
gagtataaagggcatgaacaggt
680
M11966, M11967
pgkixr
agtgctcacatggctgactttatc
XIST
xst31
ccttcagttcttaaagcgct
530
(69)
xst29r
atcagcaggtatccgatacc
XPCT
dxs128-1
cttggtttgaccttggtcaa
200
(70)
dxs128-2
ggtaaatctgctttgtcaga
XPCT
xpct675U
actcccatgtctgcatgcctgaggt
250
U05316
xpct928L
agtcgggagccgaagctcagttcaa
1(Reference number) or Genbank accession number of parent sequence used for the development of unpublished sequence tagged sites.
Cells were labeled with 5-bromo-2'-deoxyuridine (BrdU) and processed for flow cytometry as previously described (29 ,30 ). BrdU-labeled cells were separated into different phases of the cell cycle on an EPICS Elite cell sorter and BrdU-containing DNA (BrdU-DNA) was isolated from these cells by antibody precipitation as described previously (29 ). BrdU-labeled CHO-YH21 DNA was added to sorted human cells and BrdU-labeled Drosophila DNA was added to sorted human-hamster hybrid cells to provide controls for recovery and amplification of newly-replicated BrdU-DNA (29 ,30 ).
Replication timing for human genes was determined using sequence tagged site (STS) primer sets described in Table 2 . Hamster (28S and Aprt genes) or Drosophila (Gbe gene) STSs used to analyze added control BrdU-DNA were described previously (29 ,30 ). All PCR reactions contained 2.5 U AmpliTaq (Perkin Elmer Cetus) and a portion of antibody-purified BrdU-DNA corresponding to 1000 sorted cells in 100 [mu]l standard reaction buffer (Perkin Elmer Cetus). Reactions were started from ice by placing in a water-cooled thermocycler (Ericomp) maintained at the denaturation temperature (94oC or 95oC). PCR parameters have previously been described for several of the STSs listed in Table 2 (29 ,30 ). Additional reaction conditions are as follows: idsE:idsD, 95oC 5 min, 25 cycles of 95oC 30 s, 60oC 30 s, 72oC 2 min, final extension at 72oC for 7 min; pgkix:pgkixr, 95oC 5 min, 27 cycles of 95oC 1 min, 60oC 1 min, 72oC 2 min, final extension at 72oC for 7 min;xpct675U:xpct928L, 95oC 5 min, 25 cycles of 95oC 1 min, 65oC 1 min, 75oC 2 min, final extension at 75oC for 7 min; dxs128-1:dxs128-2, 94oC 5 min, 25 cycles of 94oC 1 min, 58oC 1 min, 72oC 2 min, final extension at 72oC for 7 min.
Agarose gel electrophoresis of PCR products, Southern transfer, probe isolation, probe labeling, hybridization conditions, and membrane washing procedures were as previously described (21 ,29 ). Quantitation of hybridization signals was performed by PhosphorImager analysis (Molecular Dynamics; Phosphor- Imager Analysis Facility of the Markey Molecular Medicine Center at the University of Washington). Replication of a sequence in a particular cytometry fraction is expressed as a percentage of the sum of that in all fractions.
Prior to anti-BrdU antibody purification, BrdU-DNA from certain replication profiles was digested with either HpaII or SacII as previously described (30 ,33 ). BrdU-DNA was then isolated by antibody precipitation.
Total RNA was isolated by the acid guanidinium thiocyanate method (16 ,65 ). First strand synthesis of cDNA was generated with random hexamers as previously described (16 ) from 0.25, 0.50, and 1.0 [mu]g RNA. Control reactions that contained no reverse transcriptase were performed with 0.5 [mu]g RNA. Ten percent of the cDNA product was used for PCR amplification using the standard reaction constituents described for amplification of BrdU-DNA. Amplification ofcDNA was performed using the following primer sets: XIST, xst30:29r (30 ) or xst8:9r [sequences in Figure 2 of reference (50 )]; XPCT, xpct-A3.2:A5.2 (66 ); PGK1, pgk1-R3:R4 (16 ); HPRT, hprt-1:3 (16 ); FMR1, fmr-4925 (67 ) and fmr20 (5'-catttctcattatcagagacaa-3'; human-specific primer derived from sequence HSFMR1A, Genbank no. X69962); IDS, idsE:D (68 ); G6PD, PD9:PD13R (16 ); MIC2, XMIC:XMIC2R (16 ). PCR parameters were similar to those previously published [see references above and (30 )], limiting amplification to between 18 and 21 cycles so as to remain in an exponential range. Fifteen percent of the PCR product was subject to gel electrophoresis, Southern analysis, and PhosphorImager quantification. Hybridization per ng of input RNA was calculated and averaged for PCR products obtained from 0.5 and 1.0 [mu]g RT samples and these values were normalized for cDNA recovery and amplification potential assuming equal expression of MIC2 among the various hybrids (normalization factors were obtained from values for XMIC:XMIC2R products derived from the same RT reactions). RT-PCR analysis of F9 expression in somatic cell hybrids and in human liver was performed similarly, using 0.2 [mu]M exon 4:exon 8 primers (IX-7, 5'-ttaaatggcggcagttgca-3' and chen-110, 5'-gaataattcgaatcacatt-3'; sequences derived from HUMFIX, Genbank no. J00136). One quarter of the RT product (derived from 0.125 or 0.25 [mu]g RNA) was amplified under the following conditions: 94oC 5 min, 40 cycles of 94oC 1 min, 50oC 1 min, 72oC 2 min, final extension at 72oC for 7 min. The PCR reactions were analyzed by agarose electrophoresis and ethidium bromide staining. Liver RNAs were positive for the 600 bp F9 band, whereas this product was not observed from hybrids with an active or inactive X chromosome, or from inactive X hybrid clones such as 19AS2 that had been treated with 5aC.
S. Chong Kim performed the flow cytometry fractionation of cells. Shi Han Chen provided RT-PCR primers for F9 analysis. We thank Lester Goldstein, Bonny Brewer, and Charles Laird for helpful comments on the manuscript.
This work was supported by National Institutes of Health grant HD16659.
1 Comings, D.E. (1966) The inactive X chromosome. Lancet, 2, 1137-1138.
2 Lyon, M.F. (1961) Gene action in the X-chromosome of the mouse. Nature, 190, 372-373.
3 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.MEDLINE Abstract
4 Sager, R. and Kitchin, R. (1975) Selective silencing of eukaryotic DNA. Science, 189, 426-433.MEDLINE Abstract
5 Holliday, R. and Pugh, J.E. (1975) DNA modification mechanisms and gene activity during development. Science, 187, 226-232.MEDLINE Abstract
6 Riggs, A.D. (1975) X chromosome inactivation, differentiation and DNA methylation. Cytogenet. Cell Genet., 14, 9-25.MEDLINE Abstract
7 Benedict, W.F., Banerjee, A., Gardner, A. and Jones, P.A. (1977) Induction of morphological transformation in mouse C3H/10T1/2 clone 8 cells and chromosomal damage in hamster A(T1)C1-3 cells by cancer chemotherapeutic agents. Cancer Res., 37, 2202-2208.MEDLINE Abstract
8 Constantinides, P.G., Jones, P.A. and Gevers, W. (1977) Functional striated muscle cells from non-myoblast precursors following 5-azacytidine treatment. Nature, 267, 364-366.MEDLINE Abstract
9 Jones, P.A. and Taylor, S.M. (1980) Cellular differentiation, cytidine analogs and DNA methylation. Cell, 20, 85-93.MEDLINE Abstract
10 Graves, J.A. (1982) 5-azacytidine-induced re-expression of alleles on the inactive X chromosome in a hybrid mouse cell line. Exp. Cell Res., 141, 99-105.MEDLINE Abstract
11 Lester, S.C., Korn, N.J. and DeMars, R. (1982) Derepression of genes on the human X chromosome in a hybrid mouse cell line: Evidence for differences in locus-specific rates of derepression and rates of transfer of active and inactive genes after DNA-mediated transformation. Somat. Cell Genet., 8, 265-284.
12 Hors-Cayla, M.C., Heuertz, S. and Frezal, J. (1983) Coreactivation of four inactive X genes in a hamster * human hybrid and persistence of late replication of reactivated X chromosome. Somat. Cell. Genet., 9, 645-657.MEDLINE Abstract
13 Beggs, A.H., Axelman, J. and Migeon, B.R. (1986) Reactivation of X-linked genes in human fibroblasts transformed by origin-defective SV40. Somat. Cell Mol. Genet., 12, 585-594.MEDLINE Abstract
14 Grant, S.G. and Worton, R.G. (1989) Differential activation of the hprt gene on the inactive X chromosome in primary and transformed Chinese hamster cells. Mol. Cell. Biol., 9, 1635-1641.MEDLINE Abstract
15 Wolf, S.F. and Migeon, B.R. (1982) Studies of X chromosome DNA methylation in normal human cells. Nature, 295, 667-671.MEDLINE Abstract
16 Sasaki, T., Hansen, R.S. and Gartler, S.M. (1992) Hemimethylation and hypersensitivity are early events in transcriptional reactivation of human inactive X-linked genes in a hamster * human somatic cell hybrid. Mol. Cell. Biol., 12, 3819-3826.MEDLINE Abstract
17 Hornstra, I.K. and Yang, T.P. (1994) High-resolution methylation analysis of the human hypoxanthine phosphoribosyltransferase gene 5' region on the active and inactive X chromosomes: correlation with binding sites for transcription factors. Mol. Cell. Biol., 14, 1419-1430.MEDLINE Abstract
18 Hansen, R.S. and Gartler, S.M. (1990) 5-Azacytidine-induced reactivation of the human X chromosome-linked PGK1 gene is associated with a large region of cytosine demethylation in the 5' CpG island. Proc. Natl Acad. Sci. USA, 87, 4174-4178.MEDLINE Abstract
19 Pfeifer, G.P., Steigerwald, S.D., Hansen, R.S., Gartler, S.M. and Riggs, A.D. (1990) Polymerase chain reaction-aided genomic sequencing of an X chromosome-linked CpG island: methylation patterns suggest clonal inheritance, CpG site autonomy, and an explanation of activity state stability. Proc. Natl Acad. Sci. USA, 87, 8252-8256.MEDLINE Abstract
20 Toniolo, D., Martini, G., Migeon, B.R. and Dono, R. (1988) Expression of the G6PD locus on the human X chromosome is associated with demethylation of three CpG islands within 100 kb of DNA. EMBO J., 7, 401-406.MEDLINE Abstract
21 Hansen, R.S., Gartler, S.M., Scott, C.R., Chen, S.-H. and Laird, C.D. (1992) Methylation analysis of CGG sites in the CpG island of the human FMR1 gene. Hum. Mol. Genet., 8, 571-578.
22 Gartler, S.M., Dyer, K.A., Graves, J.A. and Rocchi, M. (1985) A two step model for mammalian X-chromosome inactivation. Prog. Clin. Biol. Res., 198, 223-235.MEDLINE Abstract
23 Paterno, G.D., Adra, C.N. and McBurney, M.W. (1985) X chromosome reactivation in mouse embryonal carcinoma cells. Mol. Cell. Biol., 5, 2705-2712.MEDLINE Abstract
24 Hatton, K.S., Dhar, V., Brown, E.H., Iqbal, M.A., Stuart, S., Didamo, V.T. and Schildkraut, C.L. (1988) Replication program of active and inactive multigene families in mammalian cells. Mol. Cell. Biol., 8, 2149-2158.MEDLINE Abstract
25 Hatton, K.S. and Schildkraut, C.L. (1990) The mouse immunoglobulin kappa light-chain genes are located in early- and late-replicating regions of chromosome 6. Mol. Cell. Biol., 10, 4314-4323.MEDLINE Abstract
26 Brown, E.H., Iqbal, M.A., Stuart, S., Hatton, K.S., Valinsky, J. and Schildkraut, C.L. (1987) Rate of replication of the murine immunoglobulin heavy-chain locus: evidence that the region is part of a single replicon. Mol. Cell. Biol., 7, 450-457.MEDLINE Abstract
27 Gartler, S.M., Dyer, K.A. and Goldman, M.A. (1992) Mammalian X chromosome inactivation. Mol. Genet. Med., 2, 121-160.MEDLINE Abstract
28 Boggs, B.A. and Chinault, A.C. (1994) Analysis of replication timing properties of human X-chromosomal loci by fluorescence in situ hybridization. Proc. Natl Acad. Sci. USA, 91, 6083-6087.MEDLINE Abstract
29 Hansen, R.S., Canfield, T.K., Lamb, M.M., Gartler, S.M. and Laird, C.D. (1993) Association of fragile X syndrome with delayed replication of the FMR1 gene. Cell, 73, 1403-1409.MEDLINE Abstract
30 Hansen, R.S., Canfield, T.K. and Gartler, S.M. (1995) Reverse replication timing for XIST in human fibroblasts. Hum. Mol. Genet., 4, 813-820.MEDLINE Abstract
31 Schmidt, M. and Migeon, B.R. (1990) Asynchronous replication of homologous loci on human active and inactive X chromosomes. Proc. Natl Acad. Sci. USA, 87, 3685-3689.MEDLINE Abstract
32 Torchia, B.S., Call, L.M. and Migeon, B.R. (1994) DNA replication analysis of FMR1, XIST, and factor 8C loci by FISH shows nontranscribed X-linked genes replicate late. Am. J. Hum. Genet., 55, 96-104.MEDLINE Abstract
33 Kawame, H., Gartler, S.M. and Hansen, R.S. (1995) Allele-specific replication timing in imprinted domains: absence of asynchrony at several loci. Hum. Mol. Genet., 4, 2287-2293.MEDLINE Abstract
34 Shafer, D.A. and Priest, J.H. (1984) Reversal of DNA methylation with 5-azacytidine alters chromosome replication patterns in human lymphocyte and fibroblast cultures. Am. J. Hum. Genet., 36, 534-545.MEDLINE Abstract
35 Jablonka, E., Goitein, R., Marcus, M. and Cedar, H. (1985) DNA hypomethylation causes an increase in DNase-I sensitivity and an advance in the time of replication of the entire inactive X chromosome. Chromosoma, 93, 152-156.MEDLINE Abstract
36 Penny, G.D., Kay, G.F., Sheardown, S.A., Rastan, S. and Brockdorff, N. (1996) Requirement for Xist in X chromosome inactivation. Nature, 379, 131-137.MEDLINE Abstract
37 Brockdorff, N., Ashworth, A., Kay, G.F., Cooper, P., Smith, S., McCabe, V.M., Norris, D.P., Penny, G.D., Patel, D. and Rastan, S. (1991) Conservation of position and exclusive expression of mouse Xist from the inactive X chromosome. Nature, 351, 329-331.MEDLINE Abstract
38 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 Abstract
39 Heard, E., Simmler, M.C., Larin, Z., Rougeulle, C., Courtier, B., Lehrach, H. and Avner, P. (1993) Physical mapping and YAC contig analysis of the region surrounding Xist on the mouse X chromosome. Genomics, 15, 559-569.MEDLINE Abstract
40 Hendrich, B.D., Brown, C.J. and Willard, H.F. (1993) Evolutionary conservation of possible functional domains of the human and murine XIST genes. Hum. Mol. Genet., 2, 663-672.MEDLINE Abstract
41 Lafreniere, R.G. and Willard, H.F. (1993) Pulsed-field map of Xq13 in the region of the human X inactivation center. Genomics, 17, 502-506.MEDLINE Abstract
42 Norris, D.P., Patel, D., Kay, G.F., Penny, G.D., Brockdorff, N., Sheardown, S.A. and Rastan, S. (1994) Evidence that random and imprinted Xist expression is controlled by preemptive methylation. Cell, 77, 41-51.MEDLINE Abstract
43 Yoshitake, S., Schach, B.G., Foster, D.C., Davie, E.W. and Kurachi, K. (1985) Nucleotide sequence of the gene for human factor IX (antihemophilic factor B). Biochemistry, 24, 3736-3750.MEDLINE Abstract
44 Hansen, R.S., Ellis, N.A. and Gartler, S.M. (1988) Demethylation of specific sites in the 5' region of the inactive X-linked human phosphoglycerate kinase gene correlates with the appearance of nuclease sensitivity and gene expression. Mol. Cell. Biol., 8, 4692-4699.MEDLINE Abstract
45 Cullen, C.R., Hubberman, P., Kaslow, D.C. and Migeon, B.R. (1986) Comparison of factor IX methylation on human active and inactive X chromosomes: implications for X inactivation and transcription of tissue-specific genes. EMBO J., 5, 2223-2229.MEDLINE Abstract
46 Migeon, B.R., Axelman, J. and Beggs, A.H. (1988) Effect of ageing on reactivation of the human X-linked HPRT locus. Nature, 335, 93-96.MEDLINE Abstract
47 Mohandas, T., Sparkes, R.S., Bishop, D.F., Desnick, R.J. and Shapiro, L. (1984) Frequency of reactivation and variability in expression of X-linked loci. Am. J. Hum. Genet., 36, 916-925.MEDLINE Abstract
48 Ellis, N., Keitges, E., Gartler, S.M. and Rocchi, M. (1987) High-frequency reactivation of X-linked genes in Chinese hamster * human hybrid cells. Somat. Cell. Mol. Genet., 13, 191-204.MEDLINE Abstract
49 Hornstra, I.K. and Yang, T.P. (1992) Multiple in vivo footprints are specific to the active allele of the X-linked human hypoxanthine phosphoribosyltransferase gene 5' region: implications for X chromosome inactivation. Mol. Cell. Biol., 12, 5345-5354.MEDLINE Abstract
50 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 Abstract
51 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., Monaco, A.P. and Buckle, V.J. (1994) Absence of the XIST gene from late-replicationg isodicentric X chromosomes in leukaemia. Hum. Mol. Genet., 3, 1053-1059.MEDLINE Abstract
52 Selig, S., Ariel, M., Goitein, R., Marcus, M. and Cedar, H. (1988) Regulation of mouse satellite DNA replication time. EMBO J., 7, 419-426.MEDLINE Abstract
53 Burhans, W.C. and Huberman, J.A. (1994) DNA replication origins in animal cells: a question of context? Science, 263, 639-640.MEDLINE Abstract
54 Hamlin, J.L. and Dijkwel, P.A. (1995) On the nature of replication origins in higher eukaryotes. Curr. Opin. Genet. Dev., 5, 153-161.MEDLINE Abstract
55 Tasheva, E.S. and Roufa, D.J. (1994) Densely methylated DNA islands in mammalian chromosomal replication origins. Mol. Cell. Biol., 14, 5636-5644.MEDLINE Abstract
56 Gartler, S.M. and Riggs, A.D. (1983) Mammalian X-chromosome inactivation. Annu. Rev. Genet., 17, 155-190.MEDLINE Abstract
57 Goldman, M.A. (1988) The chromatin domain as a unit of gene regulation. BioEssays, 9, 50-55.MEDLINE Abstract
58 Forrester, W.C., Epner, E., Driscoll, M.C., Enver, T., Brice, M., Papayannopoulou, T. and Groudine, M. (1990) A deletion of the human beta-globin locus activation region causes a major alteration in chromatin structure and replication across the entire beta-globin locus. Genes Dev., 4, 1637-1649.MEDLINE Abstract
59 Kitsberg, D., Selig, S., Keshet, I. and Cedar, H. (1993) Replication structure of the human beta-globin gene domain. Nature, 366, 588-590.MEDLINE Abstract
60 Aladjem, M.I., Groudine, M., Brody, L.L., Dieken, E.S., Fournier, R.E.K., Wahl, G.M. and Epner, E.M. (1995) Participation of the human [beta]-globin locus control region in initiation of DNA replication. Science, 270, 815-819.MEDLINE Abstract
61 Holliday, R. and Ho, T. (1995) Evidence for gene silencing by DNA methylation in normal human diploid fibroblasts. Som. Cell Mol. Genet., 21, 215-218.MEDLINE Abstract
62 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.MEDLINE Abstract
63 Rosenstraus, M. and Chasin, L.A. (1975) Isolation of mammalian cell mutants deficient in glucose-6-phosphate dehydrogenase activity: linkage to hypoxanthine phosphoribosyl transferase. Proc. Natl Acad. Sci. USA, 72, 493-497.MEDLINE Abstract
64 Ledbetter, S.A., Schwartz, C.E., Davies, K.E. and Ledbetter, D.H. (1991) New somatic cell hybrids for physical mapping in distal Xq and the fragile X region. Am. J. Med. Genet., 38, 418-420.MEDLINE Abstract
65 Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162, 156-159.MEDLINE Abstract
66 Lafreniere, R.G., Carrel, L. and Willard, H.F. (1994) A novel transmembrane transporter encoded by the XPCT gene in Xq13.2. Hum. Mol. Genet., 3, 1133-1139.MEDLINE Abstract
67 Pieretti, M., Zhang, F., Fu, Y.H., Warren, S.T., Oostra, B.A., Caskey, C.T. and Nelson, D.L. (1991) Absence of expression of the FMR-1 gene in Fragile X syndrome. Cell, 66, 817-822.MEDLINE Abstract
68 Flomen, R.H., Green, P.M., Bentley, D.R., Giannelli, F. and Green, E.P. (1992) Detection of point mutations and a gross deletion in six Hunter syndrome patients. Genomics, 13, 543-550.MEDLINE Abstract
69 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.MEDLINE Abstract
70 Lafreniere, R.G., Brown, C.J., Rider, S., Chelly, J., Taillon, M.P., Chinault, A.C., Monaco, A.P. and Willard, H.F. (1993) 2.6 Mb YAC contig of the human X inactivation center region in Xq13: physical linkage of the RPS4X, PHKA1, XIST and DXS128E genes. Hum. Mol. Genet., 2, 1105-1115.MEDLINE Abstract
*To whom correspondence should be addressed
This page is maintained by OUP admin. Last updated Thu Oct 31 15:27:05 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
