How does inactivation change timing of replication in the human X chromosome?

Histone acetylation controls the inactive X chromosome replication dynamics

In mammals, dosage compensation between male and female cells is achieved by inactivating one female X chromosome (Xi). Late replication of Xi was proposed to be involved in the maintenance of its silenced state. Here, we show a highly synchronous replication of the Xi within 1 to 2 h during early-mid S-phase by following DNA replication in living mammalian cells with green fluorescent protein-tagged replication proteins. The Xi was replicated before or concomitant with perinuclear or perinucleolar facultative heterochromatin and before constitutive heterochromatin. Ectopic expression of the X-inactive-specific transcript (Xist) gene from an autosome imposed the same synchronous replication pattern. We used mutations and chemical inhibition affecting different epigenetic marks as well as inducible Xist expression and we demonstrate that histone hypoacetylation has a key role in controlling Xi replication. The epigenetically controlled, highly coordinated replication of the Xi is reminiscent of embryonic genome replication in flies and frogs before genome activation and might be a common feature of transcriptionally silent chromatin.

High-resolution R-banding at the 1250-band level. III. Comparative analysis of morphologic and dynamic R-band patterns (RHG and RBG)

High-resolution human chromosomes were obtained from lymphocytes after thymidine synchronization. The block was released either with thymidine to produce GTG (G-bands by trypsin using Giemsa) and RHG (R-bands by heating using Giemsa) banding or with BrdU (5-bromo-2'-deoxyuridine) for RBG (R-bands by BrdU using Giemsa) banding. RHG and RBG band patterns are only 75 to 85% congruent. The dissimilarities increase with the band number per genome and vary from one chromosome region to another. After high-resolution RBG banding, the BrdU-substituted bands show an unequal condensation delay, which can be, according to the bands involved, very important, minimal, or even absent. The bands showing the highest degree of condensation delay are the bands replicating the latest. The GTG- and RHG-band patterns show complementary matching for about 90% of the bands. It was found that two third of the chromosome surface appears positively stained after R-banding. This suggests that more DNA is replicated during early S-phase than during late S-phase. To obtain a fully developed RBG-band pattern in 90 to 95% of harvested mitoses, a period of 4.5 hours after the removal of the blocking agent is optimal. Such a brief release period also implies that late S-phase is much shorter than early S-phase.

Expression of H19 does not influence the timing of replication of the Igf2/H19 imprinted region

Allele specific timing of replication is believed to be a hallmark of imprinted genes, however recent evidence suggests that this might not be the case for the insulin-like growth factor 2 (Igf2) and H19 locus. In this report, we assayed the timing of replication of Igf2 and H19 in two mouse embryonic cell lines expressing both H19 and Igf2, and one cell line maternally disomic for the Igf2/H19 mouse locus which expresses H19 but not Igf22. In all cell lines, Igf2 and H19 were replicated early in the S phase of the cell cycle, and both alleles replicated at the same time. This indicates that any differences in the timing of replication at the Igf2/H19 locus are of a lesser magnitude than those found in other imprinted regions.

The spreading of X inactivation into autosomal material of an x;autosome translocation: evidence for a difference between autosomal and X-chromosomal DNA

X inactivation involves initiation, propagation, and maintenance of genetic inactivation. Studies of replication timing in X;autosome translocations have suggested that X inactivation may spread into adjacent autosomal DNA. To examine the inactivation of autosomal material at the molecular level, we assessed the transcriptional activity of X-linked and autosomal loci spanning an inactive translocation in a phenotypically normal female with a karyotype of 46,X,der(X)t(X;4)(q22;q24). Since 4q duplications usually manifest dysmorphic features and severe growth and mental retardation, the normal phenotype of this individual suggested the spreading of X inactivation throughout the autosomal material. Consistent with this model, reverse transcription-PCR analysis of 20 transcribed sequences spanning 4q24-qter revealed that three known genes and 11 expressed sequence tags (ESTs) were not expressed in a somatic-cell hybrid that carries the translocation chromosome. However, three ESTs and three known genes were expressed from the t(X;4) chromosome and thus "escaped" X inactivation. This direct assay of expression demonstrated that the spreading of inactivation from the adjoining X chromosome was incomplete and noncontiguous. These findings are broadly consistent with the existence of genes known to escape inactivation on normal inactive X chromosomes. However, the fact that a high proportion (30%) of tested autosomal genes escaped inactivation may indicate that autosomal material lacks X chromosome-specific features that are associated with the spreading and/or maintenance of inactivation.

Earlier finishing of Xp21.2 subband replication of the inactive X chromosome in Rett syndrome girl but not in her 47,XXX mother

X-inactivation mosaicism has been proposed to explain the origin of Rett syndrome. We present the results of the cytogenetic analysis, including RBG dynamic replication pattern, in a girl with Rett syndrome. The late replicating X chromosome (LRX) showed the earlier replication of subband Xp21.2 in 36% of analysed cells. Unexpectedly the maternal karyotype 47,XXX was found. Replication timing of both maternal LRX chromosomes was normal. The critical region of Xp essential for RS is proposed.

A comparative study of X-inactivation in Rett syndrome probands and control subjects

X-inactivation has been studied in a series of monozygotic female twins and their female relatives by a PCR method which detects methylation at the androgen receptor locus (HUMARA). The results obtained are compared to those from an earlier study employing probe M27 beta which detects locus DXS255. Analysis of X-inactivation in girls with Rett syndrome and their mothers by four different methods did not indicate a direct relationship between non-random inactivation of the X-chromosome and the presence of the disease with the exception that any skewing detected in the probands tended to favour the preferential inactivation of the paternally inherited X-chromosome. No evidence for the involvement of uniparental disomy in the etiology of the disease was found.

Alterations in replication timing of X-chromosome bands in Rett syndrome

A cytogenetic study has been carried out on 30 girls affected with Rett syndrome, 35 of their family members and 25 unrelated healthy control females. Karyotyping at the 850 band level revealed no detectable chromosome abnormalities in either the affected girls or their families. Observations on the sequence of the appearance of early replicating bands on both the active and the inactive X-chromosome demonstrated the same replication patterns in all of the groups investigated with the exception that band Xp21 appeared with greater frequency in the Rett syndrome cells. A degree of variation was detected both between and within the subjects when the timing of the latest bands to replicate was investigated for the active X, but the same consensus order was obtained for all groups. A comparable number of elongated X-chromosomes was found in the girls with Rett syndrome (8%) when compared to their mothers (12%) when synchronized cells were treated with a short pulse of BrdU. If a disturbance in X-inactivation does contribute to the aetiology of Rett syndrome, it is at a level which is not detected by observations on the relative timing of replication of chromosome bands.

X-inactivation in girls with Rett syndrome

Cytogenetic studies have been carried out on a series of nine girls with Rett syndrome, six of their mothers and nine normal female controls. No abnormality of the X-chromosome has been observed in any subject. X-inactivation studies using various methods of detecting the timing of individual band replication were performed. The overall pattern seen was essentially the same in all subjects, but in the patients with Rett syndrome there may be an alteration in the timing of the X-inactivation process in the region Xp11.3 or 4-->Xp21.

Demonstration of replication patterns in the last premeiotic S-phase of male Chinese hamsters after BrdU pulse labeling

Chromosome replication in the last premeiotic S-phase of male mammals has been previously studied by [3H]thymidine autoradiography and by a 5-bromodeoxyuridine (BrdU)/Giemsa technique. We used a recently developed BrdU-antibody technique (BAT) in this study. The following conclusions were drawn: (1) The replication patterns observed are similar to that of somatic cells. (2) The heterochromatin starts replication in early S-phase. (3) The euchromatic part of the X chromosome of the male Chinese hamster replicates together with the autosomes and therefore behaves isocyclicly and not allocyclicly as hitherto assumed. Hence, genetic inactivity of the X chromosome may be brought about by a mechanism different from that in somatic cells.

Replication asynchrony between homologs 15q11.2: cytogenetic evidence for genomic imprinting

Replication kinetics of the Prader-Willi syndrome critical region (15q11.2) was investigated in seven normal healthy adult females using RBG replication bands. Replication asynchrony between homologs 15q11.2 was identified consistently in about 40% of cells in all individuals. It was limited to the stages in which Xp22, Xp11, Xq13 and Xq24/26 were visible in the late-replicating X chromosome. This asynchrony suggested that replication timing overlapped between 15q11.2 and the early replicating R-bands of the late X chromosome in some cells, and that the difference in replication timing between homologs was probably related to genomic imprinting; the latter has been suggested as a pathogenetic basis of Prader-Willi syndrome. As a result of an analysis of the proportions of asynchronous and synchronous cells in each replication stage, two types of cells were deduced providing 1:1 methylation mosaicism of genomic imprinting was assumed. The first type was composed of cells with normal replication in one homolog and delayed replication in the other. The second type was composed of cells with normal replication in both homologs. Our results provide cytogenetic evidence of methylation mosaicism for mammalian genomic imprinting.

Analysis of DNA replication during S-phase by means of dynamic chromosome banding at high resolution

The characteristic patterns of dynamic banding (replication banding) were analysed. Extremely high resolution (850 to 1,250 bands per genome) G- and R-band patterns were obtained after 5-bromo-2'-deoxyuridine (BrdUrd) incorporation either during the early or the late S-phase. We synchronized human lymphocytes with high concentrations of thymidine or BrdUrd as blocking agents, followed by low concentrations of BrdUrd or thymidine respectively as releasing agents, and obtained R- or G-band patterns respectively. The dynamic R- and G-band patterns were complementary for all chromosomes, even for the late-replicating X chromosome. There was no overlapping and every part of each chromosome was positively stained by one of the two banding procedures. The complementarity of the two patterns shows that both high thymidine and high BrdUrd concentrations blocked S-phase progression near the R-band to G-band replication transition in the middle of S-phase. Some bands of the inactive X chromosome replicate before this transition concurrently with R-band replication. The 48 different telomeric regions could be classified into 5 distinct morphotypes based upon the distribution of early and late-replicating DNA in each telomeric region. The dynamic band patterns are particularly useful for the study of the structural and physiological organization of chromosomes at high resolution and should prove invaluable for assessing the replication behavior of rearranged chromosomes.

Analysis of chromosome replication by a BrdU antibody technique

Chromosome replication was studied without synchronization in human lymphocyte and amniotic cell cultures visualizing very short 5-bromodeoxyuridine (BrdU) pulses by an immunologic technique (BAT). The findings agree in general with those facts known from earlier BrdU staining techniques. The very high sensitivity of BAT was shown to allow the detection of replication in a band where 1 in 200 nucleotides is replaced by BrdU. The main observations are: though the replication patterns after BAT appear strange the bands correspond to those described by the Paris Conference (1971). At the beginning of the S-phase a stepwise onset of replication in only a subset of R-bands is confirmed. There is a considerable difference in the sensitivity between early and late S (SE and SL) for the detection of BrdU pulses. This difference probably reflects a different spatial arrangement of chromatin in R-bands as compared with G-bands below the level of cytogenetic analysis. The use of short pulses did not reveal any additional subdivision of SE or SL. The correspondence between chromosomal bands and replicon clusters is discussed briefly with respect to the different time they need for replication.