Human Molecular Genetics, 2002, Vol. 11, No. 25 3145-3156
© 2002 Oxford University Press
Molecular and cytogenetic analysis of the spreading of X inactivation in X;autosome translocations
1Wessex Regional Genetics Laboratory, Salisbury District Hospital, Salisbury, Wiltshire, SP2 8BJ, UK and 2Chromatin and Gene Expression Group, University of Birmingham Medical School, Edgbaston, Birmingham, B15 2TT, UK
Received July 29, 2002; Revised September 19, 2002; Accepted October 2, 2002
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
We have performed detailed studies of the spreading of X inactivation in five unbalanced human X;autosome translocations. Using allele-specific RT–PCR we observed long-range silencing of autosomal genes located up to 45 Mb from the translocation breakpoint, directly demonstrating the ability of X inactivation to spread in cis through autosomal DNA. Spreading of gene silencing occurred in either a continuous or discontinuous fashion in different cases, suggesting that some autosomal DNA is resistant to the X inactivation signal. This spread of inactivation was accompanied by, but not dependent upon, CpG island methylation. Observations of late-replication, histone acetylation and histone methylation show that X inactivation can spread in the absence of cytogenetic features normally associated with the inactive X. However, the distribution of histone modifications which distinguish the inactive X are more accurate cytogenetic measures of the spread of X inactivation than late-replication. Overall, despite remarkable variation in the spread of X inactivation among the five cases there was good correlation between the pattern of gene silencing and the attenuation of clinical phenotype associated with each partial autosomal trisomy. We discuss our observations in the context of hypotheses which address the spread of X inactivation.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
X inactivation is a mechanism of dosage compensation which silences the majority of genes on one X chromosome in somatic cells of female mammals. The inactive X is distinguished by a number of heterochromatic features, such as hypoacetylation of histones H2A, H3 and H4 (1), depletion of H3 dimethylated at lysine 4 (2), late-replication (3), and CpG island methylation (4). However, while each of these features is undoubtedly important in the establishment and/or maintenance of the inactive state, their exact role and the way in which the X inactivation signal is propagated along the chromosome remains unclear. X inactivation requires the presence in cis of the X inactivation centre (XIC ) located at Xq13.2 (5) and is thought to be mediated by transcripts from the XIST gene which specifically coat the inactive X during much of the cell cycle (6).
Using coat colour variegation as a marker of genetic silencing, Russell (7) first observed that in mice carrying X;autosome translocations X inactivation can spread in a variable and limited fashion into autosomal DNA. These early observations have been confirmed and extended by other studies (Reviewed in 8,9). However, in all but a few cases evidence for the extent of spread of X inactivation into autosomal DNA has been based upon either replication-timing studies or the associated clinical phenotype in carrier individuals. Recently we demonstrated that late-replication is a poor correlate of the spread of gene silencing (10), thus forcing a reassessment of studies which utilised this parameter to determine the spread of X inactivation.
While observations in four X;autosome rearrangements have noted the inactivation of single autosomal genes (11–14), only one other study has directly measured the spread of X inactivation through autosomal DNA by gene expression analysis (15). White et al. found that the spread of inactivation in an X;4 translocation was discontinuous, with active genes interspersed among inactive regions of chromatin. This mirrors similar observations of discontinuous spreading of late-replication (12,16), suggesting that some autosomal chromatin lacks features important in the spread and/or maintenance of X inactivation.
More recent studies of several mouse and human X;autosome rearrangements have found little or no spreading of XIST/Xist RNA and histone hypoacetylation into autosomal chromatin (17,18). Extending a model first put forward by Gartler and Riggs (19), Duthie et al. (17) therefore proposed that the X inactivation signal is spread by the binding of Xist RNA to sites which are present throughout the genome, but are enriched specifically on the X chromosome.
We report a detailed study of the spreading of X inactivation in four X;autosome translocations associated with varying severity of phenotype. We have used allele-specific RT–PCR and CpG island methylation analysis to determine the expression status of individual genes in the translocated segment of autosome and immunofluorescence to localize the distribution of acetylated and methylated histones and late-replicating chromatin. We also extend previous observations made in a fifth case (10). Our results directly demonstrate that X inactivation is capable of spreading through different regions of autosomal chromatin in either a continuous or discontinuous fashion. In addition we find that silencing of autosomal genes by a spread of X inactivation can occur in the absence of cytogenetic features normally associated with the inactive X.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Clinical details, parental origin and breakpoint analysis of five unbalanced X;autosome translocations
Case 1—SP, 46,X,der(X)t(X;11)(q26.3;p12) de novo (pat).
Aged 5 years, SP had mild developmental delay, large stature with growth on the 97th centile, a very long tongue not considered typical macroglossia, short neck and minor facial dysmorphisms including mild hypertelorism, mild epicanthus and slightly downward slanting eyes. PCR analysis of microsatellite markers demonstrated the der(X;11) to be of paternal origin, with breakpoints between DXS1187 and DXS1062 and D11S4102 and D11S1355, respectively (data not shown). Parental karyotypes were normal.
Case 2—SR, 46,X,der(X)t(X;7)(q27.3;q22.3) mat.
At the age of 16, SR displayed a number of severe phenotypic abnormalities with profound motor and developmental delay and severe mental retardation. She has no speech and shows a variety of orthopaedic disorders with scoliosis and dislocated hips. PCR analysis of microsatellite markers demonstrated the der(X;7) to be of maternal origin, with breakpoints between DXS998 and DXS1684 and D7S2420 and D7S523, respectively (data not shown). Consistent with this finding, parental karyotyping showed SR's mother to have the balanced form of the translocation.
Case 3—AL0044, 46,X,der(X)t(X;6)(p11.2;p21.1) mat.
Studies in AL0044 have been reported previously (18). Briefly, these found that hypoacetylation of histone H4, late-replication and XIST RNA were coincident and apparently excluded from the translocated segment of 6p on the der(X;6). AL0044 has mild developmental delay, learning difficulties (IQ=75), and short stature. PCR analysis of polymorphic markers demonstrated the der(X;6) to be of maternal origin, with breakpoints between DXS1058 and DXS8083 and ZNF76 and D6S269, respectively (data not shown). Consistent with this finding, parental karyotyping showed AL0044's mother to have the balanced form of the translocation.
Case 4—BO0566, 46,X,der(X)t(X;6)(q28;p12) de novo (pat).
BO0566 failed to thrive as a baby, with poor feeding and diarrhoea. At 16 months, her head circumference was on the 3rd centile, with a small anterior fontanelle, mild hypertelorism, thin lips, low set ears and a left ear lobe crease. At 3 years, she could sit unaided with babbling speech. At 4
years she could stand and was fed via an NG tube. She developed epilepsy at 8 years of age. Aged 10, she has reasonable motor skills, no speech or sign language and can take a few steps but mostly ‘bottom shuffles’. She has some behavioural problems, nipping and scratching, with inappropriate laughter. Generally her development is similar to a 9–12 month old. She also has a heart murmur, severe swallowing difficulties and experiences recurrent chest infections. PCR analysis of polymorphic markers demonstrated the der(X;6) to be of paternal origin, with breakpoints between DXS1073 and DXS1108 and D6S269 and GCLC, respectively (data not shown). Parental karyotypes were normal.
Case 5—AH, 46,X,der(X)t(X;10)(q26.3;q23.3) mat.
Phenotype data, replication timing analysis and results of expression analysis of 5 genes within the translocated segment 10q23.3–qter in this case have been published previously (10). Briefly, these demonstrated an apparently continuous but incomplete spread of X inactivation covering the majority of the translocated 10q chromatin. In contrast to this spread of inactivation and the patients normal phenotype, no spreading of late-replication into the translocated 10q was observed. PCR analysis of additional microsatellite markers refined the X chromosome breakpoint between DXS1073 and DXS1108 and the chromosome 10 breakpoint distal to D10S583 (data not shown). Additionally, the transcriptional status of three of the genes tested (HPS, MXI1 and ABLIM) was analysed using RNA from peripheral blood. Results were concordant with those obtained previously using RNA from EBV-transformed lymphoblasts (10).
X inactivation ratios
Methylation analysis at the androgen receptor (AR) locus demonstrated completely skewed X inactivation in EBV-transformed lymphocytes of all five cases. In every case, analysis of parental DNA showed that the der(X) was exclusively inactive (data not shown). Identical results were obtained using DNA from peripheral blood, where available (Cases 1, 2 and 5).
Gene expression
Heterozygous polymorphisms were identified in 27 genes within the translocated autosomal segments in Cases 1–4 and were used for allele-specific RT–PCR of RNA extracted from EBV-transformed lymphoblasts. Example results are shown in Figure 1. For every gene tested, analysis of cDNA from control individuals indicated normal biallelic expression in the cell types analysed.
|
In Case 1 (SP), eleven genes were studied, demonstrating an apparently continuous spread of gene silencing across nearly the entire translocated 11p segment (Table 1). The transcriptional status of five of these genes (HRAS, TSSC3, TAF2H, LMO2 and PDX1) was also analysed using RNA extracted from peripheral blood of SP, with concordant results. For HRAS, in cDNA of SP the relative intensity of one paternally-derived allele was
50% that seen in the DNA of SP and in the DNA and cDNA of normal controls (Fig. 1B). This indicates either an approximate halving of the level of transcription of HRAS on the der(X;11) in every cell, or alternatively mosaic silencing of this gene, with inactivation occurring in some 50% of cells. As previous studies of the nearby imprinted locus H19 in SP have shown it to be methylated at levels intermediate between normal controls and individuals with paternal uniparental disomy of chromosome 11 (20), it seems likely that both HRAS and H19 display a mosaic pattern of inactivation.
|
In Case 2 (SR), results of allele-specific RT–PCR of three genes within the translocated segment 7q22.3–qter are summarised in Table 2. For CNTNAP2, in cDNA of SR the relative intensity of one maternally derived allele was
30% that seen in the DNA of SR and in the DNA and cDNA of normal controls, consistent with a mosaic pattern of inactivation.
|
In Case 3 (AL0044), results of allele-specific RT–PCR of nine genes demonstrate a discontinuous spreading of gene silencing across the entire translocated segment of 6p, with active genes interspersed among inactive genes (Table 3). For SCA1, in cDNA of AL0044 the relative intensity of one maternally derived allele was
30% that seen in the DNA of AL0044 and in the DNA and cDNA of normal controls, consistent with a mosaic pattern of inactivation.
|
In Case 4 (BO0566), results of allele-specific RT–PCR of seven genes within the translocated segment 6p12–pter also show a discontinuous spreading of gene silencing across the translocated segment of 6p (Table 3).
CpG island methylation
CpG islands were identified in the 5' regions of PDX1, TAF2H, HRAS and PSMD13 within 11p12–pter, and HLA-F, HCR and MICA within 6p21–pter and their methylation status in SP, AL0044 and BO0566 investigated by PCR following HpaII or CfoI digestion of genomic DNA (Fig. 2). Analysis of control individuals showed that the CpG island of each gene was normally unmethylated.
|
In Case 1 (SP), analysis of DNA from peripheral blood and lymphoblasts showed the presence of high levels of methylation at the CpG islands of PDX1 (inactive) and TAF2H (inactive). However, no methylation was detected at the CpG islands of HRAS (
50% inactive) or PSMD13 (active) by either HpaII or CfoI analysis (Fig. 2). Analysis of DNA from lymphoblasts of both Case 3 (AL0044) and Case 4 (BO0566) showed the presence of high levels of methylation at the CpG islands of HLA-F (inactive in AL0044, non-informative in BO0566) and MICA (inactive in AL0044, non-informative in BO0566) in both individuals. No methylation was detected at the CpG island of HCR (inactive in AL0044, active in BO0566) by either HpaII or CfoI analysis. However, in SP, AL0044 and BO0566, the presence of multiple recognition sites for HpaII and CfoI within the amplicons of HRAS, PSMD13 and HCR does not allow us to exclude the presence of partial methylation (data not shown).
Histone modifications
Immunolabelling of histone H3 acetylated at lysine 14 (H3AcK14), H4 acetylated at lysine 8 (H4AcK8), or H3 dimethylated at lysine 4 (H3Me2K4), was combined with in situ hybridization using whole chromosome paint to determine the extent of spread of these histone modifications on the der(X) chromosome in EBV-transformed lymphoblasts of Cases 1–5. In every case, antisera specific to each histone epitope showed that the der(X) was pale-staining, concordant with results of the AR assay. Additionally, within each case, results gained using each antibody were concordant with one another.
In Case 1 (SP), in all 36 cells examined, the der(X;11) was clearly depleted of histone acetylation and H3 lysine 4 dimethylation along almost its entire length (Fig. 3A–I). However, in every cell examined a small punctate region of H3/H4 acetylation and H3 lysine 4 dimethylation was clearly visible at the distal tip of the translocated segment of 11p. In a proportion of cells, similar punctate staining was also apparent at the distal tip of Xp, in Xp11.2 and on the long arm of the X at Xq22–25, consistent with previous observations (1,2).
|
In Case 2 (SR), in all 27 cells examined there was a partial and continuous spread of histone hypoacetylation and depletion of H3 lysine 4 dimethylation across approximately one third of the 7q segment (Fig. 4A, E and I). However, the remaining distal portion of the translocated 7q segment was indistinguishable from the corresponding regions on the two normal chromosome 7 homologues.
|
In Case 3 (AL0044), in all 23 cells examined there was a discontinuous spread of histone hypoacetylation and depletion of H3 lysine 4 dimethylation across the translocated 6p segment (Fig. 4B, F and J). This region depleted of histone acetylation and H3 lysine 4 dimethylation varied in size between cells, in some instances covering up to approximately one third of the translocated autosome, but always confined to the distal end of the translocated 6p chromatin. In every cell the more proximal translocated portion of 6p appeared indistinguishable from the corresponding regions on the two normal chromosome 6 homologues.
In Case 4 (BO0566), in all 24 cells examined there was a complete absence of spreading of histone hypoacetylation and depletion of H3 lysine 4 dimethylation into the translocated 6p segment, with the transition in antibody staining appearing to define the X;autosome boundary (Fig. 4C, G and K).
In Case 5 (AH), in all 32 lymphoblastoid cells examined there was a continuous and almost complete spread of depletion of histone acetylation and H3 lysine 4 dimethylation across the translocated 10q segment. However, in every cell examined a small region of H3/H4 acetylation and H3 lysine 4 dimethylation was clearly visible at the distal end of the translocated segment of 10q (Fig. 4D, H and L).
Replication timing
A fluorescent late-pulse BrdU assay combined with in situ hybridization using whole chromosome paint was performed to determine the extent of spread of late-replication on the der(X) chromosome in Cases 1–4. In every case, the derivative chromosome was late replicating, concordant with results of the AR assay and histone immunofluorescence.
In Case 1 (SP), in all 50 cells examined in both peripheral blood and EBV-transformed lymphoblasts, a partial spreading of late-replication into the 11p segment was observed. In most cases the late replicating region extended continuously from the X chromatin across approximately half to two-thirds of the autosomal segment (Fig. 5A). While variations in the extent of spread of late-replication occurred, there was good concordance in the extent of spread of late-replication between peripheral blood and lymphoblasts.
|
In Case 2 (SR), the spread of late replication into the 7q segment varied among cells. 37/54 peripheral blood cells and 30/35 lymphoblastoid cells showed a partial, continuous spread of late-replication across approximately one third of the 7q segment (Fig. 5B). However, in 12/54 peripheral blood cells where the resolution of the metaphases was higher there was a discontinuous spread of late-replication. In addition to the late replicating region on the proximal portion of 7q, there was an isolated focus of BrdU staining at the telomere (Fig. 5C). In the remaining 5/54 blood cells and 5/35 lymphoblasts the late-replicating region extended only as far as the X;autosome boundary (data not shown).
In Case 3 (AL0044), 20/50 lymphoblasts examined showed a complete absence of spreading of late replication into the translocated segment of 6p (Fig. 5D). However, in the remaining 30/50 cells there was a discontinuous spread of late replication into 6p, with a variably sized region of BrdU staining on the translocated 6p telomere, with no staining on the intervening segment (Fig. 5E).
In Case 4 (BO0566) in all 50 cells examined there was a complete absence of spreading of late replication into the translocated 6p segment (Fig. 5F).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
We have directly studied the spreading of X inactivation in five X;autosome translocations and in each case demonstrated the long-range silencing of autosomal genes located up to 45 Mb from the translocation breakpoint. Thus whatever factors are responsible for the spread of X inactivation are not unique to the X chromosome. However, in each case this spreading has occurred in either an incomplete or discontinuous fashion, suggesting that autosomal chromatin does not either transmit or maintain the inactivation signal as efficiently as the X chromosome. Furthermore, our results show that spreading of X inactivation through autosomal chromatin can occur in the absence of cytogenetic features normally associated with the inactive X, such as late-replication, histone hypoacetylation and coating with XIST RNA (18).
Previously we demonstrated that late-replication is a poor cytogenetic correlate of the spread of X inactivation in an X;10 translocation (10). In contrast, observations of depletion of histone acetylation and H3 lysine 4 dimethylation in this same case show good correlation with the pattern of gene silencing, demonstrating that these histone modifications are distinct from and independent of late-replication (21). Therefore, while observations in SP and SR show some association between the spread of late-replication and gene silencing, we conclude that the distribution of histone modifications which distinguish the inactive X, such as H3AcK14, H4AcK8 and H3Me2K4, are superior cytogenetic measures of the spread of X inactivation. However, where the spread of gene silencing occurs discontinuously, as observed in the two X;6 translocations, we did not observe a corresponding distribution of depletion of histone acetylation and H3 lysine dimethylation. This may simply reflect the low resolution of immunofluorescence and histone states at the gene level may differ from that of the surrounding chromatin, as is observed for X-linked genes which escape X inactivation (2,22).
We note however that in every case, autosomal genes located within cytogenetically late-replicating regions were inactive. We propose that this late-replicating chromatin represents domains in which the spread of X inactivation is maintained in a more stable fashion. This suggestion is supported by the observation that the autosomal genes in SP, SR and AL0044 which showed partial inactivation were located in regions which demonstrated variation in the spread of late-replication.
We have shown that the CpG islands of autosomal genes silenced by the spread of X inactivation become highly methylated, as occurs on the inactive X (4). CpG island methylation probably represents an important component of the spread and/or maintenance of inactivation through autosomal DNA, as in both AL0044 and BO0566 methylation at MICA and HLA-F represents the only difference we detected between the inactive translocated copy of these genes and their active normal homologues. However, sequence analysis did not detect CpG islands adjacent to or within several other inactivated autosomal genes, suggesting that possession of a CpG island is not a prerequisite for silencing by the spread of X inactivation.
Perhaps the most surprising observation to emerge from this study is that the spread of X inactivation in each of the five cases showed different characteristics (summarised in Fig. 6), reflecting the complexity of the X inactivation cascade. In Case 1 (SP) we observed a continuous spread of gene silencing over almost the entire 11p segment, accompanied by a partial but variable spread of late-replication and an almost complete spread of depletion of histone acetylation and H3 lysine 4 dimethylation. In Case 2 (SR) there was a partial and apparently continuous spread of gene silencing which correlated well with both the spread of late replication and histone modification. However, here we also observed a discontinuous spread of late-replication in a proportion of cells. In contrast, both of the X;6 translocations (Cases 3 and 4, AL0044 and BO0566) showed a discontinuous spread of gene silencing, but cytogenetic observations were discordant between the two. AL0044 had a discontinuous and variable spread of late-replication and depletion of histone acetylation and H3 lysine 4 dimethylation which were confined to the distal portion of the 6p segment. Remarkably, the more proximal translocated region of 6p was cytogenetically indistinguishable from the corresponding regions on the two normal chromosome 6 homologues, emphasising the ability of X inactivation to ‘jump’ large regions of chromatin. The apparent absence or minimal spread of these cytogenetic features in many cells may explain why they were not reported in a previous study of this same case (18). Observations of XIST RNA distribution in this case showed little or no spread into the 6p segment (18). BO0566 showed no spread of late-replication or histone modifications. Finally, Case 5 (AH) exhibited an apparently continuous spread of gene silencing covering the majority of the 10q segment. This spread of inactivation correlated with the pattern of depletion of histone acetylation and H3 lysine 4 dimethylation but not with the spread of late-replication (10).
|
Despite these variations, in every case we observed a good correlation between the pattern of gene silencing and the attenuation of clinical phenotype associated with each partial autosomal trisomy. This suggests that our in vitro observations in lymphoblastoid cells are a good reflection of that in vivo, although we cannot exclude the possibility that the pattern of inactivation observed is a result of selection removing those cells in which dosage-sensitive autosomal genes remained active. Additionally, it is possible that the spread of gene inactivation, late-replication and/or histone modifications is a result of inefficient maintenance of the X inactivation signal in autosomal chromatin. Immediately after the onset of X inactivation in the developing embryo, the spread of inactivation through each translocated autosome may have been more complete but subsequently receded due to a lack of appropriate maintenance. A failure of maintenance could be characterized by the progressive or differential loss of features of X inactivation such as coating with XIST RNA, late-replication and histone hypoacetylation prior to the reactivation of autosomal genes, which could account for our observations of the distribution of inactive autosomal genes in relation to cytogenetic features of X inactivation. Although we cannot discount this possibility, we studied gene expression, CpG island methylation and late-replication in AH, SP and SR using both EBV-transformed lymphoblasts and peripheral blood and were unable to detect any differences between the two cell types, suggesting that the spread of X inactivation is maintained in a relatively stable fashion in both systems and is not adversely influenced by EBV-transformation. Unfortunately we were unable to obtain fibroblast cells for these studies.
Previous observations of the distribution of XIST RNA in lymphoblasts of AL0044 showed it to be localized specifically to the inactive X with little or no spread into the adjacent 6p chromatin (18). These observations are consistent with those made in murine X;autosome rearrangements (17) and suggest that XIST/Xist RNA has a reduced affinity for autosomal chromatin. Our observations in AL0044 clearly demonstrate that silencing of autosomal genes by a spread of X inactivation can occur in the absence of coating by XIST RNA. We were unable to study the distribution of XIST RNA on the der(X) in other cases as FISH studies showed an aberrant distribution of XIST transcripts in the lymphoblastoid cell lines available. In Cases 1 (SP) and 2 (SR), where XIST expression was detectable it showed a diffuse distribution which was poorly localized compared to controls, while in many cells XIST was apparently not expressed (L. Hall and J. Lawrence, unpublished data). Given that XIST transcripts were apparently normally expressed and localized in EBV-transformed lymphoblasts of AL0044, it is unclear why they showed an aberrant distribution in other similarly transformed cell lines. Although it was not possible to perform XIST RNA FISH studies in fresh samples, we believe that this sporadic failure of expression and localization of XIST transcripts in some cell lines might be an artefactual consequence of EBV-transformation, similar to that seen in transgenic lines (23) and neither had any appreciable influence on the spread of inactivation observed nor represents the situation in vivo.
We did not observe any relationship between the spread of inactivation and the location of translocation breakpoints or G-bands, as has been suggested (8,17). Positional assignments by the Human Genome Browser indicate that inactive genes were contained within both G-light and G-dark bands at similar frequency. Neither did the extent of spread of inactivation appear to be related to absolute distance from the XIC (located at Xq13.2), although in SP, SR and AH there were clear ‘gradient effects’, with inactivation diminishing in effect with increasing distance from X chromatin. It has also been suggested that on the human X the separation of the short arm from the XIC by the centromere may be responsible for the high density of genes escaping X inactivation in Xp (24,25). However, comparison of AL0044 and BO0566 suggests that centromeric heterochromatin does not represent a barrier to the spread of X inactivation.
In three of the five cases studied we observed that late-replication and/or histone hypoacetylation and H3 lysine 4 dimethylation sharply define the boundary between the X and autosomal chromatin, strongly suggesting that autosomal DNA has distinct properties from the X and is resistant to the X inactivation signal. The discontinuous spread of silencing observed in AL0044 and BO0566 suggests that certain regions of chromosome 6p in particular lack features important for the spread and/or maintenance of X inactivation. Recently Mary Lyon proposed that Long Interspersed Nuclear Elements (LINEs) might function to promote the spread of X inactivation in cis (26). In the context of their distribution on the X chromosome (27), LINEs represent good candidates for these ‘booster sequences’. However, the differences in the spread of late-replication, histone modification and the discordant inactivation of HCR between the two cases indicates that sequence-specific factors are not the sole determinants of X inactivation and that the relative position of translocation breakpoints also influences the extent of its spread in X;autosome rearrangements.
In summary we have characterized the spread of X inactivation through autosomal chromatin in five unbalanced X;autosome translocations. Our study demonstrates that transcriptional silencing by a spread of X inactivation can occur in the apparent absence of features normally associated with the inactive X, raising further questions about the mechanism by which the X inactivation signal is transmitted and maintained in these cases. We emphasize that transcription studies are necessary to accurately determine the spread of X inactivation into autosomal chromatin and cytogenetic characteristics such as late-replication and histone hypoacetylation are not necessarily associated with inactivation of autosomal genes. However, where X inactivation spreads in a continuous fashion, we suggest that cytogenetic features such as depletion of histone acetylation and H3 lysine 4 dimethylation provide more reliable indicators of the extent of spread of X inactivation than replication-timing studies.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Cell culture
EBV-transformed lymphoblastoid cell lines (ID codes DD1289, DD0003, AL0044, BO0566, DD1550 for cases 1–5 respectively) were obtained from the European Collection of Animal Cell Cultures, Porton Down, Wiltshire, UK and grown in RPMI-1640 (Sigma) supplemented with 10% fetal calf serum, 2 mM L-glutamine and 100 U/ml each of penicillin and streptomycin.
Determination of X inactivation ratios
DNA was isolated from peripheral blood samples (SP and SR) and lymphoblastoid cell lines (SP, SR, AL0044 and BO0566) using a modified salt-precipitation technique (28). X inactivation ratios were determined using the AR gene PCR assay, as described previously (29).
Identification of transcribed polymorphisms
Putative transcribed single-nucleotide and sequence-length polymorphisms contained within 11p12–pter, 7q22–qter and 6p12–pter were isolated from The Genome Database (http://gdbwww.gdb.org/), Locuslink (http://www.ncbi.nlm.nih.gov/LocusLink/) and HGBase (http://hgbase.interactiva.de/). Webcutter 2.0 (http://www.firstmarket.com/firstmarket/cutter/cut2.html) was used to identify those SNPs which altered a restriction site and primers to amplify these polymorphisms were designed from mRNA sequences in Genbank (http://www.ncbi.nlm.nih.gov/Web/Genbank) using Primer3 (http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi). In total, 167 putative polymorphisms in 103 genes were initially screened by PCR, restriction digestion and agarose gel electrophoresis for heterozygosity in the 4 probands SP, SR, AL0044 and BO0566. Polymorphisms, restriction enzymes and primer details of the 27 used for allele-specific RT–PCR are available on request.
Allele-specific quantitative RT–PCR
Allele-specific quantitative RT–PCR was performed as described previously (10). Briefly, RNA was extracted from lymphoblastoid cell lines (SP, SR, AL0044 and BO0566) and peripheral blood (SP and AH) using Trizol (Gibco BRL), DNAse-treated (Promega) and cDNA synthesized using M-MLV RTase (Gibco BRL) with gene-specific reverse-strand primers. An initial PCR was performed using unlabelled primers and subsequently used as template in a second reaction incorporating fluorescently labelled primers which was subjected to a single cycle of PCR. Use of this method avoided heteroduplex formation, thus facilitating accurate allele quantification by restriction digest (30). This second PCR reaction was then restricted (New England Biolabs) and the products resolved using an ABI 377 sequencer for quantification of alleles by peak height.
For 21 of the 27 genes tested, comparison of results gained using proband DNA showed one allele to be approximately double the intensity of that in normal controls, demonstrating their inclusion in the translocated segment of the autosome. For the remaining 6 genes (RRM1, MRPL23, HCR, CSNK2B, HLA-DRA and PSMB9) the primers used spanned large introns and were thus cDNA specific. Their inclusion in the translocated segments of the autosome was confirmed using physical and radiation-hybrid maps (http://genome.ucsc.edu/ and http://www.ncbi.nlm.nih.gov/genemap99/).
A gene was scored as inactive when the allele ratio gained using cDNA from the proband was significantly different from that obtained using DNA from the proband and almost identical to those seen using DNA and cDNA from normal controls. Such results indicate that the gene is transcribed from only two alleles in the proband, implying that the copy on the der(X) is inactive (Fig. 1A). In every case, analysis of parental DNA/RNA gave results that were consistent between the known origin of the der(X) and the transcriptionally silent allele. Similarly, a gene was scored as active when results obtained using cDNA from the proband were similar to those gained using DNA of the proband, and significantly different from those gained using both DNA and cDNA from heterozygous controls. Such results indicate that the gene is transcribed from all three alleles in the proband, implying that the copy on the der(X) remains active.
CpG island methylation analysis
NIX analysis (http://www.hgmp.mrc.ac.uk/Registered/Webapp/nix/) of genomic sequence deposited in GenBank was used to identify CpG islands located within the 5' region of genes within 11p12–pter and 6p12–pter. Primers spanning each CpG island were designed using Primer3 and each amplicon restriction mapped using Webcutter 2.0. 50 ng of DNA which had been double-digested overnight with 20 U DraI and 20 U HpaII, CfoI or MspI (New England Biolabs) was co-amplified with individual CpG island primers and control primers spanning the CpG island of PGK1, which is methylated on the inactive X and unmethylated on the active X (22). Because of the GC-rich nature of the target sequences, amplifications were performed with 10% DMSO and 7' deaza-dGTP (Roche) and products resolved through 3% agarose.
Detection of late-replicating chromatin and in situ hybridization
Late-replicating chromatin was detected as described previously (10). Briefly, lymphoblastoid cells (SP, SR, AL0044 and BO0566) and peripheral blood cultures (SP and SR) were exposed to BrdU for 5 hours prior to fixing in 3:1 methanol:acetate. Prepared metaphases were denatured, dehydrated and blocked. Incorporated BrdU was then immunolabelled using mouse monoclonal anti-BrdU (Sigma), detected with anti-mouse IgG-FITC (Sigma), post-fixed in 4% paraformaldehyde and hybridized with the appropriate TRITC-labelled whole chromosome paint (Oncor). Probe preparation, hybridization and post-hybridization washes were performed according to manufacturers instructions. Slides were counterstained with DAPI/Antifade (Vector) and images captured using Macprobe 4.1 software (PSI).
Histone immunofluorescence and in situ hybridization
Antisera to histone H3 dimethylated at lysine 4, histone H3 acetylated at lysine 14 and histone H4 acetylated at lysine 8 were raised and immunolabelling performed as described previously (1). The antibody to H3 methylated at lysine 4 was raised against H3 peptides dimethylated at lysine 4 and does not recognize H3 trimethylated at lysine 4 (B. Turner and L. O'Neill, unpublished data). Briefly, 2 hours after the addition of colcemid, lymphoblasts were cytospun onto glass slides and permeabilized in KCM Buffer (120 mM KCl, 20 mM NaCl, 10 mM Tris–HCl pH8.0, 0.5 mM EDTA, 0.1% Triton X-100). Slides were then blocked (KCM, 1% BSA), immunolabelled, washed and detected with anti-rabbit IgG-FITC prior to fixing in 4% formaldehyde and mounting in DAPI/antifade. Metaphases were then captured using Macprobe 4.1 software (PSI), post-fixed in 3:1 methanol:acetate, incubated in RNAse/pepsin, denatured and in situ hybridization subsequently performed using TRITC-labelled whole chromosome paints (Oncor) according to manufacturers instructions. Slides were then mounted in DAPI/antifade, metaphases recaptured using England finder co-ordinates and immunofluorescence and in situ hybridization images overlaid in Photoshop.
|
ACKNOWLEDGEMENTS |
|---|
We would like to thank the patients and their families involved in this study for their helpful co-operation, Dr Trevor Cole, Dr Evan Reid and especially Dr Nick Dennis for their generous provision of clinical information and patient samples, Prof. Jeanne Lawrence and Dr Lisa Hall for performing XIST RNA FISH studies and helpful discussion and Dr Andrew Barlow and Nicola Savage for advice with the replication timing assay. We would also like to thank Salisbury Hospitals Foundation for the provision of financial assistance towards the costs of presenting this work at the 51st meeting of the American Society of Human Genetics, San Diego 2001. Andrew Sharp and Hugh Spotswood are Wellcome Trust Prize Students (Refs. 058387 and 057710).
|
FOOTNOTES |
|---|
* To whom correspondence should be addressed. Tel: +44 1722336262 ext. 2047; Fax: +44 1722338095; Email: asharp{at}hgmp.mrc.ac.uk
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Belyaev, N., Keohane, A.M. and Turner, B.M. (1996) Differential underacetylation of histones H2A, H3 and H4 on the inactive X chromosome in human female cells. Hum. Genet., 97, 573–578.[Web of Science][Medline]
2 Boggs, B.A., Cheung, P., Heard, E., Spector, D.L., Chinault, A.C. and Allis, D. (2001) Differentially methylated forms of histone H3 show unique association patterns with inactive human X chromosomes. Nat. Genet., 30, 73–76.[Medline]
3 Willard, H.F. and Latt, S.A. (1976) Analysis of deoxyribonucleic acid replication in human X chromosomes by fluorescence microscopy. Am. J. Hum. Genet., 28, 213–227.[Web of Science][Medline]
4
Tribioli, C., Tamanini, F., Patrosso, C., Minalesi, L. and Villa, A. (1992) Methylation and sequence analysis around EagI sites: identification of 28 new CpG islands in Xq24–q28. Nucleic Acids Res., 20, 727–733.
5 Brown, C.J., Lafreniere, R.G., Powers, V.E., Sebastio, G., Ballabio, A., Pettigrew, A.J., Ledbetter, D.H., Levy, E., Craig, I.W. and Willard, H.F. (1991) Localization of the X inactivation centre on the human X chromosome in Xq13. Nature, 349, 82–84.[Medline]
6
Clemson, C.M., Mcneil, J.A., Willard, H.F. and Lawrence, J.B. (1996) XIST RNA paints the inactive X-chromosome at interphase: Evidence for a novel RNA involved in nuclear/chromosome structure. J. Cell Biol., 132, 259–275.
7
Russell, L.B. (1963) Mammalian X-chromosome action: inactivation limited in spread and region of origin. Science, 140, 976–978.
8 Camargo, M. and Cervenka, J. (1984) DNA replication and inactivation patterns in structural abnormality of sex chromosomes. I. X-A translocations, rings, fragments, isochromosomes, and pseudo-isodicentrics. Hum. Genet., 67, 37–47.[Web of Science][Medline]
9 Mattei, M.G., Mattei, J.F., Ayme, S. and Giraud, F. (1982) X-autosome translocations: Cytogenetic characteristics and their consequences. Hum. Genet., 61, 295–309.[Web of Science][Medline]
10 Sharp, A., Robinson, D. and Jacobs, P. (2001) Absence of correlation between late-replication and spreading of X inactivation in an X;autosome translocation. Hum. Genet., 109, 295–302.[Web of Science][Medline]
11 Couturier, J., Dutrillaux, B., Garber, P., Raoul, O., Croquette, M-F., Fourlinnie, J.C. and Maillard, E. (1979) Evidence for a correlation between late replication and autosomal gene inactivation in a familial translocation t(X;21). Hum. Genet., 49, 319–326.[Web of Science][Medline]
12 Mohandas, T., Crandall, B.F., Sparkes, R.S., Passage, M.B. and Sparkes, M.C. (1981) Late replication studies in a human X/13 translocation: correlation with autosomal gene expression. Cytogenet. Cell Genet., 29, 215–220.[Web of Science][Medline]
13 Mohandas, T., Sparkes, R.S. and Shapiro, L.J. (1982) Genetic evidence for the inactivation of a human autosomal locus attached to an inactive X chromosome. Am. J. Hum. Genet., 34, 811–817.[Web of Science][Medline]
14
Taysi, K., Sparkes, R.S., O'Brien, T.J. and Dengler, D.R. (1982) Down's syndrome phenotype and autosomal gene inactivation in a child with presumed (X;21) de novo translocation. J. Med. Genet., 19, 144–148.
15 White, W.M., Willard, H.F., Van Dyke, D.L. and Wolff, D.J. (1998) The spreading of X inactivation into autosomal material of an X;autosome translocation: Evidence for a difference between autosomal and X-chromosomal DNA. Am. J. Hum. Genet., 63, 20–28.[Web of Science][Medline]
16 Keitges, E.A. and Palmer, C.G. (1986) Analysis of the spreading of inactivation in eight X autosome translocations utilizing the high resolution RBG technique. Hum. Genet., 72, 231–236.[Web of Science][Medline]
17
Duthie, S.M., Nesterova, T.B., Formstone, E.J., Keohane, A.M., Turner, B.M., Zakian, A.M. and Brockdorff, N. (1999) Xist RNA exhibits a banded localization on the inactive X chromosome and is excluded from autosomal material in cis. Hum. Mol. Genet., 8, 195–204.
18
Keohane, A.M., Barlow, A.L., Waters, J., Bourn, D. and Turner, B.M. (1999) H4 acetylation, XIST RNA and replication timing are coincident and define X;autosome boundaries in two abnormal X chromosomes. Hum. Mol. Genet., 8, 377–383.
19 Gartler, S.M. and Riggs, A.D. (1983) Mammalian X-chromosome inactivation. Ann. Rev. Genet., 17, 155–190.[Web of Science][Medline]
20
Reik, W., Brown, K.W., Slatter, R.E., Sartori, P., Elliott, M. and Maher, E.R. (1994) Allelic methylation of H19 and IGF2 in the Beckwith-Wiedemann syndrome. Hum. Mol. Genet., 3, 1297–1301.
21 Keohane, A.M., O'Neill, L.P., Belyaev, N.D., Lavender, J.S. and Turner B.M. (1996) X-inactivation and H4 acetylation in embryonic stem cells. Dev. Biol., 180, 618–630.[Web of Science][Medline]
22
Gilbert, S.L. and Sharp, P.A. (1999) Promoter-specific hypoacetylation of X-inactivated genes. Proc. Natl Acad. Sci. USA, 96, 13825–13830.
23
Clemson, C.M., Chow, J.C., Brown, C.J. and Lawrence, J.B. (1998) Stabilization and localization of Xist RNA are controlled by separate mechanisms and are not sufficient for X inactivation. J. Cell Biol., 142, 13–23.
24
Disteche, C.M. (1999) Escapees on the X chromosome. Proc. Natl Acad. Sci. USA, 96, 14180–14182.
25
Carrel, L., Cottle, A.A., Goglin, K.C. and Willard, H.F. (1999) A first-generation X-inactivation profile of the human X chromosome. Proc. Natl Acad. Sci. USA, 96,14440–14444.
26 Lyon, M.F. (1998) X-chromosome inactivation: A repeat hypothesis. Cytogenet. Cell Genet., 80, 133–137.[Web of Science][Medline]
27
Bailey, J.A., Carrel, L., Chakravarti, A. and Eichler, E.E. (2000) Molecular evidence for a relationship between LINE-1 elements and X chromosome inactivation: The Lyon repeat hypothesis. Proc. Natl Acad. Sci. USA, 97, 6634–6639.
28
Miller, S.A., Dykes, D.D. and Polesky, H.F. (1988) A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res., 16, 1215.
29 Sharp, A., Robinson, D. and Jacobs, P. (2000) Age- and tissue-specific variation of X-chromosome inactivation ratios in normal women. Hum. Genet., 107, 343–349.[Web of Science][Medline]
30 Uejima, H., Lee, M.P., Cui, H. and Feinberg, A.P. (2000) Hot-stop PCR: a simple and general assay for linear quantification of allele ratios. Nat. Genet., 25, 375–376.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. A. Blauwkamp and G. Csankovszki Two Classes of Dosage Compensation Complex Binding Elements along Caenorhabditis elegans X Chromosomes Mol. Cell. Biol., April 15, 2009; 29(8): 2023 - 2031. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Chow, L. L. Hall, S. E. L. Baldry, N. P. Thorogood, J. B. Lawrence, and C. J. Brown Inducible XIST-dependent X-chromosome inactivation in human somatic cells is reversible PNAS, June 12, 2007; 104(24): 10104 - 10109. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Heard and C. M. Disteche Dosage compensation in mammals: fine-tuning the expression of the X chromosome Genes & Dev., July 15, 2006; 20(14): 1848 - 1867. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Popova, T. Tada, N. Takagi, N. Brockdorff, and T. B. Nesterova Attenuated spread of X-inactivation in an X;autosome translocation PNAS, May 16, 2006; 103(20): 7706 - 7711. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. McNeil, K. P. Smith, L. L. Hall, and J. B. Lawrence Word frequency analysis reveals enrichment of dinucleotide repeats on the human X chromosome and [GATA]n in the X escape region Genome Res., April 1, 2006; 16(4): 477 - 484. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. WEINBERG, L. M. VILLENEUVE, A. EHSANI, M. AMARZGUIOUI, L. AAGAARD, Z.-X. CHEN, A. D. RIGGS, J. J. ROSSI, and K. V. MORRIS The antisense strand of small interfering RNAs directs histone methylation and transcriptional gene silencing in human cells RNA, February 1, 2006; 12(2): 256 - 262. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Sharp, K Kusz, J Jaruzelska, W Tapper, M Szarras-Czapnik, J Wolski, and P Jacobs Variability of sexual phenotype in 46,XX(SRY+) patients: the influence of spreading X inactivation versus position effects J. Med. Genet., May 1, 2005; 42(5): 420 - 427. [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. West and P. Fraser Remote control of gene transcription Hum. Mol. Genet., April 15, 2005; 14(suppl_1): R101 - R111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. P. Chadwick and H. F. Willard Multiple spatially distinct types of facultative heterochromatin on the human inactive X chromosome PNAS, December 14, 2004; 101(50): 17450 - 17455. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T Mononen, A Sharp, M Laakso, R-L Meltoranta, A-K Valve-Dietz, and K Heinonen Partial trisomy 10q with mild phenotype caused by an unbalanced X;10 translocation J. Med. Genet., May 1, 2003; 40(5): e61 - 61. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||