UBC Faculty Research and Publications

Epigenetic predisposition to expression of TIMP1 from the human inactive X chromosome Anderson, Catherine L; Brown, Carolyn J Sep 29, 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12863_2005_Article_254.pdf [ 1.18MB ]
JSON: 52383-1.0221607.json
JSON-LD: 52383-1.0221607-ld.json
RDF/XML (Pretty): 52383-1.0221607-rdf.xml
RDF/JSON: 52383-1.0221607-rdf.json
Turtle: 52383-1.0221607-turtle.txt
N-Triples: 52383-1.0221607-rdf-ntriples.txt
Original Record: 52383-1.0221607-source.json
Full Text

Full Text

ralssBioMed CentBMC GeneticsOpen AcceResearch articleEpigenetic predisposition to expression of TIMP1 from the human inactive X chromosomeCatherine L Anderson and Carolyn J Brown*Address: Department of Medical Genetics, University of British Columbia, 2350 Health Sciences Mall, Vancouver, BC, CANADA V6T 1Z3Email: Catherine L Anderson - catherine.anderson@shaw.ca; Carolyn J Brown* - cbrown@interchange.ubc.ca* Corresponding author    AbstractBackground: X inactivation in mammals results in the transcriptional silencing of an Xchromosome in females, and this inactive X acquires many of the epigenetic features of silentchromatin. However, not all genes on the inactive X are silenced, and we have examined the TIMP1gene, which has variable inactivation amongst females. This has allowed us to examine the featurespermitting expression from the otherwise silent X by comparing inactive X chromosomes with andwithout TIMP1 expression.Results: Expression was generally correlated with euchromatic chromatin features, includingDNA hypomethylation, nuclease sensitivity, acetylation of histone H3 and H4 and hypomethylationof H3 at lysines 9 and 27. Demethylation of the TIMP1 gene by 5-azacytidine was able to induceexpression from the inactive X chromosome in somatic cell hybrids, and this expression was alsoaccompanied by features of active chromatin. Acetylated histone H3 continued to be observedeven when expression was lost in cells that naturally expressed TIMP1; while acetylation was lostupon TIMP1 silencing in cells where expression from the inactive X had been induced bydemethylation. Thus ongoing acetylation of inactive X chromosomes does not seem to be simplya 'memory' of expression.Conclusion: We propose that acetylation of H3 is an epigenetic mark that predisposes to TIMP1expression from the inactive X chromosome in some females.BackgroundStudies have shown considerable individual variability inthe level of expression of genes (e.g. [1,2]). In general,however, humans cannot tolerate imbalances for expres-sion of substantial numbers of genes, as demonstrated bythe lethality of the majority of chromosomal aneuploi-dies. Aneuploidy for the sex chromosomes is better toler-ated, being observed in approximately 1/500 births [3],presumably because all but one X chromosome is inacti-two X chromosomes and males who have a single X chro-mosome and the sex-determining Y chromosome [4].However, more than 15% of human X-linked genesescape inactivation, being expressed from both the activeand inactive X chromosome [5]. While such an escapefrom inactivation may maintain dosage equivalence for X-linked genes with Y homologs, the majority of humangenes that escape inactivation no longer have functional Yequivalents, and thus may show relative overexpression inPublished: 29 September 2005BMC Genetics 2005, 6:48 doi:10.1186/1471-2156-6-48Received: 04 April 2005Accepted: 29 September 2005This article is available from: http://www.biomedcentral.com/1471-2156/6/48© 2005 Anderson and Brown; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 14(page number not for citation purposes)vated. X chromosome inactivation ensures the dosageequivalence of X-linked genes between females who havefemales (reviewed in [6]). Substantially fewer genes havebeen shown to escape inactivation in mice. Although thisBMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48species difference in expression could reflect less extensivemurine expression surveys, a reduced number of genesescaping inactivation is supported by the less drastic phe-notype caused by monosomy of the X chromosome inmice (reviewed in [7]). In humans, the over or under-expression of genes that escape inactivation is a majorcontributor to the phenotypes associated with X chromo-some aneuploidies, but may also contribute to expressiondifferences between chromosomally normal males andfemales (e.g. [8]).The study of genes that escape X inactivation can provideinsight into such phenotypes, as well as contributing toour understanding of epigenetic silencing mechanisms.Inactivation occurs early in mammalian development,and the stable silencing of the X chromosome involves theacquisition of many features of heterochromatin. It is notknown if escape from inactivation is a resistance to the ini-tial silencing event, or rather reflects a high frequency ofreactivation of an initially silenced gene, as data appear tosupport both possibilities. Many genes that escape inacti-vation in humans are clustered together, which may beindicative of regions that are resistant to the initial signal(e.g. [9]). However, analysis of Smcx, one of the fewmouse genes expressed from the murine inactive X, hasshown reactivation of the gene during early development[10]. Surprisingly, recent results have shown that Smcx hasa histone modification pattern suggested to demarcatebiallelically rather than monoallelically-expressed(imprinted and other X-linked) genes [11], suggestingthat the gene is committed to escape inactivation prior toundergoing inactivation.In addition to the genes that are subject to, or escape from,inactivation, there are some human genes that show het-erogeneous X chromosome inactivation, being expressedfrom the inactive X in some females, but silenced on theinactive X in others [5,12,13]. Such genes provide anopportunity to study the same region when silent or activeon an inactive X chromosome; and can thus provideinsights into the features allowing expression from theinactive X chromosome. The human inactive X chromo-some maintains its silent status when isolated in a mouse/human somatic cell hybrid, providing a model system tostudy the inactive X chromosome apart from its activecounterpart. The largest survey of gene expression fromthe inactive X chromosome [5] analysed expression in apanel of nine inactive-X containing hybrids. That studydefined heterogeneous inactivation as expression in threeto six of the nine hybrids, which was observed for 60 ofthe 624 X-linked genes analysed. This variable expressionis not restricted to the hybrid system, but has also beendemonstrated in cells from females [5,12,13].We now report the further characterization of one of thesegenes, the X-linked tissue inhibitor of metalloproteinases,TIMP1, located in Xp11.23. Our previous studies havedemonstrated that TIMP1 is variably expressed from theinactive X in both somatic cell hybrids and humanfemales. When TIMP1 is expressed from the inactive X,flanking genes (including ARAF1 and ELK1 which lie ~20and 55 kb from TIMP1, respectively) are not expressed,suggesting that expression is being controlled in a gene-specific rather than regional fashion [12]. QuantitativeRNase protection assays showed substantial variability inexpression levels from the active X, precluding usingexpression levels to determine inactive X expression infemales. Further studies in hybrids demonstrated thatTIMP1-expressing clones were unstable and that methyla-tion does not appear to be the principle controlling fea-ture allowing variable expression of TIMP1 from theinactive X chromosome [14]. In this study we report theanalysis of other features characteristic of an inactive X todetermine which features might predispose TIMP1 toexpression from the inactive X chromosome in a subset offemales.The inactive X acquires many of the general features ofheterochromatin (reviewed in [15]), and we have nowexamined replication timing, nuclease sensitivity, and his-tone modifications for TIMP1. Late replication of the inac-tive X at the chromosome level is observed after Giemsastaining following bromodeoxyuridine incorporation[16]. Regions such as the distal and proximal short armthat contain a large proportion of genes that escape inac-tivation are not delayed in their replication, supporting aregional basis to escape from inactivation. Replication ofindividual X-linked genes has been analysed by fluores-cent in situ hybridization (FISH) or amplification ofBuDR incorporated DNA after flow cytometry. Althoughcorrespondence is not complete between different tech-niques [17], such methods have generally shown thatgenes that escape inactivation are early replicating(reviewed in [18]). DNase sensitivity is a general feature ofactive chromatin, and promoters of genes subject to inac-tivation have been seen to be less available for digestionby nucleases (e.g. [19]). Modifications to the histonesassociated with the inactive X are reflective of both generalheterochromatic changes and ones specific to the faculta-tive heterochromatin of the inactive X. Using antibodiesto acetylated histones the inactive X chromosome stainsvery palely [20], while antibodies to histone H3 methyl-ated at lysine (K) 9 (H3mK9) are generally associated withheterochromatin and those to methylated lysine 27(H3mK27) specifically mark the inactive X chromosome[21,22]. Chromatin immunoprecipitation (ChIP) hasrevealed that genes subject to inactivation show limitedPage 2 of 14(page number not for citation purposes)acetylation and elevated histone methylation (H3mK9/BMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/4827) at their promoters relative to genes escaping inactiva-tion [23,24].ResultsReplication timing of TIMP1 in human cellsTo evaluate the influence of replication timing uponexpression of TIMP1, cell lines from a female with and afemale without TIMP1 expression from the inactive X [12]were analysed by DNA FISH using probes for the TIMP1and HPRT loci. As shown in the schematic of Figure 1,cells that have not replicated either X chromosome willshow two single signals, and almost one half of the cellswere of this type. Of the cells with at least one double sig-nal, indicating replication of the locus, considerable asyn-chrony was observed for the HPRT locus, as 37% of allcells analysed showed single-double (SD) signals and17% showed the double-double (DD) signals. The TIMP1locus generally showed lower asynchrony of replicationthan HPRT (28% of cells were S/D vs 37% for HPRT, P <0.0001). The cell line with TIMP1 expression from theinactive X (cell line 2 – HSC593) showed a small trendtowards having a lower percentage of cells that replicatedasynchronously than the GM07059 cell line which doesnot have TIMP1 expression from the inactive X (25% vs29%); however this difference was not significant (P =0.17). The observed shift in replication timing, while notstatistically significant, may be a contributing factor in theexpression of TIMP1 from the inactive X. However, it islikely that additional changes are also involved, as TIMP1is within 20 kb of the ARAF1 gene and 55 kb from ELK1(see Figure 2A), and thus at least one of these genes islikely to share a replication domain with TIMP1 despiteremaining silent when TIMP1 is expressed [12].DNA methylation of TIMP1 in somatic cell hybridsAnalysis of alterations to the inactive X chromosome iscomplicated by the presence of the active X chromosomein female cells, so we have analysed features of inactive Xchromosomes isolated in mouse/human somatic cellhybrids. We have previously described the characteriza-tion of DNA methylation status and expression levels in anumber of these hybrids [14] and Figure 2B presents anoutline of the hybrids analysed in this study. In additionto hybrids retaining the active X chromosome (t60-12(t60) and AHA11aB1 (AHA)), in which TIMP1 wasunmethylated and expressed, three classes of inactive X-containing hybrids were previously described. Those thatwere methylated and did not have TIMP1 expression (Xi/-/M – t11-4Aaz-5 (t11) and t48-1a-1Daz4a (t48)) orexpressed TIMP1 and were unmethylated (Xi/+/U – t75-2maz34-1a (t75)) were stable, with no gain or loss ofexpression. In contrast, the two hybrids that expressedTIMP1 but were methylated at the TIMP1 promoter (Xi/+/and expressing (Xi/+/M) or methylated and silent (Xi/-/M) as well as occasionally unmethylated and expressing(Xi/+/U). Consistently the Xi/-/M and Xi/+/U subclonestended to be stable while Xi/+/M subclones were unstable.As the methylated-expressing clones (Xi/+/M) are unsta-ble and give rise to a mixed population of methylatedsilent and unmethylated expressing clones, these clonesdo not indicate if loss of methylation follows, or predis-poses to, expression.To further test the role of methylation in regulatingexpression in the TIMP1 region, an inactive X-containinghybrid that did not express TIMP1 (t11) was treated with5-azacytidine, which is known to induce reactivation of X-linked genes in somatic cell hybrids. As there is not aselectable marker for TIMP1 reactivation, clones were ini-tially selected for HPRT reactivation to ensure that theclones had been demethylated and potentially increasethe frequency of reactivation, as co-reactivation of differ-ent X-linked genes has been observed after 5-azacytidinetreatment [25]. 15 HPRT+ clones were selected in HATmedia, and examined by RT-PCR for TIMP1 as well asflanking gene expression (Table 1). After three passages,six clones showed expression of TIMP1. A number offlanking genes were also observed to reactivate with 5-aza-cytidine treatment; with one clone expressing ARAF1, fiveexpressing ELK1, three expressing ZNF41 and eightexpressing ZNF157. The single ARAF1-expressing clone,and all five ELK1-expressing clones, were observed inTIMP1+ clones, and all three ZNF41 positive clones wereZNF157 positive. The TIMP1 and ZNF157 genes lack aCpG island (see Figure 2A), and TIMP1 and ZNF157showed the highest (6/15 and 8/15) reactivation frequen-cies after 5-azacytidine treatment. However, ARAF1showed the lowest reactivation frequency (1/15) despitehaving a smaller CpG island than either ELK1 or ZNF41,so reactivation frequency was not simply a reflection ofCpG density.Despite 5-azacytidine treatment and subsequent expres-sion in some clones, DNA methylation continued to beobserved at 5' end of ARAF1 and TIMP1 in these clones, asshown in Figure 2C. Given our previous results demon-strating instability of expression in the presence of meth-ylation, four clones were subcloned after 7 weeks ofculture. Analysis of expression and DNA methylation forTIMP1 (Table 2) showed that, as expected, all subclonesof a non-expressing clone (t11-az-4) remained silent. Forboth t11-az-8 and 9, the majority of clones lacked TIMP1expression. Only two weakly positive clones were identi-fied for each line, and even these subclones lost expres-sion by 12 weeks in culture. Three subclasses of hybridswere observed for the t11-az-10 subclones, reminiscent ofPage 3 of 14(page number not for citation purposes)M – t86-B1maz1b-3a (t86) and t81-az1D (t81)) demon-strated instability as subclones could be either methylatedthe situation seen in hybrids from females who spontane-ously express TIMP1 from the inactive X. t11-az-10 wasBMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48the only reactivated clone that expressed ARAF1, and nineof the 14 TIMP1-expressing subclones expressed ARAF1.The single subclone that was unmethylated at TIMP1 (t11-az-10-10) expressed TIMP1, and also expressed ARAF1,which was also unmethylated. t11-az-10-7 was methyl-ated yet expressing for both TIMP1 and ARAF1. Theseclones (t11-az-10-10 and t11-az10-7) were further sub-cloned. The TIMP1 expression level of the subclones waslevel of TIMP1 than the unmethylated t11-az-10-10 sub-clones (Figure 2D).DNA methylation analysis was routinely based on meth-ylation-sensitive restriction enzyme digestion followed byPCR that examined 4 sites, 3 of which were over 50 bpdownstream of the transcription start site (Figure 2E). Toexamine methylation of additional CpG dinucleotidesReplication asynchrony for the TIMP1 (T – white bars) and HPRT (H – grey bars) genes assessed by FISHFigure 1Replication asynchrony for the TIMP1 (T – white bars) and HPRT (H – grey bars) genes assessed by FISH. The bars show the frequency of nuclei exhibiting unreplicated (single/single); asynchronously replicated (single/double) and completely replicated (double/double) signals as shown in the schematic below. The TIMP1 and HPRT probes were individually hybridized to inter-phase nuclei of the same preparations. Two human female lymphoblast cell lines were examined for the approximate degree of replication asynchrony, one that inactivated TIMP1 (1 – GM07059) and one that expressed TIMP1 from the Xi (2 – HSC593). For TIMP1, 313 cells were counted from 3 separate cell harvests for cell line 1 (T1) and 217 cells for cell line 2 (T2) from two separate harvests. 144 cells were examined for HPRT from 2 slides of one harvest for cell line 1 (H1) and 97 cells from a single harvest for cell line 2 (H2). Between different cell harvests of the same cell line the maximal difference in frequencies of nuclei in each of the three replication classes was 3%.Frequency(%)0102030405060T1 T2 H1 H2 T1 T2 H1 H2 T1 T2 H1 H2single/single single/double double/doubleUnreplicated Asynchronous FullyReplicatedReplicationPage 4 of 14(page number not for citation purposes)determined by RNase protection. The t11-az-10-7 sub-clones retained DNA methylation, and expressed a lowercloser to the promoter, methylation analysis by bisulfitemodification followed by sequencing of PCR productsBMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48Methylation of TIMP1Figure 2Methylation of TIMP1. A. The genes in the region surrounding TIMP1 are shown with the gene name written above the line dia-gram for the gene (with vertical lines representing exons and small arrows showing transcriptional orientation). The number listed on the line below at the 5' end of the genes indicates the CpG island density. The TIMP1 and ZNF147 genes do not have enough CpG sites to qualify as a CpG island so there is no number listed. The SYN1, and PFC genes are listed in grey because these genes (which are expressed in a tissue-specific manner) were not examined in this study. The presence of repeat ele-ments (SINE, LINE and LTR) is indicated by the vertical lines below the genes. This figure is based on data generated by the UCSC browser http://www.genome.ucsc.edu/ hg17 of NCBI Build 35. B. Somatic cell hybrids analysed for TIMP1 activity. In addition to the stable active X chromosomes that express TIMP1 and are unmethylated at the 5' end of the gene (Xa/+/U) there are several categories of inactive X chromosomes. Most inactive X chromosomes previously analysed do not express TIMP1 and are methylated (Xi/-/M). Other inactive X chromosomes express TIMP1 and are unmethylated (Xi/+/U), while an intermediary class of hybrids showed both DNA methylation and lower expression levels (Xi/+/M). Subcloning of these latter cells showed that they were unstable, giving rise to additional methylated expressing clones as well as methylated silent clones and expressing unmethylated clones. The arrows show approximate proportion of cells of each class derived from subcloning. Subclones further characterized are listed below, as are clones derived by 5-azacytidine treatment and their subclones (see Tables 1 and 2). C. DNA methylation of clones from four of the demethylated clones listed in Table 1. DNA from each clone was digested with EcoRI alone (U), EcoRI plus HpaII (II) or EcoRI plus HhaI (I). Primers for the 5' end of TIMP1 and ARAF1 that flank HpaII or HhaI methylation-sensitive restriction enzyme sites were used to assess methylation, while amplification of MIC2, which is unmethylated on both active and inactive X chromosomes, served as a control for complete digestion with the meth-ylation-sensitive enzymes. D. Comparison of expression levels in subclones of two 'sibling' subclones of t11-az-10 differing in methylation states. The t11-az-10-7 and t11-az-10-10 clones are striped, with their subclones shown to their right. Methylation (dark fill) was observed for t11-az-10-7 and its subclones while t11-az-10-10 and its subclones were unmethylated (unfilled). All subclones continued to express both TIMP1 and ARAF1 as assayed by RT-PCR. Despite the relative stability of the methylated TIMP1+ culture, the TIMP1 expression level was significantly lower in the methylated cultures (p < 0.01). E. Methylation analysis by bisulphite treatment. The 5' end of the TIMP1 gene was sequenced after bisulfite conversion, which changes unmethylated Cs to Us but leaves methylated Cs unchanged. Therefore, the presence of a C indicates that the CpG was methylated. The fol-lowing CpG sites were analyzed: HhaI sites (stars) at -3 and +31; and three other sites not analyzed by methylation-sensitive enzymes (circles) at +11, +17, +20. The HpaII sites (triangles) at +61 and +81 were used in methylation-sensitive assays but were not reliably analysed by bisulphite sequencing. The open circles indicate unmethylated CpGs whereas the filled circles represent methylated CpGs. The shaded circles designate that both converted and unconverted bases were seen after sequencing, indicating that both methylated and unmethylated CpGs were present. Cell lines are listed, the male cells were 1 - a z - 1 0 - 71 0 - 7 a1 0 - 7 b1 0 - 7 c1 0 - 7 d1 0 - 7 e1 0 - 7 it1 1 - a z - 1 0 - 1 01 0 - 1 0 b1 0 - 1 0 d1 0 - 1 0 eTIMP1 expressionZNF157 ZNF41ARAF1SYN1 PFCELK180 62 106 2967At11-az-4     t11-az-8     t11-az-9    t11-az-10U    II     I    U     II     I    U    II    I     U   II    ITIMP1ARAF1MIC2Cmalefemalet60 (Xa)t11 (Xi/-/M)t86-6J (Xi/+/U)t11-az-10-10 (Xi/+/U)t11-az-10-7 (Xi/+/M)t11-az-9-3 (Xi/-/M)25 kbTIMP1EDB Xa Xi Xi XiHybrids in study (+/U) (-/M) (+/M) (+/U)     Original t60 t11 t86 t75 (described in [12,14]) AHA t48 t81Subclones  t86-6P t86-6K t86-6J   t86-1U     t81-4   Demethylated  t11-az-4 t11-az-8     t11-az-9     t11-az-10  Demethylated subclones  t11-az-9-10 t11-az-10-7 t11-az-10-10 Page 5 of 14(page number not for citation purposes)GM7057 and the female cells were GM7059. t11-az-9-3 is a TIMP1- subclone of t11-az-9.BMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48was performed. Three additional CpGs near the TIMP1transcription start yielded the same methylation patternas had been observed at the restriction enzyme sites.Direct sequencing of the PCR product was performed toallow the identification of heterogeneous populations.This approach is complementary to the methylation-sen-sitive restriction enzyme approach, which gave a positivesignal (amplification) for low levels of methylated DNA.The high background corresponding to the unconverted(methylated) Cs, presumably due to their under-represen-tation in the sequence, prohibits identification of low lev-els of methylated bases by direct sequencing. However, nobackground of the converted base (A) was present andthus this technique is sensitive to the presence of low lev-els of unmethylated DNA. If the peak corresponding tothe converted base (A) was demonstrably higher than theunconverted (G), the site was called unmethylated (whitecircles in Figure 2E). Completely methylated sites (as seenfor the inactive X hybrid) are indicated by filled in blackcircles. Sequencing also yielded a mix of unmethylated(unconverted – G) peak height was greater than or equalto the unmethylated (converted – A) peak height. Such amix, demarcated by the grey circles, was also observed forthe t11-az-10-7 cells, although in this case the unmethyl-ated peak was consistently lower than the methylatedpeak.DNase sensitivity of TIMP1 in somatic cell hybridsNuclease sensitivity of the TIMP1 promoter, the promoterof the nearby ARAF1 gene, as well as the TIMP1 gene bodyand the anonymous DNA marker DXS8037 was assessedin a series of these clones (Figure 3). The promoters ofTIMP1 and ARAF1 were sensitive to nuclease on theactive, but not the inactive X chromosome, while the genebody or intergenic regions were generally resistant onboth active and inactive X chromosomes. The TIMP1 pro-moter was also sensitive to digestion in the unmethylatedexpressing (Xi/+/U) TIMP1 clones, while it was insensitivein both the TIMP1-expressing, and silent methylatedclones (Xi/+/M and Xi/-/M). A similar effect was observedTable 1: Expression of TIMP1 and surrounding genes following 5-azacytidine induced reactivation of HPRTClone * TIMP1 ARAF1 ELK1 ZNF41 ZNF157t11-az-4 - - - - +t11-az-5 - - - - +t11-az-6 + - + - -t11-az-7 + - + - -t11-az-8 + - + - -t11-az-9 + - + - +t11-az-10 + + - + +t11-az-11 - - - + +t11-az-14 - - - - +t11-az-16 - - - - +t11-az-17 - - - - -t11-az-18 - - - + +t11-az-19 + - + - -t11-az-20 - - - - -t11-az-21 - - - - -*The clones in bold were ones that were further subcloned to assess stability (see Table 2).Table 2: Stability of methylation and expression of TIMP1 in subclones of four 5-azacytidine induced HPRT reactivantsSubclones t11-az-4 (-/M) t11-az-8 (+/M)a t11-az-9 (+/M)a t11-az-10 (+/M)TIMP-, methylated (-/M) 10 10 15 3TIMP+, Methylated (+/M) 0 2a 2a 13TIMP+, Unmethylated (+/U)0 0 0 1aExpression in these branches was initially weak and became silent after time in culture. ELK1 expression was initially retained in one of the t11-az-9 TIMP1+ clones but was not analyzed after TIMP1 expression was lost. The expression of ZNF41/157 was not examined in these subclones.Page 6 of 14(page number not for citation purposes)(converted) and methylated (unconverted) sites. This wasobserved in a female cell line, where the methylatedfor the demethylated clones, with both TIMP1 andARAF1, being expressed in both the t11-az-10-10 and t11-BMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48az-10-7 clones, but only sensitive to nuclease digestion inthe t11-az-10-10 clone, where they were unmethylated.Histone modificationsA growing number of specific modifications to the histonetails have been associated with both constitutive and fac-ultative heterochromatin. To assess the potential role ofsuch modifications in the escape from inactivation ofTIMP1, we analysed the hybrids by ChIP with antibodiesto acetylated histone H3 (antibody recognizes acetylatedK9 and 14), acetylated histone H4 (antibody recognizesacetylated K 5, 8, 12, and 16) and methylated histone H3(antibody to di-methyl K9, although the antibody willcross-react with tri-methyl K27 see:http:www.upstate.com/browse/productdetail.asp?Prothe inactive X chromosome) were analysed (shown in Fig-ure 4). Additional analyses of the promoters of the PGK1gene that is expressed only from the active X chromo-some, and the ZFX gene that is expressed from both activeand inactive X chromosomes, gave the anticipated resultsfor active and inactive X chromosomes (data not shown).Antibody to acetylated H3 immunoprecipitated theTIMP1, ARAF1 and ELK1 promoters for the active, but notinactive X-containing hybrids, while the reverse wasobserved for the XIST promoter. ELK1 was not expressedin the inactive X hybrids examined, and was not immuno-precipitated. However for TIMP1 and ARAF1 associationwith acetylated histone H3 was observed for all expressingclones whether on the active or inactive X chromosome.In addition, immunoprecipitation of the TIMP1 promoterNuclease sensitivity of TIMP1Figure 3Nuclease sensitivity of TIMP1. Nuclei from various cell lines (as indicated on the left) were treated with increasing amounts of DNase I (0, 0.1, 0.25, 0.5, 1.0 units) as represented by the increasing breadth of the triangle. The DXS8037 primers flank a non-coding region and were used to check that equal digestion and PCR amplification occurred across all cell lines. The cell line designations indicate X activity, TIMP1 expression status, and methylation status as shown in Figure 2B.t60 (Xa)t75 (Xi/+/U)t48 (Xi/-/M)TIMP1gene bodyTIMP1promoterARAF1promoterDXS8037t86-6K (Xi/+/M)t11 (Xi/-/M)t86-6J (Xi/+/U)t86-6P (Xi/-/M)t11-az-10-10 (Xi/+/U)t11-az-10-7 (Xi/+/M)DNasePage 7 of 14(page number not for citation purposes)ductId=07-212). The promoter regions of TIMP1, ARAF1,ELK1 and the XIST gene (the latter is expressed solely fromwas observed for Xi/-/M clones (t86-6P/t86-1U and t81-4). Immunoprecipitation was completely concordantBMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48with expression for the acetylated histone H4 antibody atall of the genes examined. Methylation of H3 at K9 corre-sponds to silent chromatin, and thus, as anticipated,results were generally opposite to those seen for acetyla-tion. Immunoprecipitation was observed when the genewas silent – i.e. XIST on the active X and TIMP1, ARAF1and ELK1 on the inactive X. However, in all cases wherethere was expression in the presence of ongoing promotermethylation (for both ARAF1 and TIMP1) the promoterwas immunoprecipitated by the methylated H3 antibody.Thus we detect a distinctive pattern of histone acetylationand methylation that does not correspond simply withthe expression or DNA methylation status of the TIMP1gene.DiscussionThe facultative heterochromatin of the inactive X chromo-some is a fascinating system to study the epigeneticmodifications associated with silent chromatin as both anactive and inactive version of most X-linked genes exist infemale cells. However, not all genes on the 'inactive' Xchromosome are subject to silencing [26], and these genesthat escape inactivation must somehow maintain anactive state on an otherwise silent chromosome. TIMP1 isof inactive chromatin are assembled – promoter DNAhypermethylation, nuclease insensitivity and histonemethylation and hypoacetylation. In other females, how-ever, we have previously demonstrated that TIMP1 con-tinues to be expressed from the inactive X chromosome,and in this study we have exploited such chromosomes toanalyze the epigenetic features that are associated withexpression from the inactive X. By comparing expressionof TIMP1 that occurs naturally with that induced by thedemethylating agent 5-azacytidine we are also able to pro-pose which feature may predispose to expression ofTIMP1 in otherwise silent chromatin. We observed thatexpression of TIMP1 was associated with an active chro-matin structure, despite the presence of the gene on theinactive X chromosome, except in three situations.First, in female cell lines with or without TIMP1 expres-sion from the inactive X there was a very similar extent ofreplication asynchrony, suggesting that the expression ofTIMP1 was ocurring from a late-replicating region of the Xchromosome. This is not surprising, as the genes flankingTIMP1, which are not variable in their inactivation, arelocated within 55 kb, so it is likely that at least one ofARAF1 or ELK1 shares a replication origin with TIMP1. WeChIP analysis of TIMP1 expressing or non-expressing hybrid clonesFigure 4ChIP analysis of TIMP1 expressing or non-expressing hybrid clones. PCR products were amplified from DNA at various X-linked gene promoters after ChIP using antibodies to the modifications listed below the panels for the cell lines listed across the top of each panel (see Figure 2B for derivation of lines). t11-az-10-7a is a subclone of t11-az-10-7 (see Figure 2D). The DNA template for 'no Ab' lanes was prepared following the ChIP procedure without an antibody. A. PCR amplification prod-ucts after ChIP with acetylated histone H3. Similar analysis to A, using antibody to acetylated histone H4 (panel B), or methyl-ated histone H3 (C).AHA(Xa)t48(Xi/-/M)t60(Xa)t11(Xi/-/M)noAbt75(Xi/+/U)t86-6P(Xi/-/M)t86-6K(Xi/+/M)t86-6J(Xi/+/U)t86-1U(Xi/-/M)t81-4(Xi/-/M)t11-az-10-10(Xi/+/U)t11-az-10-7a(Xi/+/M)t11-az-9-10(Xi/-/M)ATIMP1ARAF1TIMP1ARAF1ELK1TIMP1ARAF1ELK1XISTBAHA(Xa)t11(Xi/-/M)noAbt75(Xi/+/U)t86-6J(Xi/+/U)t86-6K(Xi/+/M)t86-6P(Xi/-/M)t11-az-10-10(Xi/+/U)t11-az-10-7a(Xi/+/M)t11-az-9-10(Xi/-/M)ELK1XIST XISTCAcetylated H3 Acetylated H4 Methylated H3AHA(Xa)t11(Xi/-/M)noAbt75(Xi/+/U)t86-6J(Xi/+/U)t86-6K(Xi/+/M)t86-6P(Xi/-/M)t11-az-10-10(Xi/+/U)t11-az-10-7a(Xi/+/M)t11-az-9-10(Xi/-/M)Page 8 of 14(page number not for citation purposes)subject to inactivation in many females, and we now showthat when subject to inactivation the anticipated featuresassessed replication timing by FISH, and detected afrequency of replication asynchrony for the HPRT locus inBMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48close agreement with the frequencies previously reportedfor this gene [27]. It has been suggested that the 'doubledot' versus 'single dot' pattern may reflect chromatid asso-ciation in addition or instead of replication asynchrony[17]. Regardless of the underlying cause, the results werehighly reproducible and consistently showed less asyn-chrony for the TIMP1 locus relative to the HPRT locus,although still in the range that is generally consideredasynchronous [28]. This reduced asynchrony at TIMP1could reflect, or contribute to, weaker epigenetic silencingbeing established in this region of the X chromosome.There was a slight, but not statistically significant, trendfor the female showing expression of TIMP1 from theinactive X chromosome to show even less replicationasynchrony for TIMP1, however it seems unlikely thatsuch a small difference could contribute substantially toTIMP1 expression from the inactive X, and thus otherfactors must be involved in permitting the expression ofTIMP1 from the inactive X.The other two situations in which expression and activechromatin features were not concordant were found in thesomatic cell hybrids, and these results are summarized inFigure 5. As shown in the grey box, DNA methylation andadditional features of silent chromatin were detectedalong with expression in the Xi/+/M (t86, t81) cells. Weattribute this to heterogeneity in the population of cells,as subcloning yielded both methylated/silent andunmethylated/expressing clones. In addition to DNAmethylation, these clones were observed to be insensitiveto DNase and have methylated H3K9 residues near theTIMP1 promoter. As the assays for these features relied onPCR, a positive signal could be obtained from a subpopu-lation of cells with a silent chromatin structure, whileanother population of cells could be positive for expres-sion and active chromatin modifications (histone acetyla-tion). We also suggest that heterogeneity accounts forexpression in the presence of DNA methylation in severalof the demethylated clones, including the t11-az-10-7clone. However, unlike the mixed population of sub-clones obtained with all other Xi/+/M cells, eight of eightsubclones of t11-az-10-7 were methylated and expressing,and six of these subclones examined by RPA showed aconsistently reduced level of expression relative to theirunmethylated counterparts. While it was surprising thatsubcloning did not isolate distinct subpopulations, thepresence of an unmethylated subpopulation was detecta-ble by bisulphite sequencing. Thus heterogeneity againseems the most likely explanation for DNA methylation,DNase insensitivity and histone H3K9 methylation inthese expressing cells. However, without single cell assaysit is not possible to rule out that there are methylated cellsthat express TIMP1 at reduced levels and show additionalase sensitivity and gene expression at the promoters of theTIMP1 and ARAF1 genes; and previous studies of thenucleosomal organization of the HPRT1 gene promoterhave shown that methylation does not directly affect thedifferential positioning of nucleosomes on active andinactive X-linked promoters [19]. Thus we believe that themethylated/expressing/nuclease insensitive clones reflectthe presence of a subpopulation of silent cells. This heter-ogeneity demonstrates the unstable nature of silencing forTIMP1 in these cells.The third exception is the most interesting, and is high-lighted in Figure 5 with a dark grey fill. Acetylation of his-tones is generally seen for active genes, and acetylatedhistone H4 showed complete concordance withexpression for all genes. However, the Xi/-/M clones (t86-6P and 1 U as well as t81-4) showed ongoing acetylationat H3 despite having lost TIMP1 expression from the inac-tive X chromosome. The Xi/-/M subclones were derivedfrom two different unstable, naturally expressing inactiveX chromosomes (in t86 and t81), however expression ofTIMP1 in these hybrids was now stably silenced, and thusthis result was not a reflection of a mixed population ofcells. Furthermore, acetylation cannot simply reflect a fail-ure of this region to reset chromatin structure, as thedemethylated t11-az-9-10 hybrid which had lost expres-sion had also lost acetylation. Methylation of H3 K9 wasobserved in these silent clones, so the Xi/-/M cells, despitebeing a homogeneous population, appear to show bothmethylation and acetylation of histone H3. This mayreflect the lack of specificity of the antibodies used forthese experiments, as the acetylated H3 antibody recog-nizes acetylation at residues 9 and 14, while the antibodyto methylated K9 can cross-react with methylation of K27.Thus immunoprecipitation by both antibodies mayreflect modification of specific lysines. It is also possiblethat only a particular set of nucleosomes show the acetyla-tion mark, and as the sonicated fragments immunoprecip-itated in the assay averaged ~600 bp, modifications onseveral different nucleosomes flanking the primers usedcould result in immunoprecipitation. Regardless of thespecific site of modification, acetylation of histone H3was the one feature that was consistently associated withX chromosomes that were naturally predisposed toexpression from the inactive X chromosome, regardless ofexpression status. Thus we propose that histone H3acetylation differs at the TIMP1 genes in females, predis-posing some females to expression from the inactive X.Unfortunately this hypothesis is difficult to test, asfemales generally show acetylation at TIMP1, due to thepresence of the active X chromosome. It would be neces-sary to study clonal populations of cells from a femalewith a polymorphism close enough to the promoter to bePage 9 of 14(page number not for citation purposes)features of silent chromatin. These two classes of cloneswere the only exceptions to concordance between nucle-analysed by ChIP, and currently no such polymorphismsare known. Our previous work did not show anBMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48association between expression of TIMP1 and a down-stream polymorphism in exon 5 [12].A different predisposing epigenetic mark for expressionfrom the inactive X has previously been reported in mice.Methylation of H3 at K4 was restricted to the promoter inundifferentiated embryonic stem cells for genes thatwould subsequently be expressed monoallelically (i.e.genes subject to X inactivation or imprinting), while auto-somal genes or genes that escape inactivation had H3 K4methylation in the gene body as well as the promoter[11]. Such methylation was observed in the body of theSmcx gene which is initially silenced, and then reactivates,reminiscent of the H3 acetylation mark and instability ofsilencing that we observe for TIMP1. While many genesescape inactivation in humans, TIMP1 provides us with ainteresting to study additional chromatin modificationsfor TIMP1, including H3 K4 methylation, as well as addi-tional modifications recently associated with the inactiveX chromosome such as H4 K20 methylation [29] or H2Aubiquitinylation [30,31]. Even more interesting would beto determine if the changes seen are observed for addi-tional X-linked genes that are variable in their inactivationstatus, if suitable model systems could be developed.Previous examinations of genes that escape inactivationhave shown an active chromatin structure (reviewed in[7]), and recent results further demonstrate that epige-netic modifications seem to be heterogeneous in their dis-tribution along the X [32]. These 'flavours' of inactive Xchromatin may correspond to clusters of genes that escapeinactivation. While genes that escape inactivation tend toSummary of chromatin features observed in somatic cell hybrid clones for the TIMP1 geneFigure 5Summary of chromatin features observed in somatic cell hybrid clones for the TIMP1 gene. Most subclones and demethylated clones follow the patterns seen for the active and inactive X hybrids that are outlined in bold. A positive (+) designates the presence of the feature, while a negative (-) depicts the absence of the feature, while ND means that the feature has not been examined in that class of clones. The assays used are PCR-based and would detect a small population of cells. DNase sensitivity is listed as the inverse – DNase resistance – as it is the presence of undigested (resistant) DNA that will yield a 'positive' PCR signal. In each category of clones not all clones have been examined for all features. For the clones shaded in grey the results are an amalgamation of results anticipated from an active and inactive X, and we suggest these clones represent a mixed popu-lation. This suggestion is generally supported by sub-cloning experiments (see text for discussion). The ongoing H3 acetylation of the Xi/-/M hybrids (highlited in darker grey) cannot be attributed to heterogeneous cell populations, and since it is not seen for demethylated hybrids that have lost TIMP1 expression (t11-az/-/M) we suggest that this modification reflects a predisposing feature of inactive X chromosomes that express TIMP1.Hybrids instudyExpression H3AcetylationH4AcetylationDNaseResistanceDNAMethylationH3MethylationXa t60AHA+ + + - - -Xi/+/U t75t86-6J+ + + - - -t11-az/+/U t11-az-10-10+ + + - - -t11-az/+/M t11-az-8t11-az-9t11-az-10t11-az-10-7+ + + + + +Xi/+/M t86t81t86-6K+ + + + + +Xi t11t48- - - + + +t11-az/-/M t11-az-4t11-az-9-10- - - nd + +Xi/-/M t86-6Pt86-1Ut81-4- + - + + +Page 10 of 14(page number not for citation purposes)unique opportunity to examine the gene in its expressedand silent state in hybrid somatic cells. Thus, it will bebe clustered in blocks and enriched on the short arm ofthe X chromosome, genes such as TIMP1 that are variableBMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48in their inactivation are more evenly distributed along theX chromosome ([5], reviewed in [6]), and thus may reflectdifferent mechanisms leading to expression from the inac-tive X. The CTCF boundary factor has recently been shownto be found between the domain of escape and that ofinactivation [33]. It is not known whether such bounda-ries flank the variably inactivated genes. The five clonesthat showed reactivation of ELK1 were all TIMP1+, sug-gesting a co-ordinate reactivation between the two genes.Only one ARAF1 reactivant was identified, and this clonewas also TIMP1+. However, this co-ordinate reactivationof TIMP1 with flanking gene(s) is different from theTIMP1-specific expression seen for the inactive Xnaturally. The ZNF157/41 genes did not seem to show co-ordinate reactivation with the TIMP1 region, however allthree ZNF41 positive clones were ZNF157 positive, whichsuggests they may be in a separately controlled domain.While association with a CpG island does not differbetween genes subject to or escaping from inactivation,genes heterogenous in their inactivation are more likely tolack an island [5], consistent with the TIMP1 promoterbeing the least CpG dense of the region. Although TIMP1lacks a CpG island, our analysis suggests that methylationof the few CpGs present near the promoter generally cor-relates well with silencing. Demethylation resulted in atleast transient expression of all genes examined, confirm-ing the importance of DNA methylation in stable mainte-nance of X chromosome silencing. Examination of theELK1, TIMP1 and ARAF1 promoters after demethylationshowed the presence of DNA methylation, even when theclones were expressing the genes. The majority of theseclones subsequently resilenced these genes, as has beenobserved for other genes following demethylation [34],perhaps reflecting only partial initial demethylation or thespread of silencing from adjacent silent regions thatretained DNA methylation or other epigenetic marks ofsilencing.It has been proposed that the evolutionarily recent addi-tion of the X short arm may predispose genes located thereto escape from silencing [35]. Interestingly, TIMP1 is veryclose to the evolutionary breakpoint between the regionadded to the X after marsupial divergence. ARAF1 is foundon the marsupial X while TIMP1 and its surrounding geneSYN1 are autosomal in marsupials, suggesting that theyare part of the eutherian addition to the human X [36].Since genes closely flanking TIMP1 are normally subjectto X inactivation, even when TIMP1 is expressed, evolu-tionary history alone cannot explain the escape from inac-tivation, although it may result in a different genomiccontext that contributes to expression from the inactive X.However, inspection of genomic features in the regionrelative to the flanking ARAF1 and ELK1 genes (Figure2A).ConclusionSeveral factors, including reduced replication asynchrony,lower CpG density, and more recent evolutionary addi-tion to the X, may contribute to less stringently controlledinactivation for TIMP1. However, we propose that there isa difference between females for a feature unique toTIMP1 that predisposes some females to expression of thegene. This mark appears to be at least reflected in a differ-ence in acetylation of histone H3 on the inactive X infemales predisposed to expression of TIMP1 from theinactive X chromosome. While histone acetylationappears to be a predisposing mark for expression ofTIMP1, there may be a different hierarchy of epigeneticmodifications permitting expression of other genes fromthe inactive X. Elucidation of these mechanisms isimportant not only as a model for epigenetic gene regula-tion but because genes that escape inactivation contributeto the phenotype of X chromosome anueploidies, andmay also result in differential male/female expression lev-els and disease susceptibilities.MethodsCell cultureLymphoblast cell lines were grown in RPMI 1640 media(Stem Cell Technologies) supplemented with 15% fetalcalf serum (Cansera), L-glutamine (Invitrogen) and peni-cillin/streptomycin (Invitrogen). Cells were harvested 12–26 hours after addition of fresh media by centrifugation.The human/rodent somatic cell hybrids were grown inalpha minimal essential media (Invitrogen) supple-mented with 7.5% fetal calf serum, L-glutamine,penicillin/streptomycin, and non-essential amino acids(Invitrogen) to sub-confluence before harvesting withtrypsin-EDTA (0.25%). Cell lines have been previouslydescribed [14]. To generate single cell clones, the hybridcultures were plated to a final concentration of 3 to 17cells/60 mm plate. After 5 to 10 days in culture, well-sep-arated colonies were isolated by trypsinization in cloningcylinders and transferred to new 60 mm plates. To inducedemethylation, an inactive X-containing hybrid that hadnever expressed TIMP1 (t11-4Aaz-5) was treated with 5-azacytidine as previously described [37]. The cells weregrown in media supplemented with HAT (Invitrogen) toselect for HPRT reactivants and then single cell cloned totest for TIMP1-positive cultures. To remove any confound-ing effects of HPRT selection, the cells were transferredback to alpha minimal essential media after 2 weeks ofselection. The expression of genes in the TIMP1 regionwas determined by RT-PCR as described previously [12].Page 11 of 14(page number not for citation purposes)surrounding TIMP1 does not show a substantial differ-ence in the frequency of repetitive elements near TIMP1BMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48Replication timingApproximately 4 × 106 lymphoblast cells were harvestedone day after subculture to ensure that they were activelygrowing. The cell pellet was resuspended in 8 ml of pre-warmed hypotonic (0.75 M KCl) and then incubated at37C for 10 minutes. 2 to 3 ml of fixative (3:1 metha-nol:glacial acetic acid) was slowly added before spinningat 200 g for 10 minutes. The cell pellet was washed threemore times with 5 ml fixative and then resuspended in 10ml fixative for storage at -20C for up to a week. The nucleipreparations were dropped onto slides and left at roomtemperature overnight. The slides were then incubated in2X SSC at 37C for 30 minutes followed by 2 minute roomtemperature incubations in each of: 70%, 85%, and 95%ethanol, and then allowed to air dry. To denature, theslides were incubated in fresh 70% formamide/2 × SSC at74C for 2 minutes followed by an ice cold ethanol seriesof rinses (70%, 85%, 95% for 2 minutes each) and air dry-ing. Probes for both the TIMP1 locus (lambda phageTIMP-3.9X) and the HPRT locus (Hulambda4x-8, ATCC57236) were labelled with dUTP-digoxigenin (Roche1745816) by nick translation (Roche 0976776). Unincor-porated nucleotides were removed with a PCR clean-upkit (Qiagen), and 1 ul (approximately 100 ng) of labelledprobe was mixed with 10 ul of 70% formamide hybridi-zation buffer and 40 ng of human Cot-1 DNA (Invitrogen15279011). The probes were denatured at 74C for 10minutes, pre-annealed at 37C for one hour and thenadded to the prepared slide to hybridize overnight in ahumidified chamber at 37C. The slides were washed in50% formamide/2 × SSC for 15 minutes at 43C, followedby two washes in 2 × SSC for 4 minutes at 37C and then 3washes at room temperature for 2 minutes each in 1 ×PBD (0.1 M NaH2PO4, 0.1 M Na2HPO4, 0.1% Triton X).To visualize the probe, slides were incubated with 500 ngof anti-DIG (sheep – Roche 1207741) conjugated to fluo-rescein, for 5 minutes at 37C, followed by three two-minute washes in 1 × PBD. To amplify the signal, FITC-anti-sheep (IgG FI-6000 (Vector Laboratories)) was incu-bated and washed as above. The slides were then counter-stained with DAPI mixed with antifade (VectorLabs).Cells were scored for nuclei with single-single, single-dou-ble, or double-double TIMP1 or HPRT signals on a ZiessAxioplan II microscope by a single individual who wasblinded as to the cell line being analysed.Methylation analysesFor methylation-sensitive restriction enzyme analysis,genomic DNA was pre-digested with EcoRI at 37C over-night, followed by incubation with 2 ul of RNase at 37Cfor 15 minutes. After phenol extraction and ethanol pre-cipitation, the DNA was quantified by spectrophotome-try. Pre-digested DNA (2 ug) was then incubatedHhaI. An aliquot of 1 ul (100 ng) was then used as a tem-plate in the PCR reactions as described previously [14]. Allprimers flanked a region of genomic DNA that did notcontain EcoRI restriction enzyme sites and contained 1–2HpaII or HhaI sites. For bisulfite analysis, 500 ng ofgenomic DNA was first denatured with 3 ul of 3 M NaOHat 37C for 10 minutes. After 15 ul of 20 mM hydroqui-none and 255 ul of 3.9 M sodium bisulfite were addedand mixed well, reactions were left at 50C for 16 hours toallow the unmethylated cytosines to convert to uracil.DNA was purified using the DNA Wizard Clean-Up Kit(Promega). After amplification with the TIMP-S primers(see Table 3), the 3' reverse oligonucleotide was used asthe primer for sequencing of the PCR product.RNase Protection quantitation of TIMP1 levelsRNase protection analysis was performed with Ambion'sRPA II kit, following the manufacturer's directions. RNAprobes were isolated after in vitro transcription with 32P-UTP. After solution hybridization overnight of excessantisense radiolabelled probe to 10 ug of total RNA, anyunhybridized probe and sample RNA was removed byRNase digestion. The hybridized product was then sepa-rated on a native 5% polyacrylamide gel, visualized byautoradiography, and bands quantified byphosphoimager (BioRad FX). The intensity of the TIMP1fragment was compared to the intensity of the banddetected for MIC2 used to control for the amount of inputRNA. All RPA results were normalized to one stock RNAto decrease variability between gels (see [14]).DNaseI sensitivityThe protocol for nuclease sensitivity was adapted prima-rily from [38]. The somatic cell hybrids were grown to75% confluence on 60 mm plates before harvesting with0.25% trypsin-EDTA. The cell pellets were resuspended in300 ul ice-cold DNase buffer (0.3 M sucrose, 60 mM KCl,15 mM NaCl, 5 mM MgCl2, 0.1 mM EGTA, 0.5 mM DTT,and 15 mM Tris-HCL pH7.5). The suspensions were thensplit into 6 tubes of 50 ul each and another 50 ul of ice-cold DNase buffer with 0.4% Nonidet P40 was added. Thetubes were mixed gently and placed on ice for four to fiveminutes before 100 ul of freshly diluted DNaseI (Invitro-gen 18047-019) was added, such that 2.5 U, 1 U, 0.5 U,0.25 U, 0.1 U, and no enzyme were used. The reaction wasthen incubated at 25C for 5 minutes, followed by 95C for15 minutes. The DNA was isolated with a standard phe-nol/chloroform extraction protocol, quantified with spec-trophotometry, and diluted to 60 ng/ul, 12 ng/ul, and 4ng/ul. 1 ul was used as a PCR template with primers listedin Table 3. The diluted DNAs showed product intensityinversely proportional to concentration, indicating thatthe PCRs at 25 cycles of amplification were in the linearPage 12 of 14(page number not for citation purposes)overnight at 37C in a total volume of 20 ul with 20 unitsof one of the following: mock enzyme (uncut), HpaII, orrange of amplification, and the results shown in figure 3are for the 12 ng/ul template.BMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48Chromatin Immunopreciptation Assay (ChIP Assay)The hybrid cells were grown in a t25 flask and were har-vested after treatment with 5 drops of 0.25% trypsin-EDTA for 2 minutes at room temperature, and thenwashed in 1 × PBS. The cells were placed in a 1.5 ml tubewith one ml of 0.37% formaldehyde in minimal essentialmedia (Invitrogen) and incubated at 37C for ten minutes.From this point, the cells were kept on ice. The cells werewashed twice with 1/100 proteinase inhibitor cocktail(Sigma) in 1 × PBS and then centrifuged at 2500 rpm for4 minutes at 4C. The pellet was resuspended in 200 ul SDSlysis buffer (Upstate) with 1/100 proteinase inhibitorcocktail and placed on ice for 10 minutes. The suspensionwas drawn up with a 25 gauge needle and then sonicated.Immunoprecipitation was performed (Upstate Biochemi-cals catalogue number 17-295) with the following anti-bodies: Anti-acetyl H3 against lysine 9 and 14 (cataloguenumber 06-599); Anti-acetyl H4 against lysine 5, 8, 12,and 16 (catalogue number 06-866); and Anti-dimethyl-H3 against lysine 9 (catalogue number 07-212).AbbreviationsH3 – histone H3; H4 – histone H4; K – lysine; ChIP –chromatin immunoprecipitation; X – X chromosome; Y –Y chromosome; Xi – inactive X; Xa – active X; M –methylated; U – unmethylated; + expressing; - notexpressing; S – single signal; D – double signal.Authors' contributionsCA carried out the molecular genetic studies and dataanalysis and drafted the manuscript. CB conceived of thestudy, participated in the experimental design and dataanalysis and wrote the final manuscript. Both authorsread and approved the final manuscript.AcknowledgementsThis study was supported by a Canadian Institutes of Health Research oper-ating grant (MOP13690).References1. Cheung VG, Conlin LK, Weber TM, Arcaro M, Jen KY, Morley M,Spielman RS: Natural variation in human gene expressionassessed in lymphoblastoid cells.  Nat Genet 2003, 33:422-425.2. Whitney AR, Diehn M, Popper SJ, Alizadeh AA, Boldrick JC, RelmanDA, Brown PO: Individuality and variation in gene expressionpatterns in human blood.  Proc Natl Acad Sci U S A 2003,100:1896-1901.3. Thomson MW, McInnes RR, Willard HF: Genetics in Medicine.5th edition. Philadelphia, W.B. Saunders Co.; 1991. 4. Lyon MF: Gene action in the X-chromosome of the mouse(Mus musculus L.).  Nature 1961, 190:372-373.5. Carrel L, Willard HF: X-inactivation profile reveals extensivevariability in X-linked gene expression in females.  Nature2005, 434:400-404.6. Brown CJ, Greally JM: A stain upon the silence: genes escapingX inactivation.  Trends Genet 2003, 19:432-438.7. Disteche CM, Filippova GN, Tsuchiya K: Escape from Xinactivation.  Cytogenet Genome Res 2002, 99:36-43.8. Craig IW, Harper E, Loat CS: The genetic basis for sex differ-ences in human behaviour: role of the sex chromosomes.Ann Hum Genet 2004, 68:269-284.9. Tsuchiya KD, Greally JM, Yi Y, Noel KP, Truong JP, Disteche CM:Table 3: Primers for PCR analysesPrimer pair Sequence UseTIMP1 5' (promoter) [14] 5'A: CCCTTGGGTTCTGCACTGA*5'B: CCAAGCTGAGTAGACAGGCMethylationChIPDNase sensitivityTIMP1 CA (gene body) [12] CA1: GGGTTCCAAGCCTTAGGGGACA2: AGGCTGTTCCAGGGAGCCGCDNase sensitivityTIMP1 S (bisulfite) 5S: GttttTTGGtTTtTGtAtTGATGGT3S: CCAAaCTaAaTAaACAaaCATCTAaC**Bisulfite sequencingARAF1 M1:M4 (promoter) [14] M1: TGCCAAAGCCCTAAGGTCAM4: CGCTGTCGACGATGGTCTM3: GTGAGGAAACAAGAAGAGAGMethylationChIPDNase sensitivityXIST 3':5' (gene body) [39] 3':GAAGTCTCAAGGCTTGAGTTAGAAG5': TTGGGTCCTCTATCCATCTAGGTAGMethylationDNase sensitivityXIST A5:29r (promoter) [37] A5: TTTCTTACTCTCTCGGGGCT29r: ATCAGCAGGTATCCGATACCChIPELK1 5' (promoter) [14] A: GCACAGCTCTGTAGGGAAB: AGCTCACCTGTGTGTAGCGMethylationChIPSTA A:B (intergenic) A: CACCTGTGTGTCATGTATACB: CCAGTATTGGTCTTCCAGTTDNase sensitivity8037 A:B (intergenic) A: GAGGCAAGACATCCATTCCB: TGACTTTGAGCGAGCAGGTReference region* There is a mismatch in the TIMP 5'A primer, the underlined G should be C.** The lower-case letters in the primers are the bases modified by the bisulfite reaction. All C nucleotides should have been converted because there were no CpG pairs with possible protective methylation.Page 13 of 14(page number not for citation purposes)Comparative sequence and X-inactivation analyses of aPublish with BioMed Central   and  every scientist can read your work free of charge"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime."Sir Paul Nurse, Cancer Research UKYour research papers will be:available free of charge to the entire biomedical communitypeer reviewed and published immediately upon acceptancecited in PubMed and archived on PubMed Central BMC Genetics 2005, 6:48 http://www.biomedcentral.com/1471-2156/6/48domain of escape in human Xp11.2 and the conserved seg-ment in mouse.  Genome Res 2004, 14:1275-1284.10. Lingenfelter PA, Adler DA, Poslinski D, Thomas S, Elliot RW, Chap-man VM, Disteche CM: Escape from X inactivation of Smcx ispreceded by silencing during mouse development.  Nat Genet1998, 18:212-213.11. Rougeulle C, Navarro P, Avner P: Promoter-restricted H3 Lys 4di-methylation is an epigenetic mark for monoallelicexpression.  Hum Mol Genet 2003, 12:3343-3348.12. Anderson C, Brown CJ: Polymorphic X-chromosome inactiva-tion of the human TIMP1 gene.  Am J Hum Genet 1999,65:699-708.13. Carrel L, Willard HF: Heterogeneous gene expression from theinactive X chromosome: an X-linked gene that escapes Xinactivation in some human cell lines but is inactivated inothers.  Proc Natl Acad Sci USA 1999, 96:7364-7369.14. Anderson CL, Brown CJ: Variability of X chromosome inactiva-tion: effect on levels of TIMP1 RNA and role of DNAmethylation.  Hum Genet 2002, 110:271-278.15. Plath K, Mlynarczyk-Evans S, Nusinow DA, Panning B: Xist RNA andthe mechanism of X chromosome inactivation.  Annu RevGenet 2002, 36:233-278.16. Willard HF, Latt SA: Analysis of deoxyribonucleic acid replica-tion in human X chromosomes by fluorescence microscopy.Am J Hum Genet 1976, 28:213-227.17. Azuara V, Brown KE, Williams RR, Webb N, Dillon N, Festenstein R,Buckle V, Merkenschlager M, Fisher AG: Heritable gene silencingin lymphocytes delays chromatid resolution without affect-ing the timing of DNA replication.  Nat Cell Biol 2003, 5:668-674.18. Chow JC, Brown CJ: Forming facultative heterochromatin:silencing of an X chromosome in mammalian females.  CellMol Life Sci 2003, 60:2586-2603.19. Chen C, Yang TP: Nucleosomes are translationally positionedon the active allele and rotationally positioned on the inac-tive allele of the HPRT promoter.  Mol Cell Biol 2001,21:7682-7695.20. Jeppesen P, Turner B: The inactive X chromosome in femalemammals is distinguished by a lack of histone H4 acetyla-tion, a cytogenetic marker for gene expression.  Cell 1993,74:281-289.21. Heard E, Rougelle C, Arnaud D, Avner P, Allis CD, Spector DL:Methylation of histone H3 at Lys-9 is an early mark on the Xchromosome during X inactivation.  Cell 2001, 107:727-738.22. Plath K, Fang J, Mlynarczyk-Evans SK, Cao R, Worringer KA, Wang H,de la Cruz CC, Otte AP, Panning B, Zhang Y: Role of histone H3lysine 27 methylation in X inactivation.  Science 2003,300:131-135.23. Gilbert SL, Sharp PA: Promoter-specific hypoacetylation of X-inactivated genes.  Proc Natl Acad Sci U S A 1999,96(24):13825-13830.24. Rougeulle C, Chaumeil J, Sarma K, Allis CD, Reinberg D, Avner P,Heard E: Differential histone H3 Lys-9 and Lys-27 methylationprofiles on the X chromosome.  Mol Cell Biol 2004, 24:5475-5484.25. Hors-Cayla MC, Heuertz S, Frezal J: Coreactivation of four inac-tive X genes in a hamster x human hybrid and persistence oflate replication of reactivated X chromosome.  Somat CellGenet 1983, 9:645-657.26. Carrel L, Cottle AA, Goglin KC, Willard HF: A first-generation X-inactivation profile of the human X chromosome.  Proc NatlAcad Sci U S A 1999, 96:14440-14444.27. Boggs BA, Chinault AC: Analysis of replication timing proper-ties of human X-chromosomal loci by fluorescence in situhybridization.  Proc Natl Acad Sci 1994, 91:6083-6087.28. Gribnau J, Luikenhuis S, Hochedlinger K, Monkhorst K, Jaenisch R: Xchromosome choice occurs independently of asynchronousreplication timing.  J Cell Biol 2005, 168:365-373.29. Kohlmaier A, Savarese F, Lachner M, Martens J, Jenuwein T, Wutz A:A chromosomal memory triggered by Xist regulates histonemethylation in X inactivation.  PLoS Biol 2004, 2:E171.30. de Napoles M, Mermoud JE, Wakao R, Tang YA, Endoh M, AppanahR, Nesterova TB, Silva J, Otte AP, Vidal M, Koseki H, Brockdorff N:Polycomb Group Proteins Ring1A/B Link Ubiquitylation ofHistone H2A to Heritable Gene Silencing and XInactivation.  Dev Cell 2004, 7:663-676.is involved in Initiation of X-inactivation.  J Biol Chem 2004,279:52812-5.32. Chadwick BP, Willard HF: Multiple spatially distinct types of fac-ultative heterochromatin on the human inactive Xchromosome.  Proc Natl Acad Sci U S A 2004, 101:17450-17455.33. Filippova GN, Cheng MK, Moore JM, Truong JP, Hu YJ, Nguyen DK,Tsuchiya KD, Disteche CM: Boundaries between chromosomaldomains of X inactivation and escape bind CTCF and lackCpG methylation during early development.  Dev Cell 2005,8:31-42.34. Hansen RS, Gartler SM: 5-Azacytidine-induced reactivation ofthe human X chromosome-linked PGK1 gene is associatedwith a large region of cytosine demethylation in the 5' CpGisland.  Proc Natl Acad Sci U S A 1990, 87:4174-4178.35. Lahn BT, Page DC: Four evolutionary strata on the human Xchromosome.  Science 1999, 286:964-967.36. Wilcox SA, Watson JM, Spencer JA, Graves JA: Comparative map-ping identifies the fusion point of an ancient mammalian X-autosomal rearrangement.  Genomics 1996, 35:66-70.37. Tinker AV, Brown CJ: Induction of XIST expression from thehuman active X chromosome in mouse/human somatic cellhybrids by DNA demethylation.  Nucl Acids Res 1998,26:2935-2940.38. Gregory RI, Feil R: Analysis of chromatin in limited numbers ofcells: a PCR-SSCP based assay of allele-specific nucleasesensitivity.  Nucleic Acids Res 1999, 27:e32.39. Brown CJ, Ballabio A, Rupert JL, Lafreniere RG, Grompe M, Ton-lorenzi R, Willard HF: A gene from the region of the human Xinactivation centre is expressed exclusively from the inactiveX chromosome.  Nature 1991, 349:38-44.yours — you keep the copyrightSubmit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.aspBioMedcentralPage 14 of 14(page number not for citation purposes)31. Fang J, Chen T, Chadwick B, Li E, Zhang Y: Ring1b-mediated H2Aubiquitination associates with inactive X chromosomes and


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items