Immunodeficiency, Centromeric Instability, Facial Anomalies (ICF) syndrome is a rare autosomal recessive disorder that is characterized by a marked immunodeficiency, severe hypomethylation of the classical satellites 2 and 3 associated with disruption of constitutive heterochromatin, and facial anomalies. Sixty percent of ICF patients have mutations in the DNMT3B (DNA methyltransferase 3B) gene, encoding a de novo DNA methyltransferase.
In the present study, we have shown that, in ICF lymphoblasts and peripheral blood, juxtacentromeric heterochromatic genes undergo dramatic changes in DNA methylation, indicating that they are bona fide targets of the DNMT3B protein. DNA methylation in heterochromatic genes dropped from about 80% in normal cells to approximately 30% in ICF cells. Hypomethylation was observed in five ICF patients and was associated with activation of these silent genes. Although DNA hypomethylation occurred in all the analyzed heterochromatic genes and in all the ICF patients, gene expression was restricted to some genes, every patient having his own group of activated genes. Histone modifications were preserved in ICF patients. Heterochromatic genes were associated with histone modifications that are typical of inactive chromatin: they had low acetylation on H3 and H4 histones and were slightly enriched in H3K9Me3, both in ICF and controls. This was also the case for those heterochromatic genes that escaped silencing. This finding suggests that gene activation was not generalized to all the cells, but rather was restricted to a clonal cell population that may contribute to the phenotypic variability observed in ICF syndrome. A slight increase in H3K27 monomethylation was observed both in heterochromatin and active euchromatin in ICF patients; however, no correlation between this modification and activation of heterochromatic genes was found.
Citation: Brun M-E, Lana E, Rivals I, Lefranc G, Sarda P, Claustres M, et al. (2011) Heterochromatic Genes Undergo Epigenetic Changes and Escape Silencing in Immunodeficiency, Centromeric Instability, Facial Anomalies (ICF) Syndrome. PLoS ONE 6(4): e19464. https://doi.org/10.1371/journal.pone.0019464
Editor: Brian P. Chadwick, Florida State University, United States of America
Received: February 1, 2011; Accepted: March 30, 2011; Published: April 29, 2011
Copyright: © 2011 Brun et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the Fondation Jérôme Lejeune (http://subvention.fondationlejeune.org/subvention/index.aspx), by the Association pour la Recherche sur le Cancer (http://www.arc-cancer.net/) and by grants from the Saint Joseph University. E.L. was supported by the Fondation Jérôme Lejeune and the Fondation pour la Recherche Médicale (http://www.frm.org/chercheurs.htm). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
ICF (Immunodeficiency, Centromeric Instability, and Facial Anomalies; OMIM #242860), is a rare autosomal recessive disorder. Up to now, less than fifty cases have been reported. Most of the ICF patients were born from consanguineous marriages and about 60% had mutations in the DNMT3B (DNA methyltransferase 3B) gene that maps to the 20q11.2 region –. The DNMT3B enzyme catalytically methylates cytosines in CpG pairs, which are later bound by repressor complexes.
Several mutations have been described in ICF syndrome, most of them (79%) lying in the gene portion encoding the C-terminal catalytic domain. These mutations induce a decrease of the enzymatic activity. The less frequent N-terminal mutations are mostly nonsense mutations that arise as compound heterozygous mutations. Forty percent of clinically diagnosed ICF patients do not carry any mutation of this gene, suggesting mutational heterogeneity or involvement of other genes. Consequently, it has been postulated that there are two types of ICF patients: type 1 patients bearing mutations in the DNMT3B gene and type 2 patients without known mutations . In murine models, mutations in the Dnmt3b gene negatively affect proper embryonic development , .
ICF syndrome is characterized by a marked immunodeficiency: patients tend to have low levels of immunoglobulins and may exhibit low levels of B and T cells. The B cell defects associated with agammaglobulinemia or hypogammaglobulinemia in ICF type 1 are characterized by only naïve and no memory B cells in peripheral blood (PB) . Negative selection is impaired and newly generated and immature B cells accumulate in PB due to B cell maturation blockage. Because of chronic respiratory and gastrointestinal infections, many patients die at an early age. Facial anomalies are a heterogeneous trait in ICF syndrome and mainly include hypertelorism, epicanthal folds, abnormally low-set ears, and macroglossia. Stature is often reduced. Centromeric instability is the most typical feature of the disease. The juxtacentromeric heterochromatin of chromosome 1, 9, and 16 is markedly undercondensed and is involved in chromosome rearrangements and multiradiate associations. The instability correlates with a severe hypomethylation of the classical satellites 2 and 3, which are the major components of constitutive heterochromatin. ICF syndrome was the first genetic disease to be associated with a constitutional methylation defect, mainly affecting heterochromatin. The methylation anomalies of ICF syndrome can involve other genomic sequences such as α satellites, the centromeric component of constitutive heterochromatin , Alu sequences , D4Z4 and NBL2 repeats , and imprinted genes , .
Since it was rather unlikely that hypomethylation of non-coding repetitive sequences accounted for the severe clinical features that characterize ICF patients, several attempts were made in the past to search for genes that lost DNA methylation and, subsequently, became transcriptionally active in these patients. Hansen et al.  analyzed eleven genes that are normally subject to X inactivation, including SYBL1, that is also present in chromosome Y. In ICF patients, the inactive alleles of these genes were hypomethylated and some of them escaped silencing. Escape from silencing in ICF cells was associated with a marked advance in replication timing from a late to an active X-like pattern . More recently, ICF-specific changes in RNA levels of genes critical for immune function, development, and neurogenesis were identified through the analysis of global expression profiles –. These studies showed that changes in gene expression occurred across the whole genome and in both directions (under- and overexpression) in ICF compared to control lymphoblast cells. In two of these studies, DNA methylation did not change in differently expressed genes , ; the third study identified subtle but significant changes .
In previous reports, we showed that in tumor cells and cancer cell lines juxtacentromeric genes became hypomethylated and escaped silencing , . Hypomethylation was very striking (DNA methylation dropped from 80% to less than 40%) and also frequent. In this work, we investigated whether heterochromatic genes escape silencing in ICF cells too.
Table 1 lists patients and controls analyzed in this study. Four patients (ICF1, ICF2, ICF4, and ICF5) had ICF-specific clinical and karyotypic features. Among them two patients had previously described mutations in the DNMT3B gene (type 1 patients) and two siblings had no mutations in the DNMT3B gene (type 2 patients). Patient ICF3 presented the cytogenetic rearrangements that are commonly associated with ICF syndrome (chromosome breakages and multiradial configurations); however, he did not suffer from immunodeficiency and had no facial anomalies. Genomic sequencing showed this patient had a missense mutation (K770E) in the DNMT3B gene that had never been described before. He was homozygous for this mutation, his parents, who were first cousins, being heterozygous. The missense mutation resulted in an aminoacid change in the catalytic domain. This amino acid is phylogenetically conserved in various animal species.
Heterochromatic genes were hypomethylated in ICF cells
A large majority of juxtacentromeric regions have been left out from the genome sequencing project as segmental duplications made clone assembly and sequencing troublesome in these chromosome regions. Thus, to determine whether juxtacentromeric heterochromatic genes were hypomethylated in ICF patients, we chose chromosome 21 for which well annotated sequences bridging the chromosome arms with centromeric and juxtacentromeric satellites are available –. A physical map of the 2.3-Mb region that was analyzed in this work is provided in Fig. 1a. Using the CpG island searcher program , we scanned the genomic sequence of human chromosome 21  and searched for CpG islands that were associated with the 5’ region of the genes. We analyzed DNA methylation in seven genes mapping to the juxtacentromeric region of human chromosome 21. Three genes (TPTE, BAGE2, and POTED) mapped to the duplication-rich heterochromatic domain and four genes (RBM11, ABCC13, STCH, and NRIP1) mapped to the duplication-free euchromatic domain , . Two euchromatic genes LIPI and SAMSN1 were excluded from the methylation analysis because the former gene shared a CpG island with RBM11 (the two genes are head-to-head with the CpG island in the middle) and the latter gene had no associated CpG island. For each gene, we chose a set of nested primers allowing us to amplify a 500-bp DNA stretch in the 5’ CpG island and we analyzed DNA methylation by bisulphite and genomic sequencing in ICF patients and in controls. An example of methylation patterns is shown in Fig. 1d. Since heterochromatic genes belonged to families comprising several loci that have >90% nucleotide sequence identity , we selected primers that specifically amplified the gene copy located in chromosome 21. Primer specificity was checked using DNA from a monochromosome hybrid panel. Heterochromatic genes were hypermethylated in controls and hypomethylated in ICF patients (Fig. 1 and Table 2). Hypomethylation was striking (DNA methylation dropped from about 80% in normal cells to approximately 30% in ICF cells) and frequent (present in five analyzed patients). Interestingly, BAGE2 was hypomethylated only in type 2 patients (χ2 test, p-value <0.001). DNA methylation of BAGE genes was also analyzed by COBRA as previously described . In this latter case, we measured the global methylation of twelve BAGE loci mapped to the juxtacentromeric regions of chromosomes 9, 13, 18, and 21 , . BAGE loci were hypomethylated in all the ICF patients, regardless of the type, and were hypermethylated in controls. This finding shows that loss of DNA methylation is not restricted to the juxtacentromeric region of chromosome 21, but is common to other chromosomes.
a) Physical map of the 2.3-Mb juxtacentromeric region of human chromosome 21 analyzed in this study. The centromere (cen) is not shown to scale. Genes are bold typed; truncated genes are indicated by plain text and retrotransposed pseudogenes are italicized. b) Histograms represent the percentages of DNA methylation in the analyzed genes either in a pool of controls (black bars) or in a pool of ICF patients (white bars). c) DNA methylation in satellite 2. d) DNA methylation pattern of TPTE, a heterochromatic gene, in a control (left) and an ICF patient (right): full circles are 5mCpG and hollow circles are unmethylated CpG.
Genes located in the euchromatic duplication-free domain were hypomethylated both in ICF patients and controls. One exception was represented by the ABCC13 gene, which was slightly methylated in controls (35.2% 5mCpG) and significantly hypomethylated in ICF patients (17.5% 5mCpG) (χ2test, p-value <0.001).
To achieve our analysis, we analyzed DNA methylation of satellite 2 using the bisulphite and genomic sequencing method. Satellite 2 was much less methylated in ICF patients than in controls (χ2test, p-value <0.001) (Fig. 1c).
In patient ICF3, satellite 2 was as methylated as in controls; in contrast, two heterochromatic genes (TPTE and POTED) and the normally methylated euchromatic gene ABCC13 had a significant, but less striking loss of DNA methylation compared to the other patients (Table 2).
Heterochromatic genes escaped silencing in ICF cells
Afterwards, we decided to investigate whether DNA hypomethylation was associated with the activation of heterochromatic genes in ICF patients. Gene expression was first analyzed by reverse transcription and standard PCR on total RNA in five ICF patients and eleven controls. Either TPTE, or POTE, or both genes were expressed in four ICF patients, but were silent in controls. No heterochromatic genes were expressed by patient ICF3. The BAGE genes were expressed neither in ICF patients nor in controls. The expression levels of TPTE and POTE, which were specifically activated in ICF cells, was measured by real-time PCR (Fig. 2). The level of expression of TPTE was correlated with the degree of hypomethylation (Fig. 2a). Primers used for the expression analysis of heterochromatic genes were not specific for the gene copies located on chromosome 21. Thus, to determine which POTE and which TPTE were expressed by the ICF cells, we cloned and sequenced the RT-PCR products. The ICF cells expressed POTEF from chromosome 2, POTEG from chromosome 14, but they did not express POTED from chromosome 21. ICF patients also expressed TPTE from chromosome 21 and TPTE2 from chromosome 13. Overall, every patient had his own group of activated genes and activation of heterochromatic genes was not restricted to chromosome 21, but occurred in several chromosomes (Table 3). Genes located in the euchromatic duplication-free domain had the same expression pattern in ICF patients as in controls: SAMSN-1 and STCH were strongly expressed, NRIP1 was weakly expressed, and LIPI, RBM11, and ABCC13 were silent.
Black bars represent gene expression of two heterochromatic gene families that were expressed in ICF patients: TPTE (a) and POTE (b). The ABL gene was used to normalize cDNA amounts. I, ICF patients; N, controls. Type 1, 2 are ICF patients with and without known mutations in the DNMT3B gene, respectively. For TPTE (a) we superposed the percentage of DNA methylation (white bars).
Activation of heterochromatic genes was not associated with chromatin changes in ICF cells
Histone modifications were analyzed in three ICF (ICF3, ICF4, and ICF5) and three control (N9, N10, and N11) lymphoblast cell lines. Two patients (ICF1 and ICF2) could not be analyzed, because their cell lines were lost during the course of this project. ChIP (Chromatin Immunoprecipitation) experiments were done using antibodies against four histone isoforms (H3K9Me3, H3K27Me, H3K9Ac, and H4Ac) (Fig. 3).
Symbols represent single experiments either in an ICF (diamond) or in a control (circle) cell line. Filled markers denote expressed genes. The significance of the differences between ICF patients and controls was evaluated for each modification and for each gene using Wilcoxon’s rank sum test (p).
Active and silent heterochromatic genes had similar histone modifications. We analyzed TPTE and TPTE2, but not POTE genes. In this gene family, given the presence of thirteen copies dispersed on eight chromosomes , we failed to design chromosome-specific primers for real-time PCR.
Then, we compared histone modifications in ICF patients versus controls in the whole set of juxtacentromeric genes. Histone acetylation and H3K9 trimethylation were similar in ICF patients and in controls; in contrast, monomethylation of H3K27 in three genes (TPTE, TPTE2 and STCH) was significantly higher in ICF patients than in controls.
Finally, we analyzed histone modifications in the three chromatin domains that are present in the juxtacentromeric region of human chromosome 21: heterochromatin, active euchromatin and silent euchromatin. The heterochromatic genes TPTE and TPTE2 were characterized by very low levels of acetylation both in H4 and in H3K9 histones and had monomethylation of H3K27 and trimethylation of H3K9. In the euchromatic domain, histone modifications correlated with gene expression. STCH and SAMSN-1, which were highly expressed, were highly acetylated and had low levels of H3K9Me3. STCH was also highly enriched in H3K27Me. ABCC13, RBM11 and NRIP1, three euchromatic genes that were either silent or very weakly expressed, were characterized by very low levels of acetylation and were slightly enriched in H3K9Me3. In silent euchromatic genes, H3K27Me was very low.
Juxtacentromeric heterochromatic genes lose DNA methylation and escape silencing in ICF cells, without apparent chromatin changes
In the present study, we have shown that juxtacentromeric heterochromatic genes lost DNA methylation and escaped silencing in ICF cells. Such a finding was not emphasized by previous cytogenetic analyses of constitutive heterochromatin that detected hypomethylation and chromatin decondensation only in the satellite-rich domains. We analyzed chromosome 21 for which genomic sequences bridging the chromosome arms with centromeric and juxtacentromeric satellites are available –. However, having looked at paralogous genes scattered in various chromosomes, we showed that loss of methylation in juxtacentromeric heterochromatic genes occurred in many chromosomes. Hypomethylation was observed both in type 1 and type 2 patients. Only BAGE2 was specifically hypomethylated in type 2 ICF patients. A similar finding was reported for α satellite sequences whose hypomethylation was also restricted to type 2 patients . Alpha satellite sequences are located within centromeres and BAGE2 maps very close to the beginning of α satellite sequences . Centromeric chromatin is structurally and functionally distinct from the flanking juxtacentromeric heterochromatin: it contains blocks of histone H3 nucleosomes interspersed with blocks of CENP-A (an H3 variant) nucleosomes . Centromeric chromatin is also characterized by a pattern of histone modifications that is different from that of the flanking heterochromatin . So, it is tempting to suggest that loss of DNA methylation in ICF patients of type 2 might result from impairment of one of the proteins responsible for the centromeric epigenetic pattern. Interestingly, CENP-C, one of the proteins contributing to centromeric chromatin organization, interacts with DNMT3B and modulates DNA methylation both in centromeric and juxtacentromeric regions .
Euchromatic genes were unmethylated both in ICF patients and controls, with the exception of ABCC13. In normal cells, this gene was slightly methylated, but less than the heterochromatic genes. The presence of methylation in ABCC13 is consistent with the notion that it is undergoing a pseudogenization process . ABCC13 was significantly hypomethylated in ICF cells, showing that hypomethylation occurs also in the euchromatic domain that flanks the constitutive heterochromatin, but is much less striking.
In four ICF patients, heterochromatic genes escaped silencing. Transcription was detected in various paralogs, suggesting that, similar to DNA hypomethylation, gene activation was not restricted to chromosome 21, but occurred in several chromosomes.
Although DNA hypomethylation was generalized to all the heterochromatic genes and to all ICF patients, gene expression was restricted to some genes and every patient had his own group of activated genes. The activated genes differed even in two brothers (ICF1 and ICF2). This finding is consistent with what we previously observed in cancer, where heterochromatic genes became hypomethylated in almost all the tumors, but were expressed only in some of them and every tumor had a specific set of activated genes . The precise mechanism driving transcription of heterochromatic genes in ICF cells is unknown. Clearly, impairment of DNMT3B results in hypomethylation that, however, is not sufficient to allow transcription. Other epigenetic changes might occur, i.e. at the chromatin level. Thus, to better understand gene regulation in human heterochromatic regions, we next analyzed histone modifications by ChIP. Genes located in the euchromatic flanking domain had either active or repressive histone modifications, depending on their expression status. Heterochromatic genes, on the contrary, were associated with histone modifications that were typical of inactive chromatin: they were hypoacetylated and slightly enriched in H3K9Me3 and H3K27Me. This was also the case for expressed heterochromatic genes. Since it is unlikely that hypoacetylated genes are transcribed, we suggest that in ICF cells, gene activation is not generalized to all the cells, but rather is restricted to a small percentage of them. Clonal changes in chromatin may result from epigenetic modifications occurring during lymphocyte differentiation. If few cells express heterochromatic genes, transcripts can be detected by RT-PCR; on the other hand, if chromatin changes occur stochastically in a small percentage of cells, ChIP analysis on the whole cell population will not detect them. DNA hypomethylation, in contrast, occurred in all the cells, as the bisulphite and sequencing patterns of different colonies from a same tissue were rather homogeneous (Fig. 1d). It is likely that similar events take place in the inactive X chromosome of ICF cells, where genes subject to inactivation are hypomethylated, escape silencing, but exhibit normal histone modification patterns . Finally, in ICF cells, hypomethylation of subtelomeric regions was associated with aberrant transcription of these sequences, but no information on the chromatin structure was provided . Previous genome-wide transcriptomic analyses identified more than 700 differently expressed genes in ICF patients; however, none of them identified aberrant expression of these heterochromatic genes , . One possible explanation is that these previous studies focused on genes expected to be relevant for ICF phenotype, whereas we analyzed specifically heterochromatic genes. In addition, Jin et al.  selected genes that had a high expression variability between ICF and controls, but not within these two groups of individuals. Yet, the aberrant expression of heterochromatic genes is highly variable within ICF patients. Ehrlich et al.  did not observe significant changes in differently expressed genes, whereas Jin et al.  reported significant, but subtle DNA methylation differences in approximately half the analyzed upregulated genes. Our work shows that heterochromatic genes undergo dramatic changes in DNA methylation, indicating that they are bona fide targets of the DNMT3B protein.
H3K27 monomethylation is increased in ICF lymphoblast cell lines
Much less characterized than trimethylation, monomethylation of H3K27 was first described as a repressive mark that is preferentially localized in heterochromatin, but is also found in euchromatic regions . Recent genome-wide analyses have shown that this mark is frequently associated with the promoter of active genes . Our results are consistent with both statements, as we found H3K27 monomethylation in heterochromatin and in active euchromatin, but not in silent euchromatin. Interestingly, H3K27 monomethylation was higher in ICF lymphoblasts than in controls. More patients should be analyzed to corroborate these data. Alteration of this histone mark may be relevant to ICF syndrome, as H3K27 methylation plays an important role during the development of the B lymphoid lineage  and ICF patients have immature B cells in the blood .
Aberrant transcription of heterochromatic genes and ICF syndrome
A common feature of heterochromatic genes is their expression pattern mostly restricted to germ cells. TPTE and TPTE2 genes are strongly expressed in testis, in secondary spermatocytes and/or very early spermatids . POTE genes are expressed in a few normal tissues (namely testis, prostate, ovary and placenta), in embryonic stem cell lines, and in various cancers , . Immunochemistry showed that POTE proteins localized in primary spermatocytes in the seminiferous tubules . The high expression of POTE in spermatocytes suggests a role in spermatogenesis. Finally, BAGE genes are exclusively expressed in cancer and testis, encode small putative proteins that have no homology to known domains, and have no orthologs in other mammalian species , . Very interestingly, in hypomorphic Dnm3b mutant mice, genes that are normally expressed in testis are aberrantly expressed in somatic embryonic tissues . It is therefore tempting to suggest that the aberrant transcription in somatic tissues of genes that are normally expressed in germ cells may contribute to the phenotypic variability observed in ICF syndrome.
Incomplete phenotype of an ICF patient
One of the analyzed patients showed an incomplete ICF phenotype (he had neither immunodeficiency, nor the facial anomalies that are typical of ICF patients) and suffered from severe obesity. His karyotype contained multibranched and rearranged chromosomes and, at the molecular level, he was homozygous for a missense mutation resulting in an aminoacid change in the region encoding the catalytic domain of the DNMT3B protein. Loss of DNA methylation was less pronounced than in the other patients and was never associated with activation of expression. It is likely that this previously unreported missense mutation slightly affected the methylation activity of the DNMT3B protein, resulting in genome instability and a less pronounced DNA hypomethylation, without producing the classical ICF phenotype. Other ICF patients having a similar incomplete phenotype may be under-diagnosed.
To conclude, the striking loss of DNA methylation in heterochromatic genes, together with their variable pattern of aberrant expression and the clinical features of the patients involving numerous systems, make ICF syndrome an ideal model to investigate DNA methylation changes and their molecular and pathological consequences.
Materials and Methods
Patients and controls
Patients and controls used in this study are listed in Table 1. This study was performed on DNA and RNA extracted from human lymphoblast cell lines and peripheral blood in the context of routine molecular diagnostic service. Prof. André Megarbané received approval from the Institutional Review Board of the Saint Joseph University. Written informed consent to genetic testing was obtained from adult probands or parents in the case of minors after explanation of the nature and its possible implications to the patient and his family.
Three lymphoblast cell lines (ICF5, N10 and N11) were purchased at the Coriell Cell Repositories. The lymphoblast cell line ICF4 was established at the Genethon laboratory (Evry, France). Lymphoblast cell lines ICF1, ICF2, ICF3 and N9 were kindly provided by Dr. Marie-Genevieve Mattei (INSERM UMR 910, Marseille, France).
DNA and RNA extraction
DNA and RNA were extracted from lymphoblast cell lines and from peripheral blood. DNA was treated with 100 µg/ml proteinase K at 55°C overnight, then extracted with phenol/chloroform, ethanol precipitated and treated with 20 µg/ml RNAse. Total RNA was extracted with Trizol (GIBCO-BRL) according to the manufacturer recommendations.
The chromosome 21 genomic  sequence was analyzed with the “CpG island searcher” software  to identify gene-associated CpG islands. The following criteria were used: CG% >55% and Observed CpG/Expected CpG >0.65.
DNA methylation analysis
DNA methylation was analyzed by bisulphite genomic sequencing as described by Frommer et al.  and by Grunau et al.  with the following modifications. Three hundred ng of genomic DNA (extracted from peripheral blood when available, otherwise from lymphoblast cell lines) were treated with bisulphite and then purified using the Microcon YM-100 filter device (Millipore) to separate DNA from bisulphite. PCR amplifications were done in two consecutive reactions with a set of nested primers. The PCR program was: 94°C for 2 min; (94°C 1 min, annealing temperature 2 min, 72°C 3 min) for 5 cycles; (94°C 30 sec, annealing temperature 2 min, 72°C 1 min 30 sec) for 24 cycles; 72°C 10 min. Specific primers and annealing temperatures for each gene are listed in Table S1. The PCR products were purified on a low melting agarose gel using the QIAquick gel extraction kit (Qiagen), cloned into a pGEM-T-Easy (Promega) vector and transformed into E. Coli JM109 (Promega) cells. Ten colonies per gene per individual were sequenced with T7 and/or SP6 universal primers using a DYEnamic ET Terminator Cycle Sequencing kit (Amersham). The obtained sequences were aligned with the genomic sequence using the GAP4 computer package . Methylation patterns were analyzed with the MethTools software . DNA methylation of BAGE genes was also analyzed by COBRA as previously described . Briefly, using bisulphate treated genomic DNA, we amplified simultaneously 12 BAGE loci in a single PCR. The obtained product was digested with MboI and with HphI in two independent assays and methylation patterns were analyzed as previously described .
Standard PCR-Two µg of total RNA (extracted from peripheral blood when available, otherwise from lymphoblast cell lines) were reverse transcribed using the Omniscript kit (Qiagen) according to the manufacturer protocol. Reversed DNA was resuspended in 50 µl of 10 mM Tris. One µl of cDNAs was used for PCR and 10 µl of the reaction were analyzed in ethidium bromide-stained agarose gel. The PCR program was: 94°C 2 min; (94°C 30 sec, 68°C 2 min, 72°C 2 min) for 5 cycles; (94°C 30 sec, annealing temperature 30 sec; 72°C 2 min) for 25 cycles. Specific primers used for expression analysis are listed in Table S2. The housekeeping GAPDH gene was used as control.
TPTE and POTE RT-PCR products were purified on a low melting agarose gel using the QIAquick gel extraction kit (Qiagen), cloned into a pGEM-T-Easy (Promega) vector, and transformed into E. Coli JM109 (Promega) cells. Twelve colonies were sequenced with the T7 universal primer using a DYEnamic ET Terminator Cycle Sequencing kit (Amersham).
Real-Time PCR- Two µg of total RNA were reverse transcribed using the QuantiTect Reverse Transcription kit (#205311 Qiagen) according to the manufacturer protocol. Reversed DNA was purified using the Qiaquick (#28704 Qiagen). Expression was quantified using a MyIQ PCR Biorad thermal cycler and the QPCR Rox-&Go Green kit (#EPQON480 MP Biomedicals). PCR reactions were done in 25-µl final volume containing 400 nM of each primer and 1 µl of each template DNA (equivalent to 33 ng of RNA). PCR conditions were the following: 95°C for 15 min, followed by 50 cycles of (95°C 30 s, 60°C 30 s and 72°C 1 min). We used the housekeeping ABL gene to normalize cDNA amounts. Primers are listed in Table S2. Each analysis was done in triplicate and assay efficiencies were calculated using cDNA dilution series. To calculate gene expression we used the relative standard curve method.
ChIP experiments were done in 2–3 biological replicates as previously described . Antibodies were purchased from Upstate (Millipore): anti-trimethylated H3K9 (#07-442), anti-monomethylated H3K27 (#07-448), anti-hyperacetylated histone H4 (penta) (#06-946), and anti-acetylated H3K9 (#07-352).
Real-Time PCR was done in a MyIQ PCR thermal cycler (Biorad) using the QPCR Rox-&Go Green kit (#EPQON480 MP Biomedicals). To amplify genomic DNA we designed primers located in the 3’ end of the CpG islands (Table S2). The euchromatic GAPDH locus and a heterochromatic STS (JB 10) served as references to normalize the amount of precipitated DNA (their primers are given in Table S2). For a detailed description of normalization see . Briefly, ratios of precipitated DNA in each gene to that in the reference sequence (either GAPDH or STS JB10) were calculated as follows: enrichment factor = [ng GENE(B)/ng GAPDH(B)]/[ng GENE(I)/ng GAPDH(I)] with (B) for antibody bound and (I) for input. The unbound fraction of mock-treated chromatin (i.e. precipitation in parallel with the other ChIP but without antibody) was considered as input. DNA immunoprecipitated with anti-trimethylated H3K9 and anti-monomethylated H3K27 was normalized with STS (JB 10); DNA immunoprecipitated with anti-hyperacetylated histone H4 (penta) and anti-acetylated H3K9 was normalized with GAPDH.
The analysis of the methylation was performed for each gene using χ2 tests at three different levels. First, we checked the homogeneity of the methylation pattern between the cells of each individual on the 2 (5mCpG/CpG) x number of cells contingency table. Aside from a few exceptions, the patients had a homogeneous methylation pattern. Thus, the data corresponding to each individual were pooled, leading to the methylation frequencies for each individual gathered in Table 2. Then, we tested the homogeneity of the ICF patients and that of the healthy controls. The ICF group was always very heterogeneous, whereas the healthy controls showed much more similarity. We nevertheless pooled the data of all ICF patients on one side, and that of all controls on the other side, in order to establish the significance of the methylation difference between ICF and controls, using a χ2 test on the corresponding 2×2 contingency table (one degree of freedom). The p-values corresponding to this final test are given in Table 2, as well as the overall methylation frequencies for each group (control and ICF). The significance of the difference between chromatin changes in ICF versus controls was tested separately for each gene and for each histone modification using Wilcoxon’s non-parametric rank sum test.
Primers used for DNA methylation analysis.
The authors thank Marie-Genevieve Mattei (INSERM UMR 910, Marseille, France) for having provided lymphoblast cell lines and for critical reading of the manuscript.
Conceived and designed the experiments: AD. Performed the experiments: MEB EL AD. Analyzed the data: IR AD. Contributed reagents/materials/analysis tools: AM PS GL MC. Wrote the paper: IR GL AD.
- 1. Hansen RS, Wijmenga C, Luo P, Stanek AM, Canfield TK, et al. (1999) The DNMT3B DNA methyltransferase gene is mutated in the ICF immunodeficiency syndrome. Proc Natl Acad Sci U S A 96: 14412–14417.
- 2. Okano M, Bell DW, Haber DA, Li E (1999) DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99: 247–257.
- 3. Xu GL, Bestor TH, Bourc'his D, Hsieh CL, Tommerup N, et al. (1999) Chromosome instability and immunodeficiency syndrome caused by mutations in a DNA methyltransferase gene. Nature 402: 187–191.
- 4. Jiang YL, Rigolet M, Bourc'his D, Nigon F, Bokesoy I, et al. (2005) DNMT3B mutations and DNA methylation defect define two types of ICF syndrome. Hum Mutat 25(1): 56–63.
- 5. Ueda Y, Okano M, Williams C, Chen T, Georgopoulos K, Li E (2006) Roles for Dnmt3b in mammalian development: a mouse model for the ICF syndrome. Development 133: 1183–1192.
- 6. Blanco-Betancourt CE, Moncla A, Milili M, Jiang YL, Viegas-Pequignot EM, et al. (2004) Defective B-cell-negative selection and terminal differentiation in the ICF syndrome. Blood 103: 2683–2690.
- 7. Miniou P, Jeanpierre M, Bourc'his D, Coutinho Barbosa AC, Blanquet V, et al. (1997) alpha-satellite DNA methylation in normal individuals and in ICF patients: heterogeneous methylation of constitutive heterochromatin in adult and fetal tissues. Hum Genet 99: 738–745.
- 8. Miniou P, Bourc'his D, Molina Gomes D, Jeanpierre M, Viegas-Péquignot E (1997) Undermethylation of Alu sequences in ICF syndrome: molecular and in situ analysis. Cytogenet Cell Genet 77: 308–313.
- 9. Kondo T, Bobek MP, Kuick R, Lamb B, Zhu X, et al. (2000) Whole-genome methylation scan in ICF syndrome: hypomethylation of non-satellite DNA repeats D4Z4 and NBL2. Hum Mol Genet 9: 597–604.
- 10. Hansen RS, Stöger R, Wijmenga C, Stanek AM, Canfield TK, et al. (2000) Escape from gene silencing in ICF syndrome: evidence for advanced replication time as a major determinant. Hum Mol Genet 9: 2575–2587.
- 11. Ehrlich M, Buchanan KL, Tsien F, Jiang G, Sun B, et al. (2001) DNA methyltransferase 3B mutations linked to the ICF syndrome cause dysregulation of lymphogenesis genes. Hum Mol Genet 10: 2917–2931.
- 12. Jin B, Tao Q, Peng J, Soo HM, Wu W, et al. (2008) DNA methyltransferase 3B (DNMT3B) mutations in ICF syndrome lead to altered epigenetic modifications and aberrant expression of genes regulating development, neurogenesis and immune function. Hum Mol Genet 17: 690–709.
- 13. Ehrlich M, Sanchez C, Shao C, Nishiyama R, Kehrl J, et al. (2008) ICF, an immunodeficiency syndrome: DNA methyltransferase 3B involvement, chromosome anomalies, and gene dysregulation. Autoimmunity 41: 253–271. Review.
- 14. Grunau C, Sanchez C, Ehrlich M, van der Bruggen P, Hindermann W, et al. (2005) Frequent DNA Hypomethylation in the Human Juxtacentromeric BAGE loci. Genes Chromosomes and Cancer 43: 11–24.
- 15. Grunau C, Buard J, Brun ME, De Sario A (2006) Mapping of the juxtacentromeric heterochromatin-euchromatin frontier of human chromosome 21. Genome Research 16: 1198–1207.
- 16. Brun ME, Ruault M, Ventura M, Roizès G, De Sario A (2003) Juxtacentromeric region of human chromosome 21: a boundary between centromeric heterochromatin and euchromatic chromosome arms. Gene 312: 41–50.
- 17. Hattori M, Fujiyama A, Taylor TD, Watanabe H, Yada T, et al. (2000) The DNA sequence of human chromosome 21. Nature 405: 311–319.
- 18. Takai D, Jones PA (2002) Comprehensive analysis of CpG islands in human chromosomes 21 and 22. Proc Natl Acad Sci U S A 99: 3740–3745.
- 19. Ruault M, van der Bruggen P, Brun M-E, Boyle S, Roizès G, et al. (2002) New BAGE (B melanoma antigen) genes mapping to the juxtacentromeric regions of human chromosomes 13 and 21 have a cancer/testis expression profile. Eur J Hum Genet 10: 833–840.
- 20. Ruault M, Ventura M, Galtier N, Brun M-E, Archidiacono N, et al. (2003) BAGE, a gene family generated by juxtacentromeric reshuffling in the Hominidae lineage, is under selective pressure. Genomics 81: 391–399.
- 21. Walker SM, Downes CP, Leslie NR (2001) TPIP: a novel phosphoinositide 3-phosphatase. Biochem J 360(Pt 2): 277–283.
- 22. Hahn Y, Bera TK, Pastan IH, Lee B (2006) Duplication and extensive remodeling shaped POTE family genes encoding proteins containing ankyrin repeat and coiled coil domains. Gene 366: 238–245.
- 23. Sullivan BA, Karpen GH (2004) Centromeric chromatin exhibits a histone modification pattern that is distinct from both euchromatin and heterochromatin. Nat Struct Mol Biol 11(11): 1076–83.
- 24. Gopalakrishnan S, Sullivan BA, Trazzi S, Della Valle G, Robertson KD (2009) DNMT3B interacts with constitutive centromere protein CENP-C to modulate DNA methylation and the histone code at centromeric regions. Hum Mol Genet 18(17): 3178–93.
- 25. Gartler SM, Varadarajan KR, Luo P, Canfield TK, Traynor J, et al. (2004) Normal histone modifications on the inactive X chromosome in ICF and Rett syndrome cells: implications for methyl-CpG binding proteins. BMC Biol 2: 21.
- 26. Yehezkel S, Segev Y, Viegas-Péquignot E, Skorecki K, Selig S (2008) Hypomethylation of subtelomeric regions in ICF syndrome is associated with abnormally short telomeres and enhanced transcription from telomeric regions. Hum Mol Genet 17: 2776–2789.
- 27. Peters AH, Kubicek S, Mechtler K, O'Sullivan RJ, Derijck AA, et al. (2003) Partitioning and plasticity of repressive histone methylation states in mammalian chromatin. Mol Cell 12(6): 1577–89.
- 28. Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, et al. (2007) High-resolution profiling of histone methylations in the human genome. Cell 129(4): 823–37.
- 29. Su IH, Basavaraj A, Krutchinsky AN, Hobert O, Ullrich A, et al. (2003) Ezh2 controls B cell development through histone H3 methylation and Igh rearrangement. Nat Immunol 4(2): 124–31.
- 30. Tapparel C, Reymond A, Girardet C, Guillou L, Lyle R, et al. (2003) The TPTE gene family: cellular expression, subcellular localization and alternative splicing. Gene 323: 189–199.
- 31. Bera TK, Saint Fleur A, Lee Y, Kydd A, Hahn Y, et al. (2006) POTE paralogs are induced and differentially expressed in many cancers. Cancer Res 66: 52–56.
- 32. Bera TK, Saint Fleur A, Ha D, Yamada M, Lee Y, et al. (2008) Selective POTE paralogs on chromosome 2 are expressed in human embryonic stem cells. Stem Cells Dev 17(2): 325–332.
- 33. Ise T, Das S, Nagata S, Maeda H, Lee Y, et al. (2008) Expression of POTE protein in human testis detected by novel monoclonal antibodies. Biochem Biophys Res Commun 365: 603–608.
- 34. Velasco G, Hubé F, Rollin J, Neuillet D, Philippe C, et al. (2010) Dnmt3b recruitment through E2F6 transcriptional repressor mediates germ-line gene silencing in murine somatic tissues. Proc Natl Acad Sci U S A 107(20): 9281–9286.
- 35. Frommer M, McDonald LE, Millar DS, Collis CM, Watt F, et al. (1992) A genomic sequencing protocol that yields a positive display of 5-methylcytosine residues in individual DNA strands. Proc Natl Acad Sci U S A 89: 1827–1831.
- 36. Dear S, Staden R (1991) A sequence assembly and editing program for efficient management of large projects. Nucleic Acid Res 19: 3907–3911.
- 37. Grunau C, Schattevoy R, Mache N, Rosenthal A (2000) MethTools—a toolbox to visualize and analyze DNA methylation data. Nucleic Acids Res 28: 1053–1058.