Sanaa Choufani

Discovery of cross-reactive probes and polymorphic CpGs in the Illumina Infinium HumanMethylation450 microarray

DNA methylation, an important type of epigenetic modification in humans, participates in crucial cellular processes, such as embryonic development, X-inactivation, genomic imprinting and chromosome stability. Several platforms have been developed to study genome-wide DNA methylation. Many investigators in the field have chosen the Illumina Infinium HumanMethylation microarray for its ability to reliably assess DNA methylation following sodium bisulfite conversion. Here, we analyzed methylation profiles of 489 adult males and 357 adult females generated by the Infinium HumanMethylation450 microarray. Among the autosomal CpG sites that displayed significant methylation differences between the two sexes, we observed a significant enrichment of cross-reactive probes co-hybridizing to the sex chromosomes with more than 94% sequence identity. This could lead investigators to mistakenly infer the existence of significant autosomal sex-associated methylation. Using sequence identity cutoffs derived from the sex methylation analysis, we concluded that 6% of the array probes can potentially generate spurious signals because of co-hybridization to alternate genomic sequences highly homologous to the intended targets. Additionally, we discovered probes targeting polymorphic CpGs that overlapped SNPs. The methylation levels detected by these probes are simply the reflection of underlying genetic polymorphisms but could be misinterpreted as true signals. The existence of probes that are cross-reactive or of target polymorphic CpGs in the Illumina HumanMethylation microarrays can confound data obtained from such microarrays. Therefore, investigators should exercise caution when significant biological associations are found using these array platforms. A list of all cross-reactive probes and polymorphic CpGs identified by us are annotated in this paper.

Altered gene expression and methylation of the human chromosome 11 imprinted region in small for gestational age (SGA) placentae

Imprinted genes are known to be crucial for placental development and fetal growth in mammals, but no primary epigenetic abnormality in placenta has been documented to compromise human fetal growth. Imprinted genes demonstrate parent-of-origin-specific allelic expression that is epigenetically regulated i.e. extrinsic to the primary DNA sequence. To undertake an epigenetic analysis of poor fetal growth in placentae and cord blood tissues, we first established the tissue-specific patterns of methylation and imprinted gene expression for two imprinting clusters (KvDMR and H19 DMR) on chromosome 11p15 in placentae and neonatal blood for 20 control cases and 24 Small for Gestational Age (SGA) cases. We confirmed that, in normal human placenta, the H19 promoter is unmethylated. In contrast, most other human tissues show paternal methylation. In addition, we showed that the IGF2 DMR2, also paternally methylated in most human tissues, exhibits hypomethylation in placentae. However, in neonatal blood DNA, these two regions maintain the differential methylation status seen in most other tissues. Significantly, we have been able to demonstrate that placenta does maintain differential methylation at the imprinting control regions H19 DMR and KvDMR. Of note, in one SGA placenta, we found a methylation alteration at the H19 DMR and concomitant biallelic expression of the H19 gene, suggesting that loss of imprinting at H19 is one cause of poor fetal growth in humans. Of particular interest, we demonstrated also a decrease in IGF2 mRNA levels in all SGA placentae and showed that the decrease is, in most cases, independent of H19 regulation.

EBV transformation and cell culturing destabilizes DNA methylation in human lymphoblastoid cell lines

Recent research suggests that epigenetic alterations involving DNA methylation can be causative for neurodevelopmental, growth and metabolic disorders. Although lymphoblastoid cell lines have been an invaluable resource for the study of both genetic and epigenetic disorders, the impact of EBV transformation, cell culturing and freezing on epigenetic patterns is unknown. We compared genome-wide DNA methylation patterns of four white blood cell samples, four low-passage lymphoblastoid cell lines pre and post freezing and four high-passage lymphobastoid cell lines, using two microarray platforms: Illumina HumanMethylation27 platform containing 27,578 CpG sites and Agilent Human CpG island Array containing 27,800 CpG islands. Comparison of genome-wide methylation profiles between white blood cells and lymphoblastoid cell lines demonstrated methylation alterations in lymphoblastoid cell lines occurring at random genomic locations. These changes were more profound in high-passage cells. Freezing at low-passages did not have a significant effect on DNA methylation. Methylation changes were observed in several imprinted differentially methylated regions, including DIRAS3, NNAT, H19, MEG3, NDN and MKRN3, but not in known imprinting centers. Our results suggest that lymphoblastoid cell lines should be used with caution for the identification of disease-associated DNA methylation changes or for discovery of new imprinted genes, as the methylation patterns seen in these cell lines may not always be representative of DNA methylation present in the original B-lymphocytes of the patient.

Methylation of the TERT promoter and risk stratification of childhood brain tumours: an integrative genomic and molecular study

BACKGROUND Identification of robust biomarkers of malignancy and methods to establish disease progression is a major goal in paediatric neuro-oncology. We investigated whether methylation of the TERT promoter can be a biomarker for malignancy and patient outcome in paediatric brain tumours. METHODS For the discovery cohort, we used samples obtained from patients with paediatric brain tumours and individuals with normal brain tissues stored at the German Cancer Research Center (Heidelberg, Germany). We used methylation arrays for genome-wide assessment of DNA. For the validation cohort, we used samples obtained from several tissues for which full clinical and follow-up data were available from two hospitals in Toronto (ON, Canada). We did methylation analysis using quantitative Sequenom and pyrosequencing of an identified region of the TERT promoter. We assessed TERT expression by real-time PCR. To establish whether the biomarker could be used to assess and predict progression, we analysed methylation in paired samples of tumours that transformed from low to high grade and from localised to metastatic, and in choroid plexus tumours of different grades. Finally, we investigated overall survival in patients with posterior fossa ependymomas in which the identified region was hypermethylated or not. All individuals responsible for assays were masked to the outcome of the patients. FINDINGS Analysis of 280 samples in the discovery cohort identified one CpG site (cg11625005) in which 78 (99%) of 79 samples from normal brain tissues and low-grade tumours were not hypermethylated, but 145 (72%) of 201 samples from malignant tumours were hypermethylated (>15% methylated; p<0.0001). Analysis of 68 samples in the validation cohort identified a subset of five CpG sites (henceforth, upstream of the transcription start site [UTSS]) that was hypermethylated in all malignant paediatric brain tumours that expressed TERT but not in normal tissues that did not express TERT (p<0.0001). UTSS had a positive predictive value of 1.00 (95% CI 0.95-1.00) and a negative predictive value of 0.95 (0.87-0.99). In two paired samples of paediatric gliomas, UTSS methylation increased during transformation from low to high grade; it also increased in two paired samples that progressed from localised to metastatic disease. Two of eight atypical papillomas that had high UTSS methylation progressed to carcinomas, while the other six assessed did not progress or require additional treatment. 5-year overall survival was 51% (95% CI 31-71) for 25 patients with hypermethylated UTSS posterior fossa ependymomas and 95% (86-100) for 20 with non-hypermethylated tumours (p=0.0008). 5-year progression-free survival was 86% (68-100) for the 25 patients with non-hypermethylated UTSS tumours and 30% (10-50) for those with hypermethylated tumours (p=0.0008). INTERPRETATION Hypermethylation of the UTSS region in the TERT promoter is associated with TERT expression in cancers. In paediatric brain tumours, UTSS hypermethylation is associated with tumour progression and poor prognosis. This region is easy to amplify, and the assay to establish hypermethylation can be done on most tissues in most clinical laboratories. Therefore the UTSS region is a potentially accessible biomarker for various cancers. FUNDING The Canadian Institute of Health Research and the Terry Fox Foundation.

A novel approach identifies new differentially methylated regions (DMRs) associated with imprinted genes

Imprinted genes are critical for normal human growth and neurodevelopment. They are characterized by differentially methylated regions (DMRs) of DNA that confer parent of origin-specific transcription. We developed a new strategy to identify imprinted gene-associated DMRs. Using genome-wide methylation profiling of sodium bisulfite modified DNA from normal human tissues of biparental origin, candidate DMRs were identified by selecting CpGs with methylation levels consistent with putative allelic differential methylation. In parallel, the methylation profiles of tissues of uniparental origin, i.e., paternally-derived androgenetic complete hydatidiform moles (AnCHMs), and maternally-derived mature cystic ovarian teratoma (MCT), were examined and then used to identify CpGs with parent of origin-specific DNA methylation. With this approach, we found known DMRs associated with imprinted genomic regions as well as new DMRs for known imprinted genes, NAP1L5 and ZNF597, and novel candidate imprinted genes. The paternally methylated DMR for one candidate, AXL, a receptor tyrosine kinase, was also validated in experiments with mouse embryos that demonstrated Axl was expressed preferentially from the maternal allele in a DNA methylation-dependent manner.

Growth Regulation, Imprinted Genes, and Chromosome 11p15.5

Genomic imprinting refers to parent-of-origin–specific gene expression. Human chromosome band 11p15.5 houses a large cluster of genes that are imprinted. Dysregulation of this gene cluster is associated with the overgrowth and tumor predisposition syndrome, Beckwith-Wiedemann syndrome. Several genes in this imprinted cluster encode proteins involved in growth regulation, e.g. the paternally expressed IGF2 and the maternally expressed cell-cycle regulator cyclin dependent kinase inhibitor, CDKN1C. Disruption of imprinted gene expression can result from genetic or epigenetic alterations. Genetic alterations such as duplication, deletion, translocation, inversion, and mutation in imprinted regions have been shown to cause disease. In addition, epimutations that are extrinsic to the primary DNA sequence have also been shown to cause disease. These epimutations usually involve gain or loss of methylation at regulatory differentially methylated regions. Recently, several human diseases in addition to Beckwith-Wiedemann syndrome have been reported to have molecular alterations at chromosome 11p15.5. These include isolated hemihyperplasia, Russell-Silver syndrome, and transient neonatal diabetes mellitus. These molecular alterations and their phenotypic effects on growth are discussed.

Molecular findings in Beckwith-Wiedemann syndrome.

Our understanding of Beckwith–Wiedemann syndrome (BWS) has recently been enhanced by advances in its molecular characterization. These advances have further delineated intricate (epi)genetic regulation of the imprinted gene cluster on chromosome 11p15.5 and the role of these genes in normal growth and development. Studies of the molecular changes associated with the BWS phenotype have been instrumental in elucidating critical molecular elements in this imprinted region. This review will provide updated information on the multiple new regulatory elements that have been recently found to contribute to in cis or in trans control of imprinted gene expression in the chromosome 11p15.5 region and the clinical expression of the BWS phenotype.

Sequence overlap between autosomal and sex-linked probes on the Illumina HumanMethylation27 microarray

The Illumina Infinium HumanMethylation27 BeadChip (Illumina 27k) microarray is a high-throughput platform capable of interrogating the human DNA methylome. In a search for autosomal sex-specific DNA methylation using this microarray, we discovered autosomal CpG loci showing significant methylation differences between the sexes. However, we found that the majority of these probes cross-reacted with sequences from sex chromosomes. Moreover, we determined that 6-10% of the microarray probes are non-specific and map to highly homologous genomic sequences. Using probes targeting different CpGs that are exact duplicates of each other, we investigated the precision of these repeat measurements and concluded that the overall precision of this microarray is excellent. In addition, we identified a small number of probes targeting CpGs that include single-nucleotide polymorphisms. Overall, our findings address several technical issues associated with the Illumina 27k microarray that, once considered, will enhance the analysis and interpretation of data generated from this platform.

Basic concepts of epigenetics.

Several types of epigenetic marks facilitate the complex patterning required for normal human development. These epigenetic marks include DNA methylation at CpG dinucleotides, covalent modifications of histone proteins, and noncoding RNAs (ncRNAs). They function in a highly orchestrated manner, regulating mitotically heritable differences in gene expression potential without altering the primary DNA sequence. In germ cells and the developing embryo, genome-wide epigenetic reprogramming drives the erasure and reestablishment of correct epigenetic patterns at critical developmental time periods and in specific cell types. Two specific types of epigenetic regulation established in early development include X-chromosome inactivation and genomic imprinting; they regulate gene expression in a dosage-dependent and parent-of-origin-specific manner, respectively. Both genetic and environmental factors impact epigenetic marks, generating phenotypic variation that ranges from normal variation to human disease. Aberrant epigenetic patterning can lead to a variety of human disorders, including subfertility and imprinting disorders.

Cell specific patterns of methylation in the human placenta

Epigenetic processes, such as DNA methylation, are known to regulate tissue specific gene expression. We explored this concept in the placenta to define whether DNA methylation is cell-type specific. Cytotrophoblasts and fibroblasts were isolated from normal midtrimester placentas. Using immunocytochemistry, we demonstrated 95% purity for cytotrophoblasts and 60-70% for fibroblasts. We compared DNA methylation profiles from cytotrophoblasts, fibroblasts and whole placental villi using bisulfite modified genomic DNA hybridized to the Illumina Methylation27 array. Euclidean cluster analysis of the DNA methylation profiles showed 2 main clusters, one containing cytotrophoblasts and placenta, the other fibroblasts. Differential methylation analysis identified 442 autosomal CpG sites that differed between cytotrophoblasts and fibroblasts, 315 between placenta and fibroblasts and 61 between placenta and cytotrophoblasts. Three candidate methylation differences were validated by targeted pyrosequencing assays. Pyrosequencing assays were developed for CpG sites less methylated in cytotrophoblasts than fibroblasts mapping to the promoter region of the beta subunit of human chorionic gonadotropin 5 (CGB5), as well as 2 CpG sites mapping to each of 2 tumor suppressor genes. Our data suggest that epigenetic regulation of gene expression is likely to be a key factor in the functional specificity of cytotrophoblasts. These data are proof of principle for cell-type specific epigenetic regulation in placenta and demonstrate that the methylation profile of placenta is mainly driven by cytotrophoblasts.