Javier Suela

A Comprehensive Microarray-Based DNA Methylation Study of 367 Hematological Neoplasms

Background Alterations in the DNA methylation pattern are a hallmark of leukemias and lymphomas. However, most epigenetic studies in hematologic neoplasms (HNs) have focused either on the analysis of few candidate genes or many genes and few HN entities, and comprehensive studies are required. Methodology/Principal Findings Here, we report for the first time a microarray-based DNA methylation study of 767 genes in 367 HNs diagnosed with 16 of the most representative B-cell (n = 203), T-cell (n = 30), and myeloid (n = 134) neoplasias, as well as 37 samples from different cell types of the hematopoietic system. Using appropriate controls of B-, T-, or myeloid cellular origin, we identified a total of 220 genes hypermethylated in at least one HN entity. In general, promoter hypermethylation was more frequent in lymphoid malignancies than in myeloid malignancies, being germinal center mature B-cell lymphomas as well as B and T precursor lymphoid neoplasias those entities with highest frequency of gene-associated DNA hypermethylation. We also observed a significant correlation between the number of hypermethylated and hypomethylated genes in several mature B-cell neoplasias, but not in precursor B- and T-cell leukemias. Most of the genes becoming hypermethylated contained promoters with high CpG content, and a significant fraction of them are targets of the polycomb repressor complex. Interestingly, T-cell prolymphocytic leukemias show low levels of DNA hypermethylation and a comparatively large number of hypomethylated genes, many of them showing an increased gene expression. Conclusions/Significance We have characterized the DNA methylation profile of a wide range of different HNs entities. As well as identifying genes showing aberrant DNA methylation in certain HN subtypes, we also detected six genes—DBC1, DIO3, FZD9, HS3ST2, MOS, and MYOD1—that were significantly hypermethylated in B-cell, T-cell, and myeloid malignancies. These might therefore play an important role in the development of different HNs.

Comparative genome profiling across subtypes of low-grade B-cell lymphoma identifies type-specific and common aberrations that target genes with a role in B-cell neoplasia

Low-grade B-cell lymphomas are very heterogeneous. This study of comparative genome profiling identifies type-specific aberrations that target genes with a role in B-cell lymphoid neoplasms. See related perspective article on page 641. Background Low-grade B-cell lymphomas are a very heterogeneous group of tumors, whose differential diagnosis is frequently compromised by the lack of specific cytogenetic or molecular features. Our objective was to search for genomic features that allow a better molecular identification of the different types of lymphoma studied. Design and Methods We selected a panel of 87 low-grade B-cell lymphoma tumor samples that were unambiguously diagnosed (clinically and cytogenetically) as: follicular, splenic marginal zone, nodal marginal zone, lymphoplasmacytic, mantle cell, extranodal marginal zone MALT-type lymphoma or B-cell chronic lymphocytic leukemia. All samples were subjected to the same high-resolution genomic DNA analysis (array-based comparative genomic hybridization): a whole genome platform that contained 44000 probes distributed across the genome. Genomic imbalances were recorded, compiled and analyzed. Results Eighty percent of analyzed cases showed genomic imbalances (deletions and gain/amplifications) but the frequency of these imbalances ranged from 100% in mantle cell lymphomas to 33% in MALT lymphomas. A total of 95 new genomic imbalances affecting all lymphoma subtypes, were defined. We evaluated the extension of the genomic instability, detecting distinct patterns of genomic instability within subtypes. Specific pathways, such as nuclear factor kB (gains of REL and BCL11A, and losses of COMMD3, BIRC1, IKK1 and NFKB2), Polycomb group proteins (gain of BMI1 and deletion of PCGF6), DNA repair checkpoint pathways (deletion of 16q24 involving CDT1), or miRNA with a role in B-cell lymphoma pathogenesis (MIRN15A, MIRN16-1), were targeted by this genomic instability. Conclusions Although all subtypes of lymphomas showed gains and losses of DNA, the analysis of their genomic profiles indicated that there are specific aberrations in almost every subtype as well as frequent aberrations that are common to a large number of lymphoma types. These common aberrations target genes that are important in B-cell lymphomagenesis.

DNA Methylation Profiles and Their Relationship with Cytogenetic Status in Adult Acute Myeloid Leukemia

Background Aberrant promoter DNA methylation has been shown to play a role in acute myeloid leukemia (AML) pathophysiology. However, further studies to discuss the prognostic value and the relationship of the epigenetic signatures with defined genomic rearrangements in acute myeloid leukemia are required. Methodology/Principal Findings We carried out high-throughput methylation profiling on 116 de novo AML cases and we validated the significant biomarkers in an independent cohort of 244 AML cases. Methylation signatures were associated with the presence of a specific cytogenetic status. In normal karyotype cases, aberrant methylation of the promoter of DBC1 was validated as a predictor of the disease-free and overall survival. Furthermore, DBC1 expression was significantly silenced in the aberrantly methylated samples. Patients with chromosome rearrangements showed distinct methylation signatures. To establish the role of fusion proteins in the epigenetic profiles, 20 additional samples of human hematopoietic stem/progenitor cells (HSPC) transduced with common fusion genes were studied and compared with patient samples carrying the same rearrangements. The presence of MLL rearrangements in HSPC induced the methylation profile observed in the MLL-positive primary samples. In contrast, fusion genes such as AML1/ETO or CBFB/MYH11 failed to reproduce the epigenetic signature observed in the patients. Conclusions/Significance Our study provides a comprehensive epigenetic profiling of AML, identifies new clinical markers for cases with a normal karyotype, and reveals relevant biological information related to the role of fusion proteins on the methylation signature.

Oligonucleotide Array-CGH Identifies Genomic Subgroups and Prognostic Markers for Tumor Stage Mycosis Fungoides

Mycosis fungoide (MF) patients who develop tumors or extracutaneous involvement usually have a poor prognosis with no curative therapy available so far. In the present European Organization for Research and Treatment of Cancer (EORTC) multicenter study, the genomic profile of 41 skin biopsies from tumor stage MF (MFt) was analyzed using a high-resolution oligo-array comparative genomic hybridization platform. Seventy-six percent of cases showed genomic aberrations. The most common imbalances were gains of 7q33.3q35 followed by 17q21.1, 8q24.21, 9q34qter, and 10p14 and losses of 9p21.3 followed by 9q31.2, 17p13.1, 13q14.11, 6q21.3, 10p11.22, 16q23.2, and 16q24.3. Three specific chromosomal regions, 9p21.3, 8q24.21, and 10q26qter, were defined as prognostic markers showing a significant correlation with overall survival (OS) (P=0.042, 0.017, and 0.022, respectively). Moreover, we have established two MFt genomic subgroups distinguishing a stable group (0-5 DNA aberrations) and an unstable group (>5 DNA aberrations), showing that the genomic unstable group had a shorter OS (P=0.05). We therefore conclude that specific chromosomal abnormalities, such as gains of 8q24.21 (MYC) and losses of 9p21.3 (CDKN2A, CDKN2B, and MTAP) and 10q26qter (MGMT and EBF3) may have an important role in prognosis. In addition, we describe the MFt genomic instability profile, which, to our knowledge, has not been reported earlier.

Deregulated Expression of the Polycomb-Group Protein SUZ12 Target Genes Characterizes Mantle Cell Lymphoma

Polycomb proteins are known to be of great importance in human cancer pathogenesis. SUZ12 is a component of the Polycomb PRC2 complex that, along with EZH2, is involved in embryonic stem cell differentiation. EZH2 plays an essential role in many cancer types, but an equivalent involvement of SUZ12 has not been as thoroughly demonstrated. Here we show that SUZ12 is anomalously expressed in human primary tumors, especially in mantle cell lymphoma (MCL), pulmonary carcinomas and melanoma, and is associated with gene locus amplification in some cases. Using MCL as a model, functional and genomic studies demonstrate that SUZ12 loss compromises cell viability, increases apoptosis, and targets genes involved in central oncogenic pathways associated with MCL pathogenesis. Our results support the hypothesis that the abnormal expression of SUZ12 accounts for some of the unexplained features of MCL, such as abnormal DNA repair and increased resistance to apoptosis.

Detection of the first gross CDC73 germline deletion in an HPT-JT syndrome family

Hereditary primary hyperparathyroidism (HPT) may develop as a solitary endocrinopathy (FIHP) or as part of multiple endocrine neoplasia Type 1, multiple endocrine neoplasia Type 2A, or hereditary HPT‐jaw tumor syndrome. Inactivating germline mutations of the tumor suppressor gene CDC73 account for 14 and 50% of all FIHP and HPT‐JT patients, respectively, and have also been found in almost 20% of apparently sporadic parathyroid carcinoma patients. Although more than 60 independent germline mutations have been described, to date no rearrangement affecting the CDC73 locus has been identified. By means of multiplex‐PCR we found a large germline deletion affecting the whole gene in a two‐generation HPT‐JT family. Subsequently array‐CGH and specific PCR analysis determined that the mutation spanned ∼ 547 kb, and included four additional genes: TROVE2, GLRX2, B3GALT2, and UCHL5. Although no clear mutation‐specific phenotype was found associated to the presence of the mutation, further studies are needed to assess whether the loss of the neighboring genes could modify the phenotype of carriers. There was complete absence of nuclear staining in the two HPT‐JT‐related tumors available. The finding of the first rearrangement affecting the CDC73 gene warrants screening for this tumor suppressor gene inactivation mechanism not only in high‐risk CDC73 point mutation‐negative HPT‐JT families, but also in FIHP patients.

Primary Cutaneous CD30+ Anaplastic Large-Cell Lymphomas Show a Heterogeneous Genomic Profile: An Oligonucleotide ArrayCGH Approach

TO THE EDITOR Primary cutaneous CD30-positive lymphoproliferative disorders are the second most common group of primary cutaneous T-cell lymphomas (Willemze et al., 2005), after the mycosis fungoides/Sézary syndrome group. The clinical and histological features of primary cutaneous anaplastic large-cell lymphoma (C-ALCL) have been well characterized, but little is known about its underlying pathogenetic and genetic alterations. Previous comparative genomic hybridization (CGH) studies focusing on C-ALCL that included a limited number of samples yielded nonhomogeneous results (Böni et al., 2000; Mao et al., 2003; Prochazkova et al., 2003; Fischer et al., 2004; Zettl et al., 2004). Recently, three studies that were based on array CGH (aCGH) and included a small number of C-ALCL patients were published (Mao et al., 2003; Laharanne et al., 2010; van Kester et al., 2010). We have investigated the genomic profile of 19 C-ALCL patients using a 60-mer 44K oligonucleotide-arrayCGH platform and compared our results with those of previous aCGH studies. C-ALCL patients were selected according to the World Health Organization–European Organization for Research and Treatment for Cancer (EORTC) classification for cutaneous lymphomas (Willemze et al., 2005). This EORTC multicenter study was conducted in the departments of pathology and dermatology of six European centers in Spain and Switzerland. The local ethics committees approved the study, and written informed consent was obtained from all patients, in accordance the Declaration of Helsinki Principles. Clinical characteristics are detailed in Supplementary Table S1 online. The study was performed with 20 10 mm snap-frozen C-ALCL samples to ensure the high quality of the DNA. Hematoxylin–eosin staining of a frozen section of each sample was performed tumor cell infiltration of at least 70%. DNA was isolated using a commercial kit as described in manufacturer’s instructions (Dneasy Blood and Tissue Kit; Qiagen, Hilden, Germany). Genomewide analysis of patient samples was conducted using the Human Genome CGH 44K microarrays (G4410B and G4426B; Agilent Technologies, Palo Alto, CA), a whole-genome platform containing 44,000 probes along the entire human genome with a mean resolution of ±75 kb. Hybridization was performed according to the manufacturer’s protocols. Data analysis was conducted as previously described (Salgado et al., 2010). Fluorescence in situ hybridization with noncommercial probes of bacterial artificial chromosome DNA clones from the CHORI bacterial artificial chromosome/PAC resource (http://bacpac. chori.org) was performed to confirm chromosomal abnormalities previously detected by aCGH in cases for which a paraffin-embedded tissue biopsy was available. Chromosomal abnormalities were detected in 17 of 19 analyzed C-ALCL samples (89.5%). Losses were more frequently detected than gains (78.9 vs. 68.4%). Mao et al. (2003) and van Kester et al. (2010) found gains more frequently than losses, whereas Laharanne et al. (2010) detected losses more frequently. The highest frequencies of chromosomal aberrations were 60% (Mao et al., 2003) and 45% (Laharanne et al., 2010; van Kester et al., 2010), in contrast to 36.8% in our present study. Regarding the smallest overlapping regions of imbalance, 15 corresponded to losses and 9 to gains. The results are summarized in Figure 1 and detailed in Supplementary Table S2 online. The specific chromosomal regions and candidate genes mapped in these regions are detailed in Table 1. The most frequent abnormalities observed were deletions located on 16q, 13q, 17p13, and 20q13. Genomic losses of 13q34 (ING1) and 16q22.11 (CTCF) detected by aCGH were confirmed by fluorescence in situ hybridization in three patients. No significant correlation between the observed clinical features and the presence of chromosomal aberrations could be demonstrated. Furthermore, no data regarding the prognostic significance of the observed genetic results were obtained. In agreement with studies by van Kester et al. (2010) and Laharanne et al. (2010), two regions were lost in our study, at 13q33.3 and 16p11.2. These regions were not detected in the first aCGH study (Mao et al., 2003), probably because they may not have been among the 57 oncogenic regions of the AmpliOnc platform. Similar to the findings of van Kester et al. (2010), we observed losses at 3p26.3, 6q21, 8p22, 13q12.11, 13q13.1, 16p11.2-16q11.2, 17p13.1, and 17p13.3 (Supplementary Table S3 online). The main concordance between our results and those of van Kester et al. (2010) was a deletion at 16q11.2. However, differences were observed for a higher frequency of 16q losses in our series, including seven genomic regions located between 16q11.2 and 16q24.3. The most Abbreviations: aCGH, array comparative genomic hybridization; C-ALCL, primary cutaneous anaplastic large-cell lymphoma; CGH, comparative genomic hybridization

Identification of prefoldin amplification (1q23.3-q24.1) in bladder cancer using comparative genomic hybridization (CGH) arrays of urinary DNA

BackgroundArray-CGH represents a comprehensive tool to discover genomic disease alterations that could potentially be applied to body fluids. In this report, we aimed at applying array-CGH to urinary samples to characterize bladder cancer.MethodsUrinary DNA from bladder cancer patients and controls were hybridized on 44K oligonucleotide arrays. Validation analyses of identified regions and candidates included fluorescent in situ hybridization (FISH) and immunohistochemistry in an independent set of bladder tumors spotted on custom-made tissue arrays (n = 181).ResultsQuality control of array-CGH provided high reproducibility in dilution experiments and when comparing reference pools. The most frequent genomic alterations (minimal recurrent regions) among bladder cancer urinary specimens included gains at 1q and 5p, and losses at 10p and 11p. Supervised hierarchical clustering identified the gain at 1q23.3-q24.1 significantly correlated to stage (p = 0.011), and grade (p = 0.002). The amplification and overexpression of Prefoldin (PFND2), a selected candidate mapping to 1q23.3-q24.1, correlated to increasing stage and tumor grade by means of custom-designed and optimized FISH (p = 0.013 and p = 0.023, respectively), and immunohistochemistry (p ≤0.0005 and p = 0.011, respectively), in an independent set of bladder tumors included in tissue arrays. Moreover, PFND2 overexpression was significantly associated with poor disease-specific survival (p ≤0.0005). PFND2 was amplified and overexpressed in bladder tumors belonging to patients providing urinary specimens where 1q23.3q24.1 amplification was detected by array-CGH.ConclusionsGenomic profiles of urinary DNA mirrowed bladder tumors. Molecular profiling of urinary DNA using array-CGH contributed to further characterize genomic alterations involved in bladder cancer progression. PFND2 was identified as a tumor stratification and clinical outcome prognostic biomarker for bladder cancer patients.

Molecular Characterization of a New Patient With a Non-Recurrent Inv Dup Del 2q and Review of the Mechanisms for This Rearrangement

We report on newborn baby with microcephaly, facial anomalies, congenital heart defects, hypotonia, wrist contractures, long fingers, adducted thumbs, and club feet. Cytogenetic studies revealed an inverted duplication with terminal deletion (inv dup del) of 2q in the patient and a paternal 2qter deletion polymorphism. Microsatellite markers demonstrated that the inv dup del was maternal in origin and intrachromosomal. Intra or interchromosomal rearrangements may cause this aberration either by a U‐type exchange (end‐to‐end fusion), an unequal crossover between inverted repeats (non‐allelic homologous recombination: NAHR), or through breakage‐fusion‐bridge (BFB) cycles leading to a sister chromatid fusion by non‐homologous end joining (NHEJ). A high‐resolution oligo array‐CGH (244 K) defined the breakpoints and did not detect a single copy region with a size exceeding 12.93 Kb in the fusion site. The size of the duplicated segment was 38.75 Mb, extending from 2q33.1 to 2q37.3 and the size of the terminal deletion was 2.85 Mb in 2q37.3. Our results indicate that the inv dup del (2q) is likely a non‐recurrent chromosomal rearrangement generated by a NHEJ mechanism. The major clinical characteristics associated with this 2q rearrangement overlap with those commonly found in patients with 2q duplication reported in the literature.

Guidelines for genomic array analysis in acquired haematological neoplastic disorders.

Genetic profiling is important for disease evaluation and prediction of prognosis or responsiveness to therapy in neoplasia. Microarray technologies, including array comparative genomic hybridization and single‐nucleotide polymorphism‐detecting arrays, have in recent years been introduced into the diagnostic setting for specific types of haematological malignancies and solid tumours. It can be used as a complementary test or depending on the neoplasia investigated, also as a standalone test. However, comprehensive and readable presentation of frequently identified complex genomic profiles remains challenging. To assist diagnostic laboratories, standardization and minimum criteria for clinical interpretation and reporting of acquired genomic abnormalities detected through arrays in neoplastic disorders are presented.