The altered DNA methylation landscape in chronic lymphocytic leukemia emerges early and persists after treatment
Ontology highlight
ABSTRACT: Here we measured genome-wide DNA methylation in patients with chronic lymphocytic leukemia as it provides an opportunity to capture the emergence of altered methylation landscapes as well as track their dynamics over disease progression including after treatment. Specifically, our cohort includes 21 CLL cases with up to six different time points, 20 precursor states of monoclonal B cell lymphocytes (MBL) and 5 matched samples from MBL and CLL states of the same patients. We find that across all CLL cases, a highly aberrant methylation state was present consistently already at the first time point and maintained with remarkable stability over disease progression. To improve our resolution and address heterogeneity we sequenced the methylome of single cells from CD5 positive and negative naïve and memory B cells as well as unsorted B cells and CLL cells. Methylation levels were highly similar within groups pointing to a rather homogeneous population and contrasted by pronounced differences between naïve and memory B cells that do not distinguish CD5 positive and negative subgroups. Our 20 patients with MBL further confirmed the early emergence of this landscape and surprisingly, the chemotherapy did not have a notable impact on the altered methylome despite a strong depletion of white blood cells.
Project description:Chronic lymphocytic leukemia (CLL) is a biologically and clinically heterogeneous disease. The somatic hypermutation status of the immunoglobulin heavy chain variable (IGHV) genes has been identified as one of the most robust prognostic markers in CLL. Patients with unmutated IGHV status (U-CLL) typically experience an inferior outcome compared to those whose clones express mutated IGHV genes (M-CLL). We conducted a genome-wide DNA methylation analysis in CD19+ B-cells from a group of 43 CLL patients using reduced representation bisulfite sequencing (RRBS). Using base-pair resolution methylation sequencing, 2323 differentially methylated regions between CLL and normal B-cells (CLL-specific DMRs) and 569 between M-CLL and U-CLL samples (IGHV-specific DMRs) were identified in the CLL genomes. The IGHV-specific DMRs are mostly unique when compared to the CLL-specific DMRs. Less than 10% of the IGHV-specific DMRs are located in promoter regions; however, more than half of these overlap with known DNase I hypersensitive sites, enhancer regions marked by histone modification (H3K4Me1 and H3K27Ac), and transcription factor binding sites in the ENCODE datasets, which indicates that these DMRs contain regulatory sequences. Distinctive DNA methylation patterns were observed in M-CLL and U-CLL samples. Overall, U-CLL was found to contain 50% more hypermethylated regions than M-CLL samples. The hypermethylated loci observed in the U-CLL samples also appear to be hypermethylated in normal naïve B-cells as compared memory B-cells, suggesting that M-CLL and U-CLL differ in differentiation status corresponding to normal B-cell differentiation stages. RNA-seq analysis performed using matched samples (n=34), in which both DNA methylation and gene expression data were available, demonstrated excellent correlation between DNA methylation and gene expression. Several genes whose expression status was previously shown to be associated with CLL prognosis such as ZAP70, CRY1, LDOC1, SEPT10, LAG3, and LPL were differentially methylated in the promoter regions between M-CLL and U-CLL samples indicating that DNA methylation plays an important role in defining the gene expression patterns of these prognostic genes. We further validated 9 genes with IGHV-specific DMRs in the promoter regions using bisulfite pyrosequencing, and the results demonstrated excellent correlation between differential methylation and IGHV mutation status. These novel differentially methylated genes could be developed into biomarkers for CLL prognosis. In addition, DNA hypomethylation was observed in a significant number of genes involved in lymphocyte activation such as PDCD1, NFAT1, and CD5. DNA hypomethylation was observed in the proximal promoter and far up-stream enhancer regions of CD5, an important cell surface marker that uniquely identifies CLL. Overall, the DNA methylation landscape in CLL patients indicates that CLL B cells possess an active B-cell phenotype; at the same time, U-CLL and M-CLL are faithfully committed to their lineage resembling either naïve or memory B-cells. In summary, this comprehensive DNA methylation analysis has identified a large number of novel epigenetic changes in CLL patients. The results from this study will further advance our understanding of the epigenetic contribution to molecular subtypes in CLL. To perform a transcriptome analysis in CLL, we generated sequencing libraries from total RNA isolated from purified B-cells of CLL patients and healthy donnors. The RNA-seq libraries were sequenced using Illumina HiSeq2000 sequencer with a read length of 100bp. 11 CLL B-cell samples, 3 normal control samples including one each of normal CD19+ B cells were studied. We generated 20-30 million Illumina sequencing reads for each sample.
Project description:The cellular origin of chronic lymphocytic leukemia (CLL) is debated. Transcriptome analysis of CLL and normal peripheral blood and splenic B cell subsets displayed highest similarity of CLL to mature CD5+ B cells. We identified a distinct CD5+CD27+ post-germinal center B cell subset, and revealed that immunoglobulin V gene mutated CLL are more similar to mutated CD5+ B cells, whereas unmutated CLL are more related to unmutated CD5+ B cells. Stereotyped immunoglobulin V gene rearrangements were significantly enriched among CD5+ B cells, providing further genetic evidence for a derivation of CLL from CD5+ B cells. Moreover, we identified deregulated expression patterns providing novel insights into the pathophysiology of CLL, including downregulation of EBF1 and KLF family members. Transcriptome profiling of CLL and healthy human blood and splenic mature B cell subsets. Identification of deregulated transcription patterns with implications on CLL pathobiology.
Project description:Several studies have demonstrated an impaired function of the microenvironment in chronic lymphocytic leukemia (CLL), contributing to immune evasion of tumor cells and disease progression. However, in CLL-like monoclonal B cell lymphocytosis (MBL) studies are scarce. Herein, a comprehensive characterization of the microenvironment in 59 MBL, 56 early stage CLL and 31 healthy controls was conducted. Gene expression arrays and qRT-PCR were performed on RNA from CD4+ peripheral blood cells; serum cytokines were measured by immunoassays and proteomic studies; and flow cytometry was applied to evaluate peripheral blood cytotoxic, Th1, exhausted and effector CD4+ T cells, besides monocytic CD14, CD4 and HLA-DR expression. MBL and early stage CLL showed a similar upregulation of cytotoxic and Th1-related genes, expanded perforin+ and CXCR3+ CD4+ T cells as well as PD1+ CD4+ T cells compared to controls. However, a strong inflammatory response was only identified in MBL: enhanced phagocytosis, pattern recognition receptors, IL8, HMGB1, TREM1 and acute response signaling pathways, along with increased levels of proinflammatory cytokines (remarkably IL8, IFN? and TNF?). Of note, this inflammatory drive was decreased in early stage CLL: diminished proinflammatory cytokines including IFN?, decreased IL8 signaling pathway and lower monocytic HLA-DR expression compared to MBL. Besides, this inflammation was especially reduced in IGHV mutated CLL, involving a decrease of the proinflammatory HMGB1 signaling pathway. These novel findings reveal a different pathophysiology between MBL and CLL, paving the way for the development of pre-emptive immunotherapies with optimal benefits at MBL and early stage CLL, before intense immune exhaustion.
Project description:Chronic lymphocytic leukemia (CLL) is a biologically and clinically heterogeneous disease. The somatic hypermutation status of the immunoglobulin heavy chain variable (IGHV) genes has been identified as one of the most robust prognostic markers in CLL. Patients with unmutated IGHV status (U-CLL) typically experience an inferior outcome compared to those whose clones express mutated IGHV genes (M-CLL). We conducted a genome-wide DNA methylation analysis in CD19+ B-cells from a group of 43 CLL patients using reduced representation bisulfite sequencing (RRBS). Using base-pair resolution methylation sequencing, 2323 differentially methylated regions between CLL and normal B-cells (CLL-specific DMRs) and 569 between M-CLL and U-CLL samples (IGHV-specific DMRs) were identified in the CLL genomes. The IGHV-specific DMRs are mostly unique when compared to the CLL-specific DMRs. Less than 10% of the IGHV-specific DMRs are located in promoter regions; however, more than half of these overlap with known DNase I hypersensitive sites, enhancer regions marked by histone modification (H3K4Me1 and H3K27Ac), and transcription factor binding sites in the ENCODE datasets, which indicates that these DMRs contain regulatory sequences. Distinctive DNA methylation patterns were observed in M-CLL and U-CLL samples. Overall, U-CLL was found to contain 50% more hypermethylated regions than M-CLL samples. The hypermethylated loci observed in the U-CLL samples also appear to be hypermethylated in normal naïve B-cells as compared memory B-cells, suggesting that M-CLL and U-CLL differ in differentiation status corresponding to normal B-cell differentiation stages. RNA-seq analysis performed using matched samples (n=34), in which both DNA methylation and gene expression data were available, demonstrated excellent correlation between DNA methylation and gene expression. Several genes whose expression status was previously shown to be associated with CLL prognosis such as ZAP70, CRY1, LDOC1, SEPT10, LAG3, and LPL were differentially methylated in the promoter regions between M-CLL and U-CLL samples indicating that DNA methylation plays an important role in defining the gene expression patterns of these prognostic genes. We further validated 9 genes with IGHV-specific DMRs in the promoter regions using bisulfite pyrosequencing, and the results demonstrated excellent correlation between differential methylation and IGHV mutation status. These novel differentially methylated genes could be developed into biomarkers for CLL prognosis. In addition, DNA hypomethylation was observed in a significant number of genes involved in lymphocyte activation such as PDCD1, NFAT1, and CD5. DNA hypomethylation was observed in the proximal promoter and far up-stream enhancer regions of CD5, an important cell surface marker that uniquely identifies CLL. Overall, the DNA methylation landscape in CLL patients indicates that CLL B cells possess an active B-cell phenotype; at the same time, U-CLL and M-CLL are faithfully committed to their lineage resembling either naïve or memory B-cells. In summary, this comprehensive DNA methylation analysis has identified a large number of novel epigenetic changes in CLL patients. The results from this study will further advance our understanding of the epigenetic contribution to molecular subtypes in CLL.
Project description:Chronic lymphocytic leukemia (CLL) is a biologically and clinically heterogeneous disease. The somatic hypermutation status of the immunoglobulin heavy chain variable (IGHV) genes has been identified as one of the most robust prognostic markers in CLL. Patients with unmutated IGHV status (U-CLL) typically experience an inferior outcome compared to those whose clones express mutated IGHV genes (M-CLL). We conducted a genome-wide DNA methylation analysis in CD19+ B-cells from a group of 43 CLL patients using reduced representation bisulfite sequencing (RRBS). Using base-pair resolution methylation sequencing, 2323 differentially methylated regions between CLL and normal B-cells (CLL-specific DMRs) and 569 between M-CLL and U-CLL samples (IGHV-specific DMRs) were identified in the CLL genomes. The IGHV-specific DMRs are mostly unique when compared to the CLL-specific DMRs. Less than 10% of the IGHV-specific DMRs are located in promoter regions; however, more than half of these overlap with known DNase I hypersensitive sites, enhancer regions marked by histone modification (H3K4Me1 and H3K27Ac), and transcription factor binding sites in the ENCODE datasets, which indicates that these DMRs contain regulatory sequences. Distinctive DNA methylation patterns were observed in M-CLL and U-CLL samples. Overall, U-CLL was found to contain 50% more hypermethylated regions than M-CLL samples. The hypermethylated loci observed in the U-CLL samples also appear to be hypermethylated in normal naïve B-cells as compared memory B-cells, suggesting that M-CLL and U-CLL differ in differentiation status corresponding to normal B-cell differentiation stages. RNA-seq analysis performed using matched samples (n=34), in which both DNA methylation and gene expression data were available, demonstrated excellent correlation between DNA methylation and gene expression. Several genes whose expression status was previously shown to be associated with CLL prognosis such as ZAP70, CRY1, LDOC1, SEPT10, LAG3, and LPL were differentially methylated in the promoter regions between M-CLL and U-CLL samples indicating that DNA methylation plays an important role in defining the gene expression patterns of these prognostic genes. We further validated 9 genes with IGHV-specific DMRs in the promoter regions using bisulfite pyrosequencing, and the results demonstrated excellent correlation between differential methylation and IGHV mutation status. These novel differentially methylated genes could be developed into biomarkers for CLL prognosis. In addition, DNA hypomethylation was observed in a significant number of genes involved in lymphocyte activation such as PDCD1, NFAT1, and CD5. DNA hypomethylation was observed in the proximal promoter and far up-stream enhancer regions of CD5, an important cell surface marker that uniquely identifies CLL. Overall, the DNA methylation landscape in CLL patients indicates that CLL B cells possess an active B-cell phenotype; at the same time, U-CLL and M-CLL are faithfully committed to their lineage resembling either naïve or memory B-cells. In summary, this comprehensive DNA methylation analysis has identified a large number of novel epigenetic changes in CLL patients. The results from this study will further advance our understanding of the epigenetic contribution to molecular subtypes in CLL.
Project description:The cellular origin of chronic lymphocytic leukemia (CLL) is debated. Transcriptome analysis of CLL and normal peripheral blood and splenic B cell subsets displayed highest similarity of CLL to mature CD5+ B cells. We identified a distinct CD5+CD27+ post-germinal center B cell subset, and revealed that immunoglobulin V gene mutated CLL are more similar to mutated CD5+ B cells, whereas unmutated CLL are more related to unmutated CD5+ B cells. Stereotyped immunoglobulin V gene rearrangements were significantly enriched among CD5+ B cells, providing further genetic evidence for a derivation of CLL from CD5+ B cells. Moreover, we identified deregulated expression patterns providing novel insights into the pathophysiology of CLL, including downregulation of EBF1 and KLF family members. Transcriptome profiling of CLL and healthy human blood and splenic mature B cell subsets. Identification of deregulated transcription patterns with implications on CLL pathobiology. Human mature B cell subsets and CLL with mutated (mCLL) and unmutated V gene status (uCLL) were purified from peripheral blood and spleen. Samples of 5 to 7 donors each were collected and processed in three batches in a two rounded in vitro transcription protocol. Retrieved data were batch corrected and subjected to analysis. Human mature CD5+ B cell subsets and CLL with mutated (mCLL) and unmutated V gene status (uCLL) were purified from peripheral blood.
Project description:Chronic lymphocytic leukemia (CLL) is characterized by the accumulation of malignant CD5+ B lymphocytes (CLL cells) in the peripheral blood, and their progressive infiltration in lymphoid organs. Despite the efforts made, CLL remains an incurable disease. It is therefore crucial to continue searching for new therapeutic agents and targets. Arsenic trioxide (ATO) induces apoptosis in all CLL cases and it could constitute an efficient therapy for this disease. We used Affimetrix microarray technology to characterize the global gene expression profile underlying CLL cell response to ATO, with the aim of identifying the ATO mechanism of action as well as new therapeutic targets in CLL.
Project description:B cell chronic lymphocytic leukemia (CLL) is often preceded by a benign monoclonal or oligoclonal CD5+ B cell lymphocytosis. We have generated transgenic mice expressing a catalytically inactive, dominant-negative recombination activating gene 1 (dnRAG1 mice) in the periphery. These animals develop an early-onset indolent CD5+ B cell lymphocytosis, caused in part by a defect in secondary V(D)J rearrangements initiated to alter autoreactive B cell receptor specificity. Hypothesizing that the CD5+ B cells accumulating in dnRAG1 mice represent a CLL precursor, we crossed dnRAG1 mice with CLL-prone Eµ-TCL1 mice to determine whether dnRAG1 expression in Eµ-TCL1 mice accelerates the onset of CLL-like disease. We find that CD5+ B cell expansion and CLL progression occurs more rapidly and uniformly in double-transgenic mice (DTG mice) compared to Eµ-TCL1 mice, but with similar phenotypic and leukemogenic features. To gain insight into genes or pathways responsible for CD5+ B cell accumulation in the transgenic mice, we performed comparative gene expression profiling studies using normal and CD5+ B cells isolated from wild-type and transgenic mice at either 12 weeks of age (pre-leukemia) or at CLL onset in DTG mice (using age-matched wild-type and single-transgenic mice as controls). These analyses confirm the upregulation of tolerogenic genes in CD5+ B cells and reveal a possible role for prolactin signaling in the regulation of receptor editing. This study suggests that a failure to remodel B cell antigen receptor genes in response to autoreactivity may promote the benign accumulation of CD5+ B cells, which may then be subjected to secondary genetic lesions that promote CLL progression.
Project description:B-cell chronic lymphocytic leukemia (B-CLL) is a heterogenous disease with a highly variable clinical course and analysis of ZAP-70 and CD38 expression on B-CLL cells allowed for identification of patients with good (ZAP-70-CD38-), intermediate (discordant expression of ZAP-70 and CD38) and poor (ZAP-70+CD38+) prognosis. In an attempt to identify a molecular basis that may underly this diverse clinical behaviour DNA microarray technology was employed to compare eight ZAP-70+CD38+ with eight ZAP-70-CD38- B-CLL cases. We used microarrays to detail the global programme of gene expression distinguising B-CLL from patient with good (samples 1 to 8) and poor prognosis (sample 9 to 16) and identified distinct classes of up- and down-regulated genes. Experiment Overall Design: To compare the transcriptosomes of good prognosis CLL cases (ZAP-70-CD38-) to poor prognosis cases (ZAP-70+CD38+), we purified CD19+ cells from peripheral blood samples by immunomagnetic isolation using MidiMacs, resulting in >95% purity of leukemic cells as detected by FACS analysis of CD19+CD5+ cells. The leukemic cells were freshly purified from untreated patients and RNA was directly isolated from fresh cells without further ex vivo treatment of the cells. Eight immunomagnetically purified peripheral blood derived ZAP-70+CD38+ CLL cases were compared with eight ZAP-70-CD38- B-CLL cases.
Project description:We conducted a genome-wide DNA methylation analysis in CD19+ B-cells from CLL patient and normal control samples using reduced representation bisulfite sequencing (RRBS). The methylation status of 1.8-2.3 million CpGs in the CLL genome was determined; about 45% of these CpGs were located in more than 23,000 CpG islands (CGIs). While global CpG methylation was similar between CLL and normal B-cells, 1764 gene promoters were identified as being differentially methylated between the two groups. Aberrant hypermethylation was found in all HOX gene clusters and a significant number of WNT signaling pathway genes. The genes that were frequently hypermethylated were typically associated with histone H3 lysine 27 tri-methylation or bivalent domains in normal B-cells. An additional 152 genes were found to be differentially methylated between normal naïve and memory B-cells. Of these 152 genes, 123 were hypomethylated in memory B-cells when compared to naïve B-cells. Overall, CLL B-cells had methylation patterns more similar to memory B-cells than naïve B-cells. Cluster analysis showed that the tissue-specific methylated genes separated CLL samples into two groups with differential ZAP70 methylation status. Hypomethylation occurred more frequently in the gene body including introns, exons, and 3'-UTRs in CLL. The hypomethylation in the NFATc1 P2 promoter and first intron correlated with up-regulation of both NFATc1 RNA and protein expression levels in CLL suggesting that an epigenetic mechanism is involved in the constitutive activation of NFAT activity in CLL cells. This comprehensive DNA methylation map will further our understanding of the epigenetic contribution to cellular dysfunction in CLL. To perform a genome-wide analysis of DNA methylation in CLL, we applied the Reduced Representation Bisulfite Sequencing (RRBS) to CD19+ B-cells isolated from normal control and CLL peripheral blood samples. The genomic DNA from each sample was digested with the methylation-insensitive restriction enzyme MspI (restriction site, CCGG) and ligated to Illumina sequencing adaptors containing methylated cytosine residues. The ligated MspI fragments were size-selected, treated with sodium bisulfite, and amplified by PCR. The PCR products were purified and sequenced using Illumina GAIIx sequencer with a read length of 52 or 76bp. 11 CLL B-cell samples, 3 normal control samples including one each of normal CD19+, CD19+/ IgD+ naïve, and CD19+/CD27+ memory B-cell sample and three CLL cell lines (Mec-1, Mec-2, and Wac-3) were used. We generated 20-30 million Illumina sequencing reads for each sample.