Report
Teaser, summary, work performed and final results
Periodic Reporting for period 3 - EpiTALL (Dynamic interplay between DNA methylation, histone modifications and super enhancer activity in normal T cells and during malignant T cell transformation)
Teaser
T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive hematological cancer that arises from the malignant transformation of T-cell progenitors. Currently, T-ALL patients are still treated with specific combinations of harmful chemotherapeutic agents. Indeed, studies of...
Summary
T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive hematological cancer that arises from the malignant transformation of T-cell progenitors. Currently, T-ALL patients are still treated with specific combinations of harmful chemotherapeutic agents. Indeed, studies of the long-term effects of chemotherapy in patients with T-ALL showed that recent gains in leukemia-free survival have been achieved at the cost of significant increases in the rates of life-threatening and debilitating toxicities. Thus, further advances in the treatment of T-ALL require the development of effective and highly specific molecularly targeted antileukemic drugs.
In this project, we aim to study the chromatin architecture during normal and malignant T cell differentiation in order to define how DNA methylation can drive oncogenic gene expression. Indeed, we hypothesize that T-ALL patients are characterized by a unique DNA methylation signature and that specific targeting of the DNA methylation machinery might serve as a promising new therapeutic strategy in the treatment of this aggressive haematological malignancy.
Work performed
Aim 1: Genome-wide DNA methylation in normal and malignant T-cells.
Here, by studying DNA methylation patterns in normal and malignant T cells, we identified 2 main categories of T-ALL that differ based on their CpG island and Open Sea DNA methylation levels, ie. COSMe type I and type II (Figure 1). These epigenetic subtypes divide T-ALL in 2 categories which, at least in part, overlap with known genetic features of this disease. Furthermore, we identified a link between aberrant DNA methylation and treatment outcome, with COSMe type I tumors showing an increased frequency of early relapses.
Aim 2: The effect of DNMT inhibition on DNA methylation and chromatin architecture
Here, we used targeted bisulfite sequencing to specifically interrogate DNA methylation patterns at single nucleotide levels at specific regions of interest such as promoters, enhancers, CTCF binding site and transcription factor binding sites. In addition, we are specifically investigating the putative contributions of DNMT3A and/or DNMT3B to the observed phenotypes. For this, we have validated different functional shRNAs against both DNA methyltransferase and study their putative effect on the specific DNA methylation changes identified above.
Aim 3: Study the functional role of DNA methylation during normal development
Here, we aimed to use the OP9-DL1 co-culture system to study the effect of DNMT inhibition on normal T cell differentiation. However, and given that decitabine also affects normal stromal cells, we optimised a recently established stromal free DLL4 culture system in which early T cell precursors can also be generated from HSCs through exposure to the immobilized Notch ligand DLL4 with appropriate cytokines. This system is operational in the lab, and is used to study the effect of DNMT inhibition on normal T cell development.
Aim 4: Synergism between DNMT and chemotherapeutic agents in human T-ALL
Here, we performed preclinical in vivo Decitabine treatment experiments to show that primary T-ALL xenografts are uniformly sensitive to this DNA methylation inhibitor (Figure 2). Next, and given that we failed to observe in vivo synergy between Decitabine and the BRD4 inhibitor JQ1, we are now focused on the combination therapy of Decitabine with other chemotherapeutic agents which are currently used in the clinic for the treatment of human T-ALL.
Aim 5: Study the oncogenic role of Dnmt3b in T-ALL in vivo
In this aim, we showed that Dnmt3b over expression initially antagonises LMO2 driven murine T-ALL (Figure 3). However, at the stage of full blown tumor development, we have also shown that some tumours become dependent on Dnmt3b for their leukaemia survival. We are trying to understand the molecular mechanisms that might enable us to decipher these paradoxical findings.
Final results
Despite significant insights in T-ALL biology and major advances in leukemia treatment, a significant proportion of T-ALL patients will suffer from relapses or may ultimately die from their disease. In this project, we study genome-wide patterns of DNA methylation in human T-ALL with the ultimate goal of identifying a subset of leukemia patients that might eventually benefit from treatment with DNMT inhibitors, such as decitabine. Therefore, the potential socio-economical implications of this project are largely reflected in the creation of unique opportunities to address the urgent clinical need for new therapeutic opportunities in this aggressive haematological malignancy.
In addition, this study puts particular emphasis on the oncogenic role of the DNA methyltransferase B (DNMT3B) in the pathogenesis of human T-ALL. Indeed, we generated a novel DNMT3B conditional knockin mouse model to study the oncogenic role of DNMT3B in vivo. However, recent literature has also shown that DNMT3B might have a role in other tumor entities including acute myeloid leukemia as well as melanoma. Beyond the state of the art of this project, we are also using this model to study these other pathologies.
As mentioned above, the current study is focused on the analysis of genome-wide 5mC patterns in normal and malignant T-cells. However, we are currently extending our analysis towards other DNA modifications, more specifically 5-hmC (hydroxymethylation). We are using both Mass Spec based technology to identify the total amount of mC and hmC in specific samples of interest, or using hMedIP to study the genome wide patterns of 5hmC in normal and malignant T cells. Indeed, TET and IDH enzymes, which are critically involved in the determination of the 5hmC levels in a cell, are mutated in a fraction of human T-ALLs, suggesting that studying hydorxymethylation in the context of T-ALL might provide additional insights in the biology of this disease.