Abstract
Peripheral T cell lymphomas (PTCL) of γδ T cell origin including subsets of hepatosplenic T cell lymphoma (HSTL) and enteropathy associated T cell lymphoma (EATL) are characterized by short survival with multiagent chemotherapy programs compared to other PTCLs. Additionally, the molecular and genetic events underlying the development of HSTL and EATL are poorly understood, thus hampering development of more effect novel therapies in these lymphoma subtypes. Given this problem, we sought to define and characterize genetic events associated with HSTL/EATL.
Through whole exome sequencing of 150 HSTL/EATL cases along with paired normal control tissues we found the histone methyltransferase SETD2 to be the most frequently mutated tumor suppressor in HSTL and EATL (>40% of cases in both entities). Conversely, SETD2 mutation was found to be rare in other PTCL subtypes. HSTL/EATL SETD2 mutations frequently occurred as nonsense and frameshift mutations and more than half had evidence of biallelic mutation, suggesting these mutations constitute a loss of function phenotype. Accordingly, we found that overexpression of representative mutant SETD2 (c.T4361C/F1544S and c.7571_7572delGT/C2524*) in HEK293 cells resulted in decreased downstream H3K36me3 compared to wild type SETD2 controls. We further investigated the ramifications of SETD2 functional loss in the only known HSTL cell line, DERL2. DERL2 cells were transduced with lentiviral constructs bearing inducible SETD2 (N=2) or non-silencing control shRNAs. SETD2 shRNA induction in DERL2 cells resulted in >75% loss of SETD2 mRNA expression compared to non-silencing controls by quantitative PCR and resulted in potent depletion of both SETD2 and H3K36me3 protein by Western blotting. We then performed RNA sequencing of DERL2 cells after SETD2 knockdown compared to non-silencing controls. Gene expression analysis and gene set enrichment analysis (GSEA) revealed that SETD2 knockdown was associated dysregulation of multiple oncogenic pathways involved in cell cycle, mitotic, G2/M and NF-kB pathways (FDR <0.1 for each). DERL2 cells post SETD2-knockdown also revealed increased cellular proliferation in in vitro culture and in methylcellulose colony forming assays as compared to non-silencing controls (p<0.01). This effect was not noted in T lymphoblast Jurkat cells engineered in similar fashion for SETD2 loss, suggesting SETD2 loss favors γδ T cell expansion specifically in a context-specific manner. Overall these experiments thus confirm an association between the acquisition of oncogenic gene expression pathways with SETD2/H3K36me3 loss and enhanced cell survival and proliferation in malignant γδ T cells.
To understand effects of SETD2 loss in vivo, we quantitated progenitor and mature T cell subsets in T cell specific SETD2 (Setd2fl/fl, lck-cre) knockout mice. Sorted T cells from 6-8 week old mice but not other tissues derived from Setd2fl/fl, lck-cre mice showed lack of SETD2 protein and low H3K36me3/2 ratios consistent with lineage-specific SETD2 loss. In this model, we found 2.2 fold expansion (p=0.016) of γδ T cells in mesenteric lymph nodes and Peyer patches of Setd2fl/fl, lck-cre compared to controls. Analysis of gene expression in murine T cell progenitors in normal mice without SETD2 loss revealed upregulation of SETD2 mRNA in thymic double positive cells, prior to αβ lineage commitment. Taken together with the results of our in vitro experiments, this suggests that SETD2/H3K36me3 may control normal T cell progenitor fate and γδ T cell expansion through regulation of specific gene expression programs.
In summary, we describe SETD2 functional loss resulting in attenuation of H3K36 trimethylation as a frequent event in gamma delta T cell derived lymphomas. By gene expression analysis and functional studies in murine and human cancer cell line models, we show that SETD2 is important both in the regulation of γδ versus αβ T lymphocyte developement and has a direct role in fostering proliferation of malignant cells derived from the γδ T cell compartment. This appears in part due to loss of epigenetic control of diverse gene expression programs. This work has direct translational importance as we define a novel role of the SETD2/H3K36me3 axis in controlling global T cell differentiation and we identify a potential target for synthetic lethal approaches in malignancies such as HSTL and EATL.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal