Demosthenous, Christos

Abstract Mantle cell lymphoma (MCL) is an aggressive B-cell non-Hodgkin lymphoma characterized by rapid disease progression. Although the clinical outcome of MCL patients has improved in the recent years, relapses and progressive emergence of resistance to treatment are common. Therefore, alternative therapeutic strategies are in great need for the treatment of MCL patients. Polycomb repressive complex 2 (PRC2), the mammalian enzymatic complex plays crucial roles in regulation of normal and malignant hematopoiesis. Enhancer of zeste homologue 2 (EZH2), the catalytic subunit of PRC2 complex is frequently amplified or overexpressed in most solid tumor types. However, EZH2 has been reported to harbor mutations of tyrosine 641 to asparagine (Y641N) in 22% of GCB diffuse large B-cell lymphoma (DLBCL) and 7% of follicular lymphoma (FL), but not in MCL. In the current study, we sought to elucidate the role of PRC2 in MCL and its implications for epigenetic therapy. We first assessed the expression of PRC2 components EZH2, SUZ12 and EED in several MCL cell lines (Jeko, Mino, Granta, JVM2 and Z138) and normal B-cells. Western blot analysis demonstrated that SUZ12 and EED along with EZH2 are overexpressed in proliferating MCL cell lines but not in normal B-cells. Furthermore, aberrant expression of H3 trimethylation (H3K27me3), di-methylation (H3K27me2) and mono-methylation (H3K27me1) was observed in all five MCL cell lines tested as compared to normal B cells. In order to understand if MCL cells have stable and enzymatically active PRC2 methyltransferase complex, we performed co-immunoprecipitation assay using EZH2 antibody and IgG as control. Significant amount of EED and SUZ12 enrichment was observed in the EZH2 immunoprecipitates from Granta and Mino MCL cells. Collectively this data demonstrates that in the MCL cells, PRC2 complex is aberrantly active and is associated with EZH2 upregulation. To study the role of EZH2 and EED in MCL cells, we generated expression constructs of the full-length and deletion fragments EZH2-F (full length EGH2), EZH2-del (EZH2 with deletion of SET domain), EED-F (full length EED) and EED-del (EED with deletion of WD40 domain), respectively. Interestingly, overexpression of EEDWT and EZH2WT led to a significantly (p=0.05) increased cell proliferation resulted in increased H3K27 trimethylation as compared to cells transfected with empty vector. Co-transfection with EZH2-del (SET domain) with the EZH2-F attenuated EZH2-F induced H3K27me3 and cell proliferation, suggesting oncogenic role of EZH2 enzymatic activity. Interestingly, co-transfection of EED-del (deficient in WD40 domain) had no effect on EED-F induced H3K27Me3, and the proliferation was less pronounced as compared with EED-F alone. EZH2 gene inhibition with EZH2-shRNA inhibited the growth of the MCL cells. Pharmacological inhibition of global methyltransferase activity by DZNep or EZH2 specific inhibitor GSK343 greatly suppressed the H3K27me3 level with little effect on H3K27me1 in Granta cells. However, treatment with GSK343 (but not DZNep) inhibited interaction between EZH2 and EED. Furthermore, treatment with EZH2 inhibitor, GSK343, produced significant growth inhibition (more than 90%) at 20uM concentration in Mino, Granta and Z138 cells. Taken together, these results suggest wild-type EZH2 is an epigenetic target for MCL, and MCL cells are highly sensitive to the EZH2 inhibition regardless of EZH2 mutations. Our findings indicate the use of PRC2 and EZH2 epigenetic inhibitors for the treatment of MCL patients. Overall, this study implies that overexpression of wild-type EZH2 has an oncogenic role in MCL, and suggest that wild-type EZH2 may confers similar active PRC2 addiction as mutant-EZH2 does in DLBCLs, FLs and other malignancies. Disclosures No relevant conflicts of interest to declare.

Abstract Diffuse large B-cell lymphoma (DLBCL) is the most common type of non-Hodgkin Lymphoma and is characterized by deregulation of several signal transduction pathways. Approximately 50% patients with DLBCL are shown to have aberrant activation of the signal transducer and activator of transcription (STAT) pathway. However, mechanism of aberrant STAT3 signaling in DLBCL is not well understood. Protein tyrosine phosphatases (PTPs) are important enzymes that control the activity of multiple signaling pathways downstream of tyrosine kinases; mutations in PTPN11 have been found associated with the development of myeloproliferative disorder. In the current study, we sequenced a set of 38 DLBCL tumor samples for genetic mutations of PTPN6. Bi-directional sequencing reactions were performed on genomic DNA, and all exons of PTPN6 were amplified and analyzed by Sanger sequencing. We identified 2 novel heterozygous missense mutations in PTPN6 gene in 2 separate patients (2/38; 5.2%). The first missense mutation occurred in the exon 7, resulted in an Asp226 to lysine substitution (N226K), while other missense mutation found on exon 15 resulted in an Ala552 to valine substitution (A552V). In order to elucidate the functional significance of these mutations, we performed site directed mutagenesis to mutate wild type (WT) PTPN6 at codon 226 and 552. Both, wild-type PTPN6 (PTPN6WT) or those containing the N226K (PTPN6N226K) and A552V (PTPN6A552V) mutations were cloned and stably expressed in the HEK-293T cells with a lentiviral expression vector and PTPN6 phosphatase activity was measured. Our results showed marked reduction (70%) in the PTPN6 phosphatase activity in both the PTPN6 mutants as compared to the WT control. Interestingly both the PTPN6 mutants (PTPN6N226K and PTPN6A552V) promoted cell proliferation of HEK-293 cells as compared to WT PTPN6. In order to evaluate that whether or not PTPN6 mutations modulate STATs signaling, we analyzed STAT1, STAT3, STAT5 and STAT6 phosphorylation at tyrosine residue in cells stably transfected with WT or mutant PTPN6. While overexpression of WT PTPN6 reduced STAT3 phosphorylation, however no effect was observed on STAT1, STAT5 and STAT6 phosphorylation. Unlike WT-PTPN6, none of the PTPN6 mutants was able to dephosphorylate STAT3. In addition, PTPN6 A552V mutation was found to decrease the ability of PTPN6 to dephosphorylate STAT3 after induction with interleukin-10. Furthermore JAK2/JAK1 inhibition through the pharmacological inhibitor ruxolitinib (5μM) dephosphorylated STAT3 in WT PTPN6 as well as in PTPN6 mutants within 4 hours. Inhibition of JAK3 through a pharmacological inhibitor WHI-P154 (5μM) was able to completely dephosphorylated STAT3, however both the PTPN6 mutants were found resistant to JAK3 inhibition. Moreover, in a luciferase reporter assay compared to WT PTPN6, expressing either of the PTPN6 mutants (N226K and A552V) enhanced transcriptional activity of STAT3. To determine the effect of the PTPN6 mutations on the downstream targets of the JAK/STAT pathway, we demonstrated that Mcl-1, survivin and Bcl-2 protein level was significantly elevated in both mutant cells than that seen in the WT PTPN6 cells. Overall these results demonstrate that the N226K and A552V PTPN6 mutations occur in 5% of DLBCL tumors. Mutations we have found are both able to cause loss-of-function of PTPN6 that leads to increased STAT3 phosphorylation and increased transcriptional activity resulting in accumulation of Mcl-1, Bcl-2 and survivin proteins. Our data suggest that clinical trials of JAK1/JAK2 inhibitors need not be restricted to tumors with PTPN6 mutations but rather should focus on tumors with demonstrated aberrant STAT3 activation. Taken together, our data suggest that PTPN6 mutations may guide the clinical use of inhibitors of JAK3 kinase in DLBCL. Disclosures No relevant conflicts of interest to declare.