Protein arginine methyltransferase 5 (PRMT5) is a highly conserved enzyme that catalyzes monomethylation and symmetric dimethylation of arginine residues on many target proteins, including histones and non-histone proteins (P53, P65/NFκB, E2F1, AKT and spliceosome proteins). (Shailesh, 2018) PRMT5 is overexpressed in numerous solid and hematologic malignancies. (Yang, 2013) It is associated with several oncogenic mechanisms, from driving the malignant phenotype to promoting tumor cell survival and invasiveness. While preliminary data has shown PRMT5 to be upregulated in T-ALL/LBL, the mechanism by which PRMT5 promotes T-ALL/LBL has yet to be established. (Mei, 2019) Preclinical transgenic models of leukemia and lymphoma driven by a variety of oncogenes (MYC, NOTCH) have identified PRMT5 activity as a critical component for the driver phenotype. (Li, 2015) To test the hypothesis that overexpression of this enzyme is capable of malignant driver activity, we developed a transgenic (Tg) mouse model designed to over-express PRMT5 in the lymphoid compartment. (Figure 1) Tg mice were created by injecting a linearized pBH-Eμ-PRMT5 vector into FVB/N pronuclei and implanted into pseudo-pregnant FVB/N mice. We obtained five founder lines crossed with wild-type FVB/N mice to obtain the F1 generation. RT-PCR for PRMT5 showed preferential overexpression in Tg B lymphocytes compared to control FVB/N mice (p<0.01) but also overexpression in T and NK cell compartments due to leaky Eμ phenotype. We observed a high percentage of F1 progeny from all five founder lines to develop spontaneous lymphoproliferative disease compared to control mice (p<0.001). Immunophenotypic and pathologic evaluation was performed by flow cytometry, immunohistochemistry, and clonality by IgH and TCR Vβ PCR analysis. Pathologic features showed a range of lymphoid neoplasms, with pre-T lymphoblastic immunophenotype being the most common. Tumor-bearing animals demonstrated diffuse involvement of all organs, peripheral blood, and bone marrow. Pre-T lymphoblastic lymphomas (LBL) showed expression of cytoplasmic CD3, surface CD4 (dim), TdT, and variable Ki67 with constitutive overexpression of PRMT5, MYC, CYCLIN D1, Ki67, and CD99. We developed a T-LBL cell line (Tg813) derived from a thymic tumor that displayed clonal TCR Vβ.17 gene rearrangement, intracellular TdT and CD3 expression with dim surface CD4 expression, immunophenotypic characteristics consistent with T-ALL/LBL. This cell line was maintained in culture and successfully passaged in immune-competent FVB mice, which led to disseminated disease, suggesting that PRMT5 drives an immune escape program in vivo. Preliminary data demonstrate that tumor-bearing animals display a prominent population of normal and neoplastic T-cell subsets with strong PD1 expression. PD1 expression in T-ALL/LBL has been linked with oncogenic function (Yang, 2019) and, in normal T-cell subsets, is a marker for exhaustion. (Lee, 2015) In addition, treatment of our Tg813 in vitro model with PRMT5 knockdown (CRISPR CAS9) and selective inhibition with a SAM-competitive inhibitor (C220 and PRT382, Prelude Therapeutics) showed a reduction of global symmetric dimethyl arginine, loss of the PRMT5 epigenetic mark H4R3(Sme2) and decreased proliferation with cell cycle arrest by day 6, findings consistent with PRMT5-dependent driver activity. Engraftment of Tg813 into SCID and immunocompetent FVB/N mice led to disseminated lymphomas (mean: 21 days). Tg813 engrafted animals treated with C220 (10mg/kg) showed improved survival compared to vehicle-treated mice (p<0.001). This preclinical model supports the hypothesis that PRMT5 over-expression provides cancer driver activity and a unique opportunity to study the role of this oncogene in an immune-competent, in vivo model system. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal
Read full abstract