Exploring PRMT5 As a Potential Therapeutic Target in Canine Lymphomas

Non-Hodgkin lymphoma (NHL) represents approximately 4 percent of all cancer diagnoses in the United States, with diffuse large B-cell lymphoma (DLBCL) accounting for approximately 40 percent of all new cases. There are numerous histologic (GC, NGC) and genetic (double hit/expressor) subtypes of DLBCL and the overall outcome of patients who receive standard therapy is heterogeneous. Lymphoma is also a common, malignancy in dogs, and while initial responses to combination chemotherapy show clinical responses, remission time is short and cures rare. Protein arginine methyltransferase 5 (PRMT5) is a type II protein arginine methyltransferase (PRMT) enzyme capable of driving symmetric demethylation of arginine residues on histones (H3(me2)R8; H4(me2)R3) and other proteins such as P53, and the NFkB subunit p65. PRMT5 is overexpressed and dysregulated in both solid and hematologic tumors and plays a key role in the driver activity of MYC, CYCLIND1 and NOTCH in lymphoma models. Experimental therapeutic strategies aimed at targeting PRMT5 activity have led to numerous, highly selective inhibitors that are currently being translated into the clinic with several clinical trials underway or in development. Here, we characterized patterns of PRMT5 expression and correlated these with histologic subtype in canine lymphoma tissue microarrays (TMAs, n=337 lymphoma specimens). We characterized expression of PRMT5 and its symmetric dimethylation histone marks in three canine lymphoma-derived cell lines, CLBL-1, 17-71, and OSW. Treatment of PRMT5 with a highly selective small molecule PRMT5 inhibitor (CMP220) was performed to determine the dose-dependent effects (1nM-10uM range) on proliferation (MTS assay), viability (annexin V/PI flow cytometry) and biomarkers of PRMT5 activity (SDMA, ADMA, histone symmetric demethylation). CMP220 showed PRMT5 inhibitory activity that was highly selective in a methyltransferase screening assay with 37 enzymes/complexes. TMAs of all histologic subtypes of lymphoma showed over-expression of PRMT5 in 96% of specimens. Canine DLBCL showed variable over-expression in the cytoplasmic compartment (48.8% strong, 50.0% weak, n = 165) compared to negative or weak staining in normal and hyperplastic lymph nodes (n = 40). Nuclear staining was observed primarily in lymphomas of T cell lineage (PTCL, and pre T lymphoblastic lymphomas). Primary and cell line lymphoma samples showed PRMT5 over expression by Western blot and RT-PCR. PRMT5 inhibition showed a dose-dependent decrease in symmetric dimethylarginine (SDMA) and symmetric demethylation of histone H4 arginine-3 (H4(me2s)R3) with no effect on asymmetric dimethylarginine (ADMA). Interestingly, in contrast to PRMT5 inhibition in human lymphomas, symmetric demethylation of histone H3 arginine-8 (H3(me2s)R8) showed no change with PRMT5 inhibition. The PRMT5 small molecule inhibitor CMP220 inhibited growth of CLBL-1 (IC₅₀ of 123.2 nM at 6d) and 17-71 (IC50 of 100 nM at 6d). The OSW PTCL cell line showed profound resistance to PRMT5 inhibition with little cell death observed at 10 uM. Ex-vivo canine patient samples showed suppression of growth in a time and dose-dependent fashion. We have demonstrated that PRMT5 is expressed in canine B cell primary lymphomas and cell lines and that PRMT5 inhibition leads to growth suppression and induction of apoptosis. We are currently exploring genome-wide recruitment of PRMT5 on chromatin and examining chromatin and whole transcriptome changes that occur in canine lymphoma cell lines treated with PRMT5 inhibitors. This data provides justification for incorporating the canine lymphoma model into the preclinical development of PRMT5 inhibitors.