P327: loss of polycomb repressive complex 2 function causes asparaginase resistance in t-acute lymphoblastic leukaemia through decreased wnt pathway activity

Background: While mutations in epigenetic factors predict poor treatment response in T-acute lymphoblastic leukaemia (T-ALL), a mechanistic explanation of resistance to specific therapies is lacking. Polycomb Repressive Complex 2 (PRC2) alterations often co-occur with activating kinase mutations, but the molecular interplay between epigenetic and signalling pathways is poorly understood. Aims: To assess molecular interactions between epigenetic and signalling pathway mutations in T-ALL, and to determine the implications for treatment response. Methods: We first leveraged clinical mutational and transcriptional data to identify potential functional links between epigenetic and signalling factors in T-ALL. We then used CRISPR-Cas9 editing to create an isogenic model of EZH2 knockout (KO) in the Jurkat T-ALL line and performed RNA-seq, proteomics and phospho-proteomics to decipher the effects of PRC2 depletion on signalling pathways. Treatment responses and resistance mechanisms were investigated in EZH2 KO cells and orthogonal models. Results: A stringent statistical method (DISCOVER) identified genetic interactions between epigenetic and signalling alterations in a T-ALL patient cohort. Significant hits included co-occurrence of PRC2 mutations with MAPK and JAK-STAT factors mutations and novel associations with WNT pathway alterations (e.g., NLK). Comprehensive molecular profiling of the isogenic model allowed us to explore these associations further. Using the PROGENy tool to interrogate transcriptional readouts in WT and EZH2 KO cells revealed links between PRC2 loss and altered activity of several signalling pathways, including markedly reduced WNT pathway activity. These transcriptional changes correlated with protein measurements by mass spectrometry and immunoblotting. Analysis of phospho-proteome profiles by kinase set enrichment analysis (KSEA) confirmed significant changes in multiple signalling pathways, again identifying reduced WNT activity. We used the CausalPath tool to integrate transcriptomic, proteomic, and phospho-proteomic profiles from WT and EZH2-deficient cells, identifying key nodes mediating signalling changes in EZH2-null cells and confirming a central role for aberrant WNT activity in PRC2-depleted T-ALL. Transcriptional analysis of a patient cohort identified similar patterns of downregulated WNT pathway targets in PRC2-altered cases as in our cell line model, suggesting that these findings reflect the T-ALL molecular landscape in vivo. To link these results to the clinic, we tested responses of WT and EZH2-null T-ALLs to standard leukaemia therapies, finding markedly reduced asparaginase sensitivity in PRC2-depleted cells. This led us to explore the activity of the non-canonical WNT-dependent stabilization of proteins (WNT/STOP) pathway, as this has been shown to mediate asparaginase resistance by providing a reservoir of ubiquitinated proteins that bolster asparagine reserves. We confirmed that reducing PRC2 function by independent methods (gene editing, enzymatic inhibition and PRC2 degraders) in different cell types caused marked increases in cellular ubiquitination in line with decreased WNT/STOP activity. Finally, we found that proteasome inhibition abrogated asparaginase resistance in EZH2-deficient cells, confirming a direct link with WNT/STOP-mediated ubiquitination in PRC2-altered T-ALL, and providing a potential therapeutic avenue to improve asparaginase sensitivity in these cases. Summary/Conclusion: We provide a holistic analysis of signalling pathway changes in PRC2-altered T-ALL, identify a mechanistic explanation of resistance to a key blood cancer treatment, and suggest immediate solutions to overcome this important clinical problem. Keywords: Epigenetic, Polycomb, Asparaginase, T cell acute lymphoblastic leukemia