Loss of Phospholipase C Gamma 1 (PLCG1) Impairs AML1-ETO Induced Leukemia Stem Cell Self-Renewal

Chromosomal translocations found in acute myeloid leukemia (AML) can generate oncogenic fusions with aberrant epigenetic and transcriptional functions. However, direct therapeutic targeting of leukemia fusion proteins has not been accomplished so far. Although high remission rates can be induced in patients diagnosed with AML1-ETO/t(8;21)-positive AML only half of them achieve long-term disease-free survival (Papaemmanuiel et al., NEJM, 2016). In the other half of these patients, the disease maintaining leukemia stem cell (LSC) clone is not eliminated by chemotherapy. A functional characteristic of LSCs is unlimited self-renewal capacity and several signaling pathways have been identified that maintain stem cell self-renewal. Targeting the oncogene induced self-renewal capacity of LSCs has great potential to eliminate the malignant clone and prevent relapse.

To identify oncogenic cellular functions with relevance for LSC self-renewal, we performed global proteome profiling in murine AML1-ETO9a (AE) compared to MLL-AF9 (MA9) driven LSCs. Gene set enrichment analyses revealed a significant enrichment of calcium-dependent cellular functions and Phospholipase C (PLC)-signaling in AE LSCs. These data could be confirmed in sorted CD34+ blasts from AE-positive AML when compared to non-AE-AML. All PLC family members are regulators of Ca²⁺ homeostasis. However, when analyzing published AML gene expression datasets we found exclusively PLCG1 to be highly expressed in t(8;21) AML. Conditional activation of AE in embryonic stem cells resulted in induction of PLCG1 expression and PLCG1 was identified as a direct target of the AE fusion by ChIP-sequencing in AE-positive Kasumi-1 cells.Here, PLCG1 depletion resulted in reduced Ca²⁺ release, impaired proliferation and reduced colony formation in vitro. In a xenograft model, inactivation of PLCG1 resulted not only in delayed disease development (median survival shNT vs. shPLCG1: 135 days vs. not reached, p=0.02) but also in reduction of disease penetrance by 87%. Consistent with these results, transcriptome analysis revealed strong induction of gene sets related to myeloid differentiation and down-regulation of gene sets linked to proliferation, stemness and c-Myc targets.

To confirm the functional role of PLCG1-signaling in AE-driven LSCs, we generated a new conditional knockout mouse model for Plcg1 and induced leukemia using the oncogenes AE and KRAS-G12D (AE/K). Genetic inactivation of Plcg1in vivo after engraftment of leukemic cells resulted in significant reduction of LSC numbers (p=0.04) and a reduction of disease penetrance by 67% in primary recipients. Isolated LSCs revealed induction of differentiation, loss of cell cycle activity and failed to re-establish disease in secondary recipients (Plcg1+/+ vs. Plcg1-/-: median survival 12 days vs. not reached; p=0.0001). In contrast, genetic deletion of Plcg1 appeared to be dispensable for normal murine HSC function during primary and secondary transplantation. Primary human t(8;21) AML cells (derived from 4 different donors) showed impaired colony forming capacity following PLCG1 inactivation in vitro irrespective of co-occurring mutations while colony formation of human CD34+ BM cells was not affected to a major extent.

As Ca²⁺ signaling appeared deregulated in t(8;21) AML, we aimed to investigate the effects of pharmacologic Ca²⁺ inhibition as a tractable target downstream of PLCG1. To assess specifically for LSC function, we treated primary recipient mice with established AE/K-driven leukemia with the clinically approved calcineurin inhibitor ciclosporin (CsA), a compound that blocks intracellular Ca²⁺ release. CsA-treated animals showed reduction in total leukemic burden (spleen weight diluent vs. CsA, p=0.01) and LSC numbers (p=0.02). This resulted in increased survival of secondary recipient hosts (diluent vs. CsA: median 15 vs. 29 days, p=0.0002). These effects could not be observed for other oncogenes (e.g. MA9), confirming its specificity for AE-induced disease. Consistently, CsA treated primary human t(8;21)-positive AML blasts failed to form colonies in methylcellulose.

In summary, our findings identified PLCG1-dependent Ca²⁺ signaling as a critical pathway for t(8;21) LSC maintenance and self-renewal. Most importantly, as PLCG1 is dispensable for maintenance of normal HSPCs, PLCG1 could serve as a novel therapeutic target in t(8;21) AML.