The RNA Binding Protein Musashi-2 (MSI2) Sustains the Growth and the Leukemogenic Potential of MLL-Rearranged Acute Lymphoblastic Leukemia, and It Is Involved in Glucocorticoid Resistance

Abstract

Acute Lymphoblastic Leukemia occurring in infants and carrying the rearrangement of KMT2A, aka MLL (MLLr infant ALL) is a rare but very aggressive leukemia typically associated with poor outcome. Patients suffer from a dismal prognosis (overall survival: 20-40%) mainly due to the resistance to conventional therapy and a high incidence of relapse. The biological mechanisms involved in the pathogenesis and disease progression of infant MLLr leukemia are not completely understood and the identification of druggable genes is urgently needed for the development of novel therapeutic strategies. The RNA-binding protein Musashi-2 (MSI2) is a post-transcriptional regulator of mRNA translation to proteins. MSI2 has a pivotal role in normal hematopoiesis and in a variety of tumors, including leukemia, where it is involved in cell proliferation, differentiation and maintenance of the (cancer) stem cell pool. Although several studies strikingly demonstrated the involvement of MSI2 in chronic and acute myeloid leukemia, with regards to B-ALL, few papers reported over-expression of MSI2 associated with a poor outcome in both pediatric and adult B-ALL patients; however functional data are still missing. By retrospectively analyzing the gene expression profile of a large cohort of Italian patients with pediatric leukemia (n=314) we observed that MSI2 is highly expressed across the different cytogenetic subgroups, and the level of expression is even higher in ALL compared to AML cases; thus reinforcing the hypothesis that MSI2 might have a crucial role also in ALL. To unravel the functional role of MSI2 in MLL-r ALL, we successfully generated a model of MSI2 knock-out (MSI2 KO) using CRISPR/CAS9 genome editing in the MLL::AF4+ ALL cell line SEM. In a long-term competitive assay in vitro, in which the MSI2 KO and CNTRL cells were co-cultured at the initial 90:10 ratio (KO:CNTRL), we observed that after 20 passages in culture the CNTRL counterpart overgrew the MSI2 KO cells (which indeed extinguished); thus demonstrating that the lack of MSI2 leads to a proliferation disadvantage. In agreement with this, MSI2 KO cells show an impaired leukemia-initiating capacity and a lower disease burden in vivo, and mice transplanted with MSI2 KO cells display a prolonged survival compared to controls. Also, we performed a high-throughput drug screening using a manually curated library of 174 compounds (FDA-approved, currently used in the clinic) and we observed that MSI2 KO cells show a marked sensitivity to Glucocorticoids (GCs). Further validation studies in vitro demonstrated that MSI2 KO cells were significantly (100-fold) more sensitive to Prednisolone or Dexamethasone than their MSI2-expressing counterpart. In addition, the treatment of SEM cells (typically resistant to GCs) with the MSI2 inhibitor Ro 08-2750 phenocopied the sensitization to GCs observed in CRISPR-edited MSI2 KO cells, showing synergistic effects in combination with Prednisolone or Dexamethasone. As infant MLLr ALL patients are typically resistant to GCs, our findings suggest that targeting MSI2 might represent a potential novel approach to circumvent the therapy resistance. Finally, by further investigating the biological mechanisms of GC-sensitization, we unraveled a potential role for MSI2 in the regulation of the energetic metabolism of leukemic cells. Indeed, the confocal analysis of mitochondrial membrane potential and the profiling of cellular bioenergetics by Seahorse technology revealed that MSI2 KO cells display mitochondrial hyperactivity at steady-state compared to controls. Upon GCs exposure (which inhibits glycolysis in both MSI2 KO and CNTRL cells) mitochondrial respiration is increased in CNTRL cells to compensate ATP production, while it is severely impaired in MSI2 KO, leading to a mitochondrial crisis and energy failure. The importance of metabolic plasticity of tumor cells in adaptation to stress conditions is emerging as a crucial aspect of cancer biology. The putative role of MSI2 in regulating the energy metabolism of leukemia is a novel finding not previously investigated, which may pave the way for targeting the metabolic vulnerabilities of ALL as a novel therapeutic approach. In conclusion, our study demonstrates that MSI2 sustains the growth in vitro and the leukemogenic potential in vivo of MLLr ALL cells, and it is involved in GC-resistance by regulating the energy metabolism of leukemia.