A Genome-Wide RNAi Screen to Identify Novel Genes Involved in Clonal Maintenance of ALL

Acute lymphoblastic leukemia (ALL) is the most common malignancy in children and adolescents where treatment is associated with significant morbidity. Novel therapeutic approaches to improve treatment outcome and minimize side-effects are therefore required. We have previously shown a lack of stem cell hierarchy and a high frequency of leukemic stem cells in ALL. Based on these findings, we have postulated that malignant self-renewal in ALL may be similar to lymphoid, but distinct from hematopoietic stem cell (HSC) self-renewal thereby providing new therapeutic targets for diseases affecting the lymphoid system. To identify targets and pathways involved in ALL self-renewal, we have performed a genome-wide RNAi screen in the ALL cell line SEM, which carries the translocation t(4;11) and expresses the MLL/AF4 fusion gene.

SEM cells were lentivirally transduced using the Decode™ pGIPZ negative selection library consisting of 7 pools of 10,000 shRNA constructs each. The pGIPZ vector allows for Pol II-dependent shRNA expression, GFP-based cell sorting and selection for puromycin resistance. Self-renewal was addressed by replating of transduced cells on puromycin-resistant murine stromal bone marrow feeder cells (M2-10B4), which mimics a niche environment in vitro. SEM cells were selected with puromycin and seeded at low cellular-density under serum-starved conditions to ensure adherence-dependent growth. SEM cells were replated onto new feeders and the puromycin selective pressure was kept throughout the experiments to minimize silencing of shRNA expression. Cells were harvested after the second plating (F2, ~30 days after transduction) and the prevalence of shRNA constructs in F2 was compared to samples harvested 6 days after transduction (baseline, BL) using next generation sequencing of genomic DNA.

Initial sequencing results obtained from one out of seven pools show that constructs targeting 637 different genes were differentially expressed (2 –fold change, padj <0.05). For instance, shRNAs targeting genes associated with lymphoid differentiation such as IKZF3 (B-cell progenitor differentiation) and CD20 (marker of B-cell differentiation) were enriched, whereas constructs targeting genes important for proliferation and survival such as BCL2 (survival), BRD4 (epigenetic reader, therapeutic target in leukemia), MAX (partner of MYC), BMI1 (polycomb complex 1 member, required for self-renewal) or POT1 (telomere integrity) were depleted. Furthermore, constructs targeting MAPK9/JNK2 and JUN in the mitogen-activated protein kinase pathway were depleted whilst the construct targeting DUSP10 (inhibitor of MAPK9/JNK2) was enriched, further proving that our screen has functional relevance within a biological context. Interestingly, the construct targeting the aryl hydrocarbon receptor (AHR) involved in B-cell maturation was also increased, which is in line with published data showing that inhibition of AHR drives stem cell maintenance.

These results demonstrate the feasibility of our approach where we have established a robust screening protocol that can identify genes involved in survival and self-renewal. Sequencing data from the additional six pools are currently being analyzed. These analyses are expected to provide important information about the key mechanisms by which leukemic blasts maintain their “stemness” in ALL.