Intersecting High-Throughput Screens Identifies SERCA As a Target for Modulating NOTCH1 In Hematopoietic Malignancies

Abstract 555

While long-term survival rates for children with acute lymphoblastic leukemia (ALL) are approximately 80%, we have reached a plateau over the past decade with combination chemotherapy. ALL remains a leading cause of cancer-related death in children, and even for those children cured, morbidity is significant. Moreover, prognosis for adults with ALL is poor. Alternative approaches to treating this disease are needed. NOTCH1 is a well-validated target in hematopoietic malignancy, with NOTCH1 activating mutations identified in over 50% of T-cell acute lymphoblastic leukemia (T-ALL) and in 12% of chronic lymphocytic leukemias. NOTCH1 is a heterodimeric protein, normally activated by a series of proteolytic cleavages upon ligand stimulation, which forms a transcription factor complex altering gene expression. We intersected the results of two high-throughput screens focused on modulating NOTCH1 to identify new NOTCH1 inhibitors. First, we applied an unbiased, cell-based gene expression signature screening approach to identify new NOTCH1 inhibitors (Gene Expression-based High-throughput Screening (GE-HTS)) in which a gene signature serves as a surrogate for inactive NOTCH1. Among top-scoring compounds inducing the NOTCH1 inactivation signature, we identified inhibitors of SERCA (Sarco/endoplasmic Reticulum Ca⁺⁺ATPase) channels, such as the natural products thapsigargin and cyclopiazonic acid. We integrated these results with a cDNA library screen for NOTCH1 activation in which U2OS cells expressing a leukemic NOTCH1 allele, L1601PδP, were scored for NOTCH1-sensitive RBPJ luciferase reporter activity. This screen also identified the SERCA genes ATP2A2 and ATP2A3 as top candidates, whose expression activates NOTCH1, complementing our original observation of SERCA inhibition as an approach to inhibit NOTCH1.

SERCA proteins reside on the endoplasmic reticulum and are responsible for the transfer of calcium from the cytosol of the cell to the lumen of the endoplasmic reticulum. We focused our attention on the more potent molecule, thapsigargin, and confirmed that it induces the NOTCH1 off signatures across multiple T-ALL cell lines and targets NOTCH1 signaling by demonstrating its inhibitory effects on the RBPJ luciferase reporter in heterologous U2OS cells expressing the L1601PδP mutation. The anti-leukemic phenotype induced by thapsigargin was consistent with that previously reported with inhibition of gamma secretase, the enzyme involved in the final activation step of NOTCH1. Decreased viability and cell size and a G1/G0 arrest were observed upon low nanomolar thapsigargin treatment in multiple T-ALL cell lines. These effects were on-target for NOTCH1 inhibition. Overexpression of the activated form of NOTCH1, (ICN1) restored the expression of the NOTCH1 target genes c-MYC and DTX1 upon thapsigargin treatment and rescued the effects of thapsigargin on T-ALL viability.

As in gamma secretase inhibitor (GSI) treatment, we observed that thapsigargin altered the level of active ICN1 protein in T-ALL cell lines. However, in contrast to treatment with GSI, thapsigargin also decreased levels of the furin-processed transmembrane forms of NOTCH1, while the full length NOTCH1 precursor accumulated as demonstrated by western blot and immunofluorescence studies. To further confirm the role of SERCA activity in NOTCH1 maturation, we inhibited Ca⁺⁺ATPase activity with SERCA-directed shRNA in T-ALL cell lines and demonstrated that genetically-driven inhibition of ATP2A2 results in loss of ICN1 and trans-membrane NOTCH1 and the expected phenotypic consequences: reduced viability and G1/G0 arrest.

To confirm that both chemical and genetic inhibition of SERCA leads to NOTCH inactivation in vivo, we adopted a Drosophila intestinal cancer stem cell model in which the inhibition of NOTCH1 prevents differentiation of stem cell daughter cells resulting in an expansion of the stem cell population. Treating Drosophila with thapsigargin or cyclopiazonic acid resulted in an expansion of the stem cell pool consistent with NOTCH inhibition. Indeed, a transgenic RNAi hairpin directed against Ca-P60A (the Drosophila SERCA orthologue) recapitulated the NOTCH phenotype observed with chemical treatment. In summary, these pre-clinical studies suggest a role for SERCA modulation of NOTCH1 maturation and identify SERCA as a potentially druggable target to inhibit NOTCH1 in T-ALL.