A Novel Synthetic Lethal Approach Targeting SIRT6 in Acute Myeloid Leukemia

Background: Acute Myeloid Leukemia (AML) is an incurable disease characterized by a highly unstable genome, resulting in large-scale changes at diagnosis, as well as further evolution contributing to disease progression.  However, the mechanisms whereby tumor cells adapt to genomic instability are largely unknown, but recent observations have correlated these abnormalities with dysfunctional DNA damage repair (DDR) machinery. SIRT6 is an important regulator of cellular stress response and genomic integrity. Here we investigated the role of this NAD⁺-dependent deacetylase in regulating ongoing DNA damage observed in AML patients.

Methods: SIRT6 mRNA level was determined by RT-qPCR in AML patients (n=100) diagnosed at the Hematology Department of University of Genoa (Italy), compared with normal bone marrow derived CD34⁺ cells (n=5). Correlation studies with clinical and molecular characteristics of these patients were also performed. A panel of different AML cell lines and primary cells, both sensitive and resistant to conventional and novel anti-AML therapies, was used in the study. The anti-leukemic effect of DNA-damaging agents (DDAs) including idarubicin, Ara-C and fludarabine was evaluated in presence of SIRT6 depletion/inhibition by CTG assay and Annexin-V/propidium iodide staining. Mechanistic studies were performed with Western-blotting, lentivirus-mediated shRNAs and immunofluorescence assay. Analysis of DNA DSB repair was done using chromosomally integrated reporter constructs, followed by cytometer analysis.

Results: AML patients were grouped into lower and higher SIRT6 expressers according to its median mRNA level. Patients with lower expression had a higher incidence of FLT3-ITD (p=0.034, Wilcoxon signed rank test). No significant association was observed with respect to mutations of NPM1, nor with WT1 and BAALC expression. SIRT6 expression correlated also with adverse clinical outcome in term of event free and overall survival (p=0.035 and p=0.025, respectively; unpaired t test). Based on these data, we evaluated SIRT6 role in biology of AML. We found higher SIRT6 protein level in AML cell lines carrying FLT3-ITD mutation (MOLM-14 and MV 4-11) compared to cell lines harboring other mutations (OCI-AML3, THP-1, KG, NB4, HL60, Nomo1 and U937). Targeting SIRT6 by specific shRNAs weakly reduced AML cell survival compared with control-scrambled cells, by impairing DNA repair efficiency. Indeed, a restricted effect of SIRT6 impairment on DNA damage proteins (H2AX, RAD51, 53BP1, RPA32) was measured. We next examined the therapeutic relevance of SIRT6 inhibition in AML by testing effects of its depletion in combination with genotoxic agents. Remarkably, SIRT6 depletion conferred increased sensitivity of AML cells to idarubicin, Ara-C and Fludarabine. Overall, enhancing genotoxic stress while concomitantly blocking DNA double-strand breaks (DSBs) repair response, may represents an innovative strategy to increase chemosensitivity of AML cells. Further mechanistic studies revealed that SIRT6 acts as a genome guardian in leukemia cells by binding DNA damage sites and activating DNA-PKcs and CtIP by deacetylation, which in turn promotes DNA repair.

Conclusion: Genomic instability is present in all hematologic malignancies including AML. Strategies aimed to shift the balance towards high DNA damage and reduced DNA repair by SIRT6 inhibition can decrease AML growth and may benefit patients with otherwise unfavorable outcomes.