Acute myeloid leukemia (AML) is a heterogeneous disease with diverse leukemogenic driver lesions. The genetic understanding of AML has resulted in major improvements in diagnosis, classification, prognostication, and outcome prediction. However, these insights have not yet translated into molecular mechanism-based therapies in the majority of cases, because AML cells rapidly escape attempts at therapeutic targeting such as small-molecule inhibition of FLT3 internal tandem duplication (FLT3-ITD) mutants, which occur in up to 30% of AML cases and confer a poor prognosis. To identify potential new targets for combinatorial treatment approaches, we performed a series of large-scale short hairpin RNA (shRNA) screens and observed that cell lines representing various AML subtypes were dependent on expression of the RET receptor tyrosine kinase (RTK), which has not previously been implicated in AML pathogenesis. Validation experiments demonstrated that depletion of RET by shRNA knockdown or CRISPR/Cas9-mediated knockout led to cell cycle arrest in the G0/G1 phase, increased apoptosis, and reduced clonogenic activity. RTK profiling using ELISA-based antibody arrays demonstrated that RET is highly phosphorylated in RET-dependent AML cell lines. Analysis of known RET ligand/co-receptor pairs (GDNF/GFRA1, NRTN/GFRA2, ARTN/GFRA3, PSPN/GFRA4) by quantitative real-time PCR and shRNA knockdown indicated that RET signaling is facilitated mainly through NTRN/GFRA2 or ARTN/GFRA3. Interrogation of various signaling pathways known to promote myeloid leukemogenesis showed that RET knockdown resulted in decreased phosphorylation of 4E-BP1 (T37/46), p70S6K (T389), S6RP (S240/244), and ULK1 (S758), pointing to mTORC1-mediated protein synthesis and/or suppression of autophagy as important effectors of RET signaling in AML cells. Based on recent data showing that FLT3-ITD mutants can be degraded by autophagy (Larrue et al. Blood 2016), we reasoned that the RET-mTORC1 signaling axis promotes AML through protection of FLT3-ITD mutants from autophagic degradation. Consistent with this hypothesis, genetic or pharmacologic (vandetanib, danusertib) inhibition of RET predominantly affected FLT3-dependent AML cell lines and were accompanied by upregulation of autophagy and destabilization of FLT3, as evidenced by p62 degradation, LC3B turnover, increased numbers of autophagic vacuoles, and decreased FLT3 protein levels. Furthermore, we observed accumulation of STAT5, a key FLT3-ITD downstream effector, upon pharmacologic autophagy inhibition in low RET-expressing AML cells, underlining the importance of RET-mediated suppression of autophagy for leukemogenic FLT3-ITD signaling. In line with the observations in AML cell lines, preliminary data from a murine bone marrow transplantation model show that Ret is required for AML development and propagation in vivo as we observed a significant survival advantage for mice transplanted with Ret knockdown cells compared with mice transplanted with control cells. Finally, genome-wide transcriptome analysis identified elevated RET mRNA levels in 35 of 260 (13.5%) primary human AML samples. Since there are no known RET copy number alterations or mutations of the RET coding region in AML patients and cell lines, we are currently investigating whether aberrant RET expression in AML can be attributed to perturbed epigenetic regulation. To this end, we are applying chromosome conformation capture combined with high-throughput sequencing (4C-seq) technology to systematically analyze interactions of the RET promoter region with enhancer sequences in high and low RET-expressing AML cell lines. Combined, our results indicate that in a proportion of AML, RET-mTORC1 signaling promotes cell viability and proliferation through suppression of autophagy, suggesting that targeting RET or, more broadly, depletion of critical leukemogenic drivers via induction of autophagy may provide a therapeutic opportunity in this subset of patients.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution