Targeting the Proteasomal 19S Regulatory Complex Instead of the 20S Catalytic Complex As a Novel Therapeutic Strategy in FLT3+ AML

Introduction: Acute myeloid leukemia (AML) is a hematologic malignancy characterized by the expansion of deregulated hematopoietic stem cells. Despite high remission rates with standard induction chemotherapy, most patients relapse from common driver mutations, most commonly the FMS-like tyrosine kinase-3 (FLT3) gene. FLT3 tyrosine kinase inhibitors (TKIs) can improve response rates in AML patients, but up to 60% develop adverse events or drug resistance. Therefore, new therapeutic strategies are needed. An approach to circumvent drug resistance targets the ubiquitin proteasome system (UPS), which regulates protein homeostasis and cell survival. However, like FLT3 TKIs, standard proteasome inhibitors like bortezomib, which target the 20S catalytic subunit of the UPS, are associated with toxicity and drug resistance. We hypothesized that targeting 19S regulatory subunits of the UPS would inhibit AML cell growth with less toxicity. Indeed, we previously reported a role for 26S proteasome non-ATPase subunit 3 (PSMD3), a member of the 19S proteasome, in the overall survival of patients with AML (Lara et al., 2022), chronic myeloid leukemia (Bencomo-Alvarez et al., 2021), and several solid tumors (Rubio et al., 2021). However, the mechanism by which PSMD3 and other 19S proteasome subunits act as oncogenes in AML remains unknown. Here, we assessed the functional role of PSMD3 and other 19S proteasome subunits in FLT3+ AML both in vitro and in vivo.

Methods: We utilized the MOLM-13, MOLM-14, and MV4-11 cell lines to model FLT3+ AML in vitro and in vivo. AML cell lines were transduced with doxycycline (dox)-inducible lentiviral vectors harboring small-hairpin RNAs (shRNAs) targeting PSMD3 (shPSMD3) or other 19S subunits (PSMD2, PSMD9) compared with a non-targeting control (shNT). Resulting cells were assessed for changes in in vivo engraftment into NOD-scid IL2Rgammanull (NSG) mice (n=6/group), or changes in vitro using RNA sequencing and liquid chromatography (LC)/mass spectrometry (MS)-based proteomics analyses, as well as metabolic parameters such as oxygen consumption rates using the Agilent Seahorse XFp Bioanalyzer. To model TKI resistance, we created intrinsically TKI-resistant cells through long-term culture in low-dose midostaurin (final concentration 100nM), as well as extrinsic TKI resistance through co-culture with the HS-5 bone marrow stromal cell line. Drug resistance in the presence of midostaurin was confirmed by AnnexinV staining and flow cytometry.

Results: Mice receiving 3x10⁶ shPSMD3-expressing MV4-11 cells demonstrated a significant increase in overall survival (median 116 days, n=6) compared with mice receiving shNT controls (median 41 days, n=6) (p=0.0269); similar results were observed with the MOLM-13 cell line (p=0.0027). RNAseq data collected after knockdown of PSMD3 in the MOLM-13, MOLM-14, and MV4-11 cell lines resulted in upregulation of pathways involved with neutrophil degranulation, the inflammatory/immune response, and cytokine involvement, but downregulation of pathways related to vitamin and cofactor metabolism and protein phosphorylation. Other commonalities included monocarboxylic acid metabolism, generation of precursor metabolites, glycosyl compound metabolism, and small molecule biosynthesis. However, our early metabolic studies in shPSMD3 did not result in changes of oxygen consumption rates in FLT3+ AML. In scenarios of extrinsic and intrinsic TKI resistance, we observed increased activation of several critical signaling pathways compared with TKI-sensitive controls, including STAT3, STAT5, and mTOR, which were reduced upon PSMD3 knockdown. Similar to PSMD3, AML patients with high levels of PSMD2 (p=0.046) or PSMD9 (p=0.003) have a worse overall survival compared with patients having lower expression of either gene. Additionally, similar to the effects of shPSMD3 (Lara et al., 2022), shPSMD2 or shPSMD9 reduced colony formation of FLT3+ AML cell lines without affecting apoptosis.

Conclusions: We have identified an oncogenic role for PSMD3 in FLT3+ AML with roles in neutrophil degranulation, signal pathway activation, and cell cycle regulation. Future studies will establish small molecule inhibitors that can effectively block 19S proteasome function in AML but not normal myeloid stem and progenitor cells.

No relevant conflicts of interest to declare.