Discovery of an Aldo-Keto reductase 1C3 (AKR1C3) degrader

Aldo-keto reductase 1C3 (AKR1C3), a member of the aldo-keto reductase superfamily, is overexpressed in various cancers, including prostate cancer and hematological malignancies. Its overexpression contributes to tumor proliferation, aggression, and chemotherapeutic resistance. AKR1C3 plays a crucial role in androgen synthesis, particularly in hormone-dependent cancers. The AKR1C family also includes AKR1C1 and AKR1C2, which share significant sequence similarity with AKR1C3. The need for selective versus pan-AKR1C inhibition varies depending on cancer type and stage. While selective AKR1C3 inhibition is desirable in some cancers, pan-AKR1C inhibition may be beneficial in others. AKR1C3 has an emerging role in stabilizing androgen receptor (AR) and its splice variants, especially AR variant 7 (ARv7). ARv7 is a key driver of resistance to androgen receptor signaling inhibitors (ARSIs) like enzalutamide, darolutamide, and apalutamide. AKR1C3 interacts with ARv7, preventing its degradation and contributing to ARSI resistance. This study aimed to develop a novel strategy to overcome this resistance by targeting AKR1C3 for degradation using Proteolysis-Targeting Chimeras (PROTACs). PROTACs are heterobifunctional molecules composed of a ligand binding to the target protein (AKR1C3), a linker, and a ligand that binds to and recruits an E3 ubiquitin ligase, leading to target protein ubiquitination and proteasomal degradation. PROTACs offer advantages over small molecule inhibitors by enabling the targeting of 'undruggable' proteins, inducing complete protein removal, and exhibiting catalytic activity at sub-stoichiometric concentrations. The authors hypothesized that degrading AKR1C3 using a PROTAC would lead to dual degradation of AKR1C3 and ARv7, thereby countering two mechanisms of ARSI resistance.

Previous research has demonstrated the role of AKR1C3 in various cancers and its contribution to drug resistance. Studies have shown that AKR1C3 inhibitors, such as indomethacin and BMT4-158, can degrade the target protein. However, the selectivity and potency of these inhibitors varied. The group's prior work identified potent and selective AKR1C3 inhibitors, including a biphenyl derivative (inhibitor 3) with an IC50 of 43 nM and >2300-fold selectivity over AKR1C2. These inhibitors provided the foundation for the design of the PROTAC degrader in this study. The literature also highlighted the role of ARv7 in ARSI resistance and the interaction between AKR1C3 and ARv7 in stabilizing ARv7 and promoting resistance. This interaction motivated the development of a degrader targeting both AKR1C3 and ARv7. Existing research also established the principle of PROTAC-mediated protein degradation and its advantages over traditional small molecule inhibitors.

The study involved several key steps: **1. Evaluation of small molecule inhibitors:** The researchers first evaluated the ability of existing AKR1C3 inhibitors (inhibitor 3 and warhead 4) to degrade AKR1C3, ARv7, and AKR1C1/2 in 22Rv1 prostate cancer cells. Western blotting was used to assess protein expression levels at different time points (0, 24, 48, and 72 hours) and concentrations of inhibitors. **2. PROTAC design and synthesis:** Based on the structure-activity relationship (SAR) data from previous studies and molecular docking studies using the crystal structure of AKR1C3, a PROTAC degrader (PROTAC 5) was designed. This involved selecting a potent AKR1C3 inhibitor warhead, a suitable linker, and an E3 ubiquitin ligase ligand (lenalidomide). The PROTAC was synthesized through a multi-step process involving coupling of a modified AKR1C3 inhibitor warhead to lenalidomide via a click chemistry reaction. The detailed synthetic procedures and characterization data for the intermediates and PROTAC 5 are described in the supplementary methods. **3. PROTAC characterization:** The AKR1C3 inhibitory activity and selectivity of PROTAC 5 were determined using enzyme inhibition assays. The ability of PROTAC 5 to degrade AKR1C3, ARv7, and AKR1C1/2 was evaluated in 22Rv1 cells using Western blotting at various time points and concentrations. The proteasome dependence of the degradation was confirmed using the proteasome inhibitor MG132. The effect of lenalidomide alone and in combination with PROTAC 5 was also investigated to assess its contribution to the degradation. **4. Cell viability assays:** The effects of PROTAC 5 and warhead 4 on the viability of 22Rv1 cells (with varying AKR1C3 expression levels), LNCaP cells (AKR1C3 null), and LNCaP1C3 cells (stably transfected with AKR1C3) were determined using the MTT assay. The effect of combining PROTAC 5 with enzalutamide (ENZ) on cell viability was also examined to assess its ability to overcome ENZ resistance.

The study's key findings include: **1. Small molecule inhibitors induce degradation:** High concentrations of the small molecule AKR1C3 inhibitors 3 and warhead 4 led to the degradation of AKR1C3 and ARv7, along with some degradation of AKR1C1/2, in 22Rv1 cells. **2. PROTAC 5 is a potent degrader:** PROTAC 5 potently degraded AKR1C3 (DC50 = 52 nM), ARv7 (DC50 = 70 nM), and AKR1C1/C2 in 22Rv1 cells in a time- and concentration-dependent manner. The degradation was significantly greater than that observed with small molecule inhibitors at comparable concentrations. **3. Proteasome dependence:** The degradation of AKR1C3 by PROTAC 5 was shown to be proteasome dependent, further confirming the PROTAC mechanism of action. A hook effect at high concentrations was observed, providing additional evidence for the PROTAC mechanism. **4. Impact on cell viability:** PROTAC 5 reduced cell viability in 22Rv1 cells in a dose-dependent manner, particularly in cells grown in charcoal-stripped serum (high AKR1C3 expression). Importantly, PROTAC 5 sensitized 22Rv1 cells to the cytotoxic effects of enzalutamide, overcoming the resistance observed in these cells. No effect was observed in LNCaP (AKR1C3 null) cells, while the effect was restored in AKR1C3-overexpressing LNCaP1C3 cells. The data indicates that PROTAC 5 acts by a mechanism of AKR1C3 degradation and has the potential for use to overcome resistance in prostate cancer cells to clinical agents.

This study successfully demonstrated the design, synthesis, and evaluation of the first PROTAC degrader targeting AKR1C3. The results show that PROTAC 5 is a potent degrader of AKR1C3 and concomitantly reduces ARv7 levels, effectively addressing two major mechanisms of ARSI resistance in prostate cancer. The superior performance of PROTAC 5 compared to small molecule inhibitors underscores the advantages of the PROTAC strategy for targeting proteins like AKR1C3, which may be challenging to inhibit effectively with small molecules. The observation that PROTAC 5 sensitizes resistant prostate cancer cells to enzalutamide is a significant finding with substantial therapeutic implications. The ability of PROTAC 5 to degrade AKR1C1 and AKR1C2 to a lesser extent is interesting and warrants further investigation into the potential for pan-AKR1C targeting in various cancer types. The study's findings strongly support the further development of PROTAC 5 as a potential therapeutic agent for prostate cancer.

This research successfully developed and characterized the first PROTAC degrader targeting AKR1C3. PROTAC 5 effectively degrades AKR1C3 and ARv7, overcoming enzalutamide resistance in prostate cancer cells. This strategy holds significant promise for treating prostate cancer and other cancers where AKR1C3 contributes to drug resistance. Future research should focus on optimizing PROTAC 5's potency, selectivity, and pharmacokinetic properties for preclinical and clinical development.

The study primarily focused on in vitro experiments using prostate cancer cell lines. Further in vivo studies are necessary to confirm the efficacy and safety of PROTAC 5. The study also did not extensively explore the mechanism of AKR1C1/2 degradation, although observed in both small molecule and PROTAC treatments. Additional research is needed to determine the clinical relevance of this finding and to explore the potential for pan-AKR1C targeting in other cancer types. The use of a single cell line could limit the generalizability of these findings. Further testing with additional prostate cancer cell lines and other relevant cancer types would strengthen these conclusions.