Development of a dual targeting scaffold of SET7/MLL inhibitor for castration-resistant prostate cancer treatment

methyltransferase enzyme SET7 and

Histone methyltransferase enzyme SET7 and mixed-lineage leukemia protein (MLL) complex are crucial co-activators of androgen receptor (AR) and have recently emerged as potential therapeutic targets for advanced castration-resistant prostate cancer (CRPC). In this study, we described the identification of a rhodium-based hybrid complex (SM_1) as a potent blocker of AR activity via simultaneously inhibiting SET7 and MLL complex activity, which makes it a potential lead scaffold for CPRC drug development.
Two hybrid complexes SM_1 and SM_2 as dual inhibitors against SET7 and MLL complex were designed by coupling the Rh(III) (SM_1) or Ir(III) (SM_2) core with two 1-phenylisoquinoline C^N ligands and one 5,6-dmphen (where 5,6dmphen Z 5,6-dimethyl-1,10-phenanthroline) N^N ligand, which have been previously reported as SET7/9 1 and menin modulators, 2 respectively (Fig. 1A). The potential menineMLL inhibition activity of complexes was evaluated using the biomolecular fluorescence complementation (BiFC) assay to study their effects on the menineMLL interaction. To visualize the menineMLL complex in cellulo, HEK293T cells were co-transfected with MLL-VC210 and VN210menin. 3 The hybrid Rh(III) complex SM_1 showed higher potency than the Ir(III) complex SM_2 towards blocking the menineMLL interaction (Fig. S1AeC). Moreover, the binding of a small molecule fluorescent probe to the SAM-binding pocket in SET7 was tested using a fluorescence polarization (FP) assay. Encouragingly, the hybrid Rh(III) complex SM_1 strongly inhibited the SAM/SET7 interaction, with a similar potency to the positive control sinefungin, a previously reported inhibitor of SET7 (Fig. S1A, D, E). Complex SM_1 exhibited a dose-dependent inhibition of SET7 binding with an IC 50 ca. 15.5 mM in the FP assay (Fig. S1F).
To explore whether complex SM_1 could directly engage SET7 and MLL complex in cellulo, cellular thermal shift assay (CETSA) was performed in murine CRPC cell line RM-1. Menin and SET7 in cell lysates treated with 10 mM of complex SM_1 were significantly stabilized compared with DMSO-treated controls (DTm of menin treated by SM_1: 2.0 C; DTm of SET7 treated by SM_1: 3.0 C) in RM-1 cell lysates. However, there was no effect of SM_1 on the stability of MLL and the negative control GAPDH (Fig. S2A, B). The ability of complex SM_1 to engage menin and SET7 was also verified using a fluorescence-based protein thermal shift assay (FTS), which showed a clear shift of the melting temperature (ca. 3.0 C) of purified menin and SET7 in the presence of complex SM_1, but not MLL (Fig. S2C). These results suggested that SM_1 can engage menin and SET7 even in the complicated environment of cell lysates.
As a steroid hormone receptor, AR induced by its ligand 5a-dihydrotestosterone (DHT) directs the expression of target genes by binding to the androgen response elements (AREs) in PSA promoter or enhancer elements in PSA and TMPRSS2 promoters (Fig. S3A). Co-immunoprecipitation (co-IP) experiment was performed to investigate the mechanism of action of complex SM_1 on AR activity. Incubating DHT-induced LNCaP cells with complex SM_1 for 6 h resulted in a reduction of the amount of AR co-precipitating with MLL ( Fig. 1B) or SET7 (Fig. 1C). To further evaluate the effect of SM_1 towards the transcription of genes induced by AR, a dual luciferase reporter assay was performed. Complex SM_1 significantly decreased the ARdirected luciferase intensity by nearly 80% at 3 mM, compared to only 27% and 34% reduction of luciferase intensity for 3 mM of MI-2 and 20 mM of sinefungin respectively (Fig. 1D). The IC 50 value for complex SM_1 in the luciferase assay was about 1.0 mM (Fig. 1E), which is 10 times lower than the value of the reported AR inhibitor MDV3100 (Fig. S3B). We hypothesize that complex SM_1 reduces ARdirected transcriptional activity by inhibiting the interactions of MLL and SET7 with AR in the treated cells.
Peer review under responsibility of Chongqing Medical University.
Available online at www.sciencedirect.com ScienceDirect journal homepa ge : www.keaipubl ishing.com/en/j ourna ls/genes-diseases The levels of MLL and SET7 in LNCaP cells were reduced significantly after siRNA treatment (Fig. S4A, B). In the luciferase assay, SM_1 was less able to further decrease ARE-related transcription activity in MLL/SET7 double knockdown cells compared with control cells (Fig. S4C). This suggests that the inhibition of ARE-related transcription activity induced by complex SM_1 requires the presence of MLL and SET7 in DHT-stimulated LNCaP cells. Meanwhile, the protein levels of PSA and TMPRSS2, which are also under the control of AR, were also measured after siRNA or SM_1 treatment (Fig. 1F, G). Similarly, complex SM_1 showed a greater ability to reduce the levels of PSA and TMPRSS2 in control cells compared to MLL/SET7 knockdown cells.
The cytotoxicity of SM_1 was determined in various prostate cancer cell lines (22RV1, DU145, LNCaP, and PC3) and normal cell lines (LO2 and HEK293T) using the MTT assay. SM_1 demonstrated potent anti-proliferative effects versus the prostate cancer cell lines, particularly the CRPC cell lines 22RV1 (IC 50 Z 0.63 mM) and LNCaP (IC 50 Z 0.74 mM), while it showed lower cytotoxicity against the non-CRPC prostate cancer cell lines DU145 (IC 50 Z 1.83 mM) and PC3 (IC 50 Z 3.92 mM) and the normal human cell line LO2 (IC 50 Z 4.30 mM) (Fig. S5). Notably, the antiproliferative activity of SM_1 towards LNCaP cells was greater than that of the reported AR inhibitor MDV3100 (IC 50 Z 19.05 mM). We suspect that the cytotoxicity displayed by SM_1 could be associated, at least in part, with the disruption of the interaction between MLL/SET7 and AR in cellulo.
To evaluate the nature of cell death induced by SM_1, the TUNEL assay was performed. SM_1 (3 mM) induced more apoptosis compared with MDV3100 (10 mM) or vehicle groups in both LNCaP and 22RV1 cells (Fig. S6). Furthermore, flow cytometry results indicated that both SM_1 and MDV3100 increased cells in the G2/M phase and lowered cell counts in the G0/G1 phase in a dose-dependent fashion, indicating that these compounds induce G2/ M arrest (Fig. S7), which is consistent with the G2/M arrest effect of MDV3100 and AR inhibitor in previous reports. 4 The effect of SM_1 on prostate cancer growth was evaluated in a mouse xenograft model of CRPC. Paclitaxel (Taxol), a clinical chemotherapeutic drug for CRPC, 5 was used as a positive control. Compared to the control group, daily intraperitoneal (IP) injection of complex SM_1 (1.0 mg/kg) led to inhibition of LNCaP tumor volume showing the same effect as positive drugs although 0.5 mg/ kg of complex SM_1 did not show the same effect (Fig. 1H). The reduction of tumor volume by SM_1 was not significantly different from paclitaxel (10 mg/kg) over the course of treatment. Besides, there was also no significant difference in weight loss between the SM_1 and paclitaxel treatment groups (Fig. 1I). These results demonstrate that complex SM_1 exhibits comparable anti-tumor activity as compared with paclitaxel for CRPC treatment.
Collectively, our study demonstrated the potential of the SET7/MLL dual-targeting inhibitor SM_1 for CRPC treatment through AR activity inhibition. To our knowledge, SM_1 represents the first metal-based dual inhibitor of MLL/SET7, and we anticipate that the hybrid complex SM_1 will serve as a potential scaffold for the development of CRPC therapeutic agents.