Dual-Inhibitors of N-Myc and AURKA as Potential Therapy for Neuroendocrine Prostate Cancer

Abstract

Resistance to androgen-receptor (AR) directed therapies is, among other factors, associated with Myc transcription factors that are involved in development and progression of many cancers. Overexpression of N-Myc protein in prostate cancer (PCa) leads to its transformation to advanced neuroendocrine prostate cancer (NEPC) that currently has no approved treatments. N-Myc has a short half-life but acts as an NEPC stimulator when it is stabilized by forming a protective complex with Aurora A kinase (AURKA). Therefore, dual-inhibition of N-Myc and AURKA would be an attractive therapeutic avenue for NEPC. Following our computer-aided drug discovery approach, compounds exhibiting potent N-Myc specific inhibition and strong anti-proliferative activity against several N-Myc driven cell lines, were identified. Thereafter, we have developed dual inhibitors of N-Myc and AURKA through structure-based drug design approach by merging our novel N-Myc specific chemical scaffolds with fragments of known AURKA inhibitors. Favorable binding modes of the designed compounds to both N-Myc and AURKA target sites have been predicted by docking. A promising lead compound, 70812, demonstrated low-micromolar potency against both N-Myc and AURKA in vitro assays and effectively suppressed NEPC cell growth.

Keywords: Myc; aurora a kinase; drug discovery; dual inhibitor; polypharmacology; prostate cancer.

Figure A1

Synthetic Procedures for Synthesis of Compounds F3409-1339.

Figure 1

Proposed binding mode of 70551 into the N-Myc-Max DNA-binding domain (DBD) pocket. (A) Homology model of N-Myc-Max with the predicted binding site from MOE site finder (top) Max: green ribbons, N-Myc: blue ribbons, and DNA duplex: white. The dummy atoms in orange are used to represent the site, and potential occupancy for inhibition (bottom). Compounds occupying the same positions would clash with the DNA duplex. Hydrophobic surface: green, polar surface: pink, and exposed surface: red. The site is the same one as proposed in our previous publication with 70063. (B) 2D representation of the N-Myc specific lead, 70551. (C) 70551 in the proposed binding site of N-Myc-Max DBD. Ligand interaction network reveals one strong H-bond with Arg215 from 70551’s nitrogen group in its hydrazide linker.

Figure 2

CD532’s binding mode to Aurora A kinase (AURKA). (A) CD532 (magenta sticks) binds to the active site of AURKA (pink ribbons). CD532 interacts with the ATP-binding hinge region, as well as with the protein-protein interactions (PPI) disruptor moiety, to induce conformation changes in AURKA (shown in the zoomed picture on right). Three-dimensional structure extracted from PDB 4J8M. Beta-sheets have been removed from the pocket for easier visualization of AURKA’s allosteric site. (B) CD532’s active scaffold can be divided into two main portions. One responsible for the N-Myc PPI disruption (encased in green), while the other acts as an ATP-competitor scaffold to bind to the hinge of AURKA (encased in yellow).

Figure 3

Merging of 70551’s scaffold with CD532’s scaffold for a novel dual-inhibitor. (A) The N-Myc specific scaffold, 70551 (left), shared a similar core with the AURKA specific scaffold, CD532 (right). Enclosed in green and yellow boxes are the sections of each scaffold responsible for the compounds’ inhibitory activity in their respective targets. The active parts of each scaffolds were merged by a commonly shared unsubstituted phenyl ring to generate 70812. (B) In N-Myc, 70812 (magenta sticks) has the critical H-bond with the backbone O of Arg215 of Max to maintain the scaffold in place in the DBD pocket. 70812’s tail is flipped towards the N-terminus of the DBD. (C) In AURKA, 70812 is docked in the ATP-binding core of AURKA and is tuck snuggly in the same position as CD532 in the AURKA pocket. Hydrophobic surface: green, polar surface: pink, and exposed surface: red. The beta-sheets have been removed for easier visualization of AURKA’s proposed binding site.

Figure 4

70812’s inhibitory profile against N-Myc and AURKA. (A) Inhibitory activity of 70551 was first determined. IMR32 cells were treated with 70551 at 10 µM and 20 µM for 120 h, and the growth was measured visually as a percentage of confluence (measure of cell proliferation on culture plate). (B) HO15.19 (Myc negative cell lines) were treated with 70551 at 10 µM and 20 µM for 120 h, and growth was determined visually as a percentage of confluence. 70551 was revealed to be active in positive control cell lines and showing acceptable toxicity in Myc negative cell lines. (C) Using an array of MTS assays, the specificity and potency of 70551 was further characterized in three N-Myc driven cell lines of LNCaP, 22RV1, and NCI-H660 at 10 µM, 5 µM, and 1 µM. (D) Inhibitory profile of 70812 in three N-Myc positive cell lines of 22RV1, LNCaP, and NCI-H660 at three concentrations of 10 µM, 5 µM, and 1 µM for 72 h as a measure of cell activity and growth. Cell density was measured using the PrestoBlue assay. (E) Reported AURKA enzyme activity following compound insertion at 30 µM, 15 µM, 10 µM, and 5 µM. 70812 had a similar inhibitory activity to CD532, at all concentrations tested. 70812 was assayed with its parental compound to determine its cell viability and profile in 22Rv1 (F) LNCaP (G), NCI-H660 (H), and HO.15.19 (I) at 10 µM, 5 µM, and 1 µM for 72 h. Cell density was measured using the PrestoBlue assay. All experiments were performed with N = 4 replicates. p-values were obtained using an unpaired t test against the vehicle control. Differences were considered significant when p < 0.005 (**).

Figure 5

70812’s IC₅₀ in N-Myc driven cell lines. The N-Myc inhibitory activity of compound 70812 in comparison to 70063, 10058-F4, and 10074-G5 in 22Rv1 (A), LNCaP (B), and HO15.19 (C), administered through serial concentration. All experiments were performed with N = 4 replicates. p-values were obtained using an unpaired t test against the vehicle control. Differences were considered significant p < 0.005 (**).

Figure 6

70812 disrupts DNA binding to N-Myc-Max DBD. (A) Inhibition of Myc-Max interaction by 70551 with biotinylated E-box was quantified by bilayer interferometry (BLI) at concentrations of 0 µM, 100 µM, and 200 µM. (B) 70812 could disrupt DNA from binding to the N-Myc-Max DBD at concentrations of 100 µM and 200 µM in BLI assays. The experiments are representative of 3 biological replicates.