Characterization of CDK(5) inhibitor, 20-223 (aka CP668863) for colorectal cancer therapy

Abstract

Colorectal cancer (CRC) remains one of the leading causes of cancer related deaths in the United States. Currently, there are limited therapeutic options for patients suffering from CRC, none of which focus on the cell signaling mechanisms controlled by the popular kinase family, cyclin dependent kinases (CDKs). Here we evaluate a Pfizer developed compound, CP668863, that inhibits cyclin-dependent kinase 5 (CDK5) in neurodegenerative disorders. CDK5 has been implicated in a number of cancers, most recently as an oncogene in colorectal cancers. Our lab synthesized and characterized CP668863 - now called 20-223. In our established colorectal cancer xenograft model, 20-223 reduced tumor growth and tumor weight indicating its value as a potential anti-CRC agent. We subjected 20-223 to a series of cell-free and cell-based studies to understand the mechanism of its anti-tumor effects. In our hands, in vitro 20-223 is most potent against CDK2 and CDK5. The clinically used CDK inhibitor AT7519 and 20-223 share the aminopyrazole core and we used it to benchmark the 20-223 potency. In CDK5 and CDK2 kinase assays, 20-223 was ∼3.5-fold and ∼65.3-fold more potent than known clinically used CDK inhibitor, AT7519, respectively. Cell-based studies examining phosphorylation of downstream substrates revealed 20-223 inhibits the kinase activity of CDK5 and CDK2 in multiple CRC cell lines. Consistent with CDK5 inhibition, 20-223 inhibited migration of CRC cells in a wound-healing assay. Profiling a panel of CRC cell lines for growth inhibitory effects showed that 20-223 has nanomolar potency across multiple CRC cell lines and was on an average >2-fold more potent than AT7519. Cell cycle analyses in CRC cells revealed that 20-223 phenocopied the effects associated with AT7519. Collectively, these findings suggest that 20-223 exerts anti-tumor effects against CRC by targeting CDK 2/5 and inducing cell cycle arrest. Our studies also indicate that 20-223 is a suitable lead compound for colorectal cancer therapy.

Keywords: CDK inhibitor; colorectal cancer (CRC); cyclin-dependent kinase (CDK).

Figure 1. 20-223 exhibits anti-tumor activity in a colorectal cancer xenograft model

(A) Growth Inhibition of GEO cells after 72 hour treatment with 20-223. (B) Schematic representation of CRC xenograft model using GEO-GFP cells. (C) Average tumor volume comparison of DMSO and 20-223 treated tumors throughout the study. (D) Average GFP Flurorescence in DMSO and 20-223 treated tumors. (E) Average tumor weight of DMSO and 20-223 treated tumors.

Figure 2. Structural and cell free analyses of 20-223 and AT7519

(A) Docking of 20-223 into CDK5 using AutoDock Vina software. (B) Overlay of 20-223 and PNU181227 in the hinge region of CDK5. (C) Overlay of 20-223 and AT7519 in the ATP binding pocket. (D) Chemical structures of 20-223, PNU-181227, and AT7519. (E) % of remaining enzymatic activity of a panel of CDKs after incubation with 0.1μM 20-223 and 30μM ATP. (F) IC₅₀ values (nM) of CDK2 and CDK5 after incubation with 20-223 or AT7519 in cell free dose-escalation study.

Figure 3. 20-223 inhibits the kinase activity of CDK5 and CDK2 in vitro

(A) Baseline expression of CDK2 and pRB (S807/811) (left), CDK5 and pFAK (S732) (right), in untreated CRC cells. (B) Representative western blots of target substrate pRB and pFAK phosphorylation levels in GEO (left), HCT116 (middle) and HT29 (right) cell lines after 6 hour incubation with 20-223. (C) Representative quantification of % inhibition of CDK2 and CDK5 kinase activity (based on substrate phosphorylation levels) in GEO cells found in Figure 3B. (D) Cell-based IC₅₀ values generated from phosphorylation levels in Figure3B of CDK2 and CDK5 in three CRC cell lines.

Figure 4. 20-223 disrupts migration of CRC cells

(A) Wound gap images taken during the 24 hour incubation of HCT116 cells with DMSO or 1.5μM 20-223. 0 and 24 hour images were further evaluated by outlining the wound area (red lines) and zooming in on the wound boundaries (yellow box). (B) Quantification of % wound closure after treatment of HCT116 cells with DMSO of 1.5μM 20-223. (C) Western blot analyses at 6 and 24 hours after stimulation with EGF and treatment with either DMSO or 1.5μM 20-223.

Figure 5. 20-223 potently inhibits cell growth in a panel of CRC cell lines

(A) IC₅₀ values from growth inhibition studies after CRC cells were treated with 20-223, AT7519 or Roscovitine for 72 hours. (B) Average IC₅₀ values across all seven CRC cell lines after treatment with 20-223, AT7519, or Roscovitine (P < 0.001). (C) Panel containing the seven CRC cell lines used in this study and mutational status of important regulatory genes.

Figure 6. 20-223 treatment in CRC cells results in cell cycle arrest in vitro

(A) % of cells in each phase of the cell cycle after treatment with DMSO, AT7519, or 20-223 for 24 (top) and 48 (bottom) hours in GEO (left) and HCT116 (right) cells. (B) Traces representative of cell cycle analysis in GEO cells after treatment with DMSO, AT7519, or 20-223 after 24 (left) or 48 (right) hours.