Ketorolac modulates Rac-1/HIF-1α/DDX3/β-catenin signalling via a tumor suppressor prostate apoptosis response-4 (Par-4) in renal cell carcinoma

Abstract

Renal cell carcinoma (RCC) is the most difficult-to-treat form of kidney cancer with a median 5-year survival of 10% under metastatic setting. In RCC, although cytoreductive nephrectomy is common, approximately 20-30% of patients will develop recurrent cancer after surgery, which highlights the need for an effective therapy. Rho-GTPases viz, Rac-1 and Cdc42 are the central regulators of cancer cell migration and invasion and thus metastasis in multiple cancer types. Hence, we elucidated the role of Ketorolac, a modulator Rho-GTPases against RCC through potentiation of tumor suppressor Par-4. The effect of Ketorolac alone and in combination on proliferation, apoptosis, cell-cycle progression, migration, tumor inhibition and their related markers were studied. Moreover, Ketorolac's impact on metastasis by influencing Rac-1/HIF-1α/DDX3/β-catenin signalling was studied with respect to its ability to modulate the expression of tumor suppressor Par-4, and this mechanism was confirmed by siRNA knockdown studies. Ketorolac induced cytotoxicity in a panel of renal cells including patient derived tumor cells with IC₅₀ 2.8 to 9.02 mM and 0.28 to 3.8 mM in monolayer and anchorage independent clonogenic assays respectively. Ketorolac caused significant down regulation of proliferation (Ki-67, Cyclin D1, pRB and DDX3), migration/invasion (Rac-1, Cdc42, and Tiam1), and angiogenesis (HIF-1α and VEGF) markers as studied by gene and protein expression. Moreover, it caused a significant upregulation of tumor suppressor Par-4 known to be downregulated in RCC. This mechanism was further confirmed by using siRNA knockdown studies where we could demonstrate a negative relation between the expression of Par-4 and Rac-1/Cdc42. Importantly, Ketorolac alone and in combination with Sunitinib showed tumor growth inhibition (TGI) of 73% and 86% respectively in xenograft model. This anti-tumor activity was further corroborated by down regulation of Rac-1/Cdc42/HIF-1α/DDX3/β-catenin signalling. This is the first report which implicates the role of Ketorolac against RCC by acting as a small molecule secretagogue causing upregulation of Par-4 in autocrine and paracrine manner. Consequently, these findings suggest that Par-4 can serve as a valuable therapeutic target and a prognostic marker for the treatment of RCC.

Figure 1

Cytotoxicity of Ketorolac in a panel of RCC cell lines. (A) In vitro and ex vivo cytotoxicity assay for Ketorolac in panel renal cells with indicated concentrations. Ex vivo assay was performed in anchorage independent semi-solid soft agar 3D assay using renal tumor xenograft-derived cell suspension or renal cell lines. Cytotoxicity is expressed as % survival (% T/C) as a ratio of treated by control. (B) Histogram represents comparison of in vitro and ex vivo activity of Ketorolac across the panel of 10 RCC cell lines.

Figure 2

Efficacy of Ketorolac in combination with Sunitinib at indicated concentrations with 5 × 5 matrix in (A) RXF-1183L and (B) A-498 cells seeded in soft agar. Similarly, effect of Ketorolac in combination with Sorafenib was evaluated in (C) RFX-1183L, (D) RFX-486, (E) 786-O, (F) A-498 and (G) SN12C. Bliss independence analysis showed synergy or additive effect. The color coding of the tiles in the heatmap show that there is consistent concentration-dependent effect pointing towards synergy (BI > 0.15) or additive effect (− 0.15 < BI < 0.15).

Figure 3

Clonogenic and wound healing assay for Ketorolac. (A) Representative colony pictures for A-498, 786-O and Caki-1 cells. (B–D) Show histogram representing quantitation of the stained colonies using Image J for respective cells. (E) Wound-healing assay in A-498 treated with Ketorolac alone or in combination with Sunitinib. (F) Histogram representing quantitation of images using Image J. Statistically significant difference between and within different groups was determined by one-way ANOVA and post hoc multiple variance by Tukey test (*p ≤ 0.05, **p ≤ 0.01 and ***p ≤ 0.001). (G) The qRT-PCR of VEGF and E-cadherin in A-498 after treatment with Ketorolac alone and in combination with Sunitinib. The gene expression was studied at 3, 6, 18 and 24 h post treatment. (H) The protein expression for VEGF at different time points 24, 48 and 72 h after treatment (K Ketorolac and S Sunitinib).

Figure 4

(A) Apoptosis effect (% sub G0/G1) in A-498 during sequence dependent Ketorolac treatment (72 h) followed by addition of Sunitinib for last 48 h. Ten thousand cells were acquired and further analyzed by BD FACS-Lyric flow cytometry. (B) Histograms represent increased sub G0/G1 levels of A-498 after treatment with Ketorolac followed by Sunitinib. (C) Western blot depicting inhibition Cyclin D1, pRB, Ki-67 and DDX3 post 24 h Ketorolac treatment and in combination with Sunitinib. (D) Heatmap depicts modulation of proliferation and EMT markers in A-498 at 3, 6, 18 and 24 h treatment with Ketorolac alone and in combination with Sunitinib. The relative expression of genes was calculated with the relative 2^−ΔΔCt method, using GAPDH as housekeeping gene for normalization (K Ketorolac and S Sunitinib).

Figure 5

Par-4 expression in Atg++ and A-498 cells (A) mRNA expression of Par-4 in Atg++ cells treated with 25, 50 and 100 µM Ketorolac for 24 h. (B) Par-4 protein expression in Atg++ post 48 h treatment with Ketorolac. (C, D) Transcript and protein expression of Par-4, Tiam1, Rac-1 and Cdc42 in A-498 cells treated with Ketorolac 1, 3, 7 mM and Sunitinib 4 µM alone and in combination for 24 h respectively. GAPDH was included as loading control. Numbers at top represent densitometric analysis of band. (E) Heatmap depicts modulation in transcript levels of Par-4, Tiam1, Rac-1 and Cdc42 in A-498 at 3, 6, 18 and 24 h treatment with Ketorolac. The relative expression of genes was calculated with the relative 2^−ΔΔCt method, using GAPDH as housekeeping gene for normalization (K Ketorolac and Su Sunitinib).

Figure 6

(A, B) Silencing of Par-4 (30 pmol) and its effect on Rac-1 expression in A-498 cells post 2, 5, 18 and 24 h treatment at transcript level and 24, 48 and 72 h post treatment at protein level. (C, D) Silencing of Rac-1 (30 pmol) and its effect on Par-4 expression post 2, 5, 18 and 24 h at transcript level and 24, 48 and 72 h at protein expression level. Scrambled siRNA served as a negative control. GAPDH served as a loading control. (E) The level of cdc42 expression post 5 and 24 h treatment after silencing of Par-4 and Rac-1 independently. (F) The level of VEGF expression post 5 and 24 h after silencing Par-4 and Rac-1 independently. Gapdh served as a loading control. Numbers at top of the blots refer to the densitometric analysis of the immunoreactive bands and represent the fold change in gene and/or protein expression normalized to GAPDH.

Figure 7

(A) In vivo efficacy of Ketorolac in A-498 xenograft model. Nude athymic mice were implanted with ~ 30 mg A-498 tumor fragment using trocar needle in the flank region. Once tumor reached to 100 mm³, animals were randomized in different groups of Ketorolac (5 and 10 mg/kg) alone and in combination with Sunitinib (20 mg/kg). Animals were weighed regularly, and tumor volume was measured every alternate day. (B) Representative picture of mice on the day of termination (Day 20th) form different regimen groups. (C) Histogram indicate tumor weight scatter plot on 20th day of the regimen. (D) Table shows complete and partial regression (CR and PR) of A-498 tumors in mice treated with Ketorolac alone and in combination with Sunitinib. (E) qRT PCR analysis in tumor samples obtained from xenograft model. The gene expression for the transcripts of HIF-1α, DDX-3, Cdc42, Rac-1 and β-Catenin were studied.

Figure 8

Proposed pathways targeted by Ketorolac in RCC.