Galeterone and its analogs inhibit Mnk-eIF4E axis, synergize with gemcitabine, impede pancreatic cancer cell migration, invasion and proliferation and inhibit tumor growth in mice

Abstract

Survival rate for pancreatic cancer (pancreatic ductal adenocarcinoma, PDAC) is poor, with about 80% of patients presenting with the metastatic disease. Gemcitabine, the standard chemotherapeutic agent for locally advanced and metastatic PDAC has limited efficacy, attributed to innate/acquired resistance and activation of pro-survival pathways. The Mnk1/2-eIF4E and NF-κB signaling pathways are implicated in PDAC disease progression/metastasis and also associated with gemcitabine-induced resistance in PDAC. Galeterone (gal), a multi-target, agent in phase III clinical development for prostate cancer has also shown effects on the aforementioned pathways. We show for the first time, that gal/analogs (VNPT55, VNPP414 and VNPP433-3β) profoundly inhibited cell viability of gemcitabine-naive/resistance PDAC cell lines and strongly synergized with gemcitabine in gemcitabine-resistant PDAC cells. In addition, to inducing G1 cell cycle arrest, gal/analogs induced caspase 3-mediated cell-death of PDAC cells. Gal/analogs caused profound downregulation of Mnk1/2, peIF4E and NF-κB (p-p65), metastatic inducing factors (N-cadherin, MMP-1/-2/-9, Slug, Snail and CXCR4) and putative stem cell factors, (β-Catenin, Nanog, BMI-1 and Oct-4). Gal/analog also depleted EZH2 and upregulated E-Cadherin. These effects resulted in significant inhibition of PDAC cell migration, invasion and proliferation. Importantly, we also observed strong MiaPaca-2 tumor xenograft growth inhibition (61% to 92%). Collectively, these promising findings strongly support further development of gal/analogs as novel therapeutics for PDAC.

Keywords: galeterone (gal) and its analogs; pancreatic cancer resistance.

Conflict of interest statement

CONFLICTS OF INTEREST Vincent C. O. Njar is the lead inventor of galeterone (gal, VN/124-1 or TOK-001), the new analog, patents and technologies thereof owned by the University of Maryland, Baltimore, and licensed to Tokai Pharmaceuticals, Inc. Puranik Purushottamachar and Andrew K. Kwegyir-Afful are co-inventors of VNPT55 and related compounds. Patent applications to protect these novel compounds and their potential utility against pancreatic cancer have been filed. The other authors declare no potential conflicts of interest.

Figures

Figure 1. Anti-proliferative activities of ARDAs and gemcitabine in PDAC cells

A. Structures of Androgen Receptor Degrading Agents (ARDA). B. Gal and analogs decrease cell viability of MiaPaCa-2 cells utilizing MTT cell viability assays. C. MTT cell viability assays in drug-naive and drug-resistant (MiaPaCa-2, MiaPaCa-GR and MiaPaCa-GTR) show a decrease in gemcitabine activity in drug-resistant cells D. Gal and analogs potentiate effects of gemcitabine and strikingly decrease GI50values in gemcitabine-resistant PDAC cells (**p < 0.001). PDAC cells were first treated with gal and analogs, then with gemcitabine. E. Gal and gemcitabine synergistically inhibit cell viability of gemcitabine-resistant PDAC cells. Cell viability assays were conducted for gal and gemcitabine individually and GI50 values calculated. Both compounds were combined at a constant ratio and utilized in cell viability assays. Fractional effects of single agents and in combination were calculated and analyzed by the calcusyn software to compute the combination indices (CI) at ED50, ED75 and ED90. CI value at ED90 reveal strong synergy between the two compounds in MiaPaCa-GR cells.

Figure 2. Gal/analogs and gemcitabine cause cell cycle arrest and induce apoptosis in PDAC cells

A. Gal and VNPP433-3B, induce G1 cell cycle arrest in MiaPaCa-2 cells. B. Gal and analogs cause accumulation of cells in the G1 phase of cell cycle in metastatic PDAC cells (S2-VP10). C. Gal and analogs deplete cell cycle regulators (cyclin B1 and cyclin D1) and upregulate p21 in MiaPaCa-2, S2-VP10 and MiaPaCa-GTR cells. D. Cell viability assays indicate that concentrations utilized (5 and 10 μM), over a 24 h time point did not compromise cell viability significantly. E. Gal/analogs and gemcitabine induce apoptosis in S2-013 cells in vitro, using the acridine orange/ethidium bromide assay to determine loss of cell membrane integrity. Cells were treated at 2.5 μM for 72 h. Arrows next to L indicate live cells; arrows pointing to A indicate apoptotic cells; and arrows pointing N indicate necrotic cells F. Gal/VNPP433-3β and gemcitabine (5 and 10 μM) were compared in their ability to induce apoptosis, analyzed by flow cytometry. Cells treated with compounds for 24 h were stained with annexin v and propidium iodide (PI). Early and late apoptotic cells were analyzed by FACS. All compounds show significant apoptotic induction (*p < 0.05, **p < 0.001). G. Gal and analogs enhanced depletion of Bcl-2 and increased expression of cleaved PARP, caspase 3 and Bax in gemcitabine-naive/resistant PDAC cells.

Figure 3. Gal and analogs deplete Mnk1/2 protein expression and downregulate eIF4E phosphorylation

A and B. ASPC1 and S2-013 cells treated with gal and VNPP433-3β show a dose dependent decrease in K-Ras, Mnk1 and peIF4E. C and D. CaPan-1 and MiaPaCa-GTR cells treated with gal and VNPP414 at 10 μM also show decrease in Ras, Mnk2 and peIF4E protein expression. E. S2-013 cells treated with gal/VNPP433-3β/gemcitabine was analyzed for phosphorylation levels of eIF4G, eIF4E and 4E-BP. Protein expression analysis reveal an enhanced expression of Mnk1 by gemcitabine treatment.

Figure 4. Gal and analogs inhibit metastatic inducing pathways

A. MiaPaCa-2 and S2-013 cells treated with gal and VNPP433-3β at increasing concentration show a dose-dependent decrease in p-p65. B. MiaPaCa-GR and MiaPaCa-GTR cells exposed to gal and VNPTP433-3β show profound depletion of EZH2 protein expression MiaPaCa-GTR (left panel) and MiaPaCa-GR (right panel). C. S2-013 cells, left panel, exposed to gal (5 – 10 μM) show downregulation of N-cadherin, and an upregulation of E-cadherin (bottom graph). Densitometric analysis suggests a profound increase in E-Cad: N-Cad ratio, suggesting loss of mesenchymal characteristics. D. S2-VP10 cells exposed to 10 μM of gal/analogs exhibit a downregulation of N-cadherin and an upregulation of E-cadherin resulting in an increase in E-cad N-cad ratio (bottom chart). E. Gal and VNPP433-3β deplete MMP-2/-9 in MiaPaCa-2 and MiaPaCa-GTR cells after 24 h treatment period. F. S2-013 cells transfected with Mnk1 siRNA show downregulation of N-cadherin, MMP-2/-9 and peIF4E.

Figure 5. Downregulation of EMT markers

after A, top panel) Media from S2-013 cells, treated with indicated compounds was separated on a zymogram gel to analyze the proteolytic activity of MMP-1/-9. (A, bottom graph) densitometric analysis of zymogram gel in top panel, reveals a significant decrease in collagenase activity of MMP-1/-9 (**p < 0.001), zymogram assays were repeated at least three times and presented as mean ± S.E.M. B. Densitometry of proteolytically digested bands on zymogram gels analyzed from gal, CGP, gemcitabine and combinations’ treated MiaPaCa-2 cultures, shows significant upregulation of MMP-9 collagenase activity in the presence of gemcitabine. Gal and CGP both decrease gemcitabine-induced MMP-9 activity (*p < 0.05, **p < 0.001). C and D. Gemcitabine-naïve, Gemcitabine-resistant and Gemcitabine/erlotinib resistant, PDAC cells, were treated with gal and VNPP433-3β for 24 h at 10 μM. Protein expression analysis show downregulation of Snail, Slug, Cox-2, CDC42 and CXCR4 in all four cell lines.

Figure 6. Gal and analogs possess anti-migratory and anti-invasive properties

A. Scratch wound healing assays reveal that gal/analogs, CGP and gemcitabine at 5 μM, inhibit Panc-1(top panel) and S2-013 (bottom panel) PDAC cell migration. B. Distance migrated by Panc-1 and S2-013 cells after treatment were quantified. Wounds were measured before and after the 12 h time point. Distance migrated were quantified by measuring the difference at time 0 and 12 h and normalized to control. (Distance migrated = Distance at time 0 hour - distance at 12 h/ Distance migrated by control) (*p < 0.05, **p < 0.001). C. Gal/analogs, CGP, gemcitabine at 5 μM, inhibit gemcitabine-naive (CaPan-1), MiaPaCa-GR and MiaPaCa-GTR PDAC cell invasion. D. Quantification of invaded PDAC cells was done by counting cells in quadrants. Only invading cells at the bottom of inserts were counted. Quantified invaded cells shows a significant inhibition of PDAC cell invasion (**p < 0.001).

Figure 7. Gal and analogs inhibit colony formation of PDAC cell

A. Gal/VNPP433-3β and gemcitabine significantly inhibit colony formation of ASPC1 and B. S2-013 cells. C. Gal/VNPP414/VNPP433-3β/CGP and gemcitabine decreased colony formation of S2-VP10 (left panel) and MiaPaCa-2 (right panel) cells. Both VNPP414 and CGP potentiated the effects of Gemcitabine and severely reduced colonies (C, right panel) in MiaPaCa-2 cells. D. Gal and VNPP433-3β significantly inhibited colony forming units (CFU) of gemcitabine-erlotinib-resistant PDAC cells and potentiated the effects of gemcitabine in further decreasing the number of colonies formed, contrary to gemcitabine alone which did not show strong inhibition of colonies. (D, right graph) Colony formation assays were repeated at least three times and colonies counted in four quadrants of the wells. Results are represented as averages with S.E.M. (*p < 0.05, **p < 0.001). Note: The numbers in parenthesis are concentrations in μM.

Figure 8. Gal and analogs downregulate stem cell factors and inhibit sphere formation in S2-013 PDAC cells

A. 200 cells resuspended in media to break up clumps and ensure single cell suspension were seeded in ultra-low adherent 24-well plates. After sphere formation, cells were treated with indicated compounds 2X in 14days and spheroids analyzed by light microscopy and images taken. B and C. Gal and VNPP433-3β deplete protein expression of putative stem cell factors in gemcitabine-naïve and gemcitabine-resistant PDAC cells. β-Catenin, Oct-4, Nanog, BMI-1 and cell proliferation factor c-Myc were downregulated after 24 h treatment period.

Figure 9. In vivo activity of gal and analogs

A. Effect of gal, VNPT55, VNPP414 and VNPP433-3β were evaluated in MiaPaCa-2 PDAC xenograft-bearing mice. Mice (n = 5) were administered with gal [0.26 mmol/kg (100.9 mg/kg)/twice daily], VNPT55 [0.26 mmol/kg (125.4 mg/kg)/twice daily], VNPP414 [0.068 mmol/kg (32.6 mg/kg)/twice daily], and VNPP433-3β [0.068 mmol/kg (30 mg/kg)/once daily], by intraperitoneal injection, 5 days per week for 32 days. Tumors were measured twice a week. Gal and analogs significantly inhibited tumor growth (** p< 0.001; %T/C: ratio of tumor volume in treated mice vs. control). Effective criteria for %T/C value according to NCI standard is ≤42% [50, 51]. TGI: Tumor Growth Inhibition index = [1-(mean volume of treated tumors)/(mean volume of control tumors)] × 100%). B. Mean body weights of mice were taken twice a week for the duration of the study. Mean body weights showed no significant toxicities to mice. C. H & E stain of liver, lung and kidney in both compound treated animals and vehicle treated animal show no gross organ abnormalities. D. Western blot analyses show a marked depletion of Mnk1/2, EZH2 and an upregulation of proapoptotic protein Bax by all four compounds in vivo.

Figure 10. Downregulation of oncogenic biomarkers in vivo

A. Immunohistochemical staining show significant depletion of Mnk1/2, peIF4E, BMI-1, SLUG, vimentin and PCNA from paraffinized sections of tumor tissue. B. Quantified IHC stains, show a significant downregulation of biomarkers in vivo (**p < 0.001).