Combined targeting of Arf1 and Ras potentiates anticancer activity for prostate cancer therapeutics

Cancer cells PC3, DU145, 22Rv1, LNCaP, MDA-MB-231, T47D, H1299 and SW60 were obtained from American Type Culture Collection (ATCC) and passage <5 were used in this study. All cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum.

Exo2, BFA and Secin H3 were obtained from Selleckchem (Houston, TX). Salirasib and β-actin antibody were purchased from Sigma-Aldrich (St Louis, MO). Antibodies that recognize p-AKT, p-ERK1/2, p-STAT3, p-Src, AKT, ERK1/2, STAT3, Src and cleaved (c)-PARP were purchased from Cell Signaling Technology (Beverly, MA). Ki67 and Arf1 antibodies were purchased from Abcam (Cambridge, MA). Cell proliferation was determined by CellTiter 96® AQueous One Solution Cell Proliferation Assay (MTS) (Promega, Madison, MI) and crystal violet staining. Apoptosis was determined by FITC Annexin V Apoptosis Detection Kit with 7-AAD (BioLegend, San Diego, CA). Western blot, Arf1 activation, and Transwell invasion analyses were carried out as described previously [21, 28‐31].

Cell migration was determined using the Radius™ 24-well from Cell Biolabs (San Diego, CA). In this assay, cells were seeded on Radius cell migration plates and allowed to form monolayers before circular gaps were generated by removing the gels. Cells were then treated with DMSO or different drugs for 24 h, and the migratory gaps were captured at the same magnification using a Zeiss LSM-510 inverted microscope (Zeiss, Germany).

Six-week-old male NOD/SCID mice were purchased from the National Cancer Institute (NCI) and all animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Augusta University. To generate xenotransplantation models, exponentially growing PC3 cells (1.5 × 10⁶ cells) were suspended in 100 μl of PBS/matrigel (1:1) and injected subcutaneously into the right flanks of mice. One week after PC3 cell implantation, mice were randomized to receive control vehicle or drug(s) (n = 5). The control mice were injected intraperitoneally (i.p.) with 100 μl of sterile saline, whereas treatment groups received equal volume treatment of Exo2 and salirasib, alone or in combination. Exo2 or/and salirasib were administered intraperitoneally 5 days for beginning 1 week and every other day for the following 2 weeks. Tumor volume was measured externally every 3 to 5 days using vernier calipers as length × width² × 0.52. The mice were sacrificed on treatment day 21, and the xenografts were removed and processed for IHC with Ki67 and p-ERK1/2 antibodies as described previously [25, 28, 31].

Loss of Arf1 leads to reduced phosphorylation levels of ERK1/2 (25). Thus, we treated DU145 and PC3 cells with Arf1 inhibitors BFA, Secin H3 or Exo2 to explore whether they affect the MAPK pathway in prostate cancer cells. Neither BFA nor Secin H3 could reduce ERK1/2 phosphorylation significantly at 20 μM concentration, compared with the control (Fig. 1a). In contrast, Exo2 exhibited strong inhibitory effects on inactivation of ERK1/2 at the same concentration in both DU145 and PC3 cells (Fig. 1a). We next determined the effective dose of Exo2 by treating DU145 and PC3 cells with concentrations ranging from 10 μM to 50 μM. After 24 h of drug exposure, Western blot analysis showed that Exo2 effectively inactivated ERK1/2 at 20 μM and the inhibitory effect is dose-dependent (Fig. 1b). The time-course analysis of Exo2 treatment showed that phosphorylation of ERK1/2 was decreased following 8 h of exposure and much lower at 12 h after treatment (Fig. 1c). We further determined the possible pathway signaling involved in mediating the effects of Exo2 treatment. In the presence of Exo2, decreased phosphorylation of AKT was only found in DU145 cells, but not PC3 (Fig. 1c). Phosphorylation of STAT3 was not detectable in PC3 cells but it was increased in Exo2-treated DU145 cells (Fig. 1c). No significant changes in Src phosphorylation were observed with or without Exo2 (Fig. 1c). MTS assays showed that Exo2 inhibited proliferation of DU145, PC3 and another androgen-independent 22Rv1 cells in a dose-dependent manner (Fig. 1d). Different from DU145, PC3 and 22Rv1, LNCaP is androgen-dependent cells. Given the fact that AR signaling is also involved in the regulation of ERK1/2 signaling as well as prostate cancer survival, we next sought to determine the inhibitory effects of Exo2 on these cells. Similar tendency was observed in LNCaP cells that Exo2 dose-dependently suppressed cell proliferation. These data were confirmed by cell growth assays with crystal violet staining (Additional file 1: Figure S1), indicating that Exo2 has an anticancer activity in two major types of prostate cancer cells regardless of AR signaling. We next examined the effect of Exo2 on blocking Arf1 activation. The decreased GTP-bound Arf1 (activated form) was positively associated with reduced ERK1/2 activation (Fig. 1e). These data indicate that Exo2-induced repression of proliferation in prostate cancer cells, at least partially, through inhibiting the Arf1-ERK1/2 signaling cascade.

The Ras inhibitor salirasib can disrupt the spatiotemporal localization of active Ras through competing with Ras for binding to Ras-escort proteins [14‐16, 32]. We treated various cancer cell lines with salirasib for 72 h and found no significant inhibitory effect on cell proliferation in the presence or absence of 20 μM salirasib (Fig. 2a). When the concentration was increased to 50 μM, salirasib inhibited proliferation of prostate cancer DU145 and PC3 cells, but not of other types of cancer cells (Fig. 2a). To study the possible mechanisms involved in salirasib-induced repression of cell proliferation, we determined multiple critical signaling pathways by Western blot which revealed decreased phosphorylation of ERK1/2 in both DU145 and PC3 cells (Fig. 2b). Similar to the effects of Exo2, decreased phosphorylation of AKT and increased phosphorylation of STAT3 were observed in salirasib-treated DU145 cells (Fig. 2b). A dramatic decrease in Src phosphorylation was detected in PC3 cells during the first 24 h after salirasib treatment, while there were no significant changes in phospho-Src levels in DU145 cells in the presence or absence of salirasib (Fig. 2b). We then evaluated the synergistic effect of Exo2 and salirasib in PC3 cells. This drug combination inactivated ERK1/2 and suppressed cell proliferation more efficiently than either drug alone (Fig. 2c and d). Addition of salirasib in the treatment of Exo2 augmented the inhibitory effect of Exo2 on AKT phosphorylation in DU145 cells, but not in PC3 and LNCaP cells (Fig. 2c), suggesting that ERK1/2 activation is a common downstream target of this combined treatment. These data support a notion that prostate cancer cell proliferation can be suppressed more efficiently by co-inhibition of Arf1- and Ras-mediated MAPK signaling.

To address the question whether Exo2 or/and salirasib affect other phenotypes of prostate cancer cells, we treated DU145 and PC3 cells with Exo2, salirasib or their combination and determined cell migration, invasion and apoptosis in response to the different treatments. Gap closure migration and Transwell invasion assays showed that either Exo2 or salirasib inhibited cell migration and invasion, and the inhibitory effects of co-treatment were much greater compared with the single drug effects (Fig. 3a and b). Additionally, both Exo2 and salirasib induced apoptosis in prostate cancer cells and co-treatment led to a higher apoptotic rate (Fig. 3c and d). Western blot showed that salirasib enhanced Exo2-mediated induction of cleaved Caspase 3 and PARP (Fig. 3e). These observations indicate that the combination of Exo2 and salirasib is more potent in suppressing migration, invasion and promoting apoptosis in prostate cancer cells than either drug alone.

To assess the impact of Exo2 on prostate tumor growth in vivo, subcutaneous prostate tumor xenografts were established by injecting PC3 cells into NOD/SCID mice. When the xenografts were established, animals were randomly assigned to treatment or control groups. For drug treatment, Exo2 at different dosages (10 mg/kg, 20 mg/kg or 30 mg/kg) was administered to tumor-bearing mice for a total of 3 weeks. A significant reduction of prostate tumor burden was revealed as smaller tumor volume and lower tumor weight following treatment with 20 mg/kg or 30 mg/kg Exo2, compared with the control group (Fig. 4a and b). The tumor burden was not altered when mice were treated with 10 mg/kg Exo2 (Fig. 4a and b). Treatment with 30 mg/kg Exo2 led to a significant body weight loss of mice (Fig. 4b). Indeed, mice treated at the lower, but effective, dose of Exo2 (20 mg/kg) had less average weight loss than those treated with the higher dose (30 mg/kg) and it was tolerated over the duration of multiple treatments (Fig. 4b). These observations suggest that Exo2 dosage higher than 30 mg/kg can induce toxicity which may cause lethality. We collected the xenografts from the different treatment groups and performed Western blot to determine the activation status of MAPK signaling, which showed that phosphorylation levels of ERK1/2 were reduced in the Exo2 treatment group (Fig. 4c and d). Consistent with in vitro data, Exo2 inhibited ERK1/2 activation in a dose-dependent fashion (Fig. 4c and d). Taken together, these findings indicate that Exo2 exhibits anticancer activity by inhibiting the MAPK pathway and the effective dose of Exo2 is 20 mg/kg in SCID mice.

Based on these encouraging data, we generated the SCID-xenograft model to evaluate the synergistic effects of Exo2 (20 mg/kg) and salirasib (20 mg/kg) on prostate tumor growth. After a 3-week treatment, tumor burden was significantly reduced in the Exo2 arm, but not in the salirasib arm (Fig. 5a and b). Although dosage optimization of salirasib treatment in the mouse model still needs to be established, these data indicate that treatment of mice with 20 mg/kg salirasib alone cannot inhibit prostate tumor growth. The reduction in tumor growth was more significant in the mice receiving combination treatment compared with those receiving Exo2 alone (Fig. 5a and b). No significant body weight loss was observed in all these treatments (Fig. 5b). To determine whether MAPK signaling was involved in this synergistic treatment, we examined ERK1/2 activation using the cell lysates collected from the xenografts treated with Exo2 and salirasib, alone or in combination. Western blot analysis showed that either Exo2 or salirasib blocked ERK1/2 activation in prostate tumor cells (Fig. 5c). However, much lower phosphorylation levels of ERK1/2 were found in the dual-treated group compared to single drug-treated groups (Fig. 5c). IHC analysis with phospho-ERK1/2 antibody confirmed that salirasib augmented Exo2-induced repression of ERK1/2 activation (Fig. 5d and f). IHC staining also showed that treatment with Exo2 in the presence of salirasib can significantly reduce a cell proliferation marker Ki67 (Fig. 5e and f). These data demonstrate that combination of Exo2 and saralisab has a superior inhibitory activity on prostate tumor growth compared to monotherapy.