A small molecule inhibitor of atypical protein kinase C signaling inhibits pancreatic cancer cell transformed growth and invasion

Abstract

Pancreatic cancer is highly resistant to current chemotherapies. Identification of the critical signaling pathways that mediate pancreatic cancer transformed growth is necessary for the development of more effective therapeutic treatments. Recently, we demonstrated that protein kinase C iota (PKCι) and zeta (PKCζ) promote pancreatic cancer transformed growth and invasion, by activating Rac1→ERK and STAT3 signaling pathways, respectively. However, a key question is whether PKCι and PKCζ play redundant (or non-redundant) roles in pancreatic cancer cell transformed growth. Here we describe the novel observations that 1) PKCι and PKCζ are non-redundant in the context of the transformed growth of pancreatic cancer cells; 2) a gold-containing small molecule known to disrupt the PKCι/Par6 interaction, aurothiomalate, also disrupts PKCζ/Par6 interaction; 3) aurothiomalate inhibits downstream signaling of both PKCι and PKCζ, and blocks transformed growth of pancreatic cancer cells in vitro; and 4) aurothiomalate inhibits pancreatic cancer tumor growth and metastasis in vivo. Taken together, these data provide convincing evidence that an inhibitor of atypical PKC signaling inhibits two key oncogenic signaling pathways, driven non-redundantly by PKCι and PKCζ, to significantly reduce tumor growth and metastasis. Our results demonstrate that inhibition of atypical PKC signaling is a promising therapeutic strategy to treat pancreatic cancer.

Keywords: atypical PKCs; aurothiomalate; invasion; pancreatic cancer; transformed growth.

Figure 1. PKCι and PKCζ play non-redundant roles in pancreatic cancer cell transformed growth

A-C) Panc-1 cells transfected with lentivirus expressing either control, non-targeting (NT) RNAi, PKCζ-targeting RNAi, or PKCι-targeting RNAi, and also carrying either control vector (pBabe), or vector expressing PKCι or PKCζ were assayed for A) expression of PKCι, PKCζ and actin by immunoblot analyses, B) anchorage-independent growth (colony formation in soft agar), and C) cellular invasion through Matrigel-coated transwell chambers. B, C) Bars = average of 3 or more replicates±SD and graph is representative of 2 or more independent experiments.*p < 0.05 vs NT-pBabe; **p < 0.05 vs corresponding RNAi + control vector. D-F) Panc-1 cells transfected with two doses of NT RNAi (control), NT and PKCι RNAi, NT and PKCζ RNAi, or PKCι and PKCζ RNAi were assessed for D) protein expression by immunoblot analysis, E) anchorage-independent growth (colony formation in soft agar), and F) cellular invasion. E, F) Data are expressed relative to NT+control±SD, n=3 or more, and graph is representative of 2 or more independent experiments. *p < 0.05 vs NT-NT; **p < 0.05 vs corresponding NT+ PKCι or NT+ PKCζ RNAi cell lines.

Figure 2. Par6 knockdown phenocopies the effect of aPKC knockdown on pancreatic cancer cell transformed growth and invasion

Panc-1 (A) and MIA PaCa-2 (B) cells stably expressing either NT or Par6-targeted RNAi constructs were evaluated for Par6, PKCι, and PKCζ mRNA expression (PKC expression presented as 2^-dCT, Par6 expression =2^-dCT x 10 for presentation on same scale). C, D) The effect of Par6 knockdown on anchorage-independent growth (soft agar colony formation) and cellular invasion was assessed in (C) Panc-1 and (D) MIA PaCa-2 cells. Bar color indicates Par6-targeting RNAi construct utilized, corresponding to the labels in A, B). Results are expressed as percent NT control±SD n=3, and graph is representative of 2 or more independent experiments. *p < 0.05 vs NT.

Figure 3. ATM inhibits PB1 domain-mediated interactions

(A) Binding of GST-PKCζ (1-113, ▼) or GST (■) to Par6. Biotinylated Par6 was bound to streptavidin-coated plates and incubated with increasing concentrations of purified protein. Bound peptide was detected using the MesoScale Discovery detection system. (B) ATM inhibits binding of GST-PKCζ to Par6. Biotinylated Par6 bound to streptavidin-coated plates was incubated with 0.4μM of GST-PKCζ in the presence of increasing concentrations of ATM. The average of 3 replicates±SD is plotted and graph is representative of 2 or more independent experiments.

Figure 4. ATM inhibits aPKC downstream signaling and transformed growth in pancreatic cancer cell lines

A) Panc-1 and MIA PaCa-2 cells were pre-treated with ATM (50 μM) for 48hrs then assayed for Rac1 activity. Immunoblot analysis of Rac1 in precipitates (active Rac1) and total cellular extracts (total Rac1) is shown. The ratio of active Rac1/total Rac1 is plotted relative to control-treated cells (±SD, n=4 independent experiments *p < 0.05). B) Panc-1 (left) and MIA PaCa-2 (right) cells were treated with ATM (50 μM) for 48hrs then subject to immunoblot analysis for expression of p-STAT3, total STAT3, PKCι, PKCζ, and β-actin as a loading control. The ratio of active p-STAT3/total STAT3 is plotted relative to control-treated cells (±SD, n=4 independent experiments *p < 0.05). C) Immunoblot analysis of PKCι, PKCζ and β-actin expression in MIA PaCa-2, Panc-1, Capan-1 and CFPAC pancreatic cancer cell lines. D) Anchorage-independent growth was assessed in Panc-1 (open circles), MIA PaCa-2 (triangle), and Capan-1 cells (filled circle) in the presence of increasing concentrations of ATM (μM). Bar=average of at least 4 replicates±SD. E) Cellular invasion was assessed in Panc-1, MIA PaCa-2, and CFPAC cells pre-treated for 48hrs with ATM and in the presence of increasing concentrations of ATM (μM). Bar=average of 3 replicates±SD. Each panel is representative of 2 or more independent experiments.

Figure 5. Effect of aPKC knockdown on pancreatic cancer cell sensitivity to ATM inhibition of the transformed phenotype

A) Anchorage-independent growth was assessed in Panc-1 (left), and MIA PaCa-2 (right), cells stably expressing NT, PKCι, or PKCζ RNAi in the presence of control diluent (filled bars) or 50 μM of ATM (open bars). Bar=average of at least 4 replicates±SD. B) Cellular invasion was assessed in Panc-1 (left), and MIA PaCa-2 (right), cells stably expressing NT, PKCι, or PKCζ RNAi in the presence of control diluent (filled bars) or 50 μM of ATM (open bars). Bar=average of 3 replicates±SD. *p < 0.05 versus the same RNAi construct with control treatment, #p < 0.05 versus NT RNAi with same treatment. C) Cellular invasion was assessed in Panc-1 (left), and MIA PaCa-2 (right), transfected with two doses of NT RNAi (control), or PKCι and PKCζ RNAi, in the presence of control diluent (filled bars) or 50 μM of ATM (open bars). Bar=average of 3 replicates±SD, *p < 0.05 versus the same RNAi construct with control treatment, NS=difference is not statistically significant. Each panel is representative of 2 or more independent experiments.

Figure 6. ATM inhibits Panc-1 tumor size and aPKC signaling pathways in vivo

A) Tumor growth was monitored by bioluminescence (total flux, photons per second) detected by IVIS imaging of orthotopic Panc-1 pancreatic tumors in live, anesthetized mice. Representative bioluminescent imaging of mice with orthotopic Panc-1 pancreatic tumors treated daily with either saline or 60mg/kg ATM is shown. Mice were sacrificed after 4 weeks of treatment and tumors were analyzed for the following characteristics: B) final tumor weight, C) tumor BrdUrd incorporation, D) cleaved caspase-3, and E) tumor necrosis as described in Materials and Methods. F) Mice carrying orthotopic Panc-1 tumors were treated with daily injections of saline or 60mg/kg ATM for 7 days and harvested 24 hours after the final injection. Tumors were assessed for p-ERK1/2 (top panels), and p-STAT3 (bottom panels) by IHC staining. Bar=100 μm. G) Mice treated for 4 weeks with ATM or saline were evaluated for metastases to distal organs as described in Materials and Methods. The percentage of mice in each treatment group with confirmed metastasis to each distal organ is plotted, n = 16 per group, Difference in metastasis to each organ was analyzed by Fisher Exact test and *p < 0.05 was considered significant.

Figure 7. Summary of ATM-mediated inhibition of aPKC signaling