IPA-3 inhibits the growth of liver cancer cells by suppressing PAK1 and NF-κB activation

Abstract

Hepatocellular carcinoma (HCC) is one of the major malignancies worldwide and is associated with poor prognosis due to the high incidences of metastasis and tumor recurrence. Our previous study showed that overexpression of p21-activated protein kinase 1 (PAK1) is frequently observed in HCC and is associated with a more aggressive tumor behavior, suggesting that PAK1 is a potential therapeutic target in HCC. In the current study, an allosteric small molecule PAK1 inhibitor, IPA-3, was evaluated for the potential in suppressing hepatocarcinogenesis. Consistent with other reports, inhibition of PAK1 activity was observed in several human HCC cell lines treated with various dosages of IPA-3. Using cell proliferation, colony formation and BrdU incorporation assays, we demonstrated that IPA-3 treatment significantly inhibited the growth of HCC cells. The mechanisms through which IPA-3 treatment suppresses HCC cell growth are enhancement of apoptosis and blockage of activation of NF-κB. Furthermore, our data suggested that IPA-3 not only inhibits the HCC cell growth, but also suppresses the metastatic potential of HCC cells. Nude mouse xenograft assay demonstrated that IPA-3 treatment significantly reduced the tumor growth rate and decreased tumor volume, indicating that IPA-3 can suppress the in vivo tumor growth of HCC cells. Taken together, our demonstration of the potential preclinical efficacy of IPA-3 in HCC provides the rationale for cancer therapy.

Figure 1. Inhibitory effect of IPA-3 on PAK1.

(A) Relative protein levels of PAK1 in various cell lines. The signal intensities of the bands were quantified and normalized by taking that level of MIHA as 1. (B) MTT assay on Day 1 and 2. H2M cells (4×10³) were seeded onto 96-well plates and treated with different dosages of IPA-3. The cells were harvested after incubation and the MTT assay was performed as mentioned in Material and Method. The graph showed the percentage of viable cells plotted against the dosage of IPA-3 (C) Western blotting analysis of the P-PAK1, total PAK1, P-JNK and total JNK. Serum-starved H2M cells were treated with either DMSO or IPA-3 at the indicated concentration for 15 minutes and followed by FBS replenishment and overnight culture. (D) H2M cells were serum-starved and treated with IPA-3 (10, 20 or 40 µM) in complete medium and allowed to grow overnight. The cell morphology images were captured at a magnification of 40X. Scale bar, 0.4 mm.

Figure 2. IPA-3 suppressed cell proliferation.

(A) The effect of IPA-3 on the cell proliferation rates of MIHA (upper, left panel), HepG2 (upper, middle panel) H2P (upper, right panel), H2M (lower, left panel) and MHCC97L (lower, right panel) cells. Statistical analysis was performed by comparing with the value of DMSO control. *P<0.05, **P<0.01 (ANOVA). (B) BrdU labeling assays of MIHA (upper, left panel), HepG2 (upper, middle panel), H2P (upper, right panel), H2M (lower, left panel) and MHCC97L (lower, right panel) cells. *P<0.05 (ANOVA) compared with the DMSO control. Error bars, mean ± SD of triplicate samples. (C) Representative plates of colony formation assay of MIHA (left panel), HepG2 (middle panel) and H2M cells (right panel). Bar chart of colony formation assay (lower panel). *P<0.001, **P<0.01 (ANOVA) compared with the DMSO control. Error bars, mean ± SD of triplicate samples.

Figure 3. The induction of apoptosis by IPA-3.

(A) Annexin V-7ADD staining assay. The fluorescence signals of annexin V-PE and 7-AAD were detected with PE-A and Per-Cy5-5-A channels, respectively (upper panel). The percentage of H2M cells stained with annexin V-PE only under different treatments was shown in a bar chart (lower panel). *P<0.001 (ANOVA) compared with the DMSO control. (B) Serum-starved H2M cells were treated with increasing dosages of IPA-3 together with serum replenishment for 24 hours. Western blotting analysis of PARP1, P-PAK1 (T423), total PAK1 and cleaved caspase 3 was performed.

Figure 4. The suppressive effect of IPA-3 on migration of H2M cells.

(A) Paxillin protein expression was detected by immunofluorescence analysis under IPA-3 treatment. H2M and H2P (upper panel) cells were serum-starved overnight and treated with either DMSO control or IPA-3 (20 µM) for 15 minutes, followed by FBS replenishment for 10 minutes. Immunofluoresence signals of phalloidin (Red), paxillin (Green) and DAPI (Blue) represent stress fiber, focal adhesion and nucleus, respectively (magnification 40X). The number of focal adhesion (paxillin) were counted in H2M (lower, left panel) and H2P (lower, right panel), and represented in the bar chart. Error bars, mean ± SD of triplicate samples. *P<0.01 (t-test) compared with DMSO control. (B) Western blotting analysis on the phosphorylation levels of PAK1 and paxillin. Serum-starved cells were treated with various concentrations of IPA-3 as indicated for 15 minutes, followed by FBS replenishment for 10 minutes. (C) Representative images of Transwell migration assay of H2M cells. Cells were treated with either DMSO or 10 µM IPA-3, and were allowed to migrate for 24 hours. Images show the cells having migrated to the lower chamber (upper panel). The number of migrated cells were counted and represented in the bar chart (lower panel). Error bars, mean ± SD of triplicate samples. *P<0.01 (t-test) compared with DMSO control.

Figure 5. The inhibitory effect of IPA-3 on NF-κB nuclear translocation.

(A) Effect of IPA-3 on subcellular localization of NF-κB was evaluated by immunofluorescence staining. After overnight serum starvation, H2M (left panel) and MIHA (right panel) cells were treated with either DMSO or IPA-3 (20 µM, 15 minutes) followed by an addition of TNF-α (20 ng/ml, 15 minutes). NF-κB was detected with a specific antibody (Green) and nucleus was stained with DAPI (Blue). (B) Western blotting analysis of P-PAK1 (T423) and total PAK1 were detected in the H2M cells stimulated by TNF-α (20 ng/ml) with or without IPA-3 pretreatment (20 µM, 15 minutes). TNF-α was included in the culture medium for 0, 0.5, 1, 2 or 4 hours. (C) Expression of quantitative real-time PCR was performed to analyze the mRNA level of MMP-9 (left panel) and COX-2 (right panel). Serum-starved H2M cells were treated with or without IPA-3 pretreatment (10 or 20 µM, 15 minutes) followed by TNF-α (10 or 20 ng/ml, 24 hours). Quantitative results of MMP-9 and COX-2 mRNA levels were normalized to β-actin. The values represented the mean ± SD of three independent experiments. *P<0.001 (ANOVA), ***P<0.05 (ANOVA) compared with the TNF-α control.

Figure 6. The suppressive effect of IPA-3 in nude mouse xenograft model.

(A) MHCC97L cells were used for the xenograft model. Mice were treated three times weekly either with DMSO or IPA-3 (2 mg/kg or 4 mg/kg, i.p.). *P<0.001 (ANOVA) compared with the DMSO control group. (B) Tumor weights were measured at the end of study. *P<0.001, **P<0.01, (ANOVA) compared with the DMSO control group. (C) Representative results of Western blotting analysis. P-PAK1 (T423), total PAK1, P-JNK and total JNK were detected. ***P<0.05 (ANOVA) compared with the DMSO control. Error bars, mean ± SD of 5 animals per group.