Effect And Mechanism Of Triptolide Derivative LB - 1 On Pancreatic Cancer Resistance

Pancreatic cancer is an aggressive human cancer characterized by disruption of molecular mechanisms involved in cell proliferation, invasion and metastases. Pancreatic cancer shows limited susceptibility or high resistance to radiation treatments and conventional classes of cytotoxic drugs, which leads to poor prognosis with short median survival. That growing evidence in clinical and preclinical studies has implicated hypoxia is widely present in pancreatic cancer and hypoxia inducible factor 1 a (HDF-1α) and signal transducer and activator of transcription-3 (Stat3) occur at a surprisingly high frequency in pancreatic cancer. Both HIF-1α and Stat3 are key nuclear transcriptional factors, which control the expression of genes involved in proliferation, angiogenesis, invasion and metastasis and cell cycle progression, et al. Efforts to develop inhibitors of HIF-la and Stat3 are under way, but, so far, there is still no antitumor drug goes on sale, whose main mode of action is targeting HIF-1α or Stat3. Therefore, it is important to develop novel medicines targeting both HIF-1α and Stat3 to treat pancreatic cancer.Triptolide (LA) is an active ingredient of the traditional medicine tripterygium wilfordii Hook F. Recent study shows that LA could inhibit the transcriptional activity of HIF-1α to exert antitumor activity in ovarian cancer. But LA does not go further to the clinical trial due to its poor water solubility and severe toxicity. Therefore, we switch our focus on LA derivatives. In a series of structural-modified compounds, LB-1 is a novel, easy synthesis and potent LA derivative with less toxicity. Therefore, more detailed researches are focused on the role and mechanism of LB-1 in this paper.We sought to address the above issue in the following two aspects:the role of HIF-1α and Stat3 in pancreatic cancer with LB-1; the effects of LB-1 in pancreatic orthotopic model on epithelial-mesenchymal transition (EMT), cell cycle and apoptosis.1. LB-1 exerts antitumor activity in pancreatic cancer by inhibiting HIF-la and Stat3 signaling.1.1. The screening of LA and its derivatives on HIF-1 inhibitory activity.We firstly characterized the effects of LB-1 on HIF-1 ctivity in HRE-U251 and pGL3-U251 cells. The results showed that LB-1 decreased hypoxia induced luciferase in a dose-dependent manner, with an ICso of 0.123μM. Parallel MTT assay dissociated the cytotoxic activity of LB-1 from the inhibition of HIF-1 activity.1.2. Differential protein expression profiling by isobaric tags for relative and absolute quantitation (iTRAQ) proteomics of LB-1 in HRE-U251 cells.We adopted iTRAQ proteomics to verify HRE-U251 screening model and evaluate the potential mechanism of LB-1. The results displayed that hypoxia promoted expression of genes (z-score>2) related to tumor progression, such as HIF-1α, TGFβ1, STAT4, NF κ B, TGFβ3, et al. LB-1 decreased the expression of genes, such as ANGPT2, HIF-1α, p38MAPK (z-score<-1). Therefore, it is credible of HRE-U251 to be a screening model to select HIF-1 a inhibitor and LB-1 probably inhibited HIF-1α transcriptional level to exert antitumor activity.1.3. Antitumor activity of LB-1 in pancreatic cancer.MTT and colony formation assays were applied to determine the efficacy of LB-1. MTT assay showed that LB-1 and LA could inhibit proliferation of solid tumor cell lines derived from different tissues and organs. The anti-proliferative activity of LB-1 is slightly lower than that of LA in Mia-PaCa2 and SW1990 cells. Colony formation assay further confirmed the result of MTT assay.We next assessed the efficacy of LB-1 in Mia-PaCa2 xenograft model. We found that LB-1 displayed significant antitumor activity in a dose-dependent manner. The inhibition rates of LB-1 (1.0,2.0,4.0 mg/kg, i.p.), based on tumor weight change, were 36.06%,57.10% and 68.39%, respectively. The inhibition rate of LB-1 (8.0 mg/kg, ip) was 95.16%, which achieved maximum efficacy in Mia-PaCa2 xenograft model. And orally administered LB-1 still displayed significant antitumor activity; the inhibition rate of LB-1 (8.0mg/kg, ig) was 80.81%. The body weight of mice only slightly reduced whether orally administered or intraperitoneal injection of LB-1. Taken together, these results indicated that LB-1 had minimal systemic toxicity compared with LA.1.4. The mechanism of LB-1 on HIF-1α and Stat3 signaling pathway.To further prove whether the effect of LB-1 on HIF-1 a is associated with its antitumor activity, HIF-1α shRNA was applied to effectively knockdown HIF-1α in SW1990 cells. MTT assay results demonstrated that the silencing of HIF-1α partially prevented the cytotoxicity of LB-1. Western blotting and real-time reverse transcription-polymerase chain reaction (Q-PCR) assays showed that LB-1 promoted degradation of HIF-1α via the ubiquitin-proteasome pathway and significantly affected HIF-1α mRNA expression under hypoxic condition. The effect of LB-1 on HIF-1α took place at transcriptional level. Further studies showed that LB-1 suppressed Stat3 phosphorylation at Tyr 705 residue. To further prove whether the inhibitory effect of LB-1 on p-Stat3 is associated with HIF-1α activity, SW1990 cells were exposed to the Stat3 shRNA which could effectively knockdown Stat3. Western blotting results depicted that Stat3 shRNA not only decreased the protein levels of p-Stat3 and Stat3, but reduced protein expression of HIF-la and common target gene vascular endothelial growth factor (VEGF). Combined treatment with both LB-1 and Stat3 shRNA resulted in a synergic inhibitory effect on HIF-1α and VEGF expression. Co-IP assay results showed that LB-1 could inhibit this association among HIF-la, p-Stat3 and p300. Q-PCR, enzyme-linked immunosorbent assay (ELISA) and human umbilical vein endothelial cell (HUVECs) tube formation assays results further demonstrated that LB-1 significantly inhibited target gene VEGF levels. In accordance with in vitro results, LB-1 (4.0mg/kg) treatment could inhibit HIF-la, p-Stat3 and Stat3 protein accumulation, and interestingly, decrease mammalian target of rapamycin (mTOR) and ribosomal S6 kinase P70 (P70S6K) phosphorylation. Moreover, LB-1 (4.0mg/kg) treatment also decreased mRNA levels of HIF-la and VEGF using Q-PCR assay.Therefore, on the one hand, prevention of Stat3 activation, which works with HIF-la, mediates antitumor and anti-angiogenesis activities of LB-1 in pancreatic cancer. On the other hand, LB-1 suppressed HIF-la activity probably via inhibition of upstream phosphatidylinositol 3-kinase (PI3K)/Akt/mT0R pathway.2. LB-1 suppresses pancreatic tumor growth and metastasis by inhibiting EMT and inducing apoptosis in SW1990-G-luc nude pancreatic orthotopic model.LB-1 could inhibit invasion and metastasis in Mia-PaCa2 and SW1990 cells in a dose dependent manner by transwell assay. And the invasion and metastasis inhibition of LB-1 was not due to cytotoxicity of LB-1. We successfully established SW1990-G-luc nude pancreatic orthotopic model. Then we employed this orthotopic model to evaluate antitumor growth and invasion and metastasis of LB-1. Results showed that LB-1 could inhibit the growth of pancreatic cancer and reduce the range of invasion and metastasis in SW1990-G-luc nude pancreatic orthotopic model.Considering that invasion and metastasis are characterized by EMT in a variety of tumors including pancreatic cancer. Western blotting and Q-PCR indicated that LB-1 significantly promoted the expression of epithelial biomarker E-cadherin and decreased mesenchymal biomarker Vimentin, β-catenin and transcriptional repressor Snail levels. Flow cytometry and confocal further confirmed the results of western blotting and Q-PCR. In order to clarify the role of Snail in reversed EMT process by LB-1, Snail over expression plasmid was transiently transfected with pancreatic cancer Mia-PaCa2 and SW1990 cells, western blotting results demonstrated that Snail overexpression did not completely eliminated epithelial cells characteristic induced by LB-1. We next adopted HIF-1α shRNA to effectively knockout HIF-1α in SW1990 cells, and the results showed that HIF-1 a shRNA could effectively reverse the EMT process. Combined treatment with both LB-1 and Stat3 shRNA did not observed the further EMT process reversion, indicating that HIF-1 a plays a crucial role in regulation of EMT by LB-1.We next focus on the effect of LB-1 on cell cycle. LB-1 could significantly arrested Mia-PaCa2 and SW1990 cells in S phase by flow cytometry.LB-1 could significantly induce cell apoptosis in Mia-PaCa2 and SW1990 cells determined by flow cytometry. And LB-1 decreased levels of Caspase 9, Caspase 8, Caspase 3 and PARP, and increased the protein expression of C-Caspase 9, C-Caspase 8, C-Caspase 3 and C-PARP. These results demonstrated that LB-1 was capable of inducing pancreatic cancer cell apoptosis through a mitochondrial-dependent endogenous and exogenous caspases apoptotic pathway, ultimately shearing PARP, causing inhibition of DNA repair and promoting degradation of DNA.In summary, LB-1, as a potent HIF-1 a and Stat3 inhibitor, suppressed HIF-1α transcriptional level and Stat3 protein accumulation, blocked the physical interactions among HIF-1α/p300/p-Stat3, decreased the combination with HRE, thus inhibited HIF-1α transactivation to reduce VEGF expression, thereby hypoxia-induced angiogenesis. Moreover, LB-1 reversed EMT process, arrested cells in S phase which blocked DNA synthesis of tumor cells to induce cell apoptosis, and ultimately inhibited the growth of pancreatic cancer cells in SW1990-G-luc nude pancreatic orthotopic model. Together, LB-1 exhibits antitumor effect and can be potentially developed as a novel drug to treat pancreatic cancer.