The antitumor effect of the novel agent MCL/ACT001 in pancreatic ductal adenocarcinoma

Pancreatic ductal adenocarcinoma (PDAC) is a major challenge in cancer therapy, there are more than four hundred thousand deaths per year, and the 5-year survival rate is less than 10%. The incidence continues to rise. Treatment with classic drugs offers limited therapeutic benefits. The aim of this study was to investigate the mechanism and effect of the new agent ACT001, the active metabolite of Micheliolide (MCL), in vitro and in vivo against PDAC. MTT assay, wound healing assay, and flow cytometry were used to assess the effects of MCL/ACT001 in vitro. DCFH-DA assay was used to assess ROS accumulation. Western blotting, immunohistochemical staining and TUNEL assay were also conducted to determine the mechanisms. PANC-1-Luc cells and bioluminescent reporter imaging were used to assess antitumor effect of ACT001 using a orthotopic xenograft model in vivo. MCL/ACT001 significantly inhibited cell growth in PDAC in a dose-dependent manner, induced cell apoptosis, cell migration and reactive oxygen species (ROS) accumulation in vitro. In vivo, ACT001 (400 mg/kg/day) inhibited PDAC tumor growth in orthotopic xenograft mice. We verified that EGFR and Akt were markedly overexpressed in PDAC cells and patient tumors. Mechanistic investigations revealed that MCL exerted its antitumor activity via regulation of the EGFR–Akt–Bim signaling pathway, thus inducing Bim expression both in vitro and in vivo. MCL/ACT001 is a highly promising agent in the treatment of PDAC patients.


Introduction
Pancreatic ductal adenocarcinomas (PDAC) make up the majority (~ 95%) of all pancreatic malignancies and are associated with poor overall survival (Connor and Gallinger 2022;Conway et al. 2019). It is the major cause of death in cancer patients. In the past 25 years, the number of cases has more than doubled around the world (Klein 2021). Most patients with PDAC present with symptomatic, surgically unresectable disease (Singhi et al. 2019). Treatment with the classic drugs gemcitabine, FOLFIRINOX or nab-paclitaxel offers limited therapeutic benefits, and the overall survival did not improve significantly because of drug resistance . Even though many molecular targeted agents are being tested in clinical trials, there are still very few new drugs that can be used to treat PDAC (Neoptolemos et al. 2018). Hence, it is urgent to find new agents that are safe and effective for the treatment of PDAC.
Epidermal growth factor receptor type 1 (HER1/EGFR) is overexpressed in PDAC and is associated with tumorigenesis and poor prognosis. Ardito, C. M. and Navas, C. showed that the development of PDAC, which was driven by K-Ras oncogenes, is totally dependent on EGFR signaling (Ardito et al. 2012;Navas et al. 2012). Bruns, C. J. showed that EGFR-Akt signaling inhibitors could inhibit pancreatic tumor growth and metastasis in an animal model. Moore, M. J. revealed that the EGFR inhibitor erlotinib combined with gemcitabine could improve anticancer effects and prolong the progression-free survival compared to gemcitabine alone (Bruns et al. 2000;Moore et al. 2007). Furthermore, Lihua Huang suggested that the EGFR-Akt signaling pathway played an important role in the progression of cell apoptosis (Huang and Fu 2015). Chen F revealed that EGFR-Akt 1 3 signaling has a close relationship with Bim, a critical mediator of apoptosis. His group found that that EGFR activation suppressed Bim expression . Chen, H. found that low expression of Bim is related to EGFR inhibitor resistance (Chen et al. 2017). Coloff JL revealed Bim's key role in Akt-mediated cell survival in cancer (Coloff et al. 2011). Reynolds C and Yue D found that inhibition of Akt promoted Bim-induced apoptosis in hepatocellular carcinoma and leukemia (Reynolds et al. 2014;Yue and Sun 2018). Reactive oxygen species (ROS) also play an essential role in the process of tumor development (Gorrini et al. 2013;Moloney and Cotter 2018). ROS overload could accelerate cell death.
Micheliolide (MCL) is a kind of guaianolide sesquiterpene lactone. It is extracted from Michelia compressa or synthesized from parthenolide (PTL), a natural product with many antitumor effects (Ghantous et al. 2013;Zhai et al. 2012). The half-life of MCL is 2.64 h in mouse plasma, while that of PTL is 0.36 h, which makes MCL a promising candidate for cancer therapy (Jin et al. 2007;Zhang et al. 2012). ACT001, the pro-drug of MCL, was synthesized and developed by Professor Yue Chen and could sustain the release of MCL over 8 h in humans (Xi et al. 2019). Studies have found that MCL or ACT001 has antitumor effects in many cancers, such as colitis-associated cancer (Viennois et al. 2014), rhabdomyosarcoma , breast cancer (Jia et al. 2015a;Jin et al. 2018), leukemia (Ji et al. 2016;Li et al. 2018) and glioma Guo et al. 2019). A phase II clinical trial for ACT001 to treat glioma patients is in progress. However, a study of MCL/ACT001 in PDAC has not been reported.
In this study, we investigated the antitumor effect of MCL/ACT001 in PDAC. MCL, the active metabolite of ACT001, not only induced ROS generation, and mitochondrial dysfunction, but also inhibited EGFR-Akt signaling and promoted Bim-induced apoptosis in PDAC cells. These results suggest that MCL/ACT001 could be a novel potential agent for the treatment of PDAC.

Cell lines and cell culture
The human PDAC cell lines PANC-1 and Mia-PaCa-2 and the human pancreatic ductal epithelial cell line HPDE6-C7 were cultured in DMEM containing 10% fetal bovine serum (FBS) and antibiotics (penicillin and streptomycin); HEK293T and BxPC-3 cells were cultured in RPMI 1640 containing 10% FBS and antibiotics (penicillin and streptomycin); and CFPAC-1 cells were cultured in IMDM containing 10% FBS and antibiotics (penicillin and streptomycin).
They were all cultured in a 5% CO 2 incubator at a constant temperature of 37 ℃.

Cell viability assay
In the exponential growth period, the cells were counted and mixed well, and 100 μL/well cell suspension was plated into 96-well plates. After incubation for 12 h, 100 μL/well MCL or ACT001 was added to the cells. When the time was up, 20 μL MTT (5 mg/mL) reagent was added to each well, and the cells were incubated for 2 h in the incubator. Then, the supernatant was discarded, and 100 μL DMSO was added to the cells. After 10 min of incubation, the absorbance was measured using a Tecan Infinite F50 microplate reader.

Wound healing assay
In the exponential growth period, the cells were plated into 6-well plates and then incubated for 24 h in a CO2 incubator. When the cells overgrew, a 200 µL pipette tip was used to make a scratch on the cell monolayer. Then, the medium was discarded, the cells were washed twice with PBS, and fresh medium with or without MCL was added to each well. Micrographs were obtained at 0 h and 24 h post-scratch.

JC-1 analysis for mitochondrial membrane potential (MMP)
A total of 5 × 10 5 cells/well were plated into 6-well plates, incubated for 12 h and then treated with MCL for 6 h. Subsequently, the cells were collected, centrifuged and incubated with JC-1 (2 μM) for 30 min at 37 °C, washed, resuspended in PBS and analyzed with fluorescence-activated cell sorter (FACS). MMP was quantitated by the ratio of red fluorescence to green fluorescence.

Measurement of ROS levels
For the ROS measurement, the cells were plated into black 96-well plates (Thermo # 165,303). After 12 h of incubation, 30 μM MCL was added, and the cells were treated for the indicated times. Then, the supernatant was discarded, and 10 μM oxidant-sensing fluorescent probe DCFH-DA (Beyotime) was added. After 20 min of incubation at 37 °C in the dark, the fluorescence was examined using a Tecan Infinite F50 microplate reader with the green fluorescence channels (488 nm excitation/525 nm emission).

Immunoblotting
Cells or tumor tissues were lysed using cold RIPA buffer on ice for 20 min. The lysate was centrifuged for 25 min at 12,000 rpm at 4 ℃. Then, the supernatant was collected and the protein concentration was measured using a Pierce BCA protein assay kit (Thermo # SH251403). The protein sample was added to SDS-PAGE gels at 20 μg/well, electrophoresed, and then transferred onto PVDF membranes at low temperature. After that, the membranes were blocked with 5% BSA diluted in TBST for 1 h and incubated with primary antibody overnight and secondary antibody for 1 h.

Cell apoptotic assay
A total of 5 × 10 5 cells/well were plated into 6-well plates and incubated for 12 h followed by incubation with various concentrations of MCL for 24 h. Cells were collected, centrifuged, washed twice with PBS, suspended in 500 μL binding buffer, stained with an Annexin V-APC/7-AAD apoptosis kit (MultiSciences Biotech, China), and then analyzed using FACS.

Mouse xenograft studies
Female Balb/c nude mice (Vitalriver Beijing, China) at the age of 6-8 weeks were used. Surgical implantation was performed to establish an orthotopic model. A total of 100 μL of 5 × 10 6 luciferase-expressing PANC-1 cells suspended in PBS/Matrigel (1:1) (BD Biosciences) were injected into the pancreas of anesthetized mice (isoflurane, RWD Life Science Co., Ltd.). After the operation, the mice were injected with ketoprofen (Yuanye Biotechnology Co., Ltd.) (Kowolik et al. 2019). When the bioluminescence intensity reached 1 × 10 8 , the mice were randomized into two groups: vehicle control (n = 10) and ACT001 (n = 10; 400 mg/kg/day, p.o.). Bioluminescent imaging was performed weekly. Animal health was monitored daily during the experiments. After 6 weeks, the mice were euthanized and the tumors were removed for the following studies.

Bioluminescent reporter imaging
Before imaging, intraperitoneal injection with D-luciferin (120 mg/kg, i.p.; PerkinElmer Inc.; #122,799) was performed. Then the mice were turned sideways on the black board of the instrument and the bioluminescent signals of the pancreas region were recorded using the IVIS Spectrum System.

Immunohistochemistry (IHC)
The paraffin-embedded tissues were sliced into 5 µm sections, dewaxed with xylene, and rehydrated with gradient ethanol. Citrate buffer was used to repair antigens. Then, the slices were incubated with fresh 3% H 2 O 2 (diluted with methanol) for 10 min, blocked with 10% goat serum (diluted with PBS) at room temperature for 30 min, incubated with primary antibodies at 4 °C overnight and secondary antibodies at room temperature for 30 min. After that, the sections were stained with DAB solution (Solarbio, DA1010) and hematoxylin for the appropriate time.

TUNEL assay
The apoptosis of tumor tissues was determined by the terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay according to the kit (absin, abs50021). Staining was assessed using fluorescence microscopy.

Statistical analyses
Data are represented as the mean ± SEM. We used GraphPad Prism 8.0 to analyze the statistics. All experiments were repeated at least three times unless otherwise stated. We used Student's t test to compare two groups. A p value less than 0.05 indicated statistical significance.

MCL induced a decrease in MMP, ROS generation and apoptosis in PDAC cells
To investigate whether MCL induces PDAC cell death via apoptosis, we treated PDAC cells with MCL for 24 h to perform FACS analysis (Annexin V and 7-AAD) using a flow cytometer. Annexin V was used to indicate early apoptosis, and cells with positive for 7-AAD and Annexin V indicated and CFPAC-1 cells were treated with 30 μM MCL for 0, 2, 4, and 6 h; BxPC-3 cells were treated with 30 μM MCL for 0, 3, and 6 h; Mia-Paca-2 cells were treated with 30 μM MCL for 0, 1, and 2 h, and the activity of ROS was detected with DCFH-DA probes. *P < 0.05, **P < 0.01, ***P < 0.001 late apoptosis. As shown in Fig. 2, MCL significantly induced PANC-1, CFPAC-1, BxPC-3 and Mia-PaCa-2 cell apoptosis in a dose-dependent manner. The decline in MMP is considered one of the symbolic events of early cellular apoptosis. Changes in MMP could be assessed by monitoring JC-1, which exhibits red fluorescence at high membrane potential, accumulates in mitochondria and exhibits green fluorescent monomer exits in the cytosol. MMP was measured in PDAC cells. PANC-1 and CFPAC-1 cells treated with MCL showed a reduction in red fluorescence and an increase in green fluorescence (Fig. 3A). In our analysis, previous research found that MCL exerted cytotoxic effects by generating ROS (Ji et al. 2016;Jia et al. 2015b;Wang et al. 2019;Xu et al. 2019); thus, the level of ROS in PDAC cells was also detected after the cells were treated Kaplan-Meier analysis with the optimal cut-off value of EGFR and Akt expression in human PDAC samples (n (high) = 89, n(low) = 89). C Immunoblot analysis of EGFR, P-EGFR, Akt, and P-Akt in PDAC tissues compared to adjacent normal tissues (N Normal, T Tumor). D Immunoblot analysis of EGFR, P-EGFR, Akt, and P-Akt in PDAC cells compared to HPDE6-C7 cells. E and F Immunoblots of lysates from PANC-1 cells treated with 10 μM or 20 μM MCL for 24 h were used to detect EGFR, P-EGFR, Akt, P-Akt, p-ERK, p-STAT3, STAT3, apoptosisrelated Bim and cl-Caspase3 protein expression. (G and H) Western blotting assays were performed to examine the expression of EGFR, P-EGFR, Akt, P-Akt, p-ERK, p-STAT3, STAT3, apoptosis-related Bim and cl-Caspase3 in CFPAC-1 cells treated with 20 μM MCL. I Bim protein levels were detected by immunoblot assay. J PANC-1 cells were transfected with Bim-siRNA or control-siRNA and then treated with 1 μM or 5 μM MCL. Cell survival was detected by MTT assay. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 1 3 with MCL for the indicated times. Our results showed that ROS accumulated in MCL-treated PDAC cells (Fig. 3B). These data indicated that MMP and ROS might contribute to the effects of MCL on PDAC cells.

MCL inhibited EGFR/Akt/Bim signaling in PDAC cells
EGFR-Akt hyperactivation and overexpression have been found in many cancers including PDAC (Ardito et al. 2012;Li et al. 2021;Nedaeinia et al. 2014). EGFR-Akt signaling was chosen to explore the potential mechanism of MCLinduced cell apoptosis. We searched the TCGA database (http:// gepia. cancer-pku. cn/ index. html) and found that EGFR and Akt were significantly overexpressed in pancreatic tumors compared to normal tissues (Fig. 4A). We then explored whether EGFR and Akt protein expression was associated with overall survival. Kaplan-Meier survival analysis was performed in 89 high-expression PDAC patients and 89 low-expression PDAC patients and found that higher EGFR and Akt expression positively correlated with reduced overall survival (Fig. 4B). To investigate the roles of EGFR and Akt in PDAC, we performed immunoblotting with PDAC cells and human tissues. We found that EGFR-Akt signaling was hyperactivated in PDAC cells and tumors compared to HPDE6-C7 cells and the paired noncancerous tissue regions (Fig. 4C, D). Treatment with EGFR-Akt inhibitors induced Bim upregulation and apoptosis in leukemia (Reynolds et al. 2014). Researchers revealed a key role of Bim in Akt-mediated cell survival through Akt-dependent stimulation in cancer (Coloff et al. 2011). Dose MCL affect the EGFR-Akt-Bim signaling pathway? Fig. 4E, F, G, H showed that MCL treatment for 24 h had little effect on ERK or STAT3 but inhibited EGFR and Akt phosphorylation and up-regulated the expression of the apoptosis-associated proteins Bim and cl-Caspase3 in PANC-1 and CFPAC-1 cells. To evaluate the role of Bim in the process of MCL-induced cell death, cells were transfected with Bim-siRNA or control-siRNA, and then treated with MCL. We found that Bim-siRNA-transfected cells had a statistically higher survival rate than the control-siRNA group after MCL treatment (Fig. 4I, J. These data indicated that MCL-induced apoptosis by inhibiting the EGFR-Akt-Bim signaling pathway.

MCL inhibited tumor growth in a xenograft tumor model
To study the efficacy of MCL/ACT001 against PDAC in vivo, tumor xenografts were established by orthotopically injecting luciferase-tagged PANC-1 cells into nude mice. The mice were randomized to ACT001 treatment and vehicle control until bioluminescence intensity reached 1 × 10 8 . Tumor growth was observed and photographed weekly by bioluminescence imaging (Fig. 5A). Bioluminescence imaging revealed that the tumor was statistically significant (2.5fold difference) between the treatment and control groups (Fig. 5B, C). The mice were euthanized after 6 weeks of treatment. These results indicated that MCL/ACT001 inhibited tumor growth. Furthermore, the tissue sections were stained with hematoxylin-eosin, and the vehicle-treated groups exhibited high grade adenocarcinoma. IHC analysis revealed that vehicle-treated tumors contained more Ki67positive cells and fewer Bim-positive cells than ACT001treated tumors, and P-EGFR-or P-Akt-positive cells were reduced in the tumors treated with ACT001 (Fig. 5D). The TUNEL assay revealed that ACT001-treated tumors displayed increased apoptotic cells (Fig. 5E). The immunoblot assay also showed that EGFR-Akt signaling was inhibited, and the apoptosis-related proteins Bim and cl-Caspase3 were up-regulated in ACT001-treated tumors (Fig. 5F). Collectively, our findings emphasize MCL/ACT001 as a new agent that has antitumor effects on PDAC in vivo and in vitro.

Discussion
PDAC is one of the most lethal epithelial malignancies, and the incidence and overall mortality rate are rising every year. Advanced PDAC patients respond worse to conventional chemotherapy and targeted therapy because of individual differences and drug resistance. It is very urgent to find new and effective drugs. ACT001 was synthesized and developed by Professor Yue Chen at our university and has been designated an orphan drug by the FDA. Its safety and tolerability in advanced glioma patients have been evaluated. In a phase I dose-escalation study, ACT001 dose levels ranged from 100 mg BID to 1200 mg BID. Patients tolerated treatment well with no dose-limiting toxicity. It is currently in phase II clinical trials for the treatment of recurrent glioma. The clinical trials of ACT001 are not limited to glioma. A Ib/IIa clinical trial for the treatment of neuromyelitis carried out in PLA General Hospital in China and a phase II clinical Bioluminescence imaging of orthotopically implanted 5 × 10 6 PANC-1-Luc cells in live mice. The mice were treated with the pro-drug ACT001 (400 mg/kg/day, p.o.) or vehicle for 6 weeks (n = 10). B The data are the relative mean bioluminescence intensity in tumors from ten mice per group. C Tumors of PANC-1-Luc cells implanted orthotopically from mice treated with either vehicle or ACT001. D H&E staining and IHC analysis of Bim, Ki67, P-EGFR, and P-Akt in xenograft tumors from mice treated with ACT001 or vehicle. Scale bars for HE, 50 μm. Scale bars for IHC, 25 μm. E The apoptosis of tumors treated with ACT001 was measured by the TUNEL method and examined using confocal microscopy. Scale bars, 25 μm. F Immunoblot analysis of EGFR, P-EGFR, Akt, P-Akt, Bim and cl-Caspase3 in xenograft tumors from mice treated with ACT001 or vehicle. G Action model of MCL on PDAC cells. ***P < 0.001 ◂ trial for the treatment of pulmonary fibrosis carried out in Australia. In these clinical trials, the mechanisms underlying the anticancer and anti-inflammatory effects of ACT001 are its regulatory effects on inflammatory processes and the cancer immune microenvironment, such as NF-κB and STAT3 signaling pathway inhibition and T cell regulation. MCL, the active metabolite of ACT001, has specific cytotoxicity in many kinds of cancers. MCL inhibits intestinal inflammation and colitis-associated cancer cell viability (Viennois et al. 2014), induces breast cancer cell death in vitro (Jia et al. 2015b), inhibits glioma  and rhabdomyosarcoma ) cell growth and induces acute myelogenous leukemia stem cell apoptosis in vitro (Ji et al. 2016). It also exerts antitumor effects on hepatocellular carcinoma (Yao et al. 2020). However, to date, there have been no reports about the effect of MCL/ACT001 on PDAC. In this study, we revealed that the novel agent MCL/ACT001 had antitumor growth effect on PDAC in vitro and in vivo.
In our research, we compared the sensitivity to MCL between PDAC cells and human pancreatic ductal cells. The IC50s of MCL in PDAC cells ranged from 7.21 to 14.43 μM, while the IC50 of MCL in human pancreatic ductal cells was 41.30 μM, which was 2.86-to 5.73-fold that in PDAC cells. These data suggested that MCL is more sensitive to PDAC cells than normal cells and might be a potentially valuable and safe drug for PDAC therapy. What truly matters for a good drug is its low toxicity and high efficacy. In subsequent study, we found that MCL inhibited PDAC cell migration and induced PDAC cell apoptosis. What is the molecular mechanism regulated by MCL?
EGFR is frequently overexpressed, and the EGFR-Akt signaling pathway is hyperactivated in different types of human cancers. It is a classic target for antitumor drug design (Nedaeinia et al. 2014;Sigismund et al. 2018). Akt also plays an important role in cell apoptosis (Tang et al. 2016). We interrogated the TCGA database, performed an immunoblot assay, and found that EGFR and Akt were overexpressed in PDAC tissues and cells. Kaplan-Meier survival analysis revealed that high EGFR or Akt expression positively correlated with reduced overall survival in PDAC patients. Researchers have shown that EGFR inhibition is effective against KRAS-wild-type PDAC patients, alters the tumor immune microenvironment and sensitizes other chemotherapy drugs in PDAC (Li et al. 2021) (Sidaway 2017). ACT001 inhibits PD-L1 transcription and modulates the antitumor immune response (Tong et al. 2020). PDAC patients who are treated with gemcitabine combined with erlotinib have a survival benefit (NCT01303029). Combined inhibition of EGFR and C-RAF completely regressed PDAC (Blasco et al. 2019). We can see that EGFR inhibition combined with other chemical drugs will be more effective for PDAC therapy.
The underlying mechanisms associated with the EGFR-Akt signaling pathway of MCL have been investigated. We found that MCL inhibited EGFR, P-EGFR, Akt and P-Akt protein expression and increased the proapoptotic protein Bim in PDAC cells. Immunoblot assays showed a remarkable increase in Bim and cl-Caspase3 protein following MCL treatment. Bim, a tumor suppressor, is markedly reduced in human cancer (Greenhough et al. 2010;Tan et al. 2005) and negatively regulated by Akt. Akt inactivation promotes Bim accumulation (Yue and Sun 2018). Knockdown of Bim partially blocked MCL-induced cell death, indicating that Bim might partially participate in the process of MCLinduced apoptosis. Previous studies found that MCL induces cancer cell death by decreasing MMP and increasing ROS generation (Jia et al. 2015b). Consistently, MCL treatment induced ROS production and decreased MMP in PDAC cells. Most importantly, the results of an animal experiment (Kowolik et al. 2019) indicated that MCL/ACT001 could efficiently inhibit pancreatic tumor growth with a maximal inhibitory rate of 60.0%.
Taken together, MCL/ACT001 effectively inhibited PDAC cell growth and induced cell apoptosis in vitro and in vivo. The effect of MCL/ACT001 was partially mediated by the EGFR-Akt-Bim pathway, and oxidative stress might also be involved in MCL-induced apoptosis. We hope to provide a new choice for PDAC combination therapy.