Background
Pancreatic cancer is an extremely aggressive disease that develops from non-invasive precursors, pancreatic intraepithelial neoplasia (PanIN) [
1]. Due to great advances in the understanding of pancreatic cancer biology, the molecular mechanisms underlying pancreatic cancer development have been fairly well elucidated. The activating mutation of the Kirsten rat sarcoma 2 (K-Ras), a viral oncogene homolog is one of the earliest event of the PanINs, which increases in frequency as the disease progresses [
2]. Epidermal Growth Factor family ligands (TGF-α and EGF) and their receptors (ERBB2, known as HER2/neu, and ERBB3) also function at the earliest stages of pancreatic neoplasia [
3]. Mutation of tumor suppressor p53 is a later stage event that is seen in more than 50% of patients with pancreatic adenocarcinoma [
4,
5]. As the tumor develops, the SMAD4/DPC4 gene is frequently altered, which is an indicator of poor prognosis in pancreatic adenocarcinoma [
6]. Loss of wild type breast cancer gene 2 (BRCA2) also appears in the late-stage of patients who inherited the germ-line heterozygous mutations of BRCA2 [
7,
8]. More recently, mutations in Ataxia Telangiectasia Mutated (ATM) have been identified in patients with familial pancreatic cancer [
9]. We additionally reported that Chromatin modifying protein 1A (Chmp1A) of the Endosomal sorting complex required for transport (ESCRT) family member plays a role in pancreatic cancer as a tumor suppressor [
10‐
12].
Based on the structure and function, ESCRT family is classified as ESCRT- 0, I, II, and III [
13]. They collectively act to transport the membrane-associated proteins such as receptor proteins into lysosomes for degradation via the formation and sorting of multivesicular bodies (MVBs). Chmp1A (called Did2/ Vps 46–1 in yeast) is a member of ESCRT-III and functions in the sorting of MBVs like the other members of ESCRT-III complex [
14,
15]. However, Chmp1A is unique since it is the only protein of ESCRT family that contains a nuclear localization signal (NLS) at its N-terminus [
12,
16]. In addition to their function in the formation and sorting of MVBs, recent studies have linked ESCRT family with a number of human diseases [
17,
18]. Correspondingly, we have provided the first evidence showing that Chmp1A functions as a tumor suppressor in the pancreas via the activation of the tumor suppressor p53 [
1]. We next have shown that NLS of Chmp1A is required to facilitate the growth inhibitory function of all-trans retinoic acid (ATRA) [
12]. We have further demonstrated that Chmp1A inhibits pancreatic tumor cell growth via the activation of ATM, and that the NLS of Chmp1A is important for the activation of ATM and p53 [
11].
New therapeutics have been developed based on these front-line molecular insights and clinically tried on patients. However, the therapeutics have produced disappointing results in clinical trials due to severe toxicity and development of resistance to medication [
19]. Thus, there is an urgent need for complementary therapeutic interventions to improve the efficacy of drugs by minimizing drug-mediated toxicity and resistance. To overcome toxicity and resistance of the therapeutics, non-toxic dietary supplement alone or in combination with low doses of therapeutics have been successfully assessed for the treatment of various cancers [
20]. Most dietary supplements have derived from plants, which encompass a large number of current medicines [
21]. Organic compounds are the biologically active ingredients of plant medicine and these are often isolated from plants or synthesized as analogues for diverse applications. Anacardic acid (AA) is one of the dietary supplements isolated from cashew apple and nut, the fruits of
Anacardium occidentale, and cashew nutshell liquid (CNSL, products of the tree). AA is also found in mangos [
22]. Although collectively referred as AA, cashew products were shown to contain several distinct AAs based on their unique side chains [
23].
AA has been shown to exert anti-proliferative activity in cancer cells such as breast, lung, and prostate. AA regulates various signaling pathways for its activities; association with p53 in breast and prostate cancer [
20,
24], inhibition of Src/FAK/Rho GTPase to block angiogenesis in prostate cancer [
24], and correlation with ATM in squamous cell carcinoma cells of lung [
25]. As for chemotherapeutics, 5-FU was shown to activate p53 [
26] and ATM in colorectal cancer [
27], and GEM to activate p53 in breast cancer [
20]. In addition, we have shown that Chmp1A activates ATM and p53 in pancreatic cancer cells [
10‐
12]. Based on these mechanistical findings, we have proposed that AA exhibits anticancer activity via the activation of Chmp1A, ATM and p53, which we refer to as the Chmp1A - ATM - p53 signaling pathway. We also propose that AA increases the anticancer efficacy of 5-FU and GEM, possibly by activating the same signaling pathway. In summary, the results of this study will provide the functions and the underlying mechanisms of new complementary medicine, which could be applied to the prevention and treatment of pancreatic cancers.
Methods
Reagents
Dimethyl sulfoxide (DMSO) was obtained from Thermo Fisher Scientific, MA, USA, Puromycin dihydrochloride, 5-Fluorouracil, and Gemcitabine hydrochloride from R & D system, Minneapolis, MN, USA, and Anacardic acid from EMD Millipore, Germany. Polyclonal antibody against Chmp1A was generated in our laboratory and successfully used [
10‐
12]. Other antibodies were purchased from commercial sources: rabbit polyclonal antibodies against p53, and monoclonal antibody against phospho- p53 at serine 15 from Cell Signaling, USA; monoclonal antibodies against Gapdh and phospho- ATM at serine 1981 from Pierce, USA; monoclonal antibody against ATM from Sigma-Aldrich, Germany. Goat anti-rabbit or mouse HRP conjugated secondary antibody were purchased from Chemicon-EMD Millipore, Germany.
Cell culture
All cell lines were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA). BXPC-3 cells were cultured in RPMI medium containing 10% fetal bovine serum (FBS), and 10,000 U/mL of penicillin-streptomycin. CAPAN-2, BXPC-3, and PANC-1 cells were cultured respectively in McCoy 5A, RPMI-1640, and Dulbecco’s modified Eagle’s medium (DMEM) that contain 10% FBS, and 10,000 U/mL of penicillin-streptomycin. Chmp1A silenced PANC-1 cells were maintained in DMEM media supplied with 10% FBS, 10,000 U/mL of penicillin-streptomycin and 1 μg/ml puromycin as described previously [
10]. All the cells were cultured at 37 °C under 5% CO2. The antibiotics, FBS and media were obtained from Invitogen- Thermo Fisher Scientific, USA. Fluorescence coupled secondary antibodies were obtained from Molecular Probes- Thermo Fisher Scientific, USA; Alexa Flour 488 and Alexa Flour 555 for green and red fluorescence, respectively.
MTT assays
AA, GEM and 5-FU were dissolved in DMSO as 100 μM stock and kept in the dark as recommended by the manufacturer. 30,000–70,000 cells per well were seeded onto wells in 24- well plate. Twenty four hours later the media were replaced with fresh media containing AA, 5-FU or GEM alone or in combination of AA and 5-FU or GEM, and cells were incubated at 37 °C. The doses were adjusted based on the published data [
20,
25]. MTT reagent (Sigma-Aldrich, Germany) was dissolved in PBS, filtered, and kept at 4 °C in dark. MTT assays were performed following the manufacturer’s instruction with a few modifications. Briefly, 24, 48, and 72 h later, 50 μl of MTT was added to each well and incubated for 4 h at 37 °C. The MTT precipitate was dissolved by pipetting up and down after addition of 500 μl of color development reagent (isopropanol plus 0.04 N HCl), and the optical density (OD) was measured at 590 nm (620 nm as reference) with a BioTek EON™ Spectrophotometer (Thermo Fisher Scientific, USA).
To mimic a 3- dimensional (3D) in vivo tumor environment, we cultured cells onto the Nunclon™ Sphera plates (Thermo Fisher Scientific, USA) that promote spheroid formations. Following manufacturer’s instructions, same number of cells was seeded onto the plates. The next day, the cells were treated with the appropriate reagent(s). Every other day, half of the media was replaced with fresh media containing appropriate reagent(s), and photos were taken 7 days after initial treatment using an Olympus 1X71 inverted microscope. The number of spheroids was counted, and the diameter of spheroid was measured. DMSO treated samples were used as a control.
Western blot analysis
We processed cells between 1 and 2 days after treatment with AA, 5-FU or GEM for WB analysis. WB was carried out following the protocol established in the lab by lysing cells with SoluLyse-M protein extraction reagent (Genlantis, CA, USA) containing protease inhibitor cocktail (Sigma-Aldrich, Germany) [
10‐
12]. BCA assay (Pierce- Thermo Fisher Scientific, USA) was performed to measure protein concentration on Microplate Spectrophotometer (BioTek EON™, UK). TGX™ precast gels (BioRad, CA, USA) were used to separate proteins, which were transferred to nitrocellulose membranes (Pierce-Thermo Fisher Scientific, USA). After incubation with primary and secondary antibody, the membranes were processed with chemiluminescence substrate (Pierce-Thermo Fisher Scientific, USA) and protein bands were visualized using ChemiDoc™ XRS imaging system (Bio-Rad, CA, USA).
Immunocytochemical analysis
For immunocytochemical analysis, we seeded cells over coverslips onto 12- wells plate, and treated cells with appropriate reagent(s) the next day. Within 2 days after treatment, the cells were fixed with 4% para-formaldehyde for 15 min, permeabilized with series of 75%, 50% and 25% of MeOH in PBS for 5 min each, and rehydrated with PBS, 3 times for 10 min each. The cells were incubated with blocking buffer (PBS/5% serum/0.1% Tween 20) for 30 min, and with primary antibody overnight at 4 °C. Next day, the cells were washed with PBS and incubated with fluorescence labeled secondary antibody till the protein expression was apparent. The cell- bound coverslips were mounted on the slides using Vectashield mounting media with Dapi nuclei marker (Vector Laboratory, INC, USA), and analyzed with Olympus FluoView FV1000 confocal microscope.
Silencing of Chmp1A protein
We generated stable cell lines with Chmp1A- specific shRNA (KD6) to silence Chmp1A protein. KD6 is one of the clones generated and have been used to knockdown (KD) Chmp1A protein. These stable cells were cultured in media containing puromycin following a previously described protocol [
10]. In these assays, we increased the amount of AA to induce substantial growth inhibition. And we tested whether Chmp1A is required for the growth inhibitory effect of AA. If Chmp1A is essential for the growth inhibitory function of AA, we hypothesize that AA would not induce growth inhibition in the absence of Chmp1A protein. Clone of cells expressing non-specific shRNA (NS2) was used as a control.
Statistical analysis
All the assays were performed in triplicates at three biologically independent times. Results from quantitative studies were expressed as Mean ± SD (standard deviation). Most of the comparisons between experimental and control groups were performed by one-way analysis of variance (ANOVA) followed by one-sided Student’s t-test. A p-value less than 0.05 was considered statistically significant.
Discussion
In this study, we examined the potential application of Anacardic acid as complementary medicine for pancreatic cancer by treating cells with AA alone or in combination with chemotherapeutics. AA has been shown to be a potential complementary medicine for breast and prostate cancers [
31], and this study demonstrates the similar paradigm in pancreatic cancer. Our data reveals that AA inhibits pancreatic tumor growth in 2D cell growth and 3D spheroid formation assays. The effect of AA on growth inhibition of pancreatic cancer was substantial especially in 2D cell cultures. However, the effect of AA on growth was somewhat dependent on the stage of tumor progression and on the number of cells treated. To achieve the similar growth inhibition, a higher amount of AA was required for the cells derived from more progressed compared to less progressed tumors. Also, for similar growth inhibition, higher amount of AA was required for the samples with more cells compared to less cells. This might imply that AA would benefit patients the most if it is incorporated in daily diet as prevention or therapeutic regimen for early stage of pancreatic cancer.
Dietary supplement in combination with chemotherapeutics has been assessed for the enhanced effect on various cancers in vitro, pre-clinical and clinical studies. Those studies have shown promising results for slowing down tumor growth and/or increasing survival rates [
32‐
34]. Thus, we investigated whether AA potentiates the anticancer activity of 5-FU or GEM, current chemotherapeutics for pancreatic cancer. Pancreatic cancer cells exhibited significant growth inhibition when 5-FU or GEM was applied individually in 2D cell growth and 3D spheroid formation assays. The chemotherapeutic induced growth inhibition was further increased when AA was added simultaneously. In summary, this study has provided strong evidence supporting that AA might be used as complementary or alternative medicine for the prevention or treatment of pancreatic cancer.
We then investigated whether Chmp1A is mechanistically involved in the growth inhibitory action of AA, 5-FU or GEM in pancreatic cancer cells. We have shown that Chmp1A suppresses pancreatic cancer cell growth by the activation of ATM and p53 [
10‐
12]. The anticancer activity of AA is shown to be associated with ATM and/or p53 [
20,
24‐
27]. Thus, we assessed Chmp1A, ATM or p53 proteins for their protein level and/or activation in the cells treated with AA, 5-FU, and GEM. AA treated cells have shown an increase in Chmp1A protein level, which was greatly intensified with Chmp1A overexpression, demonstrating the implication of Chmp1A in AA mediated growth inhibition. Unexpectedly, 5-FU and GEM treated cells also have exhibited an increase in Chmp1A protein level, a slightly greater increase than that in AA treated cells. This is the first data to show a molecular link between Chmp1A and the anti- cancer action of 5-FU or GEM. Further, our data on Chmp1A silencing experiments demonstrate that Chmp1A is necessary target signaling molecule for the growth inhibitory action of AA. However, silencing data did not show the necessity of Chmp1A for the anticancer activity of 5-FU or GEM, indicating the difference in signaling pathway between AA and chemotherapeutics.
We also have examined the activation of ATM and p53 using phosphorylated- ATM or p53 antibodies in immunocytochemical assays. Minor activation of ATM was noticed in the subset of cells treated with AA or 5-FU, but major activation in all the cells treated with GEM. With the addition of AA, 5-FU did not show a major change in ATM activation, but GEM showed a greater activation of ATM in all the cells examined. Our data on p53 activation in immunostaining was similar to that obtained from Western blot analysis.
Acknowledgements
The authors thank UPIKE - KYCOM for their tremendous support on the present study. The authors thank Dr. Carol Beech at the Proteomic Core Facility at the University of Kentucky for the service on proteomic analysis. The authors also thank Dr. Cathryn Rehmeyer at KYCOM, and Michelle Goff at UPIKE for editing the manuscript.