Cannabigerol Induces Autophagic-Cell Death by Inhibiting EGFR-RAS-Pathways in Human Pancreatic Ductal Adenocarcinoma Cell Lines

1. Introduction

Pancreatic Ductal Adenocarcinoma (PDAC) is an infiltrating neoplasm derived from the pancreatic ductal tree, and its 5-years overall survival rate increased from 3% to 9% in the last 50 years [1]. The worst prognosis is related to the absence of specific symptoms, diagnosis biomarkers, high invasiveness, and treatment resistance [2]. The PI3K/AKT/mTOR and RAF/MEK/ERK pathways, frequently activated in PDAC, are involved in mediating autophagy, apoptosis and chemoresistance, and they are known to drive PDAC development and tumor maintenance [3]. While pharmacological approaches to block RAS signalling were clinically unsuccessful, as demonstrated by no clinical benefit as monotherapy in trials [4], efforts are focusing on downstream proteins, such as RAF, MEK and Akt [5]. Different PI3K/Akt/mTOR inhibitors are in clinical evaluation, but their antitumoral effects may be limited by the over-activation of the RAS/RAF/MEK signalling [6]. Novel strategy to combine inhibitors of these pathways could be more effective, and some preclinical studies have evidenced improved efficacy in various tumour types [7]. Regarding PDAC, this adjuvant therapy was demonstrated to improve overall survival but in a limited number of patients [8], thus new pharmacological and combination therapies are needed. Phytocannabinoids as Tetrahydrocannabinol (THC) and Cannabidiol (CBD), compounds derived from Cannabis sativa L., as well as medical cannabis, were evidenced to exert an antitumoral activity when in combination therapies in glioblastoma, as reported in one clinical trial and in one case report [9,10]. Moreover, many in vitro and in vivo preclinical studies evidenced as phytocannabinoids show antitumoral effects in different human cancers, and synergistic effects with approved chemotherapeutic drugs, as recently reviewed [11].

The non-psychomimetic phytocannabinoid, cannabigerol (CBG) was evaluated for its potential anticancer effect in human cancer cell lines, including melanoma and glioblastoma [12,13]. Although the pathways activated by CBG were not yet clarified, recently CBG was demonstrated to bind the kinase active site of EGFR and inhibit its intracellular kinase activity, inducing apoptosis in EGFR-overexpressing A549 and A431 cell lines [14]. This is particularly important because EGFR protein family is involved in initialization and progression of PDAC and EGFR inhibitors are used in clinical trials as anticancer therapy [15].

In this study, CBG anticancer effect was investigated in two human PDAC cell lines, PANC-1 and MIAPaCa-2, in order to analyse its interactions with the EGFR-RAS related pathways involved in sustaining cell viability and chemoresistance.

2. Results

2.1. CBG inhibits cell growth of PDAC cell lines

PANC-1 and MIAPaCa-2 PDAC cell lines, and immortalized NCM460D and NHFA-12 cell lines were treated with different doses of CBG (up to 31.648 μg/mL) as single administration and cell viability was determined by MTT assay at 72 h post-treatment. Results showed that CBG reduces cell viability in both cell lines with an IC₅₀ of 15.64 μg/mL for PANC-1 and 13.77 μg/mL for MIAPaCa-2 (Figure 1), while lower cytotoxicity was observed in normal human NCM460D and NHFA12 cell lines (Supplementary Figure 1). For the next experiments, we selected two doses of CBG (11.08 and 12.66 µg/mL), as lower cytotoxic doses for studying molecular and biological mechanisms in PDAC cell lines and with no cytotoxic effects on normal cells. CBG was always used as single administration.

2.2. CBG induces autophagy by inhibition of EGFR and Akt/mTOR pathway in PDAC cell lines

EGFR and Akt/mTOR protein levels were evaluated by Western blot in PDAC cell lines at 48 h post-treatment with CBG 11.08 and 12.66 µg/mL. Immunoblots evidenced a decrease of EGFR expression in PANC-1 with CBG highest dose, while for MIAPaCa-2 the reduction was significant with both doses (Figure 2). Then, total Akt and its phosphorylation form were investigated. Data showed for PANC-1 cells a slight modulation of total Akt protein and a reduction of phospho-Akt levels after the administration of CBG. In MIAPaCa-2 cells, the highest dose of CBG induced a marked reduction of total Akt and also a significant decrease of its phosphorylation. Similar results were obtained for mTOR protein expression, that was reduced especially after the treatment with the highest dose of CBG (Figure 2).

Since CBG reduced the Akt/mTOR signalling pathway, also autophagy induction was evaluated. We observed the conversion of microtubule-associated protein 1A/1B-light chain 3 (LC3-I) to LC3-II form at 48 h post-treatment with CBG (11.08 and 12.66 µg/mL). CBG induced in both cell lines a significant increase of LC3-II lipidated form suggesting the activation of the early steps of autophagy; in particular MIAPaCa-2 cells showed the highest levels of LC3-II at the highest dose of CBG (Figure 3A).

To confirm the induction of autophagy, acridine orange dye (AO) analysis, which accumulates in acidic spaces and emits bright red fluorescence related to the degree of the acidity and the volume of acidic vesicular organelles (AVOs), was applied. AVOs are indicative of autophagy, therefore we evaluated AVOs under confocal fluorescence microscope. As shown (Figure 3B), an increase in the level of red fluorescent signals reflecting an AVO formation was observed in CBG-treated cell lines, 48 h post-treatment.

2.3. CBG reduces RAS downstream pathway in PDAC cell lines

To further investigate the CBG-induced effect in inhibiting EGFR pathway, the expression of total RAS (Pan-RAS) was assessed by Western blot analysis. As evidenced in Figure 4A, Pan-RAS levels were statistically reduced at 48 hr post-CBG treatment. Due to the relevance of the RAS/RAF/MEK pathway in PDAC, we measured the signalling proteins downstream of RAS in the lysate of PDAC cells treated with CBG at the most effective dose of 12.66 mg/ml. By MILLIPLEX® RAS-RAF Oncoprotein Magnetic Bead Panel 6-plex, we evaluated changes in phosphorylated BRAF (Ser446, pBRAF), CRAF (Ser338, pCRAF), and MEK1 (Ser217/Ser221, pMEK1), as well as the relative total protein levels (Figure 4B). The results confirmed that CBG treatment reduced total Pan-RAS levels in both cell lines. It also promotes the downregulation of BRAF, CRAF and MEK1 phosphorylation in PANC-1, and of BRAF in the MIAPaCa-2 cell line. Summarizing, these data further confirm that CBG was able to interfere with the EGFR-RAS pathways by reducing also downstream effectors often aberrantly activated in PDAC.

2.4. CBG induces apoptosis in PDAC cell lines

To further investigate whether reduction of cell viability and activation of autophagy was associated to cell death, PANC-1 and MIAPaCa-2 cells were treated with CBG 11.08 and 12.66 µg/mL and Annexin V-FITC staining analysis were performed 48 h post-treatment. Results showed that CBG increases Annexin V⁺ cells in a dose-dependent mode (Figure 5A), suggesting the induction of apoptotic cell death. The activation of ProCaspase-3 (ProCasp-3) and the presence of Caspase-3 (Casp-3) cleaved form was investigated by Western blot analysis. Blots showed an increase in Casp-3/ProCasp-3 ratio after 48 h post-treatment with CBG 12.66 µg/mL, in both cell lines (Figure 5B).

2.5. The combination of CBG with PTX or GEM induces higher cytotoxicity compared to administration of single drugs in PDAC cell lines

To support the potential use of CBG as integrative therapy in PDAC, we evaluated the combination effects of CBG with PTX and GEM, by MTT assay. Both cell lines were treated with CBG 7.91, 12.66 and 15.82 µg/mL in combination with three cytotoxic doses of PTX (1.5, 3 and 6 µg/mL) or GEM (25, 50 and 100 µg/mL) and cell viability was evaluated by MTT assay at 72 h post-treatment. Results showed that the combinations with the two higher doses of CBG resulted in greater cytotoxicity when compared to PTX or GEM alone, in both cell lines (Figure 6 A, B). Then drugs interaction was evaluated with SynergyFinder, using the Bliss model. We obtained a Bliss synergy score of 4.40 for PTX and CBG, and 10.87 for GEM and CBG, on PANC-1 cell line, indicating an additive and synergistic effect, respectively. In MIAPaCa-2 cell line, the Bliss synergy score was 17.19 for PTX and CBG and 14.05 for GEM and CBG combination, suggesting a synergistic effect (Figure 6 C, D).

3. Discussion

PDAC poor prognosis is mainly due to an aggressive behaviour associated with chemoresistance. The search of integrative therapies that can ameliorate chemotherapy side effects or improve their effect are constantly under study [18]. Cannabis sativa L. contains more than 100 phytocannabinoids and, for some of them, several biological properties are well known. Beside a direct anticancer effect, mainly demonstrated for CBD, in in vitro and in vivo experiments, phytocannabinoids were suggested to ameliorate numerous important side effects induced by chemotherapeutics [19]. In this study, CBG showed a dose-dependent cytotoxicity in PDAC cells, as also evidenced in other human preclinical cancer models as in glioblastoma multiforme (GBM) [13]. Further, THC, CBD and synthetic cannabinoids reduced PANC-1 and MIAPaCa-2 cell growth and viability, as reported in several studies and, in line with our evidence, MIAPaCa-2 was more sensitive than PANC-1 to treatments with phytocannabinoids [20,21]. In several pathologies, including cancer such as in glioma cells, cannabinoids have been demonstrated to activate autophagy and apoptotic cell death, through the interaction between apoptosis and autophagy signalling mechanisms [16,22]. Several pathways are mediated through the multiprotein complex involved in EGF/EGFR, including RAS and mTOR. These pathways suppress autophagy and promote proliferation and resistance to chemotherapy [23]. The simultaneous inhibition of EGFR and RAS/mTOR was demonstrated to provide a synergistic antitumor effect in various human cancers, indeed the PI3K/AKT/mTOR axis, a frequently dysregulated pathway in PDAC, is responsible for the control of cell proliferation and resistance [24], and these pathways can be inhibited by cannabinoids [16,25]. The present data evidenced as CBG induced autophagy by reducing the Akt-mTOR pathways with consequent LC-3 conversion and autophagic vesicle formation, as previously observed with others phytocannabinoids. For example, THC inhibits AKT/mTOR [22], reducing the proliferation of glioma cells. A similar finding was observed in carcinoma cells, where THC inhibits AKT/mTORC1 through ER stress-dependent activation of AMPK [26]. In our results, CBG reduced mTOR protein expression and, in line with this, some studies demonstrated that CBD inhibits mTOR signalling pathway in breast cancer and in human glioma [25,27]. Autophagy mechanism can be mediated by EGFR and Akt/mTOR signalling axis. We investigated the modulation of pro-autophagic markers and results showed that CBG increased LC3-II expression and in particular, for the more sensitive MIAPaCa-2, the increase has been very noticeable. Autophagy has a double function: to induce cancer resistance to chemotherapy and protect cancer cells from death, or to be correlated with cancer cell death. Herein, we evidenced that CBG induced autophagy and also apoptosis in both cell lines. Cannabinoids are also involved in reducing cancer cell growth, by the EGFR-RAS-RAF-MAPK pathway. In pancreatic cancer KRAS mutated upregulates EGFR endogenous expression and hyperactivation request in acinar to ductal metaplasia [28]. Herein, we showed that CBG suppresses EGFR expression in PANC-1 and MIAPaCa-2. Up to now, there are no data about the ability of CBG to reduce the EGFR expression; however, some studies demonstrated that CBD and THC can reduce EGFR expression in A549, H460 and H1792 cells and suppress EGF/EGFR signalling pathways in breast cancer [29]. Moreover, our data evidenced as CBG was able to reduce downstream RAS signalling, suggesting a specific role in decreasing the RAS oncogenic pathways in PDAC. In our studies, Annexin V positive cells and Caspase-3 cleavage confirmed CBG induction of apoptosis. In line with our results, CBG induced Caspase-3/-7 dependent apoptosis in glioblastoma and Caco-2 cells [13,14]. Moreover, in PANC-1 and MIAPaCa-2 cells also CBD induced apoptosis and Caspase-3 activation [20]. Lastly, many studies demonstrated that cannabinoids could increase the chemotherapeutic drugs efficacy, reducing tumour growth and overcoming drugs resistance [11]. Herein, the combination of CBG with GEM or PTX, increased the cytotoxicity compared to the administration of the drug alone, also showing a synergistic effect for some combinations. In GBM cells, CBD and CBG plus temozolomide did not show additive effect, but in cholangiocarcinoma cells, CBG synergized with GEM and cisplatin [13,30]. Moreover, in PDAC, CBD showed the ability to increase GEM and PTX efficacy in in vitro tests and KPC mice treated with CBD and GEM showed a survival three times longer than mice treated with GEM [20,21]. Overall, our data evidenced the ability of CBG to induce autophagy, reduce EGFR/AKT/RAS pathways, promote apoptotic cell death, and increase the sensitivity of PDAC cell lines to chemotherapeutic drugs.

4. Materials and Methods

4.1. Cell Lines

PANC-1 and MIAPaCa-2, human pancreatic ductal adenocarcinoma cell lines purchased from Sigma Aldrich (Milan, Italy), were cultured in DMEM high glucose medium (EuroClone, Milan, Italy) supplemented with 10% of fetal bovine serum (FBS), 2 mM L-glutamine, 100 IU/mL penicillin, 100 mg streptomycin and 1 mM sodium pyruvate. Epithelial cell line derived from the human normal colon (NCM460D), and human normal fibroblast (NHFA12) were used as normal cell for cytotoxicity assay. NCM460D was cultured in RPMI1640 supplemented with 10% of fetal bovine serum (FBS), 2 mM L-glutamine, 100 IU/mL penicillin, 100 mg streptomycin. NHFA12 was cultured as for PANC-1 cell line. Cell lines were maintained at 37 °C with 5% CO₂ and 95% of humidity.

4.2. Reagents

Pure CBG (≥98% purity) was purchased (Cayman Chemical, cat. No. 36975). CBG was solubilised in ethanol 70% (Et-OH) at 50 mM (15.8 mg/mL) concentration. Paclitaxel (PTX, 6 mg/mL; cat. No. 33069-62-4) and Gemcitabine (GEM; 50 mg/mL; cat. No. 122111-03-9) supplied by (Sigma Aldrich) were dissolved in water. All the compounds were aliquoted, stored at -20°C and each aliquot was used one time.

4.3. MTT Assay

3 x 10⁴ cells/ml were seeded in 96-wells plate. The following day, compounds or vehicles were added, and six replicates were used for each treatment. At 72 h post-treatment, cell viability was analysed as previously reported [16]. The absorbance of the sample was measured at 570 nm using an ELISA reader microliter plate (BioTek Instruments, Winooski,VT).

4.4. Annexin V Staining

Annexin V-FITC staining and cytofluorimetric analysis was applied for detection of apoptotic cells on PDAC cell lines. 3x10⁴ cells/ml were treated with different doses of CBG for 48 h. After treatment, cells were stained with Annexin V-FITC (5 ml; Adipogen, No. AG-40B-0005F) and the percentage of positive cells analysed by FACScan flow cytometer using the CellQuest software.

4.5. Western Blot Analysis

PDAC cell lines were treated with CBG and harvested at 48 h post-treatments. Lysates were obtained with lysis buffer (TRIS 1M pH 7.4, NaCl 1M, EGTA 10 mM, NaF 100 mM, Deoxycholate 2%, EDTA 100 mM, TritonX-100 10%, Glycerol, SDS 10%, Na2P2O7 1M, Na3VO4 100 mM, PMSF 100 mM, Cocktail of enzyme inhibitors and H2O), separated on a SDS polyacrylamide gel, and Western blot were performed as previously reported (Marinelli et al., 2020). The following antibodies were used: mouse anti-mTOR (1:1000, GeneTex), mouse anti-EGFR (1:5000, GeneTex), rabbit anti-pAkt (1:1000, Cell Signaling), rabbit anti-Akt (1:1000, Cell Signaling), rabbit anti-LC3 (1:1000, Cell Signaling), rabbit anti-Caspase-3 (1:1000, Cell Signaling), mouse pan-RAS (1:500, Santa Cruz Biotechnology), mouse β-actin (1:500, Santa Cruz Biotechnology) and mouse anti-glyceraldehydes-3-phosphate dehydrogenase (GAPDH, 1:1000, Santa Cruz Biotechnology). Abs were incubated overnight or 1 h according to manufacturer’s protocol and then incubated with their respective HRP-cojugated anti-rabbit or anti-mouse (1:2000, Cell Signaling, Danvers, MA, USA) Abs for 1 h. Peroxidase activity was visualized with the LiteAblot®PLUS or TURBO (EuroClone, Milan, Italy) kit and densitometric analysis was carried out by a Chemidoc using the Quantity One software version 4.6 (Bio-Rad, Milan, Italy).

4.6. Acridine Orange Staining

To detect acidic vesicular organelles, the vital staining of cells with acridine orange (AO, Sigma-Aldrich) was performed. 3x10⁴ cells/mL were seeded in 12-well plates and treated with CBG at 11.08 and 12.66 µg/mL for 48 h. Then, cells were stained with 1 μg/mL AO, washed in PBS, immobilized on slides using the cytospin centrifuge and analysed with C2 Plus confocal laser scanning microscope (Nikon Instruments, Firenze, Italy). Optimized emission detection bandwidth was configured by Zeiss Zen control software. Images were processed using NIS Element Imaging Software (Nikon Instruments, Firenze, Italy). Cytoplasm and nuclei of AO-stained cells fluoresced bright green, whereas the acidic autophagic vacuoles fluoresced bright red.

4.7. Milliplex multiplex assay

The levels of Total RAS, pBRAF, pCRAF, pMEK1 in treated cells were measured using RAS-RAF Oncoprotein Panel 6-Plex Magnetic Bead Kit 96-well Plate (EMD Millipore Corporation, Billerica, MA, USA) following the manufacturer's protocol. Data were analysed using Luminex® MAGPIX® instrument with xPONENT® software (Luminex Corporation, Austin, TX, USA).

4.8. Statistical Analysis

The data presented represent the mean with standard deviation (SD) of at least 3 independent experiments. The statistical significance was determined by Two Way-Anova followed by Dunnett's or Sidak’s multicomparison test, using GraphPad software.

4.9. Drug interaction

Drugs interaction was evaluated with SynergyFinder, using the Bliss model [17]. Bliss independence reference model was used for multiplicative effect of single drugs as if they acted independently. Bliss synergy score larger than 10 is considered synergistic, from -10 to 10 is considered additive and less than -10 antagonistic.

5. Conclusions

In conclusion, our results showed that CBG, a non-psychomimetic cannabinoid from Cannabis Sativa L., can induce an anticancer effect on two human PDAC cell lines, supporting the ability of cannabinoids in interfering with several pro-tumoral pathways. Further study in pre-clinical in vivo models will be performed to better understand CBG effects on PDAC progression.