Non-small cell lung cancer (NSCLC) is the leading cause of cancer deaths worldwide; however, only limited therapeutic treatments are available. Hence, we investigated the role of cannabinoid receptors, CB1 and CB2, as novel therapeutic targets against NSCLC. We observed expression of CB1 (24%) and CB2 (55%) in NSCLC patients. Furthermore, we have shown that the treatment of NSCLC cell lines (A549 and SW-1573) with CB1/CB2 and CB2-specific agonists Win55,212-2 and JWH-015, respectively, significantly attenuated random as well as growth factor-directed in vitro chemotaxis and chemoinvasion in these cells. We also observed significant reduction in focal adhesion complex, which plays an important role in migration, upon treatment with both JWH-015 and Win55,212-2. In addition, pre-treatment with CB1/CB2 selective antagonists AM251 and AM630, prior to JWH-015 and Win55,212-2 treatments, attenuated the agonist-mediated inhibition of in vitro chemotaxis and chemoinvasion. Additionally, both CB1 and CB2 agonists Win55,212-2 and JWH-133, respectively, significantly inhibited in vivo tumor growth and lung metastasis (~50%). These effects were receptor-mediated as pre-treatment with CB1/CB2 antagonists abrogated CB1/CB2 agonist-mediated effects on tumor growth and metastasis. Reduced proliferation and vascularization along with increased apoptosis was observed in tumors obtained from animals treated with JWH-133 and Win55,212-2. Upon further elucidation into the molecular mechanism, we observed that both CB1 and CB2 agonists inhibited phosphorylation of AKT, a key signaling molecule controlling cell survival, migration and apoptosis, and reduced MMP-9 expression and activity. These results suggest that CB1 and CB2 could be used as novel therapeutic targets against NSCLC.

Cell culture

Human NSCLC cells, A549 cells and SW-1573 cells were obtained from American Type Culture Collection (ATCC). The cell lines have been characterized by ATCC and we have not done any further characterization. Cells were cultured and maintained in DMEM or RPMI 1640, supplemented with 10% heat-inactivated fetal bovine serum (FBS), 5 units/mL penicillin, and 5 mg/mL streptomycin.

Cell viability and Proliferation

Cells grown for 24 h at a density of 1 × 10⁴ cells per 100 µl of complete medium in 96-well plate were incubated with JWH-015/Win55,212-2 (1–20 µM) (Tocris Cookson, Ellisville, MO) for 24 h. Cell viability was determined by the CellTiter 96 Aqueous One Solution Reagent (MTS, Promega, Madison, WI), according to the manufacturer’s instructions.

For proliferation assay, 2 × 10⁵ cells/well were plated in six-well plates in the presence of different concentrations of JWH-015 or Win55,212-2 or vehicle in presence of EGF (100 ng/ml). Cells were trypsinized and counted using a haematocytometer after 24, 48 and 72 h (11), (17).

Cell Scattering assay

1×10⁴ cells/well were plated in 6-well culture plates. Once they formed small colonies, they were serum starved for 24 h and pretreated with JWH-015 or Win55,212-2 for 30’ prior to stimulation with EGF (10 ng/ml) and incubated for another 48 h. Cell scattering was examined by phase-contrast microscopy and photographed.

Cell migration and invasion assays

A series of assays were used to study the effect of cannabinoid receptor activation on migratory and invasive potential of NSCLC cells. In wound-healing assay, cells were plated at 70% confluence in 10% serum-DMEM. At 24 h after seeding, the monolayers were wounded by scoring with a sterile plastic 200-µl micropipette tip, washed, and fed with serum-free DMEM (0.1% FBS). After culturing the cells in presence or absence of JWH-015 and Win55,212-2 with pertussis toxin (PTX) and EGF (10 ng/ml) for 72 h, cells were fixed with 4% paraformaldehyde in PBS for 5 min at room temperature and photographed using a low-magnification phase-contrast microscope. The extent of migration into the wound area was evaluated qualitatively.

For trans-well chamber (Corning-Costar, Lowell, MA) migration assays, 100 µl of 1 × 10⁶ cells were pretreated with JWH-015 or Win55,212-2 prior to antagonist (PTX, AM630 or AM251) (Tocris Cookson, Ellisville, MO) treatment and were seeded in the upper chamber of transwells in serum-free medium as described previously (10). DMEM with 10% serum or EGF (10 ng/ml) in presence of corresponding cannabinoid agonist/antagonist was added to the lower chamber. Cells that migrated into the lower chamber were counted 12 h after stimulation with EGF.

In in vitro invasion assay, pre-coated Matrigel 24-well invasion chambers (BD Biosciences, San Jose, CA) were used. Treatment of the cells was similar to migration assay as above. After 24 hr stimulation with EGF, cells adherent to the outer surface of the filter membrane separating upper and lower chambers were fixed and stained with Harris modified Fisher hematoxylin (Fisher, Waltham, MA) and the cells counted (11, 17).

Immunoblot analysis

Extraction of proteins from cultured cells and immunoblot analysis was conducted as per standard protocols. Briefly, the cells were lysed in cell lysis buffer, extracts clarified by centrifugation at 12,000 rpm, and the supernatants subjected to immunoblot analysis. Primary antibodies directed against phospho-AKT and AKT (Cell Signaling Technology, Beverley, MA) were used in dilutions of 1:1000. After washing three times each for 10 min in 1X TBST, blots were exposed to the secondary antibody (anti-mouse or anti-rabbit IgG-HRP) (Santa Cruz Biotechnology Inc., Santa Cruz, CA) at a dilution of 1:2000 and visualized using ECL chemiluminescence detection system (Amersham, Buckinghamshire, UK).

Immunofluorescence microscopy

Cells were grown on tissue-culture treated chambered slides (BD Biosciences, San Jose, CA), overnight, treated with cannabinoids and stimulated for 3 h with EGF 10 ng/ml in serum-free medium, washed and fixed with 4% paraformaldehyde, permeabilized in phosphate-buffered saline (PBS) containing 0.2% Triton, and blocked with 3% bovine serum albumin (BSA). Cells were incubated with Texas Red-labeled phalloidin (Molecular Probes, Carlsbad, CA) and FITC-labeled antivinculin (Sigma, St. Louis, MO) for 2 h at room temperature and washed. After final washes and mounting, cells were examined using a confocal microscope.

ELISA

MMP-9 levels were measured in supernatants from cells treated with different concentrations of JWH-015 and Win55,212-2 for 24 h in serum-free growth medium. Protein levels in the supernatants were assayed using a MMP-9 ELISA Kit (R&D Systems, Minneapolis, MN) following the manufacturer’s instruction. Optical density was measured at 450 nm. MMP-9 concentration was calculated by comparing the data to the known standards for MMP-9 proteins.

Mouse Model Study for Tumor Xenograft Growth and Metastasis

Sub-cutaneous tumors were generated in immunodeficient SCID CB-17 mice (Charles River Laboratories Inc. Wilmington, MA) by s.c. injection of 3 × 10⁶viable A549 cells in PBS. To grow pulmonary tumor colonies, mice were given injections of 5 × 10⁵ viable tumor cells through the lateral tail vein. Once the tumors reached an average palpable size in subcutaneous model and 7 days after cell injection in metastatic model, animals were assigned randomly to various groups and injected daily with different doses of cannabinoid agonists JWH-133 (1mg/kg/day) and Win55,212-2 (0.1mg/kg/day) alone and in combination with CB2 specific antagonists SR144528 (Sanofi Recherche, Montpellier, France) and CB1/CB2 non-specific antagonist AM281 (analogue of AM251 for use in animal studies) at a dose 0.1mg/kg/day peritumorally or intraperitoneally for 28 days. Tumor sizes were measured with external calipers weekly in two dimensions throughout the study and calculated as: tumor volume = length × (width)²/ 2 (11, 17).

Immunohistochemistry

Lung carcinoma tissue microarray (TMA) slides were obtained from Imgenex (San Diego, CA). Antigen retrieval was performed by Heat-Induced Antigen retrieval (HIER), where slides were placed in Dako TRS solution (pH 6.1) for 25 min at 94°C and cooled for 15 min. Slides were then placed on a Dako Autostainer, immunostaining system, for use with immunohistochemistry. Slides were incubated for 60 min in CB1 or CB2 antibodies (1:200) or their respective peptides and detected using a labeled polymer system, ImmPRESS Anti-Rabbit Ig (peroxidase) Kit (Vector Laboratories, Burlingame, CA) as per manufacturer’s protocol. Staining was visualized with DAB chromogen. Slides were then counterstained in Richard Allen hematoxylin, dehydrated through graded ethanol solutions and coverslipped. Sections were examined for CB1 and CB2 expression and immunoreactivity was scored as 0 (no or weak focal expression) and 1 (strong diffuse expression).

Samples from tumor xenografts of mice were dissected and embedded in OCT (Tissue-Tek, Torrance, CA). Standard immunohistochemistry techniques were used as per the manufacturer’s recommendations (Vector Laboratories, Burlingame, CA) using the primary antibody against Ki 67 (Neomarkers, Fremont, CA) at a dilution:1/100 and CD31 (BD Pharmingen, Franklin Lakes, New Jersey) at a 1:50 dilution for 60 min at RT. Vectastain Elite ABC reagents (Vector Laboratories, Burlingame, CA) using avidin DH:biotinylated horseradish peroxidase H complex with 3,3′-diaminobenzidine (Polysciences, Warrington, PA) and Mayer’s hematoxylin (Fisher Scientific, NJ) was used for detection of the bound antibodies. The nuclei for Ki67 staining and CD31 expression in vessels was quantitated in 5 random microscopic (×10) fields per tumor. For TUNEL assay, frozen tumor sections from treated mice were analyzed by the terminal uridine deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay using the In Situ Cell Death Detection Kit following the instructions of the manufacturer (Roche, Indianapolis, IN). The number of apoptotic nuclei per area was counted in 20 visual fields per section under a fluorescence microscope. The results were expressed as the percent change with vehicle-treated samples normalized to 100% (11).

Statistical Analysis

Results were analyzed using a two-tailed Student’s t test to assess statistical significance. Values of P < 0.05 were considered statistically significant.

Cannabinoid receptors are expressed in human lung cancers

Cannabinoid receptors have been shown to be over-expressed in different cancers such as skin and breast (18, 19). However, to the best of our knowledge, their expression in lung carcinomas is not documented in the literature. Therefore, we first investigated cannabinoid receptor expression in a set of non-small cell lung carcinomas and showed that CB1 expression was seen in 24% (7/29) while CB2 was expressed in 55% (16/29) of cases. To ascertain the specificity of the cannabinoid receptor antibodies used for detection of the two receptors, antigen pre-absorption experiments were carried out with the corresponding blocking peptides. The peptides blocked anti-CB1 and anti-CB2 antibody binding, both in cancerous as well as normal tissue samples (Fig. 1).

Figure 1

Expression of cannabinoid receptors in human lung adenocarcinomas and normal lung

Synthetic cannabinoids inhibit proliferation of NSCLC cell lines in vitro

EGFR over-expression and over-activation is reportedly associated with resistance of lung cancer to conventional chemotherapy. Hence, the effect of CB1/CB2 and CB2 specific cannabinoids Win55,212-2 and JWH-015, respectively, was analyzed on direct EGF-induced as well as serum mediated proliferation in NSCLC cell lines, A549 and SW-1573 cells. CB1 and CB2 are receptors that belong to the class of G-protein coupled receptors (GPCRs). Reports suggest cross-talks between GPCRs and EGFR, where GPCRs are shown to transactivate the EGF receptors (20). In the present study, we observed dose-dependent inhibition of proliferation of NSCLC cell line, A549 cells, with both Win55,212-2 as well as JWH-015. The CB1/CB2 agonist Win55,212-2 significantly attenuated the proliferation of A549 in response to serum and EGF by 70% and 60%, respectively (Fig. 2A). A corresponding dose of the CB2-specific agonist JWH-015 also reduced the number of viable cells significantly by 40% and 50% in a dose dependent manner in serum and EGF stimulated A549 cells (Fig. 2A). Analogous inhibitory tendencies were observed on the proliferation of SW-1573 cell line (data not shown).

Figure 2

Inhibition of proliferation and migration in A549 cells with CB1 and CB2 specific cannabinoids

Synthetic cannabinoids inhibit cell migration and invasion of NSCLC

The ability of tumor cells to migrate from the site of the primary tumor and to invade surrounding tissues is a prerequisite for metastasis, which is the major cause of cancer related mortality. Hence, to evaluate the therapeutic potential of cannabinoid receptors as possible targets, the effects of CB1 and CB2 agonists on EGF-induced cell migration and invasiveness were investigated. It was observed that cannabinoid agonists pretreatment impaired the cells to generate initial protrusions and scatter in response to EGF resulting in a rounded morphology (Fig. 2B). Furthermore, different in vitro methods such as scratch wound assay, trans-well migration and invasion assays were employed to study the effect of cannabinoid treatment on EGF-induced migration of NSCLC cell lines. As shown in Fig. 2C, Win55,212-2 and JWH-015 significantly decreased EGF-induced colonization of the wound areas. The expansion of cell population upon EGF stimulation was quantified by calculating the percentage re-colonization of the scratched wound surface over 72 h. Since CB1 and CB2 belong to the class of G-protein coupled receptors (GPCRs) (8, 9), we pretreated cells with pertussis toxin (PTX), which inhibits GPCR signaling. As shown in Fig 2C, PTX pretreatment abrogated Win55,212-2 and JWH-015-mediated inhibitory effects on EGF-induced wound healing. These results implicate the involvement of cannabinoid receptors CB1 and CB2 in the inhibition of EGF-induced wound healing.

To understand the mechanism of CB1 and CB2 receptor mediated inhibition of migration, we analyzed the effect of cannabinoid receptor agonists on formation of focal adhesions induced with EGF. Focal adhesions are sites of tight adhesion and provide a structural link between the actin cytoskeleton and the extracellular matrix. Focal adhesions were visualized through immunofluorescent staining for Vinculin, a membrane-cytoskeletal protein in focal adhesion plaques. Control and experimental cells exhibit focal adhesions at the periphery upon stimulation with EGF (Fig. 3). However, treatment of the cells with JWH-015 and Win55,212-2 prior to stimulation with EGF caused significant reduction in the number of focal adhesions of the cells. Furthermore, the polymerization state of actin, visualized as stress fibers stained with phalloidin actin, another important component related to focal adhesion formation and cell migration, was also found to be decreased in the cannabinoid treated cells.

Figure 3

CB1 and CB2 receptor mediates inhibition of focal adhesions formation

To confirm the CB1 and CB2 receptor mediated inhibitory effects on EGF-induced migration and invasion, transwell migration studies were conducted with NSCLC cells pre-treated with CB1 and CB2 specific antagonists, prior to synthetic cannabinoid agonist’s exposure. We observed that synthetic cannabinoid agonists inhibited EGF-induced migration and invasion of NSCLC lung cancer cells in a dose-responsive manner (Fig. 4A). Both the agonists, Win55,212-2 and JWH-015 significantly inhibited EGF-induced trans-well migration as well invasion by ~50% in nano-molar doses in comparison to the corresponding vehicle treated cells. Furthermore, pretreatment with CB1 and CB2 specific antagonist AM281 and AM630, respectively, prior to the agonists treatment resulted in abrogation of the cannabinoid induced inhibitory effect on EGF-stimulated migration and invasion (Fig. 4C & 4D). The concentrations of cannabinoid receptor agonists and antagonists used in this study did not have any significant effect on the viability of the cell lines as was confirmed by trypan blue staining of the cells in upper chambers of the transwell inserts as well as by viability (MTS) assay (data not shown). In view of previous reports where cannabinoids have been reported to induce migration in different cell lines (21–23), we investigated the induction of chemotaxis in A549 and SW-1573 cells by JWH-015 and Win 55-212,2 and found that there was no increased migration over the spontaneous migration observed in control unstimulated cells (data not shown).

Figure 4

CB1 and CB2 receptor mediated inhibition of EGF-induced migration and invasion in NSCLC cell lines

Synthetic cannabinoids inhibit NSCLC tumor growth and metastasis in vivo

Since regression in experimental animal tumor models represents an important endpoint of clinical relevance, we evaluated the ability of cannabinoids to inhibit tumor growth and lung metastasis in vivo. To analyze the anti-tumor effects of cannabinoids in vivo, tumors generated by sub-cutaneous inoculation of the highly malignant NSCLC cell line (A549) were treated with vehicle or CB2 selective agonist JWH-133 (analogue of JWH-015 used for animal studies) and CB1/CB2 agonist Win55,212-2. Furthermore, to characterize the role of cannabinoid receptors in this inhibition of growth, CB1 selective antagonist AM281 (analogue of AM251 used for animal studies) and CB2 selective antagonist SR144528 were administered in different combinations along with the agonist. Cannabinoid administration blocked the subcutaneous growth of tumor cells in vivo by ~70% with JWH-133 (1mg/kg/day) and ~50% with Win55,212-2 (0.1mg/kg/day) (Fig. 5A). The attenuation of tumor growth with the cannabinoids was abrogated when the cannabinoids were administered in presence of antagonists, suggesting the direct implication of the cannabinoid receptors in this growth inhibitory effect. As the tumor growth is directly correlated to the proliferation of cells, which depends on adequate vascularization required for oxygen and nutrient supply to the dividing cells, we analyzed important cellular parameters like proliferation, angiogenesis and apoptosis within the tumor tissues. In tumors from animals treated with both CB1/CB2 and CB2 specific agonists, we observed a significant reduction in Ki67 immunostaining (marker for proliferation), as well as CD31 immunostaining (marker for angiogenesis). Besides this, an increase in apoptosis was observed, as analyzed with tunnel staining shown as percent changes with respect to the untreated control animals (Fig. 5B).

Figure 5

CB1 and CB2 receptor activation inhibits the xenograft growth and metastasis of lung tumors in SCID mice

Given the inhibition of migration and invasion of NSCLC cells in vitro, we evaluated the effect of cannabinoid agonist treatment on in vivo lung metastasis in SCID mice. We observed significant surface lung metastasis resulting in increased lung weight in animals inoculated with A549 cells via lateral tail vein injection (Fig. 5C). Both CB1/CB2 agonists Win55,212-2 (~60%) and CB2 JWH-133 (~50%) treatment attenuated lung metastasis in experimental animals over control animals. Furthermore, co-administration of CB1 and CB2 selective antagonists along with agonists abrogated this inhibitory effect, as evident by the surface lung metastases and reversal of lung weight in comparison to vehicle treated animal (Fig. 5C). Of interest, no significant alterations in physiological parameters like body weight and liver weight could be ascribed to cannabinoid receptor agonist or antagonist administration, except for non-significant reduction in body weight with JWH-133 (data not shown).

Synthetic cannabinoids inhibit AKT/PKB phosphorylation and secretion of MMP9 in human NSCLC cell lines

We next sought to determine CB1/CB2-mediated signaling mechanisms that inhibit tumor growth. We observed reduction in the tyrosine phosphorylation of AKT in cells pretreated with Win55,212-2 and JWH-015 prior to EGF-stimulations (Fig. 6A, B). However, cannabinoids pretreatment did not change AKT protein level. AKT is the key survival molecule which is essential for EGF-induced cell migration, invasion and proliferation (17, 24, 25).

Figure 6

CB1 and CB2 receptor specific ligands inhibit the EGF-induced phosphorylation of AKT and PMA-induced secretion of MMP-9 in NSCLC cells

In our study, we also observed that both Win55,212-2 and JWH-015 significantly inhibited the secretion and activity of MMP-9 in PMA-stimulated NSCLC cell lines, A549 measured through ELISA (Fig. 6C) and zymography (data not shown). Of importance, the reduction observed in the secretion of MMP-9 was significantly inhibited when the treatment was carried out in the presence of PTX, an inhibitor for GPCRs, directly implicating the involvement of CB1 and CB2 receptors. Matrix metalloproteinase are important proteins that play critical role in cancer growth and metastasis. A significant upregulation and co-expression of MMP-9 is observed in different cancers (26–28).

We thank Justine Curley Cohen and Wendy Beth Israel Deaconess Medical Center microscopy core facility for staining and analysis of the lung histology microscopy specimens. We would also thank Susie Jones, Department of Pathology, The Ohio State University, Columbus, Ohio for the immunostaining of TMA and tumor slides. This work is partly supported by grants from the National Institutes of Health R01CA109527 (RKG).

EGFR

epithelial growth factor receptor

CBRs

cannabinoid receptors

NSCLC

Non-small cell lung cancer

The authors declare that they have no potential conflicts of interest.