Targeting cannabinoid receptors in gastrointestinal cancers for therapeutic uses: current status and future perspectives

KEYWORDS:

• Cannabinoids
• cannabinoid receptors
• Cannabis sativa
• gastrointestinal cancer
• ∆⁹-tetrahydrocannabionol

1. Introduction

Δ⁹-tetrahydrocannabinol (THC) is the main psychoactive component of Cannabis sativa, a widely distributed plant used since antiquity to treat gastrointestinal diseases ranging from intestinal inflammation to disorder of motility/secretion, visceral pain, and nausea [1]. THC was isolated for the first time in 1964, but its pharmacological mode of action was discovered in the late 80s and early 90s, when two specific G-coupled receptors for THC, named CB₁ and CB₂ receptors, were identified and found to be responsible of many of the pharmacological action of Cannabis and THC. CB₁ receptors, mostly identified in central and peripheral nerves, including the enteric nervous system, are responsible of the well-known psychotropic effect of marijuana, while CB₂ receptors are mostly involved with the immune system. The identification of cannabinoid receptors was followed by the discovery of endogenous ligands, that is, the endocannabinoids anandamide (arachidonoyl ethanolamide) and 2-arachydonoyl glycerol (2-AG) and of the enzymes responsible for endocannabinoid biosynthesis and degradation [e.g. fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MAGL), mostly involved in anandamide and 2-AG degradation, respectively] [2]. Cannabinoid receptors, endocannabinoid, and the enzymes involved in endocannabinoid synthesis/degradation are collectively referred as the ‘endocannabinoid system’. Research led in the past 20 years has revealed the endocannabinoid system as an important regulatory system in the gastrointestinal tract, being involved in several important functions such as motility, secretion, sensation, inflammation, and carcinogenesis [1,3]. A multitude of studies have proposed a role for the endogenous cannabinoid system in the pathogenesis of colon cancer, suggesting a potential impact of cannabinoids in this disease. Therefore, we will focus here on the role and effect of the endogenous cannabinoid system in colorectal cancer, based on the studies on cell lines, animal models, and preclinical human studies.

2. Studies in gastric and colorectal cancer cells

Cannabinoids, via CB₁ and possibly CB₂ receptors, suppress proliferation and migration and stimulate apoptosis in colorectal cancer cells [4–8]. Intracellular mechanisms associated to CB₁-mediated proapoptotic effects include inhibition of RAS–MAPK and PI3K–AKT pathways [6], downregulation of the antiapoptotic protein survivin [7], and stimulation of ceramide biosynthesis [8]. Interestingly, anandamide has been shown to inhibit the growth of COX-2-expressing colorectal cancer cells by inducing cell death in a cannabinoid receptor-independent mechanism [9]. Finally, it should be highlighted that cannabinoid agonists decrease cell viability and proliferation of human gastric adenocarcinoma cell line [10], including 5-FU resistant cells [11], via G1 phase cell cycle arrest which is mediated by activation of the MAPK pathway and inhibition of pAKT [12].

Cannabinoids may also potentiate the effect of clinically used chemotherapeutic agents. For example, the cannabinoid receptor agonist HU-210 and the antitumoral drug 5-fluorouracyl have been shown to exert cytotoxic synergistic effects in the colorectal cancer cells [13]; furthermore, anandamide synergistically enhanced paclitaxel-induced apoptosis, possibly through caspase-3, -8, and -9 activation, in gastric cancer cells [14].

3. In vivo studies on experimental models of gastric and colon cancer carcinogenesis

A number of studies have consistently shown that cannabinoids are able to prevent or reduce carcinogenesis in different animal models of colon cancer. Thus, genetic and pharmacologic experiments revealed that deletion of CB₁ receptors accelerated intestinal adenoma growth in Apc mice whereas activation of CB₁ attenuated tumor growth [7]; also, peritumoral treatment with the CB2 receptor agonist CB13 reduced the growth of xenograft tumors generated by injection of colorectal cancer cells to immunodeficient mice [8]; finally, the cannabinoid receptor agonist HU-210 reduces the formation of aberrant crypt foci (preneoplastic lesions) induced by the carcinogenic agent azoxymethane [15].

Consistent with a protective effect against gastrointestinal cancers, the cannabinoid receptor agonist WIN 55,212-2 has been shown to exert antineoplastic actions on the xenograft model of gastric cancer [16].

Protection against carcinogenesis can be afforded not only by direct cannabinoid receptor activation, but also via inhibition of the main endocannabinoid degradative enzymes (i.e. FAAH and MAGL), possibly resulting in increased endocannabinoid levels in the gut. Thus, the FAAH inhibitor N-arachidonoylserotonin prevented the formation of aberrant crypt foci (notably those with four or more crypts, which best correlate with final tumor incidence), an effect associated to increased colon anandamide and 2-AG levels and normalization of cleaved caspase-3 expression [15]. More recently, it has been shown that MAGL and its main substrate 2-AG are present in the colorectal cancer tumor tissues and that a pharmacological inhibition of MAGL reduced (i) the number of preneoplastic lesions and tumors in the azoxymethane model of colon carcinogenesis and (ii) xenograft tumor volume in immunodeficient mice, the latter effect being associated to downregulation of VEGF and FGF-2, reduction in the number of blood vessels, and downregulation of cyclin D1 [17]. Such results are consistent with in vitro studies showing that colorectal cancer cells growth is inhibited via knockdown of MAGL [18] and that a pharmacological enzyme inhibition resulted in a direct antiangiogenic effect in human endothelial cells [17].

Finally, the atypical cannabinoid O-1602, that is, an analog of the phytocannabinoid cannabidiol (CBD) with little or no affinity to brain CB₁ receptors, has been shown to attenuate tumor growth in colitis-associated colon cancer. Due to its lack of central effects, this compound could represent another potential option for colon cancer [19].

4. Human studies

Although cannabinoids have been evaluated clinically for attenuating symptoms of cancer pain and as antiemetics for chemotherapy-induced nausea and vomiting [20], there are no clinical data available to date concerning the effect of cannabinoids on the disease progression. The available data come from colon biopsies from colorectal cancer patients, which have consistently shown increased endocannabinoid levels [4,21]. Because endocannabinoids reduce intestinal tumor growth, tissue endocannabinoid elevation may be seen as a protective device activated to restore homeostasis disrupted by pro-carcinogenic mechanisms. On the other hand, studies on cannabinoid receptor expression have furnished contrasting results [4,7,8,21] making the role or cannabinoid receptors less clear than that of endocannabinoids. Interestingly, cancer patients were found to have a 2.9 times greater probability of nucleotide changes in the CB₁ gene [22], which was found, in a different study, to be hypermethylated in 77% of tumor samples from colon cancer patients [7]. Finally, CB₁ high immunoreactivity was demonstrated to be a significant prognostic factor following surgery in stage IV colorectal cancer [23] and CB₂ receptor mRNA expression significantly correlated with lymph node involvement in cancer patients [24]. Collectively, such results may have theoretical clinical implications in the light of the widespread use of marijuana, potentially able to modify colorectal cancer course.

5. Non-psychotropic phytocannabinoids and experimental colorectal cancer

In addition to THC, the main psychotropic phytocannabinoid, the plant Cannabis sativa synthetizes several non-psychotropic cannabinoids (phytocannabinoids), which have been evaluated in experimental models of inflammation and colon cancer. Non-psychotropic phytocannabinoids are safe compounds and does not activate efficiently brain cannabinoid receptors. Recently, two of such phytocannabinoids, namely CBD and cannabigerol (CBG) have been shown to exert chemopreventive effects in the murine model of carcinogenesis induced by azoxymethane and to inhibit the growth of xenograft tumors [25–27]. In colorectal cancer cells, CBD protected DNA from oxidative damage, increased endocannabinoid levels – possibly via FAAH inhibition – and reduced cell proliferation in a CB₁-, TRPV1- and PPARγ-antagonists sensitive manner [25]; CBG affects the growth of cancer cells mainly via a proapoptotic mechanism and overproduction of reactive oxygen species [27].

6. Expert commentary and conclusions

Convincing scientific evidence suggests that cannabinoids, in addition to their well-known use in palliative care in oncology (e.g. improvement of appetite, attenuation of nausea associated to antitumoral medicines, alleviation of moderate neuropathic pain) can reduce, via antiproliferative and proapoptotic as well as by inhibiting angiogenesis, invasion and metastasis or by attenuating inflammation, the growth of cancer cells and hinder the development of experimental colon carcinogenesis in vivo. Preclinical studies on cells and animals (via both genetic and pharmacological approaches) have partly identified the molecular pathways involved in cannabinoid action and revealed additive and synergistic effects with conventional antineoplastic drugs. For a possible future clinical investigation, the best strategy could be the use of (i) inhibitors of enzymatic degradation, that is, FAAH and MAGL inhibitors, which, by increasing the local amount of endocannabinoids, are expected to be devoid of the unwanted central effects of direct-acting, brain CB1-mediated, cannabinoid agonists and (ii) CB₂ receptor agonists or non-psychotropic cannabinoids, such as CBD and CBG. In the light of the lack of clinical trials for cannabinoids as antineoplastic agents, the use of cannabinoids in clinical practice should be undoubtedly not recommended. Experimentally, a number of cannabinoids at various doses have been shown to be protective in animal studies. However, given the heterogeneity of such results, dose generalizations and extrapolation for humans cannot be made. It is our opinion that, in setting future clinical trials, cannabinoids should be tested at the same doses of other conditions for which they are currently under clinical investigation (e.g. epilepsy, multiple sclerosis, and chronic pain) and for which safety data are already available. Finally, due to the possible widespread use following its legalization in many countries, attention should be paid to avoid possible clinical biases deriving from the use of Cannabis by patients.

Declaration of interest

F Borrelli has performs paid research for GW Pharmaceutical that produces the phytocannabinoids discussed in the manuscript. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Additional information

Funding

This manuscript has not received any funding.