Reply to the letter to the editor “Long‐term cannabinoid receptor (CB1) blockade in obesity: Implications for the development of colorectal cancer”

In the recent letter by Wright et al.,1 it has been described that the cannabinoid receptor (CB1) antagonist/inverse agonist AM251 is able to inhibit NCM460 cell proliferation after a daily treatment of 96 hr. On the basis of this observation, obtained in cells exhibiting a nontransformed phenotype, the authors strongly suggest that rimonabant, another selective CB1 antagonist/inverse agonist, could affect negatively the integrity and maintenance of the intestinal mucosa as a consequence of its continuous intake during the pharmacotherapy of obesity. They also suggest that rimonabant-induced mucosal damage might trigger persistent regenerative processes leading to colon cancer. From this point of view, we believe that the conclusions on rimonabant, obtained from experiments carried out with another CB1 antagonist/inverse agonist, could be hazardous, because rimonabant is really a highly selective CB1 receptor antagonist but it can also act in a way independent from the receptor.2, 3 We have just demonstrated that rimonabant exhibits immunomodulatory and antiinflammatory effects in normal human lymphocytes.4 These effects were not associated to apoptosis but were characterized by a G1/S phase cell cycle arrest. In this system, rimonabant used in combination with the CB1 agonist anandamide showed synergistic/additive effect in the blockade of cell proliferation.4 On the other hand, we have to taken into account that, even if both AM251 and rimonabant are able to block functionally the CB1 receptor, they are not the same molecule and this could be responsible for different kinds of biological responses evoked by the cells. Moreover, Izzo et al.5 have recently demonstrated that the fatty acid amide hydrolase inhibitors increased colon endocannabinoid levels and reduced the formation of aberrant crypt foci (ACF) which are considered the earliest identifiable neoplastic lesions in humans.6, 7 Interestingly, no differences in ACF formation were observed between CB1 receptor-deficient and wild-type mice. It was concluded that pharmacological enhancement of endocannabinoid levels, through inhibition of endocannabinoid hydrolysis instead of CB1 modulation, reduces the development of precancerous lesions in the mouse colon.5 Concerning colon cancer pathogenesis, we have interesting data that rimonabant significantly inhibits the viability of human adenocarcinoma DLD1 cells after 24 and 48 hr of treatment in a concentration-dependent manner (Fig. 1). Moreover, in colon mucosa, oestrogens and their receptors exert protective effects and the presence of considerable amounts of CB1 receptor, as well as oestrogen receptor expression has been shown in colorectal carcinoma and in human normal colon mucosa.8 We have demonstrated that CB1 receptor is potentially an oestrogen-responsive gene in colon cancer cells DLD1 and SW620. In these cells, the oestrogen-mediated activation of CB1 receptor, at both transcriptional and posttranscriptional levels, suggests that oestrogen may modulate cancer cell growth via activation of CB1.9 In summary, although more data are needed to assess the threshold of toxicity of rimonabant in humans, this compound showing antiobesity, antiinflammatory and antitumoral effects,10 might provide insights into mechanisms relevant to the pathogenesis of colon cancer. DLD1 colon cancer cell viability after the treatment with rimonabant. Exponentially growing cells (1 × 106) were seeded in 6-well plates and cultured in RPMI 1640 medium with 10% foetal bovine serum (Sigma, Milan, Italy) at 37°C and 5% CO2. Cells were exposed to increasing concentrations of rimonabant for 24 and 48 hr prior to harvesting (trypsinization) and both floating and adherent cells were counted by using a hemocytometer. Cell viability was determined by trypan blue exclusion staining and expressed as the percentage of viable cells in the control samples at each time point (n = 6; *p < 0.05; **p < 0.01 vs. control with the Student's t test). This work was supported by Associazione Educazione e Ricerca medica Salermitana, ERMES. A.S. and S.P. were supported respectively by a fellowship from AIRC and from FIRC. Yours sincerely, Antonietta Santoro, Patrizia Gazzerro, Anna Maria Malfitano, Simona Pisanti, Chiara Laezza, Maurizio Bifulco