The study of the effect of chemopreventive agents on apoptosis of colonic epithelial cells

Conclusions and Summary Colorectal cancer (CRC) is the third most common type of cancer worldwide. The previous existence of an adenomatous polyp is almost a prerequisite for CRC development and any effort for chemoprevention at an earlier stage could be effective. Previous evidence suggests that Non Steroid-Antiinflammatory Drugs (NSAIDS) exhibit a chemopreventive effect on CRC. Fluorinated quinolones, including their major representative, ciprofloxacin, inhibit the bacterial type II topoisomerase while their mode of action on mammary cells has not been entirely elucidated. Previous evidence suggests additional properties of the fluoroquinolone family exhibiting a potent role in immunoregulation, induction of growth factors and apoptosis. TGF-β (transforming growth factor-β) is a multifunctional cytokine. The TGF-β signalling pathway has been considered as both suppressor and promoter of oncogenesis; during early stages, it has been shown, in animal models, that epithelial cells lose their growth inhibitory responsiveness to TGF-β which eventually promotes invasion and metastasis in the later stages of carcinoma development. The study of the effect of Ciprofloxacin, NSAIDs, Cycloogynase (COX)-2 inhibitors and 5-Amino-salicylate (5-ASA,) added alone or in combination with 5-fluorouracil (5-FU), on apoptosis and proliferation of 2 human colonic cancer cell lines (HT-29 and Caco-2) were the principal aims of the present study. Apoptosis was initially evaluated by 7-aminoactinomycin D (7-AAD) and annexin-V staining and was further examined by a commercially available cell death detection ELISAplus kit for analysis of DNA fragmentation. Growth and viability of cells were measured by the tetrazolium salt assay [3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenol tetrazolium bromide] (MTT). In the present study, we showed for the first time that ciprofloxacin inhibits the proliferation of HT-29 cells via the induction of TGF-β1 (confirmed by blotting of TGF-β1 mRNA expression with semiquantitative multiplex PCR and quantitative evaluation of the production of the cytokine in cell culture supernatants by ELISA) since its effect was abrogated by the addition of a neutralizing anti-TGF-β antibody in culture. Caco-2 cells do not express functional TGF-β1 receptors and thus are resistant to the inhibitory effect of ciprofloxacin. In our experimental model, the antiproliferative effect of ciprofloxacin was not attributed to apoptosis but rather to cell cycle arrest (as we showed with PI staining and Flow cytometry) and could be reversed after the removal of the quinolone from culture. Our results indicate a possible antineoplastic effect of ciprofloxacin at the early stages of carcinogenesis where the TGF-β signaling pathway remains intact; outlining an unknown immunomodulatory effect of quinolones in a well differentiated epithelial cell line, without the prerequisite presence of a triggering inflammatory cytokine. Σhe ηnduction of TGF-β1 by ciprofloxacin may be associated with a possible in vivo anti-inflammatory effect of ciprofloxacin in non-infectious colitides like Inflammatory Bowel Diseases (IBD) and this possible association is the most intriguing experimental hypothesis of our study. 124 To the best of our knowledge, in the present study we showed for the first time the effect of CAY10404, a novel hyperselective COX-2 inhibitor on HT-29 and Caco-2 human colonic cancer cell lines. CAY10404 inhibited cellular proliferation in a time- and concentration dependent manner while enhancing the antiproliferative effect of 5-FU. These results indicate a possible antineoplastic effect of CAY10404 in CRC. In our experimental model, all NSAIDS that were used (Aspirin, Indomethacin and CAY10404) exhibited a significant dose dependent effect on cellular proliferation of HT-29 and Cacν-2 cells. Overall, the antiproliferative profile of all NDAIDs that were used was similar indicating that it is not dependent on the COX-2 specificity of the agent used. In the present study, 5-ASA was used in concentrations that could be achieved in vivo but could not induce a significant antiproliferative effect in Caco-2 cells, while 5-ASA exhibited a significant dose dependent antiproliferative effect on HT-29 cells. On the other hand, 5-ASA, as well as all the rest of NSAIDs that were used, significantly induced apoptosis in HT-29 and Caco-2 cells. This effect has been proposed as a chemopreventive mechanism in CRC. Taking into consideration the alternative response of Caco-2 cells to treatment with ciprofloxacin, the differential effect of 5-ASA on Caco-2 cells is possibly cell-line specific. Caco-2 cells do not express functional TGF-β receptors, an event that apparently corresponds to the late stages of carcinoma development. This indicates an insufficiency of 5-ASA in reducing cellular proliferation during the late stages of carcinogenesis, where other NSAIDS are still efficient. This alternative mechanism of action of 5-ASA is apparently functional during the later stages of carcinoma development, as it was showed in Caco-2 and HT-29 cells, promoting apoptosis and possibly sensitizing cancer cells to the antiproliferative effect of 5-FU, but it is not sufficient enough to induce significant antiproliferative effects when 5-ASA is used solely. This experimental hypothesis could possibly explain the efficacy of 5-ASA in precancerous clinical situations as in IBD which could be related to the chemopreventive effect of long-term treatment with 5-ASA in IBD patients, which is apparently COX independen