Red meat is associated with increased risk of colorectal cancer and increases the endogenous formation of N-nitrosocompounds (NOC). To investigate the genotoxic effects of NOC arising from red meat consumption, human volunteers were fed high (420 g) red meat, vegetarian, and high red meat, highfiber diets for 15 days in a randomized crossover design while living in a volunteer suite, where food was carefully controlled and all specimens were collected. In 21 volunteers, there was a consistent and significant (P < 0.0001) increase in endogenous formation of NOC with the red meat diet compared with the vegetarian diet as measured by apparent total NOC (ATNC) in feces. In colonic exfoliated cells, the percentage staining positive for the NOC-specific DNA adduct, O 6 -carboxymethyl guanine (O 6 CMG) was significantly (P < 0.001) higher on the high red meat diet. In 13 volunteers, levels were intermediate on the high-fiber, high red meat diet. Fecal ATNC were positively correlated with the percentage of cells staining positive for O 6 CMG (r 2 = 0.56, P = 0.011). The presence of O 6 CMG was also shown in intact small intestine from rats treated with the N-nitrosopeptide N-acetyl-NV-prolyl-NV-nitrosoglycine and in HT-29 cells treated with diazoacetate. This study has shown that fecal NOC arising from red meat include direct acting diazopeptides or N-nitrosopeptides able to form alkylating DNA adducts in the colon. As these O 6 CMG adducts are not repaired, and if other related adducts are formed and not repaired, this may explain the association of red meat with colorectal cancer. (Cancer Res 2006; 66(3): 1859-65)

Cell exfoliation in the gut is an important cell renewal mechanism. To approach its investigation we applied a novel immunomagnetic technique for isolation of exfoliated cells from human stool. Exfoliated colonocytes were isolated from 168 stool samples. The cells were assessed microscopically using conventional stains and immunohistochemistry. The technique allowed us to obtain well-preserved colonocytes displaying characteristic features of well-differentiated colonic epithelium and positive immunostaining for cytokeratin 5/8. No mucin-producing cells were found. Exfoliated cells did not produce inducible nitric oxide synthase, albeit cultured colon carcinoma cells HT-29 analysed in parallel showed strong immunostaining. Analysis of exfoliated cell numbers in consecutive stool samples from the same subjects revealed considerable interindividual variation. Overall exfoliated colonocyte numbers were relatively low, isolation being unaffected by addition during the procedure of excessive amounts of HT-29 cells. Apoptosis was extremely rare among exfoliated colonocytes. Well-preserved exfoliated colonocytes can be consistently isolated from human faeces using a simple procedure. Our findings suggest that the actual process of cell exfoliation in the human colon may be much less intense than is generally accepted. Exfoliated cell isolation from human stool constitutes a convenient non-invasive approach that can be used for diagnostic and research purposes.

We previously reported that increased endogenous nitrosation in human subjects infected with the liver fluke Opisthorchis viverrini in north-east Thailand could be a risk factor for the development of cholangiocarcinoma. In the present study we examined our hypothesis that this increased endogenous nitrosation is mediated by nitric oxide (NO) synthase induced by O. viverrini infestation. Syrian golden hamsters experimentally infected with O. viverrini liver fluke excreted in the urine significantly greater amounts of nitrate, a stable oxidization product of NO, than untreated hamsters (3.64 +/- 0.86 versus 2.64 +/- 0.60 mumol/hamster/day, P < 0.001). When the rapidly nitrosatable thiazolidine 4-carboxylic acid was administered orally, the infected hamsters also excreted significantly elevated levels of N-nitrosothiazolidine 4-carboxylic acid than untreated hamsters (4.27 +/- 2.20 versus 2.33 +/- 1.13 nmol/hamster/day, P < 0.01), indicating that endogenous nitrosation is elevated in the animals with liver fluke. NO synthase activity measured in liver cytosol was about twice as high in the infected hamsters as in untreated animals. The enzyme, whose biochemical characteristics were similar to that induced in activated murine macrophages, was immunohistochemically localized in the cytoplasm of macrophages and eosinophils in the inflammation zone surrounding the parasite-containing bile ducts. These results support our hypothesis that, in fluke-infected subjects, NO synthase induction leads to excess production of NO and the observed elevated endogenous nitrosation. Since high concentrations of NO exert cytotoxic and mutagenic effects per se, excess NO produced in chronically infected/inflamed tissues may also play a role in initiation and subsequent modulation stages of cholangiocarcinoma development.