The effects of polydextrose and xylitol on microbial community and activity in a 4-stage colon simulator.

This study focused on the effects of candidate prebiotics polydextrose (PDX) and xylitol on the microbial community and its metabolic activity in a colon simulator. A semicontinuous, anaerobic culture system was used with 4 vessels mimicking the conditions in the human large intestine from proximal to distal colon. Bacterial inocula for the independent simulations were obtained from fecal samples of different donors. Synthetic medium, mimicking the contents of the small intestine, containing either 2% of the prebiotic candidate or no added carbohydrates as a control, was fed to the system. After 48 h of simulation samples were collected and analyzed. A sustained degradation of polydextrose throughout the colon model and a more rapid degradation of xylitol were observed. The fermentation of both compounds was characterized by a significantly increased production of short-chain fatty acids (SCFA). Polydextrose increased the concentrations of all SCFA, especially acetate and propionate, and xylitol especially the concentration of butyrate. Branched-chain fatty acids (BCFA) levels decreased significantly as a result of polydextrose and xylitol supplementation, whereas biogenic amine levels remained mostly unchanged. Thus, a beneficial shift in the metabolic patterns of the colon microbes was measured with both of the tested products. These in vitro studies provide evidence to the prebiotic characteristics of polydextrose; also, further beneficial properties of xylitol were demonstrated in the colon model.

Intron 1 retaining cyclooxygenase 1 splice variant is induced by osmotic stress in human intestinal epithelial cells.

The biological roles of intron 1 retaining cyclooxygenase (Cox) 1 splice variants Cox-3 and PCox-1a (Cox-1ir) are not known. In humans, Cox-3 transcription has previously been shown to occur in the brain and in the aorta. However, conclusive evidence regarding the existence of a human Cox-3 protein is lacking. We studied the expression of intron 1 retaining cyclooxygenase 1 splice variants in the human colon cancer cell line Caco-2 and in human colonic tissue samples. In Caco-2 cells, their transcription was induced up to 47-fold by osmotic stress. The corresponding protein, however, could not be detected by Western blotting. In human colonic tissue samples derived from intact and inflamed areas, a low level of Cox-1ir mRNA (1500 +/- 1280 copies per 100 ng total RNA; mean+/-standard deviation; n = 20) was also found. In Caco-2 cells, induction of Cox-1ir under osmotic stress was reversed by addition of the organic osmolyte betaine. Under hypertonic but not under isotonic conditions, splice variant-specific degradation of Cox-1ir mRNA using RNA interference resulted in increased production of fully spliced Cox-1 and Cox-2 mRNA (P = 0.002). In summary, our results indicate that the intron 1 retaining Cox-1 splice variant RNA molecules are expressed by human intestinal epithelial cells in a controlled manner, are most likely not translated and play a regulatory role in the cyclooxygenase mediated epithelial osmoregulation.