Multicompartmental nontargeted LC-MS metabolomics: explorative study on the metabolic responses of rye fiber versus refined wheat fiber intake in plasma and urine of hypercholesterolemic pigs.

A multicompartmental nontargeted LC-MS metabolomics approach was used to study the metabolic responses on plasma and urine of hypercholesterolemic pigs after consumption of diets with contrasting dietary fiber composition (whole grain rye with added rye bran versus refined wheat). To study the metabolic responses, we performed a supervised multivariate data analyses used for pattern recognition, which revealed marked effects of the diets on both plasma and urine metabolic profiles. Diverse pools of metabolites were responsible for the discrimination between the diets. Elevated levels of phenolic compounds and dicarboxylic acids were detected in urine of pigs after rye consumption compared to refined wheat. Furthermore, consumption of rye was characterized by lower levels of linoleic acid derived oxylipins and cholesterol in the plasma metabolic profiles. These results indicate that higher consumption of nonrefined dietary fiber is reflected in higher excretion of phenolic compounds and dicarboxylic acids in urine and lower levels of linoleic acid derived oxylipins and cholesterol in plasma, which can be linked to beneficial health effects of rye components. On the other hand, pro-inflammatory lipid mediators were detected in higher concentration after rye consumption compared to refined wheat, which is opposite to what would be expected. These may indicate that even though a positive lowering effect with respect to cholesterol and fatty acids was achieved, this effect of rye dietary fiber was not sufficient to prevent inflammation in pigs. Moreover, we performed an alignment of the metabolic profiles between the breads consumed by pigs, plasma, and urine with the purpose to follow the metabolic fate of the compounds and to identify their pathways. One metabolite was identified in all three compartments, 16 metabolites were similar between bread and plasma, 3 were similar between plasma and urine, and 2 were similar between bread and urine. The use of multicompartmental metabolomics offered higher order information, including intercompartment relationships, and provided novel targets for future research.