Intestinal Ketogenesis and Permeability

Abstract

Consumption of a high-fat diet (HFD) has been suggested as a contributing factor behind increased intestinal permeability in obesity, leading to increased plasma levels of microbial endotoxins and, thereby, increased systemic inflammation. We and others have shown that HFD can induce jejunal expression of the ketogenic rate-limiting enzyme mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase (HMGCS). HMGCS is activated via the free fatty acid binding nuclear receptor PPAR-α, and it is a key enzyme in ketone body synthesis that was earlier believed to be expressed exclusively in the liver. The function of intestinal ketogenesis is unknown but has been described in suckling rats and mice pups, possibly in order to allow large molecules, such as immunoglobulins, to pass over the intestinal barrier. Therefore, we hypothesized that ketone bodies could regulate intestinal barrier function, e.g., via regulation of tight junction proteins. The primary aim was to compare the effects of HFD that can induce intestinal ketogenesis to an equicaloric carbohydrate diet on inflammatory responses, nutrition sensing, and intestinal permeability in human jejunal mucosa. Fifteen healthy volunteers receiving a 2-week HFD diet compared to a high-carbohydrate diet were compared. Blood samples and mixed meal tests were performed at the end of each dietary period to examine inflammation markers and postprandial endotoxemia. Jejunal biopsies were assessed for protein expression using Western blotting, immunohistochemistry, and morphometric characteristics of tight junctions by electron microscopy. Functional analyses of permeability and ketogenesis were performed in Caco-2 cells, mice, and human enteroids. Ussing chambers were used to analyze permeability. CRP and ALP values were within normal ranges and postprandial endotoxemia levels were low and did not differ between the two diets. The PPARα receptor was ketone body-dependently reduced after HFD. None of the tight junction proteins studied, nor the basal electrical parameters, were different between the two diets. However, the ketone body inhibitor hymeglusin increased resistance in mucosal biopsies. In addition, the tight junction protein claudin-3 was increased by ketone inhibition in human enteroids. The ketone body β-Hydroxybutyrate (βHB) did not, however, change the mucosal transition of the large-size molecular FD4-probe or LPS in Caco-2 and mouse experiments. We found that PPARα expression was inhibited by the ketone body βHB. As PPARα regulates HMGCS expression, the ketone bodies thus exert negative feedback signaling on their own production. Furthermore, ketone bodies were involved in the regulation of permeability on intestinal mucosal cells in vitro and ex vivo. We were not, however, able to reproduce these effects on intestinal permeability in vivo in humans when comparing two weeks of high-fat with high-carbohydrate diet in healthy volunteers. Further, neither the expression of inflammation markers nor the aggregate tight junction proteins were changed. Thus, it seems that not only HFD but also other factors are needed to permit increased intestinal permeability in vivo. This indicates that the healthy gut can adapt to extremes of macro-nutrients and increased levels of intestinally produced ketone bodies, at least during a shorter dietary challenge.