Short chained fatty acids and the colon: how do they cause vasodilatation?

It is not a new observation that acetic acid (Bauer & Richards,1928) and other short chained fatty acids (SCFAs) such as butyric and proprionic acid cause vasodilatation in many vascular beds. This vasodilator effect is, however, interesting since these acids are produced in large amounts in the colon of non-ruminant mammals by bacterial anaerobic fermentation of polysaccharides. In the colon the concentration of SCFAs may reach 100 mmol kg⁻¹. Although it is difficult to know what the concentrations near the colonic resistance arteries are, it is not unlikely that the concentrations are high enough to have vasoactive effects. Furthermore instillation of SCFAs into the rectum has been shown to increase rectal mucosal blood flow (Mortensen et al. 1991). Important also in this context, the solutions used for peritoneal dialysis contain high concentrations of acetate or lactate (and are often hyperosmolar) and have also been shown to induce vasodilatation (Miller et al. 1981).

Interestingly it has also become apparent during the last 20–30 years that SCFAs have trophic effects in the bowel and may be important for wound healing after colonic surgery (Rolandelli et al. 1986). It has also been suggested that ulcerative colitis may result from a metabolic defect in SCFA oxidation in the colon and that colitis may be improved by SCFA enemas (Harig et al. 1989). Also effects of SCFAs on ileal motility have been reported as well as a role for SCFAs in promoting Na⁺ reabsorption. Thus there is little doubt that SCFAs play an important role in intestinal physiology and pathophysiology.

There is thus good reason to investigate the mechanisms responsible for the vasodilator effect of SCFAs, which in principle could underlie all of the above effects. This has therefore been an area, which has been actively investigated and several mechanisms have been discussed.

Strong evidence for a role of cAMP has been presented. Also the role of smooth muscle pH (pH_i) has been discussed. It has been suggested that the transient decrease in pH_i when a SCFA is added might be involved in the vasodilator effect although this is controversial. Surprisingly no measurements of membrane potential have been reported and there are only few reports on [Ca²⁺]_i. The latter measurements seem to argue against a role for [Ca²⁺]_i in the vasodilator effect of acetate. The role of the endothelium has (obviously) been addressed by all the groups looking at the mechanisms involved in dilatation to SCFAs. The responses to SCFAs have been elicited before and after removal of the endothelium. For once there has been a delightful consistency amongst everyone in concluding that the endothelium did not play a role. It was thus quite surprising when Knock et al. (2002) in this issue of The Journal of Physiology reported a 75 % inhibition of the dilatation to proprionate in isolated mesenteric small arteries after removal of the endothelium. They went on to show that the endothelium dependence was likely to be mediated through the endothelium-derived hyperpolarizing factor (EDHF). The reason why the endothelium suddenly appeared important was that the endothelium dependence previously was only tested at concentrations higher than 30 mm of butyrate and proprionate whereas Knock et al. (2002), used 10 mm. One lesson from this study therefore is the importance of examining the full concentration range of a mediator to understand all of its potential mechanisms of action.

With the work of Knock et al.(2002) we have moved towards an understanding of the vasodilator mechanism of SCFAs, but also towards greater complexity. The question now is how close we are to an understanding. There is undoubtedly still some way to go. First there is the problem that the physiologically relevant concentration of SCFAs near the mesenteric resistance arteries is not known. Secondly a very important issue is the possible differences between the three predominant SCFA. In physiological conditions acetate, butyrate and proprionate are always found in combination and the interaction between them could very well be relevant. Noteworthy here is that the endothelium dependence of SCFA-mediated vasodilaton was smaller with butyrate than with proprionate (Knock et al. 2002) and has previously been found to be absent in rat tail arteries with 4 and 16 mm acetate (Daugirdas et al. 1987). This suggests the interesting possibility that the three SCFAs have different mechanisms of action. Also the role of pH_i is unclear. The transience of the dilatation suggested to Knock et al. (2002) that a transient decrease of pH_i in the endothelial cells could be important for release of EDHF. This suggestion, however, needs experimental testing. Further it is important that under conditions where dilatation to SCFAs is physiologically relevant, hyperosmolarity is present. Hyperosmolarity is a strong vasodilator in the mesenteric circulation in its own rights, but very little is known about the interaction between SCFAs and hyperosmolarity and most experiments with SCFAs on isolated arteries are made in iso-osmotic solutions.

It is therefore important that our understanding of a physiologically very active group of substances has improved with this surprising observation of Knock et al. (2002). On the other hand it is quite obvious that there is some way to go before we understand how the action of SCFAs is mediated under physiological conditions.