Polyethylene glycol increases intestinal absorption and hepatic uptake of indole and skatole in sheep fed

Six lactating ewes were orally drenched each day with polyethylene glycol (PEG) to remove the effects of the condensed tannins in sulla (Hedysarum coronarium) and six other ewes (control) received a drench of water commencing 28 days prior to the measurement period according to a block design. At week 6 of lactation, portal absorption and hepatic uptake of indole and skatole were increased by PEG. Effi ciency of hepatic extraction of indole and skatole was not different between the treatments. The splanchnic release of indole and skatole was higher in the PEG ewes compared with control.


INTRODUCTION
The indole compounds, indole and skatole are amines that give a faecal taint to milk and dairy products and are perceived in Asian and European markets as undesirable fl avours (Keen, 1998).Studies in Australia indicated that these compounds are not present in any signifi cant quantity in forages (Conochie, 1953).The amines found in milk are most likely the result of rumen metabolism of the amino acid tryptophan (Keen, 1998).
Condensed tannins (CT) in the diet reduce the degradation of dietary protein in the rumen (McNabb et al., 1996) and the concentration of circulating indole and skatole in peripheral blood (Roy et al., 2002).Rumen microbial fermentation and liver metabolism of indole and skatole are two processes that potentially could modify their supply to the peripheral blood.
Our study tested the hypothesis that polyethylene glycol (PEG) increases the peripheral concentration of indole and skatole by reducing their absorption from the rumen into the portal vein and/or due to limited hepatic extraction.The aim was to measure their absorption into the mesenteric and portal veins and fl ux across the liver and total splanchnic tissue (TSP; PDV + liver) in lactating ewes.

Animals, surgical procedures and dietary treatments
Twelve lactating ewes were prepared with catheters in the mesenteric, portal and hepatic veins, and in the mesenteric artery (Huntington et al., 1989).Three weeks postpartum, all ewes were offered fresh sulla (Hedysarum coronarium; 2000 g DM d -1 ; 80 g CT d -1 ) for 28 days.Six ewes were orally drenched daily with PEG (160 g d -1 in water) to remove the effects of the CT (CT inactive; PEG group) whilst six ewes were kept as controls (CT active).

Infusion and sampling protocols, measurement and statistics
Six weeks postpartum, para-aminohippurate was infused for 7 h into the mesenteric vein for measuring TSP plasma fl ows (Lobley et al., 1995).Plasma indole and skatole concentrations were measured in blood collected from the vessels previously mentioned (Roy et al., 2002).Splanchnic plasma fl ow and net fl ux of indole and skatole were calculated as described by Lobley et al. (1995).Data were subjected to the GLM procedure of SAS according to a completely randomized block design.One control ewe died during the experimental period for reasons not related to the treatment.Signifi cant statistical differences between treatments were declared at a probability less than 0.05.

RESULTS
Plasma fl ows across the mesenteric-drained viscera (MDV), PDV, TSP and hepatic artery were not affected by PEG (Table 1).The concentration of indole and skatole in the mesenteric vein was similar between treatments.In the mesenteric artery, hepatic vein and portal vein, their concentration was higher in PEG ewes.Net appearance of indole and skatole in the mesenteric drainage was similar between the PEG and control groups and in the portal drainage, was higher in the PEG group.Net hepatic uptake of indole and skatole was higher in the PEG ewes, but their hepatic extraction ratio (net uptake/infl ux) was not affected by the treatments.The net release of these metabolites by the TSP was higher in the PEG group.

DISCUSSION
Our study shows that the intra-luminal addition of PEG to lactating ewes fed the CT-containing plant sulla increased the net portal fl ux of indole and skatole.Their mesenteric appearance represents less than 8% of their portal appearance.Although their gastric fl ux was not measured, the mesenteric data (small intestine only), suggest that these metabolites are primarily absorbed across the rumen epithelium, draining into the portal vein via the gastric vein.Our results also agree with a study where portal absorption of indole and skatole was correlated with their plasma concentration (Hammond et al., 1983).Our values (7.1 µg/min/kg) are within the range reported in Hammond's study (1 to 46 µg/min/kg).While the portal absorption and hepatic uptake of indole and skatole increased in the PEG ewes, their effi ciency of removal by the liver did not and consequently, the splanchnic fl ux also increased.

CONCLUSIONS
These results establish that the intra-luminal addition of PEG to lactating ewes fed the CT-containing plant sulla increased the peripheral plasma concentration of indole and skatole most likely by altering their formation in the rumen and thereby increasing their absorption to the portal drainage.

Table 1 .
Effect of polyethylene glycol (PEG) on net fl ux of skatole and indole across the mesentericdrained viscera (MDV), portal-drained viscera (PDV), liver and total splanchnic tissue (TSP) in lactating ewes 1