Résumé
Feeding horses with high-starch diet can lead to alteration of their hindgut and fecal bacteria composition, with a potential shift in gram negative and positive bacteria. Gram-negative bacteria contain LPS in their outer membrane, which can translocate into bloodstream following hindgut microbial disturbances. The aim of our study was to evaluate if diet composition (high-fiber or high-starch diets) can modulate the LPS concentration in equine feces and plasma using an innovative measurement method. Six horses were used in a longitudinal study composed of three consecutive periods. During the first period (H1; 3 weeks), they were fed 100% hay. After a gradual transition, they received during the second period (HB; 4 weeks) a diet composed of 57% hay and 43% barley. During the third period (H2; 6 weeks), they returned to the 100% hay diet without transition. Colonic, fecal and blood samples were collected on days 10 and 20 in H1 and HB periods and on days 10, 20 and 40 in H2 period. Total anaerobic, amylolytic and lactate-utilizing bacteria were enumerated in colon and feces using culture technics. Fecal and plasma LPS concentration was determined by direct quantitation of 3HM by high-performance liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS), using a method described recently in human for plasma samples (Pais de Barros et al., 2015), and by adapting this method for fecal samples. The fecal LPS concentrations increased significantly when horses changed from hay diet to hay/barley diet and significantly decreased when horses returned to the hay diet (values comparable to those found during the first hay diet). Similar variations were observed on the bacteria concentrations in colon and feces of horses. The LPS concentration did not vary in plasma samples when the hay/barley diet was fed. It was assumed that dietary stress was not important enough to disrupted the hindgut mucosa and allow LPS translocation into plasma. These data showed that fecal LPS concentrations are very sensitive to diet changes, and could reflect alteration in the hindgut bacteria composition occurring under the dietary change. The measure of LPS through the 3HM raised the question of the accurate representation of LPS. In conclusion, quantification of 3HM could be a simple method to follow the evolution of bacterial composition in the feces, and rapidly state on the balance or imbalance of the bacteria. Further researches have to be conducted to use this parameter as an indicator of bacterial disturbances.