On the basis of the isomer-specific effects of trans fatty acids (FA) on human health, and the detrimental effect of t10,c12-conjugated linoleic acid (CLA) on cows’ milk fat production, there is a need to identify factors that affect the shift from trans-11 to trans-10 pathway during ruminal biohydrogenation of FA. This experiment was conducted in vitro and aimed at separating the effects of the diet of the donor cows from those of the fermentative substrate, which is necessary to prevent this shift. A total of four dry Holstein dairy cows were used in a 4 × 4 Latin square design. They received 12 kg of dry matter per day of four diets based on maize silage during four successive periods: the control diet (22% starch, <3% fat); the high-starch diet, supplemented with wheat plus barley (35% starch, <3% crude fat); the sunflower oil diet, supplemented with 5% of sunflower oil (20% starch, 7.6% crude fat); and the high-starch plus oil diet (33% starch, 7.3% crude fat). Ruminal fluid of each donor cow was incubated for 5 h with four substrates having similar chemical composition to the diets, replacing sunflower oil by pure linoleic acid (LA). The efficiency of isomerisation of LA to CLA was the highest when rumen fluids from cows receiving dietary oil were incubated with added LA. The shift from trans-11 to trans-10 isomers was induced in vitro by high-starch diets and the addition of LA. Oil supplementation to the diet of the donor cows increased this shift. Conversely, the trans-10 isomer balance was always low when no LA was added to incubation cultures. These results showed that a large accumulation of trans-10 FA was only observed with an adapted microflora, as well as an addition of non-esterified LA to the incubation substrate.

The effects of the absence of protozoa in the rumen of lambs on animal growth, rumen fermentation, microbial diversity and fatty acid profiles in abomasal fluid and intramuscular fat were measured in ten control and ten protozoa-free (PF) lambs. PF lambs were prepared by isolating them from ewes within the first 24 h after birth. The PF and control lambs were kept for 4 months in two different fields and received a daily supplement of 250 g concentrate. The bacterial populations visualised by denaturing gradient gel electrophoresis differed between the two groups of animals and showed a higher bacterial diversity in control lambs than in PF lambs. Abomasal contents from control lambs contained higher concentrations of 22 : 5n-3 and 22 : 6n-3 and lower concentrations of vaccenic acid (trans-11–18 : 1) and 20 : 3n-6 than PF lambs. The rest of the fatty acids, including the conjugated linoleic acid (CLA) isomers, were present at the same concentrations in abomasal contents from both experimental groups. Fatty acid composition in intramuscular fat showed differences between the groups. PF lambs had higher proportions of 18 : 0, 18 : 3, trans-10, cis-12-CLA and total SFA than control lambs. Control lambs had higher proportions of cis-9–18 : 1, cis-9, cis-12–18 : 2, 20 : 3n-6, 22 : 6n-3 (DHA) and MUFA. In conclusion, rumen defaunation led to higher tissue levels of the trans-10, cis-12-CLA isomer and SFA and lower PUFA:SFA ratio and n-3 PUFA in lamb muscle.