Effect of supplementation with fish oil or microalgae on fatty acid composition of milk from cows managed in confinement or pasture systems

The objective of this study was to examine the interaction between lipid supplement (LS) and management system (MS) on fatty acid (FA) composition of milk that could affect its healthfulness as a human food. Forty-eight prepartal Holstein cows were blocked by parity and predicted calving date and deployed across pasture (PAS; n = 23) or confinement (CONF; n = 25) systems. Cows within each system were assigned randomly to a control (no marine oil supplement) or to 1 of 2 isolipidic (200 g/d) marine oil supplements: fish oil (FO) or microalgae (MA) for 125 ± 5 d starting 30 d precalving. The experiment was conducted as a split-plot design, with MS being the whole-plot treatment and LS as the subplot treatment. Cows were housed in a tie-stall barn from −30 until 28 ± 10 d in milk (DIM) and were fed total mixed rations with similar formulations. The PAS group was then adapted to pasture and rotationally grazed on a perennial sward until the end of the experiment (95 ± 5 DIM). Milk samples were collected at 60 and 90 DIM for major components and FA analyses. Milk yield (kg/d) was lower in PAS (34.0) compared with CONF (40.1) cows. Milk fat percentage was reduced with MA compared with FO (3.00 vs. 3.40) and the control (3.56) cows. However, milk fat yield (kg/d) was not affected by lipid supplements. Compared with CONF, PAS cows produced milk fat with a lower content of 12:0 (−38%), 14:0 (−28%), and 16:0 (−17%), and more cis-9 18:1 (+32%), 18:3 n-3 (+30%), conjugated linoleic acid (CLA; +70%) and trans 18:1 (+34%). Both supplements, regardless of MS, reduced similarly the milk fat content of 16:0 (−12%) and increased CLA (+28%) and n-3 long-chain polyunsaturated FA (n-3 LC-PUFA; +150%). Milk fat content of trans 18:1 (trans-6 to trans-16) was increased with FO or MA, although the effect was greater with MA (+81%) than with FO (+42%). The interaction between MS and LS was significant only for trans-11 18:1 (vaccenic acid, VA) and cis-9,trans-11 CLA (rumenic acid). In contrast to CONF, feeding FO or MA to PAS cows did not increase milk fat content of VA and rumenic acid. We concluded that compared with CONF, milk from PAS cows had a more healthful FA composition. Feeding either FO or MA improved n-3 long-chain polyunsaturated FA and reduced levels of 16:0 in milk fat, regardless of MS, but concurrently increased the trans 18:1 isomers other than VA, at the expense of VA, particularly in grazing cows.