Effects of ruminal or duodenal supply of fish oil on milk fat secretion and profiles of trans-fatty acids and conjugated linoleic acid isomers in dairy cows fed maize silage

Milk fat secretion and profiles of conjugated linoleic acids (CLA) and trans-18:1 isomers in milk fat due to a ruminal or duodenal supply of fish oil (FO; EPA = 22% of total fatty acids, DHA = 7%) were evaluated using six lactating Holstein cows with cannulas in their rumen and duodenum. Cows were fed a control diet based on corn silage (66% of DM), or the control plus 300 mL FO/d into the rumen (FO-R) or duodenum (FO-D) for 4 weeks in a replicated 3 × 3 Latin square design. Dry matter intake was higher with the control (19.8 kg/d), intermediate with FO-D (18.0 kg/d), and lower with FO-R (16.2 kg/d). Yield of milk (22.7 kg/d), lactose (1.09 kg/d), or crude protein (0.70 kg/d) was not affected by treatments. Content of crude protein, true protein, and casein was lower in response to FO-R compared with the control. Despite similar jugular concentration compared with the control, jugular-mammary venous difference for β-OH-butyrate was 54% lower with FO-R or FO-D. Milk fat content and yield were lowest with FO-R (25 g/kg, 567 g/d) compared with FO-D (32 g/kg, 737 g/d) or the control (35 g/kg, 783 g/d). Percentage of trans5-, trans6 + 7 + 8-, trans9-, trans10-, trans11-, trans12-, and trans13 + 14-18:1 in milk fat was greater due to FO-R (0.05, 0.44, 0.48, 1.8, 9.2, 0.70, and 0.82%, respectively) compared with the control or FO-D (0.02, 0.22, 0.19, 0.27, 1.1, 0.35, and 0.40%). Although 18:0 and cis9-18:1 percentage was markedly lower in response to FO-R, percentage of cis11-, cis13-, and cis15-18:1 was greater compared with the control or FO-D. Cis9,trans11-CLA (87-95% of total CLA) averaged 3.2% in response to FO-R compared with 0.51% for control or FO-D. Trans10,cis12-18:2 was not detected. Among non-conjugated 18:2 isomers, cis9,cis12-18:2 was lower but cis9,trans13-, trans9,cis12-, and trans11,cis15-18:2 greater when FO-R was compared with the control or FO-D. There was high correlation (r = 0.94-0.99) among data obtained by GLC versus HPLC for most trans-18:1 isomers and cis9,trans11-18:2, suggesting GLC alone with our chromatographic conditions provided adequate separation of isomers. Milk 20:5n-3 (EPA) and 22:6n-3 (DHA) were 0.08 and 0.04% with the control, increased to 0.36 and 0.17% with FO-R, and were 1.47 and 0.47% with FO-D. Data indicate, ruminal supply of FO led to hydrogenation of EPA and DHA and enhanced percentage of various trans-18:1, cis-18:1, non-conjugated 18:2, and cis9,trans11-18:2 in milk fat. Combined increases in trans-18:1 along with reduced availability of 18:0 for endogenous synthesis of cis9-18:1 may play a key role in reduced milk fat synthesis in cows fed fish oil.