Use of odd and branched-chain fatty acids in rumen contents and milk as a potential microbial marker.

The objectives of this study were: 1) to determine if a correlation exists between rumen odd and branched-chain fatty acids (OBCFA, i.e., C(15:0), iso C(15:0), anteiso C(15:0), C(17:0), iso C(17:0), anteiso C(17:0), and C(17:1)), uracil, and purine bases (PB), 2) to evaluate the potential of milk OBCFA secretion to predict duodenal flow of microbial protein in lactating cows, 3) to evaluate the accuracy of the latter prediction equations using an independent data set, and 4) to determine whether these predictions were more accurate than predictions based on dry matter intake (DMI) and dietary characteristics. In the first experiment, 4 lactating dairy cows arranged in a 4 x 4 Latin square were offered diets based on grass silages of different botanical composition and a standard concentrate. The relationship between rumen pool size of OBCFA and microbial matter was investigated. Rumen pool size of microbial matter (g), determined 4, 12, and 17 h after feeding, using uracil and PB as microbial markers, was closely related to OBCFA (g) [r(2) = 0.716, root mean square error (RMSE) = 4.45]. To correct for differences in marker concentrations among strains of rumen microbes, C(17:0) was included in the regression equations, resulting in an increased predictive power (r(2) = 0.780, RMSE = 3.92). The relationship between microbial flow to the duodenum and milk OBCFA yield was evaluated in a second experiment with lactating dairy cows offered diets based on grass silage and concentrates differing in starch source. Similar to observations in the rumen, milk OBCFA yield was closely related to microbial flow to the duodenum (RMSE = 4.28), but predictive power of equations did not increase when straight-chain C(17)-fatty acids were included in the regression equations (RMSE = 4.92). Evaluation of the current prediction equations with 3 independent datasets resulted in a root mean square prediction error of 20.5 and 13.4% of the observed mean for equations based on milk secretion of total OBCFA and straight-chain C(17)-fatty acids, respectively. Comparison of the accuracy of the latter equations with 2 previously published equations based on DMI and dietary characteristics suggest the former to be more accurate. This first evaluation suggests that milk OBCFA could be used as a marker for duodenal flow of microbial matter, especially when accurate measurements of DMI are not available.