Amino acid needs of lactating dairy cows: Predicting limiting amino acids in contemporary rations fed to high producing dairy cattle in California using metabolic models

The objectives were to predict amino acid (AA) profiles of intestinally delivered protein in California high group (i.e., lactating but not yet confirmed to be in calf) dairy cattle fed contemporary rations using three metabolic models of dairy cows. This was done in order to predict limiting AA in dairy rations to determine if there was enough consistency in the nutrient profiles of these rations to support a common ruminally protected (RP) AA package to supplement similar rations. Nutrient profiles of 16 commercial high group dairy cow rations were evaluated, and limiting AA predicted by the metabolic models ‘Amino Cow', ‘CPM Dairy' and ‘Shield'. Higher inclusion levels of corn products in rations increased the contribution of corn CP to the total CP content of the total mixed ration (TMR), from 0.20 to 0.40. Even though the lysine to methionine ratio decreased as more corn CP was included in the TMR, it did not have a major impact on the final predicted AA profile of metabolizable protein (MP), but MP delivery (g/d) to the duodenum was predicted to decrease with increased corn CP levels. None of the models predicted any affect of increased corn CP levels on milk components but, according to Shield, it did have an effect on milk yield which increased when the ratio of lysine to methionine in MP decreased. The sequence of AA limitation among rations was the same within model, but differed substantially among models. Methionine, isoleucine and lysine were predicted to be most limiting according to Amino Cow, CPM Dairy and Shield, respectively. There appears to be sufficient consistency in nutrient profiles among rations to support a ruminally protected AA complex to balance the model predicted AA profile in order to increase animal productivity and efficiency of utilization of nutrients. There is no absolute way to decide which model predictions are most correct. However because Shield predictions suggested a higher correlation between Lys and Met in MP and production, as well as predicted AA ratios to milk responses related to these ratios, use of the Shield predicted AA package is supported.