Amino acid analysis using multiple time point hydrolysis and non-linear regression for determination of amino acid profiles and 21-hour correction factors of biological standards, and dairy cattle milk, tissue, rumen microbes, and various feeds

Data published over the last few years has demonstrated that the amino acids (AA) concentrations of milk and other substrates was not correctly described after the standard 21–24 h hydrolysis period, used to break down protein into its constituent AA, and that longer or shorter hydrolysis times might be needed for certain AA. The study objective was to determine the AA composition of ruminant tissue, milk, microbes, and feeds using multiple time-point hydrolysis and non-linear regression to establish optimal AA profiles and develop correction factors for single time-point hydrolysis. The AA were analyzed using this approach on sixteen feeds, nine cattle tissue samples, eight milk samples, and six ruminal microbial samples to represent both supply and requirement protein sources. Substrates were analyzed by HPLC-UV and/or HILIC-TQMS following hydrolysis at 110 °C in a block heater for 10–14 different time points ranging from 2 to 360 h using acid and alkaline hydrolysis. Following hydrolysis, least-squares non-linear regression was used to determine the true AA content. Many AA of all substrates continued to be released after the 24 h endpoint and were all observed to be declining by 360 h. The branched-chain AA (BCAA), Ser, Thr and Trp were found to need the highest correction due to hydrolysis loss or incomplete recovery, but most AA had some correction depending on the protein source. This leads to the conclusion that there is currently an underestimation of the AA concentration in ruminant substrates. Correction factors varied among protein substrates and trended or were significantly different for Ala, Arg, Asp, Glu, Trp, and Val (P ≤ 0.10). Protein-specific factors could help improve AA formulation for improved predictions of nutrient supply and requirements.