Because microbial crude protein (MCP) constitutes more than 50% of the protein digested in cattle, its AA composition is needed to adequately estimate AA supply. Our objective was to update the AA contributions of the rumen microbial AA flowing to the duodenum using only studies from cattle, differentiating between fluid-associated bacteria (FAB), particle-associated bacteria (PAB), and protozoa, based on published literature (53, 16, and 18 treatment means were used for each type of microorganism, respectively). In addition, Cys and Met reported concentrations were retained only when an adequate protection of the sulfur groups was performed before the acid hydrolysis. The total AA (or true protein) fraction represented 82.4% of CP in bacteria. For 10 AA, including 4 essential AA, the AA composition differed between protozoa and bacteria. The most noticeable differences were a 45% lower Lys concentration and 40% higher Ala concentration in bacteria than in protozoa. Differences between FAB and PAB were less pronounced than differences between bacteria and protozoa. Assuming 33% FAB, 50% PAB, and 17% of protozoa in MCP duodenal flow, the updated concentrations of AA would decrease supply estimates of Met, Thr, and Val originating from MCP and increase those of Lys and Phe by 5 to 10% compared with those calculated using the FAB composition reported previously. Therefore, inclusion of the contribution of PAB and protozoa to the duodenal MCP flow is needed to adequately estimate AA supply from microbial origin when a factorial method is used to estimate duodenal AA flow. Furthermore, acknowledging the fact that hydrolysis of 1 kg of true microbial protein yields 1.16 kg of free AA substantially increases the estimates of AA supply from MCP.

Although it has been acknowledged for a long time that a single period of hydrolysis, normally 21 to 24 h, is not the optimal time for most of the AA, a single period is routinely used due to time and cost constraints. As models to balance dairy rations for proteins are evolving toward balancing for AA, it becomes critical to improve the predictions of AA supply from digested proteins. Our objective was to develop correction factors that could systematically be applied to AA concentrations obtained after a 24-h hydrolysis of proteins to account for incomplete recovery and therefore determine their true AA composition. Thirteen substrates were selected to represent different types of proteins commonly used to estimate the supply of AA in ration formulation models: feed ingredients (grass silage, corn silage, soybean meal, canola meal, high-protein corn dried distillers grains, and wheat dried distillers grains plus solubles), 16-h rumen residues (soybean meal and canola meal), digesta (duodenal digesta and feces), and rumen microorganisms (fluid-associated bacteria, particle-associated bacteria and protozoa). Each protein was hydrolyzed in 6 N HCl for multiple hydrolysis times: 13 (2, 4, 8, 12, 18, 21, 24, 30, 48, 72, 96, 120, and 168 h) for feed ingredients, rumen residues, and digesta, and 9 (2, 4, 8, 18, 24, 30, 48, 96, and 168 h) for rumen microorganisms; all analyses were conducted in triplicate. Using nonlinear regression, the AA composition in the protein before the hydrolysis (A 0 ) was derived for each AA in each protein. Two ratios were calculated as potential correction factors: A 0 /24-h concentration (A 0 /24h) and the maximal concentration/24-h concentration (max/24h). Both ratios were tested to determine if the type of proteins was affecting them. The ratios A 0 /24h were not affected by the type of proteins, whereas the ratios max/24h were also not affected by the type of proteins except for 3 nonessential AA (Ala, Glu, and Gly). In an attempt to propose correction factors, our results were combined with results from the literature reporting ratios A 0 /24h, ratios max/24h, or the ratio of the AA composition calculated from gene structure/24 h. The correction factors proposed for individual AA varied from 1.02 (Asp) to 1.12 (Thr). For the essential AA, the highest ratios were obtained, as expected, for the branched-chain AA and Thr. Formulation programs balancing dairy rations for essential AA would need to acknowledge the incomplete recovery of AA when obtained from 24-h hydrolysis and include correction factors, specific for each AA, but the same across different types of proteins, to correctly estimate the true AA supply to dairy cows.