Meta-analysis of postruminal microbial nitrogen flows in dairy cattle. I. Derivation of equations

Abstract

The objective was to summarize the literature and derive equations that relate the chemical composition of diet and rumen characteristics to the intestinal supply of microbial nitrogen (MicN), efficiency of microbial protein synthesis (EMPS), and flow of nonammonia nonmicrobial N (NANMN). In this study, 619 treatment means from 183 trials were assembled for dairy cattle sampled from the duodenum or omasum. Backward elimination multiple regression was used to derive equations to estimate flow of nitrogenous components over a large range of dietary conditions. An intercept shift for sample location revealed that omasal sampling estimated greater MicN flow relative to duodenal sampling, but sample location did not interact with any other variables tested. The ruminal outflow of MicN was positively associated with dry matter intake (DMI) and with dietary starch percentage at a decreasing rate (quadratic response). Also, MicN was associated with DMI and rumen-degraded starch and neutral detergent fiber (NDF). When rumen measurements were included, ruminal pH and ammonia-N were negatively related to MicN flow along with a strong positive association with ruminal isovalerate molar proportion. When evaluating these variables with EMPS, isovalerate interacted with ammonia such that the slope for EMPS with increasing isovalerate increased as ammonia-N concentration decreased. A similar equation with isobutyrate confirms the importance of branched-chain volatile fatty acids to increase growth rate and therefore assimilation of ammonia-N into microbial protein. The ruminal outflow of NANMN could be predicted by dietary NDF and crude protein percentages, which also interacted. This result is probably associated with neutral detergent insoluble N contamination of NDF in certain rumen-undegradable protein sources. Because NANMN is calculated by subtracting MicN, sample location was inversely related compared with the MicN equation, and omasal sampling underestimated NANMN relative to duodenal sampling. As in the MicN equation, sampling location did not interact with any other variables tested for NANMN. Equations derived from dietary nutrient composition are robust across dietary conditions and could be used for prediction in protein supply-requirement models. These empirical equations were supported by more mechanistic equations based on the ruminal carbohydrate degradation and ruminal variables related to dietary rumen degradable protein.

Invited review: Nitrogen in ruminant nutrition: A review of measurement techniques

“…Studies that have directly compared N flows using omasal and duodenal sampling indicate that they are comparable (Ahvenjärvi et al, 2000;Rotta et al, 2014), although in the study of Ahvenjärvi et al (2000) where microbial NAN was also determined, it tended to be higher in the omasal canal than in the duodenum. Roman-Garcia et al (2016) in a meta-analysis reported a greater microbial N and smaller NANMN flows for omasal sampling than for duodenal sampling. An indirect validation based on the relationships between omasal NAN flow and milk protein yield derived using meta-analysis also showed that omasal measurements were accurate .…”

“…This technique was modified by Ahvenjärvi et al (2000) and is now used widely . Most omasal sampling protocols use a triple-marker system (France and Siddons, 1986) to mathematically correct for unrepresentative digesta sampling (Ahvenjärvi et al, 2003;Broderick et al, 2010;Roman-Garcia et al, 2016), with 15 N, and to a lesser extent purine bases, employed as microbial markers . Compared with sampling from the duodenum, the omasal canal provides some advantages as a sampling site: it is less invasive, a rumen cannula suffices, secretion of endogenous N into the reticulorumen is lower than that into the duodenum, and sampling takes place before acid-pepsin hydrolysis in the abomasum (Huhtanen et al, 1997;Ahvenjärvi, 2006).…”

Meta-analysis of postruminal microbial nitrogen flows in dairy cattle. II. Approaches to and implications of more mechanistic prediction

“…Data were collected and prepared for analysis as described in Roman-Garcia et al (2016). Briefly, a literature search was conducted to identify studies that measured duodenal or omasal N flows.…”

“…To minimize bias from incorrectly specifying ingredients, the study-level nutrient residuals were calculated and used to adjust book values to ensure no mean bias within a study, as described in Hanigan et al (2013). Means and standard errors were collected for all response variables of interest, and when standard errors were not directly reported, calculated using error propagation (Roman-Garcia et al, 2016).…”

“…To more mechanistically represent the role of ruminal carbohydrate digestibility in microbial protein synthesis, it is useful to compare equation methods that directly predict MicN flow with those that predict ruminal carbohydrate (starch + NDF) digestibility as a first step to subsequently multiply by a predicted efficiency of microbial protein synthesis (EMPS; MicN/ amount of carbohydrate degraded in the rumen). In a companion paper (Roman-Garcia et al, 2016), 3 equations estimating microbial protein were developed. These equations estimated MicN flow as a function of (1) diet composition, (2) ruminal digestibility of starch and NDF, or (3) ruminal digestibility of OM and EMPS.…”