Changes in the Metagenome-Encoded CAZymes of the Rumen Microbiome Are Linked to Feed-Induced Reductions in Methane Emission From Holstein Cows.

Abstract

Enteric methane (CH4) emission from cattle is strongly linked to the feeding regime and the rumen microbial community structure. Here, we report that feed-induced CH4-reducing effects correlate with specific alterations in the profile of the microbiome-encoded carbohydrate-active enzymes predicted from the rumen fluid metagenome. Rumen microbiome samples were obtained by mouth-tube sampling from 12 lactating Holstein cows after 3–4 weeks of feeding with three different concentrate-to-forage-ratio diets, i.e., standard, high, and extremely high levels of concentrate (4 cows per group; constant dry matter intake in the three groups). Increased inclusion of concentrate involved increased starch levels in the diet at the expense of fiber. The extreme diet resulted in 48% reduction of the CH4 emission per kg dry matter intake compared to the standard diet. From metagenome sequencing of the rumen fluid samples from each cow, 561 different microbial strains (bins) could be derived from analysis of 260 billion DNA base pairs. In the cows fed, the extreme diet, the relative abundance of the majority of the bins, was significantly altered compared to the other groups. Fibrobacterota and Verrucomicrobiota were less abundant in the Extreme group. Surprisingly, no significant abundance changes were observed among Archaea and Bacteroidota, although abundance changes of individual bins of these phyla were found. For each of the 561 bins, the functions of the metagenome-encoded carbohydrate-active enzymes were predicted by bioinformatics using conserved unique peptide pattern (CUPP) analysis. By linking each of the predicted molecular functions of the enzymes to their substrates, changes were found in the predicted abundance of the different enzyme types. Notably, the decreased CH4 emission of the extreme diet group was concurrent with a profound decrease in the xylan-active enzymes, targeting the xylan backbone β-1,4-linkages, acetyl-, feruloyl-, and methyl-glucuronoyl substitutions in xylan. This work provides a first enzyme-conversion-based characterization of how extreme feeding, i.e., lowered forage, can drive rumen microbiome changes that support decreased CH4 emission via a changed carbohydrate-active enzyme profile. The data, furthermore, provide a metagenome-wide catalog of enzymes, underpinning the microbial conversion of different feed fibers (the enzymes attacking specific carbohydrate linkages) in the rumen of Holstein cows.