Estimation of the Association Between the Rumen Bacterial Abundance and Host Phenotypes of Lactating Dairy Cows Using Structural Equation Modeling

This study aimed to determine the association between the rumen bacterial abundance and the host phenotypes of dairy cows by performing structural equation modeling (SEM). Two herds of milking Holstein cows in Hokkaido and Gunma prefectures in Japan were examined. To improve the goodness-of-fit of the model, rumen bacteria were categorized into four groups based on the relative abundance in rumen microbiota and their correlation with (acetate + butyrate)/propionate ratio (NGR) as an indicator for rumen fermentation and enteric methane production. The top three most abundant archaeal operational taxonomic units were used as variables. Rumen fermentation parameters and blood metabolites were respectively transformed into three principal components (PCs). The SEM model revealed the interaction of the methane intensity with energy-corrected milk (ECM) and rumen PC1 and PC2 representing NGR and acetate concentration, and their standardized regression coefficients were comparable. Several bacteria that produce propionate and its intermediates, such as Prevotella lacticifex, Segatella baroniae, Parafannyhessea umbonata, and Dialister succinatiphilus, were detected in bacterial clusters interacting with rumen PC1 and PC2. Our findings indicated that ECM and several propionate- and its intermediate-producing bacteria contributed to reduce methane intensity; however, the archaea and blood PCs did not fit the SEM model.