Estimation of feed intake and digestibility in Zebu type Arsi steers fed natural pasture using the n-alkane technique

Correlation of Ruminal Fermentation Parameters and Rumen Bacterial Community by Comparing Those of the Goat, Sheep, and Cow In Vitro

In this study, we aimed to establish the correlation between ruminal fermentation parameters and the bacterial community by comparing those of the goat, sheep, and cow, thus illustrating the main bacteria causing the difference in rumen fermentation among goats, sheep, and cows and providing a new idea for improving the feed digestibility of ruminants. Rumen fluid from goats (Taihang White cashmere goat, n = 6), sheep (Hu sheep, n = 6), and cows (Holstein cow, n = 6) was collected using oral intubation and immediately brought back to the laboratory for a fermentation test with the same total mixed ration (TMR) feed in vitro. The rumen bacterial composition was measured by high-throughput sequencing of 16S rRNA genes in the MiSeq platform, the gas production (GP) was recorded after 2, 4, 6, 8, 10, 12, 24, 36, and 48 h of fermentation, and the feed nutrient digestibility and the rumen fluid parameters were determined after 48 h of fermentation. The results showed that the 48 h GP of the sheep group was higher than that of the cow group (p < 0.05), and the theoretical maximum GP was higher than that of the goat and cow groups (p < 0.05). The organic matter digestibility (OMD), dry matter digestibility (DMD), crude protein digestibility (CPD), and gross energy digestibility (GED) of the sheep group were higher than those of the goat and cow groups (p < 0.05). The ammonia nitrogen (NH3-N), microbial protein (MCP), and total volatile fatty acids (TVFA) concentrations of the sheep group were higher than those of the other groups (p < 0.05), and the pH of the sheep group was lower than those of the other groups (p < 0.05). The 16S rRNA gene sequencing revealed that bacterial composition also differed in the rumens of the sheep, goat, and cow groups (ANOSIM, p < 0.05). We then used a random forest machine learning algorithm to establish models to predict the fermentation parameters by rumen bacterial composition, and the results showed that rumen bacterial composition could explain most of the ruminal fermentation parameter variation (66.56%, 56.13%, 65.75%, 80.85%, 61.30%, 4.59%, 1.41%, −3.13%, 34.76%, −25.62%, 2.73%, 60.74%, 76.23%, 47.48%, −13.2%, 80.16%, 4.15%, 69.03%, 32.29%, and 89.96% for 48 h GP, a (GP of quickly degraded part), b (GP of slowly degraded part), c (GP rate), a + b (theoretical maximum GP), DMD, OMD, GED, CPD, NDFD, ANDF, pH, NH3-N, MCP, acetic acid, propionic acid, butyric acid, valeric acid, TVFA, and A:P (acetic acid–propionic acid ratio), respectively). A correlation analysis revealed that Lactobacillus, Prevotellaceae_UCG-003, Selenomonas, Peptostreptococcus, and Olsenella significantly correlated with most in vitro fermentation parameters (p < 0.05). A comprehensive analysis showed that rumen fermentation parameters and bacterial composition differed in goats, sheep, and cows. The ruminal fermentation parameters of GP, a, b, c, a + b, pH, NH3-N, propionic acid, valeric acid, and A:P could be accurately predicted by rumen bacteria (explanation > 55% of variation), and the Lactobacillus, Prevotellaceae_UCG-003, Olsenella, Selenomonas, and Peptostreptococcus were the main bacteria that affected the in vitro fermentation parameters of goats, sheep, and cows.