Rumen Sampling Methods Bias Bacterial Communities Observed

Abstract

The rumen is a complex ecosystem that plays a critical role in our efforts to improve feed efficiency of cattle and reduce their environmental impacts. Sequencing of the 16S rRNA gene provides a powerful tool to survey shifts in the microbial community in response to feed additives and dietary changes. Oral stomach tubing a cow for a rumen sample is a rapid, cost-effective alternative to rumen cannulation for acquiring rumen samples. In this study, we determined how sampling method, as well as type of sample collected (liquid vs solid), bias the microbial populations observed. The abundance of major archaeal populations was not different at the family level in samples acquired via rumen cannula or stomach tube. Liquid samples were enriched for the order WCHB1-41 (phylum Kiritimatiellaeota) as well as the family Prevotellaceae and had significantly lower abundance of Lachnospiraceae compared with grab samples from the rumen cannula. Solid samples most closely resembled the grab samples; therefore, inclusion of particulate matter is important for an accurate representation of the rumen microbes. Stomach tube samples were the most variable and were most representative of the liquid phase. In comparison with a grab sample, stomach tube samples had significantly lower abundance of Lachnospiraceae, Fibrobacter and Treponema. Fecal samples did not reflect the community composition of the rumen, as fecal samples had significantly higher relative abundance of Ruminococcaceae and significantly lower relative abundance of Lachnospiraceae compared with samples from the rumen.