Abstract #491

# 491
Comparison of noninvasive ruminal sampling techniques to standard cannula sampling method for ruminal microbial analysis.
D. Pitta*1, C. Lage2,3, S. Räisänen2, A. Melgar2, K. Nedelkov2,4, X. Chen2,5, J. Oh2, N. Indugu1, B. Vecchiarelli1, J. Bender1, A. Hristov2, 1University of Pennsylvania, University of Pennsylvania, School of Veterinary Medicine, New Bolton Center, Kennett Square, PA, 2The Pennsylvania State University, The Pennsylvania State University, University Park, PA, 3Universidade Federal de Minas Gerais, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil, 4Trakia University, Faculty of Veterinary Medicine, Trakia University, Stara Zagora, Bulgaria, 5Lanzhou University, College of Pastoral Agriculture Science and Technology, Lanzhou University, China.

Rumen microbes play an important role in the conversion of indigestible plant material to energy and protein in dairy cows. Sampling for ruminal contents via cannula is considered the gold standard technique for microbial analysis, but this technique requires ruminal cannulated animals, and specialized animal facilities. The purpose of this study was to determine if other sampling methods and locations along the digestive tract may serve as a non-invasive proxy to the cannula method. Six ruminal-cannulated lactating Holstein cows were fed a standard total mixed ration for 2 wk to allow for adaptation followed by samplings during the third week. Sampling locations and method included salivary content, bolus, rumen content via stomach tube, sampling through the rumen cannula, and sampling of feces. Samples were collected at 0 (pre-feeding), 2, 4, 6, 8 and 12 h post-feeding over 2 d. Samples were extracted for genomic DNA, PCR-amplified for the V1V2 region of the 16S rDNA bacterial gene for bacterial diversity analysis using QIIME pipeline followed by statistical analysis in R. At the community level, saliva, bolus and fecal samples clustered in separate groups (P ≤ 0.05) whereas stomach tube samples and cannulated samples clustered together (P ≥ 0.05), indicating that stomach tube and cannula samples were homogeneous. Both saliva and feces had certain bacteria that were not previously reported in the rumen, suggesting these sample types have distinct microbial fingerprints. Notably, bolus samples at 6, 8 and 12 h had similar compositions with the corresponding cannula samples. There were several bacterial lineages in bolus samples also detected in saliva samples, but not in other sample types, indicating possible contamination of the bolus sample with saliva. It can be concluded the stomach tube method may serve as a proxy to cannula method for microbial investigations. Bolus samples may also serve as a proxy; however, sampling time and removal of bacterial contaminants from saliva will be needed for microbial analysis.

Key Words: rumen bacteria, dairy cow, ruminal sampling