Meta-proteomics of rumen microbiota indicates niche compartmentalisation and functional dominance in a limited number of metabolic pathways between abundant bacteria.

The rumen is a complex ecosystem. It is the primary site for microbial fermentation of ingested feed allowing conversion of a low nutritional feed source into high quality meat and milk products. However, digestive inefficiencies lead to production of high amounts of environmental pollutants; methane and nitrogenous waste. These inefficiencies could be overcome by development of forages which better match the requirements of the rumen microbial population. Although challenging, the application of meta-proteomics has potential for a more complete understanding of the rumen ecosystem than sequencing approaches alone. Here, we have implemented a meta-proteomic approach to determine the association between taxonomies of microbial sources of the most abundant proteins in the rumens of forage-fed dairy cows, with taxonomic abundances typical of those previously described by metagenomics. Reproducible proteome profiles were generated from rumen samples. The most highly abundant taxonomic phyla in the proteome were Bacteriodetes, Firmicutes and Proteobacteria, which corresponded with the most abundant taxonomic phyla determined from 16S rRNA studies. Meta-proteome data indicated differentiation between metabolic pathways of the most abundant phyla, which is in agreement with the concept of diversified niches within the rumen microbiota.

Development of an automatic laboratory computer flagging system to identify urine albumin samples potentially affected by antigen excess ('hooking').

BACKGROUND: Immunoassays for urinary albumin are often subject to the problem of antigen excess (the 'hook' effect) at high albumin concentrations. We developed an automated protocol to identify such samples based on urinary albumin to creatinine ratio (uACR) and urinary total protein (uTP) results. METHODS: An automated flagging system was designed and written into the laboratory computer system to alert technical staff to samples potentially affected by the 'hook effect'. This flag was activated when there was a combination of an uTP of ≥2400 mg/L and an uACR of <30 mg/mmol. RESULTS: The potential rate of false-negative uACR results was approximately 0.17% in samples from primary care and diabetic clinic sources. CONCLUSIONS: Samples with falsely low uACR results were identified, allowing the vast majority of results to be authorized without intervention. The protocol prevented the reporting of false-negative uACR results which might impact on the management of patients.