Simultaneous assessment of soil microbial community structure and function through analysis of the meta-transcriptome.

• Main Authors: Tim Urich, Anders Lanzén, Ji Qi, Daniel H Huson, Christa Schleper, Stephan C Schuster
• Format: Article
• Language: English
• Published: Public Library of Science (PLoS) 2008-06-01
• Series: PLoS ONE
• Online Access: http://europepmc.org/articles/PMC2424134?pdf=render

Soil ecosystems harbor the most complex prokaryotic and eukaryotic microbial communities on Earth. Experimental approaches studying these systems usually focus on either the soil community's taxonomic structure or its functional characteristics. Many methods target DNA as marker molecule and use PCR for amplification.Here we apply an RNA-centered meta-transcriptomic approach to simultaneously obtain information on both structure and function of a soil community. Total community RNA is random reversely transcribed into cDNA without any PCR or cloning step. Direct pyrosequencing produces large numbers of cDNA rRNA-tags; these are taxonomically profiled in a binning approach using the MEGAN software and two specifically compiled rRNA reference databases containing small and large subunit rRNA sequences. The pyrosequencing also produces mRNA-tags; these provide a sequence-based transcriptome of the community. One soil dataset of 258,411 RNA-tags of approximately 98 bp length contained 193,219 rRNA-tags with valid taxonomic information, together with 21,133 mRNA-tags. Quantitative information about the relative abundance of organisms from all three domains of life and from different trophic levels was obtained in a single experiment. Less frequent taxa, such as soil Crenarchaeota, were well represented in the data set. These were identified by more than 2,000 rRNA-tags; furthermore, their activity in situ was revealed through the presence of mRNA-tags specific for enzymes involved in ammonia oxidation and CO(2) fixation.This approach could be widely applied in microbial ecology by efficiently linking community structure and function in a single experiment while avoiding biases inherent in other methods.