Lawrence Berkeley National Laboratory

Current technologies applied to environmental samples for RNA transcriptional profiling include RT-PCR and functional gene microarrays using total RNAs. However, these methods bear significant limitations that prevent their application in a high throughput manner to de novo communities. To circumvent this limitation, we are developing a method for direct sequencing of cDNA from environmental samples utilizing the HT sequence analysis system Bio454. Since about 80% of total RNA of microorganisms may consist of rRNAs, it is crucial to first remove rRNAs as completely as possible without degrading mRNA quality and quantity. We compared three methods for removing rRNA from the total RNA of Desulfovibrio vulgaris. The first method utilizes biotin modified oligos complementary to conserved regions in 16S & 23S rRNA. The second uses an exonuclease that targets rRNAs bearing a 5 monophosphate. The third method uses first strand transcription with universal primers for 16S & 23S RNAs, and subsequent removal of rRNA and cDNA. All three methods were able to significantly enrich mRNA from rRNA We evaluated the effect of these methods on mRNA quality by microarray analysis. Correlation coefficients between mRNA enriched and control samples ranged from 0.84 to 0.96. Between 0.2 and 2% of genes showed significantly altered expression levels (p=0.05) after rRNA removal with each method, however most showed increased intensity suggesting rRNA removal may increase sensitivity. Currently we are constructing cDNA libraries for HT sequencing with the 454 to further optimize and validate this approach in single species, as well as make comparisons of HT sequence based methods with existing microarrays. The developed tools will then be deployed to understand microbial responses and survival in stressed environmental systems.