Submitted to: American Society for Microbiology Conference
Publication Type: Abstract Only
Publication Acceptance Date: 6/14/2007
Publication Date: 8/26/2007
Citation: Filiatrault, M.J., Bronstein, P., Stodghill, P., Declerck, G.A., Schneider, D.J., Cartinhour, S.W. 2007. Shotgun Sequencing of the Pseudomonas syringae DC3000 Transcriptome. American Society for Microbiology Conference. p. 20. Interpretive Summary:
Technical Abstract: Pseudomonas syringae pv. tomato strain DC3000 is a bacterial plant pathogen capable of causing disease in tomatoes and Arabidopsis. The genome of this bacterium has been sequenced, however as with other genomes, accurate annotation and determination of coding vs. non-coding regions has proven to be difficult. In order to better characterize the genome and to better understand gene regulation of this pathogen, we employed high-throughput pyrosequencing technology to analyze the transcriptome of P. syringae. Total RNA was isolated from bacteria grown under iron-limited conditions and ribosomal RNA (rRNA) sequences were removed to enrich for mRNA. The enriched RNA sample was directly processed and sequenced by 454 Life Sciences (Branford, CT). Briefly, processing consisted of fragmenting the RNA, converting it to a double-stranded cDNA template, amplifying by PCR, and sequencing. Using this approach, approximately 26 million bases of sequence were obtained in one sequencing run. BLAST-based computational methods were developed to filter the large volume of sequence data generated. Using these techniques, remaining rRNA and repetitive intergenic sequences were identified and removed from the data set. The remaining sequences were assembled and mapped onto the DC3000 genome. This high-throughput sequencing approach provided us with both the 5’ and 3’ ends of the transcripts as well as internal sequences enabling us to assemble a partial draft transcript map of P. syringae DC3000. Approximately 30% of the currently annotated open reading frames were found to be expressed under the conditions analyzed. In addition, our high-throughput experimental analysis of the transcriptome provided evidence for the expression of hypothetical genes and revealed previously unannotated genes as well as candidate small RNAs. RT-PCR was used to confirm the expression of a sample of transcripts detected using 454 sequencing. These results demonstrate that pyrosequencing is an efficient approach for large-scale validation of gene expression and genome annotation. Since this technology has not been previously used to evaluate the transcriptome of a bacterium, this technique represents a novel approach for transcript mapping in bacteria.