Poster 2.087

9/1 1:00 - 5:00, Arrillaga Alumni Center, McCaw Hall

Systems Approach for Identifying Transcriptional and Metabolic Changes in Shewanella oneidensis MR-1 During Growth-Phase Transitions

Qasim K. Beg1,*, Mattia Zampieri2,5,*, Niels Klitgord2, Sara Baldwin2, Timothy Gardner1,4 and Daniel Segrè1,2,3

1.Department of Biomedical Engineering, 2.Bioinformatics Program, 3.Department of Biology, Boston University, Boston, MA. 4.Amyris Biotechnologies, Emeryville, CA. 5.International School for Advanced Studies, Trieste, Italy.

In order to study the complex regulatory network of the metabolically versatile bacterium Shewanella oneidensis MR-1 and to understand its unique respiratory and metal-reducing abilities we integrate metabolite and mRNA measurements from a bioreactor experiment under defined growth conditions with computational analyses of gene expression and flux balance analysis. The growth-phase related gene expression changes in E. coli have been studied in the past; however, not much is known about physiological and transcriptional changes in S. oneidensis when cells pass through exponential, stationary, and transition phases. To examine if transcriptional signatures can help understand these changes, we grew S. oneidensis MR-1 in Lactate-limited minimal medium. Time-course samples from various phases of S. oneidensis growth were subjected to microarray analysis (using Shewanella arrays from Affymetrix). To analyze the transcriptional data, we developed a new reverse engineering algorithm (Dynamic Network Enrichment or DNE) that integrates the causality information contained in a time-dependent gene expression data with steady-state expression profiles. A gene network model is first trained on a large compendium of gene expression measurements from M3D dataset (http://m3d.bu.edu) and then the time gene expression profile is fitted on top of the previous inferred network. The DNE algorithm is based on the assumption that during growth of an organism in a batch culture, the gene expression is continuously modulated at various levels because of changing environmental conditions (i.e. medium composition). DNE predicts the pathways primarily modulated by these external perturbations, highlighting the major transcriptional programs during transition between various growth phases. This approach identifies a total of 289 genes, which are enriched for major transcriptional factors mediating batch growth of S. oneidensis MR-1 in lactate-minimal medium. Other genes identified as significantly affected during growth phase changes include genes recently discovered to be part of lactate utilization machinery (SO1518-SO1521 and SO3460). Furthermore, three major dynamic sub-networks emerged from the condition specific gene-network reconstruction. The first sub-network is enriched for genes involved in the starch and sucrose metabolism, glycogen and nitrogen metabolism, TonB energy transduction system, and two component system genes (glnI, glnG, glnA) regulating the glutamate metabolism in low nitrogen conditions. Genes for glycogen and ABC polyamine transporter were also present in this first component. The second sub-network contains genes involved in chemotaxis signal transduction; the third sub-networks include genes that are part of the transposase system (prophage genes). The information gathered about these sub-networks about majorly perturbed genes will be integrated with metabolite measurements and a flux balance analysis model, with the goal of unveiling the sequence of regulatory mechanisms and metabolic requirements mediating the Shewanella's response to environmental changes in the growth medium.

*These authors contributed equally to this work.