(352d) Development of a Metabolite Stress-Response Model in Solventogenic Clostridia by Coupling Multiple –Omic Data with a Genome-Scale Model

The toxic-metabolite stress response is a problem of major and general importance in all microbial systems of interest to bioenergy production. Among the potential microbial hosts for biofuel production, solventogenic clostridia are of major importance because of their inherent ability to utilize a large variety of substrates (hexoses, pentoses, oligosaccharides, xylan, and starches) with minimal to no catabolite repression. Here, we focus on understanding and modeling the stress response of Clostridium acetobutylicum to two important toxic metabolites: butanol and butyrate. Butanol is the major metabolite produced during stationary phase, while butyrate is primarily produced during exponential growth.

In order to model the stress response to these metabolites, C. acetobutylicum cultures were grown to mid-exponential phase and then stressed with four levels of the metabolite of interest (no, low, medium, and high levels of stress). Samples were collected following the stress to look at the cellular response to the different stress levels. The transcriptional response was measured using both genomic microarrays and deep sequencing (RNA-seq) using the Illumina technology. The goal of using these complimentary technologies is to not only assess gene expression patterns affected by the metabolite stress, but also examine and model changes at the molecular level impacted by stress, such as changes in expressed operon mRNAs, transcriptional start cites, as well as stress-specific RNA degradation. Furthermore, proteomic changes were measured using the iTRAQ technology to quantitatively evaluate protein expression changes, especially because the key mechanism of regulation of the universal stress response system (the HSP system) is largely regulated at the protein level by protein-protein interactions. Finally, guided by the transcriptional data, metabolic flux changes of core programs within the cells were determined using ¹³C-labeled chemicals. The goal of these measurements is to quantitate how cells re-allocate its energy and carbon resources under stress, whereby a large number of programs are downregulated but a few are upregulated. These data were then integrated within a second-generation genome-scale model, which incorporates algorithms to regulate the model based on gene and protein expression, protein-protein interactions, and metabolic flux regulation. We ultimately aim to use this model to design a semi-synthetic stress response system that enables the cells to produce toxic metabolites at higher rates and titers.