2011 Annual Report
1a.Objectives (from AD-416)
The overall objective of this project is to elucidate genomic mechanisms of detoxification and tolerance of ethanologenic yeast to biomass conversion inhibitors furfural and 5-hydroxymethylfurfural (HMF), and thereafter to genome-wise manipulate and engineer more robust strains for low-cost biomass conversion to ethanol. This study will identify and characterize genes involved in pathways relevant to detoxification, biotransformation, and tolerance to furfural and HMF involved in biomass conversion to ethanol; and elucidate regulatory mechanisms of major gene interactions in relevant pathways involved in furfural and HMF detoxification and tolerance using computational prediction and mathematical modeling.
1b.Approach (from AD-416)
We plan to study genomic regulatory mechanisms of inhibitor detoxification by yeast during ethanol production from dilute acid-hydrolyzed biomass. We propose to characterize the genomic transcriptional profiling of wild-type and several improved, more inhibitor-tolerant strains in response to furfural and 5-hydroxymethylfurfural (HMF) supplied in a defined culture medium. To accomplish this, yeast cells will be sampled in a time-course study to isolate total RNA and conduct microarray experiments using two-color microarray with spiking universal external RNA quality controls. Inhibitor and inhibitor-conversion products, glucose consumption, ethanol production, and other byproducts generated during the fermentation process will also be monitored during the time-course study to establish metabolic profiles for wild-type and more tolerant strains involved in detoxification of biomass conversion inhibitors. Based on data from culture time-course studies, we will propose computational models to predict the behavior of the gene function and expression of natural and genetically engineered networks under furfural and HMF stress. A dynamic mathematical model using difference equations and estimate parameters will be applied and tested for its ability to describe gene regulatory network behavior. Based on these approaches, we will form testable hypotheses to explain molecular and genomic mechanisms of yeast detoxification and tolerance to furfural and HMF.
We have identified the first key regulatory elements and candidate genes involved in yeast tolerance to inhibitors associated with hydrolysates. Yeast tolerance to fermentation inhibitors such as furfural and 5-hydroxymethylfurfural (HMF) is key to a sustainable low cost biomass-to-ethanol industry. However, development of tolerant strains is hampered by a lack of understanding of genetic mechanisms underlying stress tolerance for ethanologenic yeast. Global profiling of tolerant and susceptible yeast under challenge of HMF was explored and more than 300 genes significantly involved in regulation of HMF tolerance and detoxification were identified. A data-driven discrete dynamic system model was used to detect gene regulatory interactions and to predict system dynamic behavior based on microarray data sets. Composed of linear difference equations, the model reconstructed transcriptional regulations in gene regulatory networks in response to HMF. The modeling identified a system of linear difference equations to represent temporal interactions among significantly expressed genes with reference to the normal condition in the absence of the inhibitor. A statistically significant model of the yeast gene regulatory network was derived from time course gene expression measurements for exposure to HMF, and it revealed several verified transcriptional regulation events. These events implicate two transcription factors consistently known for their regulatory roles by other studies or predicted by independent sequence motif analysis, suggesting their involvement in yeast tolerance and detoxification of the inhibitor. Using our model system, we identified 12 potentially significant regulatory interactions, among which, were 2 significant regulatory elements for HMF tolerance in yeast. The use of a discrete dynamic system model to detect interactions in a network is novel, and the method has potential for broad applications in biology. Based on a preliminary topology and working-zone based pathway enrichment analysis, we also approached a computation program to determine whether interactions and working zones on a known pathway have been involved differentially by the inhibitors. We have currently performed simulation studies on hypothetical networks to verify the method against a commonly used non-topology based pathway enrichment approach. Results contribute to the general understanding of stress tolerance and benefit development of more tolerant yeast strains for conversion of renewable agricultural residues to ethanol. This represents the final report of grant agreement 3620-41000-147-02G between ARS and NMSU. The Authorized Departmental Officer's Designated Representative (ADODR) monitored the activities of this agreement via monthly teleconferences and frequent e-mail contacts.