Genome-scale reconstruction and metabolic modelling of the fast-growing thermophile Geobacillus sp. LC300

Abstract

Thermophilic microorganisms show high potential for use as biorefinery cell factories. Their high growth temperatures provide fast conversion rates, lower risk of contaminations, and facilitated purification of volatile products. To date, only a few thermophilic species have been utilized for microbial production purposes, and the development of production strains is impeded by the lack of metabolic engineering tools. In this study, we constructed a genome-scale metabolic model, an important part of the metabolic engineering pipeline, of the fast-growing thermophile Geobacillus sp. LC300. The model (iGEL604) contains 604 genes, 1249 reactions and 1311 metabolites, and the reaction reversibility is based on thermodynamics at the optimum growth temperature. The growth phenotype is analyzed by batch cultivations on two carbon sources, further closing balances in carbon and degree-of-reduction. The predictive ability of the model is benchmarked against experimentally determined growth characteristics and internal flux distributions, showing high similarity to experimental phenotypes.

Keywords: Flux sampling; Genome-scale metabolic modelling; Geobacillus; Metabolic engineering; Thermophile.

Graphical abstract

Fig. 1

Composition of the biomass component in the model.

Fig. 2

Growth, substrate consumption, and acetate production of LC300 at 68 °C, with glucose (A) or xylose (B) as carbon source. The x-axis represents time from inoculation. The primary y-axis represents concentrations of substrate and acetate, and the secondary y-axis represents cell mass on a log₁₀ scale. Data points show the mean of quadruplicate bioreactor cultures, with error bars indicating standard deviation.

Fig. 3

Comparison of estimated fluxes in the central carbon metabolism by Cordova et al. (Cordova and Antoniewicz, 2016; Cordova et al., 2017) and predicted fluxes by flux sampling of iGEL604. Each point represents a reaction with its experimentally determined flux (Cordova and Antoniewicz, 2016; Cordova et al., 2017) on the x-axis and the mean of the flux sampling of iGEL604 on the y-axis. The standard deviation of each reaction is visualized as error bars. Fluxes are scaled to a substrate uptake rate of 100 mmol g_DW⁻¹ h⁻¹. (A) Flux sampling with glucose as carbon source, constraining glucose uptake and oxygen uptake to experimentally determined values. Red dots indicate reactions of metabolites at branch-points in the metabolism (Glucose-6-phosphate, Pyruvate, Acetyl-CoA and Isocitrate). (B) Flux sampling with xylose as carbon source, constraining xylose and oxygen uptake to experimentally determined values. Purple dots indicate reactions of metabolites at branch-points in the metabolism. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)