First, with a genome size of ~580 kb and approximately 480 protein coding regions, Mycoplasma genitalium is the smallest known free-living organism. The reduced genomic content of M. genitalium has led researchers to suggest that the molecular assembly represented by this organism may be a close approximation to the minimal set of genes required for bacterial growth. One challenge to its study, though, has been that it has extremely fastidious nutrient requirements. Here, we present a flux balance analysis (FBA) approach to examine the substrate uptake requirements for M. genitalium. This approach involves constructing a metabolic model that encompasses the reactions carried out by the enzymes encoded in the genome.
Once a metabolic model was constructed, it was validated by comparing FBA simulation predictions with experimental data for growth of various mutant strains. For instance, these include classifications of genes as essential or nonessential for growth. The validated model is employed to predict mathematically viable uptake environments for M. genitalium. Specifically, this process is used to systematically guide the construction of chemically defined medium based on the underlying metabolic pathways.
The second topic concerns an important tool that can be used to quantify the flux values in the network of a metabolic model: isotope labeling. Here, we present a computational framework that combines a constraint-based modeling framework with 13C isotopic label tracing. This Escherichia coli model includes 393 fluxes, 214 metabolites, and balances on cofactors such as ATP and NADH as well as the electron transport chain, full amino acid biosynthesis and degradation, and a detailed biomass equation. By including pathways such as cofactor balancing and the electron transport chain, we can ensure that the results are biologically relevant.
Two of the important considerations for analysis of the isotope model are labeled substrate choice and how isotope measurements impact the elucidation of fluxes in large-scale metabolic reconstructions. By introducing a degrees of freedom based optimization method it is possible to exhaustively identify all combinations of isotope labeling experiments and flux measurement that completely resolve all flux values in the network. These measurements, consisting of both partial or full isotope state determination, are assigned relative costs that allow the experimentalist to select the measurements that will be both sufficient and economical. Additional measurements can be taken in order to provide information about the confidence intervals of the fluxes.
Experimental results of flux elucidation are presented for an Escherichia coli strain engineered to produce amorphadiene, a precursor to the anti-malarial drug artemisinin. These include a statistical analysis of fluxes determined for the system such as the minimal and maximal values of the fluxes given measurement noise. We also introduce a degree of resolution calculation that quantifies how the isotope data restrict these fluxes within the ranges allowed by overall stoichiometry.
Once these flux ranges are determined, they can guide engineering of the system to improve product yields. For instance, if we wish to increase the production rate of a target compound, we can identify other fluxes that need to be changed beyond the ranges determined by the isotope data. These fluxes then become potential targets of manipulation. This approach is applied to the production of amorphadiene and other bio-compounds of industrial interest.