Design of Transient Isotopic Labeling Studies for the Experimental Measurement of Autotrophic Metabolic Fluxes

Avantika A. Shastri and John A. Morgan. School of Chemical Engineering, Purdue University, 480 Stadium Mall Dr., West Lafayette, IN 47907

Metabolic flux analysis enables the determination of flux-phenotypes under varying environmental and genetic conditions, therefore forming an integral part of any metabolic engineering process. Although 13C based MFA has been well developed for heterotrophic systems, limited progress has been made to extend it to purely autotrophic systems. This is due to the fact that current methods are isotopic steady state methods, and in autotrophic systems under these conditions, every single carbon atom in every downstream molecule has the same labeling percentage as the single input carbon (CO2).

In this work, we tackle this problem using a transient 13C labeling approach to identify central metabolic fluxes in a prokaryotic cyanobacterium, Synechocystis sp. PCC 6803. We have delineated the key aspects of this approach to require steady-state metabolite concentration measurements and transient isotopic labeling profiles. Metabolic measurements were made using a combination of GC-MS and LC-MS based techniques, and initial flux estimates were obtained using linear-programming based flux balancing. These were used to simulate transient labeling patterns in central metabolism. A systematic study of flux identifiability was performed based on sensitivities and measurement error in pool sizes. Different (simulated) test isotopic labeling datasets were also used to evaluate flux identifiability. The insights gained enable a comprehensive experimental design strategy for flux measurements, that includes a continuous photobioreactor equipped with a rapid sampling device, and protocols for rapid quenching and metabolite extraction. The simulations have also guided the time-scale for sampling and analysis of individual metabolites.