Metabolic Flux Analysis with a Comprehensive Isotopomer Model in Bacillus subtilis Michael Dauner, James E. Bailey, Uwe Sauer Institute of Biotechnology, ETH Zu ¨ rich, CH 8093 Zu ¨ rich, Switzerland; telephone: +41 1 6333672; fax: +41 1 6331051; e-mail: sauer@biotech.biol.ethz.ch Received 30 October 2000; accepted 11 April 2001 Abstract: Fluxes in central carbon metabolism of a ge- netically engineered, riboflavin-producing Bacillus subti- lis strain were investigated in glucose-limited chemostat cultures at low (0.11 h -1 ) and high (0.44 h -1 ) dilution rates. Using a mixture of 10% [U- 13 C] and 90% glucose labeled at natural abundance, 13 C-labeling experiments were carried out to provide additional information for metabolic flux balancing. The resulting labeling pattern in the proteinogenic amino acids were analyzed by two- dimensional [ 13 C, 1 H] nuclear magnetic resonance (NMR) spectroscopy. To account rigorously for all available data from these experiments, we developed a comprehensive isotopomer model of B. subtilis central metabolism. Us- ing this model, intracellular carbon net and exchange fluxes were estimated on the basis of validated physi- ological data and biomass composition in combination with 2D NMR data from 45 individual carbon atom spec- tra in the amino acids. Glucose catabolism proceeded primarily via glycolysis but pentose phosphate pathway fluxes increased with increasing growth rate. Moreover, significant back fluxes from the TCA cycle to the lower part of glycolysis via the gluconeogenic PEP carboxyki- nase were detected. The malic enzyme reaction, in con- trast, was found to be inactive. A thorough statistical analysis was performed to prove the reliability of the isotopomer balance model and the obtained results. Specifically, a  2 test was applied to validate the model and the chi-square criterion was used to explore the sensitivity of model predictions to the experimental data. © 2001 John Wiley & Sons, Inc. Biotechnol Bioeng 76: 144–156, 2001. Keywords: metabolic flux analysis; isotopomer model; Bacillus subtilis; NMR spectroscopy; central metabolism INTRODUCTION A deeper and quantitative understanding of cellular metabo- lism is important for both fundamental and applied biologi- cal sciences. A particularly valuable tool to gain insight into the complex responses and capabilities of whole cellular metabolism is metabolic flux balancing. The underlying fundamental principle of such analysis is the conservation of mass, which, combined with knowledge on the biochemi- cal reaction network, can be used to balance all metabolic reactions in an organism. A typical microbe such as Esch- erichia coli contains about 700 metabolic reactions (Karp et al., 1999). As the backbone of metabolism, central carbon pathways are often at the focus of interest. Consequently, all biosynthetic reactions to cellular components are lumped into a single reaction that represents the building block re- quirements for biomass generation, thereby greatly simpli- fying the considered biochemical model. Moreover, most flux balance analyses restrict themselves to (quasi-) steady- state conditions, so that metabolite concentrations and en- zyme kinetics can be neglected. The resulting linear equa- tion system can now easily be solved for the fluxes by matrix calculus (Varma and Palsson, 1994). A typical problem of metabolic flux analysis using such metabolite balances are underdetermined equation systems (Bonarius et al., 1997). These are caused by alternative pathways and redundant reactions in central metabolism, an inherent feature of biological systems (Gottschalk, 1986). To obtain unambiguous solutions, either critical assump- tions on enzyme kinetics (Vaseghi et al., 1999), cofactor balances (Sauer et al., 1996), or objective functions (Bonarius et al., 1996) are used. In many cases, reaction models can alternatively be simplified by the lumping of reactions, of course, at the expense of flux resolution (Val- lino and Stephanopoulos, 1993). An alternative is the use of isotopic tracer data, which provide additional independent information on metabolism. While many stable isotopes are present in biological sys- tems, 13 C- and 14 C- based protocols are predominantly used because they are at the core of the organic chemistry of life (Szyperski, 1998). While 14 C-labeled products of cellular metabolism can be analyzed by radiographic methods, 13 C- labeled products are usually analyzed by either nuclear magnetic resonance (NMR) spectroscopy or gas chromatog- raphy–mass spectrometry (GC–MS). The derived informa- tion is then used to augment the metabolite balances by balances on the administered label. These balances can be constructed either on the total label found in a metabolite (Szyperski et al., 1999), enrichments in specific atom posi- tions (Marx et al., 1995) or fragments (Christensen and Nielsen, 1999), or intact carbon bonds in cellular metabo- lites (Sauer, et al., 1997; Szyperski, 1995). All these balance Correspondence to: Uwe Sauer Contract grant sponsor: Boehringer Ingelheim Fonds © 2001 John Wiley & Sons, Inc.