Sarah A. Munro1,2, Leila Choe3, Kelvin H. Lee3, Stephen H. Zinder4, Larry P. Walker2


Biochemical Sciences Division, National Institute of Standards and Technology, Gaithersburg MD 20899 Department of Biological and Environmental Engineering, Cornell University, Ithaca, NY 14853 3 Chemical Engineering Department and Delaware Biotechnology Institute, University of Delaware, Newark, DE 19711 4 Department of Microbiology, Cornell University, Ithaca, NY 14853


Hydrogen (H2) could play a significant role in meeting our future energy needs. One promising species for biological hydrogen production is the hyperthermophilic bacterium, Thermotoga neapolitana. This organism converts sugars to H2 via fermentation. Research was conducted in three phases using the systems biology approach of iterative experimental and computational analysis to evaluate the metabolic capabilities of T. neapolitana.  First, experimental analysis was conducted to determine the T. neapolitana fermentation stoichiometry and response to environmental perturbations.  The hydrogen yield from glucose was established for this species and the effects of temperature, oxygen, and pH on growth and hydrogen production were evaluated.  In the second phase, a constraint-based model of T. neapolitana central carbon metabolism was built.  To accomplish this, a comparative model reconstruction method was created to convert a constraint-based model for the closely related species Thermotoga maritima into a model for T. neapolitana based on synteny between the two annotated genome sequences.  This comparative model development method is an expeditious procedure to generate a model by extending the utility of a related model and bypassing the detailed database searching and integration required for the original model reconstruction process.  Experimental data from the first physiological study was used to test the accuracy of the model and model simulations were used to generate hypotheses related to substrate utilization.  In the final phase, hypotheses generated from the T. neapolitana model were tested in physiological and proteomic experiments.  The model predicted that glycerol, L-rhamnose, and cellotetraose could not support growth.  Experimental results showed that, as expected, both glycerol and L-rhamnose did not support growth, however cellotetraose did support growth.  This underscores the value of iterative modeling and experimentation to increase our understanding of biological systems.