PLoS Comput. Biol. 9, e1003371 (2013)

Credit: PLOS COMPUT. BIOL.

Existing metabolic models of Trypanosoma brucei have aided in the search for new drug targets in this parasite but provide only a limited picture of cellular metabolism. Kerkhoven et al. now expand the existing, glycolytic-focused model to include the pentose phosphate pathway (PPP), which generates NADPH and thus links to oxidative stress. With these reactions in place, the authors discovered that phosphorylated compounds were 'leaking' from glycolysis into the PPP, disrupting the conservation of the glycolytic reactions. To resolve this, the authors tried two separate amendments: in model C (see picture), ribokinase was introduced to enzymatically restore phosphates. Experimental tests ruled out ribokinase as capable of single-handedly fixing the leak but suggested that a larger network of enzymes could provide a solution to the problem. In model D, an ATP-ADP antiporter was added to transport phosphorylated metabolites into the system; however, this mathematically complete solution could not be tested directly as the relevant transporters are not yet known. By combining these models, which explicitly incorporate uncertainty, with cellular assays, the authors were able to test a prior prediction that the enzyme 6PGDH would be a good drug target; they observed that 6PGDH inhibition was lethal, but this was unexpectedly related to the loss of the oxidative branch of the PPP rather than anticipated feedback mechanisms.