Designing metabolism: Alternative connectivities for the pentose phosphate pathway (Q1286860)

The primary concern of this paper is to develop and illustrate a new method for generating an ensemble of alternative networks in order to establish a scenario for the evolution of a real pathway under selection pressure. In particular, alternatives to the nonoxidative stage of the pentose phosphate pathway are generated, by first establishing networks that only take into account the transformations between the carbon skeletons of the key metabolites (C-nets) and then adding the functional groups to obtain possible realistic networks of changers and connectors (R-nets). The class of R-nets is far larger than the class of C-nets which is decomposed into 53 families which segregate into seven superfamilies according to the number of their modules (i.e., coupled sets of reactors that occur more than once in a network) and common patterns of linkages among the modules and remainders. The C-nets and the R-nets are all designed by algorithms and additional modifications by hand. Though the optimality criteria that are studied allow to strongly reduce the number of candidate networks for pathways that could be possibly realized in nature, this number is still very large. However, many aspects show how well the real pentose phosphate pathway fits into this ensemble.