Abstract: Since the dawn of life, evolution of novel metabolic pathways has enabled new strategies for exploiting environmental resources and manipulating competing and cooperating organisms. Bioinformatic evidence suggests that organisms evolve novel metabolic pathways by patching together existing enzymes with promiscuous side activities into new combinations. However, many questions remain. What promiscuous enzymes were available in the organism in which a new pathway evolved? Why did one pathway emerge rather than other possibilities? Did the first pathway to be discovered “win”, or did subsequent evolution allow a better solution to emerge?

We define “protopathways” as the earliest stage in the evolution of novel pathways. At this stage, promiscuous enzymes are serving new functions as well as their native functions and proper regulation has not emerged. We have developed a model system that allows us to follow the emergence of protopathways. We delete pdxB, a gene required for synthesis of the essential cofactor pyridoxal 5’-phosphate (PLP, aka vit B6) in g-proteobacteria. Complicated population dynamics occurred during laboratory evolution of ∆pdxB E. coli as competing clones rose and fell in abundance. Within 150 generations, a dominant clone (JK1) had evolved a four-step protopathway that bypasses the block in PLP synthesis caused by deletion of pdxB. We have identified the order in which four mutations arose in JK1 and the physiological effect of each. Interestingly, the second mutation created a cheater that was less fit on its own but thrived in the population by scavenging nutrients released from the fragile parental cells. The dominant lineages at the end of the experiment all derived from this cheater strain. We have also evolved ∆pdxB strains of Salmonella enterica, Alliivibrio fischeri and Pseudomonas putida. In some cases, different mutations led to the same novel protopathway. In others, a different protopathway emerged.