| Specialization is a balance of evolutionary adaptation and its accompanying costs. Here we focus on the Lenski Long-Term Evolution Experiment, which has maintained cultures of Escherichia coli in the same, defined seasonal environment for 50,000 generations. This dissertation explores the extent and means by which metabolic specialization occurs over an extended period in the same environment.;Chapter 1 provides an overview and introduction of the problem.;In Chapter 2 we investigated the acquired dependence on citrate for optimal growth of some of the adapting populations. Earlier work uncovered that one of the adapting populations gained the ability to utilize citrate as a sole carbon source. We showed that in addition to this population, three other lineages evolving in parallel began to rely on citrate as a chelator of iron for optimal growth on glucose. This specialization seemed to have occurred through loss of function, most consistent with the accumulation of mutations in iron transport genes that were obviated by abundant citrate.;In Chapter 3 we examined changes in fitness of the evolving populations on carbon sources other than the glucose on which they adapted. We demonstrated that declines in performance were much less widespread than suggested by previous results, and surprisingly accompanied by improvements on a variety of substrates. Strains with higher mutation rate exhibited significantly more declines, and these were ameliorated by growth at lower temperature. These findings suggested that specialization does not mainly result as a consequence of adaptive tradeoffs, but rather due to the gradual accumulation of disabling mutations in unused portions of the genome.;In Chapter 4 we tested the ability of Flux Balance Analysis models to predict evolved changes in central metabolism. We measured metabolic fluxes for evolved populations from the Lenski experiment, and compared them along with datasets from two other experiments to flux predictions. We found that improved growth largely derived from increased rate of substrate use. Flux predictions were more accurate for two experiments initiated with relatively sub-optimal ancestors, whereas ancestors near the optimum tended to move away from predictions over experimental evolution. |
