Dancing to a different tune: adaptive evolution fine-tunes protein dynamics
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
The molecular mechanisms that underpin adaptive evolution are not well understood. This is largely because few studies relate evolved alleles (genotype) with their physiological changes (phenotype), which move a population to better fit its environment (adaptation). The work described in this thesis provides a case study exploring the molecular changes underlying adaptive evolution in a key allosteric enzyme. It builds upon a long-term evolution experiment by Richard Lenksi, where twelve replicate populations of Escherichia coli have adapted in parallel to better fit their low-glucose environment. I focused on the allosteric enzyme pyruvate kinase type 1, since this has been shown to adapt to this environment. First, I used X-ray crystallography to determine a higher resolution structure (2.2 Å) than previously available of the wild-type PK1 enzyme for comparison with the evolved enzymes. I resolved the ambiguous space-group problem that affects these crystals, and demonstrated that the kinetic function of the recombinant enzyme is the same as previously reported. In addition, I propose a new model for allosteric activation: a combination of structural and dynamic analyses determined that the allosteric signal is transferred by a series of dynamic changes between the allosteric site, upon fructose-1,6-bisphosphate binding, and the active site for increased substrate binding. The functional analyses demonstrated that all eight evolved PK1 enzymes have a reduced activity compared to the wild-type PK1 at physiological substrate concentrations. Not only did the evolved PK1 enzymes show a parallel decrease in activity, but they all showed changes to substrate binding affinity and seven of the eight showed an altered allosteric activation mechanism. These results suggest that natural selection has selected for enzymes with a reduced activity by altering the functional mechanism of the evolved enzymes. However, in crystal and in solution structure characterisation determined that all of the evolved PK1 enzymes have maintained the same structural fold as the wild-type PK1. Although the fold is the same, substrate binding promiscuity suggested a change in the flexibility of the enzyme, allowing substrates of different sizes and shapes to bind. Computational and experimental dynamics studies determined that natural selection has selected for reduced activity by altering the dynamics in all of the evolved PK1 enzymes, and it has used altered dynamics to change the allostery of the enzymes. Therefore, this study provides the first example of adaptive evolution fine-tuning protein dynamics to alter allostery. This thesis describes the molecular mechanisms underlying one aspect of adaptation of Escherichia coli to the low-glucose environment in Lenski’s long-term evolution experiment. The adaptive mutations in Escherichia coli’s pyruvate kinase type 1 serve to increase the availability of phosphoenolpyruvate for glucose uptake. From a molecular perspective, natural selection has selected for adaptive amino acid substitutions that produce an enzyme with reduced catalytic activity at low phosphoenolpyruvate concentrations, thus decreasing phosphoenolpyruvate consumption. In addition, the adaptive mutations have altered the enzymes’ affinity for the allosteric activator (fructose- 1,6-bisphosphate), fine-tuning them to match the concentration of fructose-1,6- bisphosphate in the cell at the point of glucose re-introduction. Overall, this work describes the intricate relationship between genetic changes and the resulting phenotype and demonstrates the parallel nature of adaptation for this particular case study. Whereby, parallel changes are mapped from organismal fitness, to enzyme function and to enzyme structure. The dynamic changes, however, are not parallel thus making the prediction of specific changes in adaptive evolution difficult.