Open Access
Li, Hui
Graduate Program:
Integrative Biosciences
Doctor of Philosophy
Document Type:
Date of Defense:
October 03, 2005
Committee Members:
  • Stephen James Benkovic, Committee Chair
  • Costas D Maranas, Committee Member
  • Squire J Booker, Committee Member
  • Ming Tien, Committee Member
  • protein evolution
  • protein engineering
  • purine biosynthetic pathway
  • domain swapping
  • (β/α)8 barrel protein
One of the ultimate goals of protein engineering is the de novo design of novel proteins with desired activities and properties. However, our current knowledge of protein structures and functions are far from complete to achieve this goal. On the other hand, nature has successfully evolved an enormous number of proteins with novel functions for their hosts to fit the ever changing environments, and naturally occurring proteins present the most diverse and complicated information about protein structure-function relationships. Studying the evolutionary-related proteins with structural and functional homology will provide not only the detailed information about protein structure-function relationships, but also the insights to the strategies that nature had adopted for protein evolution. Here, we studied two pairs of enzymes with significant homology on their structures, reaction mechanisms, and the active site architectures. In our attempt to interconvert the enzymatic activities between two members in the same pair, rational methods, such as site-directed mutagenesis and rational domain swapping, were applied first on the basis of our current understanding of each protein. Further, the additional sequence spaces were explored using combinatorial methods, such as ITCHY, random mutagenesis, and DNA shuffling, to identify their potential roles in terms of protein structures and functions. The first pair of enzymes we chose are Escherichia coli purT-encoded glycinamide ribonucleotide (GAR) transformylase (PurT) and Escherichia coli N5-carboxylaminoimidazole ribonucleotide (N5-CAIR) synthetase (PurK). While both enzymes are involved in the de novo purine biosynthesis, PurT catalyzes the third reaction of the purine biosynthetic pathway, the conversion of GAR, ATP and formate to formyl GAR, ADP and inorganic phosphate (Pi); and PurK catalyzes the sixth reaction of the purine biosynthetic pathway, the conversion of 5-aminoimidazole ribonucleotide (AIR), ATP and bicarbonate to N5-CAIR, ADP and Pi. The effort to interconvert the enzymatic activities between PurT and PurK suggested that these two enzymes might evolve through domain swapping. Several crucial structural elements for catalysis were also identified in each protein, which provides value information for protein structure-function relationships. The second pair of enzymes are Escherichia coli N-acetylneuraminate lyase (NAL) and Escherichia coli dihydrodipicolinate synthase (DHDPS), two (β/α)8 barrel proteins. NAL catalyzes the degradation of N-acetylneuminate (NANA) to N-acetylmannosamine (ManNAc) and pyruvate. DHDPS catalyzes the branch-point reaction of the lysine biosynthetic pathway in plants and microbes: the condensation of L-aspartate-β-semialdehyde and pyruvate to dihydrodipicolinate (DHDP). Both enzymes were observed to be able to catalyze each other’s reaction, and this functional promiscuity between NAL and DHDPS is considered as a strong statement that they are evolutionary-related. A possible evolutionary scheme from NAL to DHDPS through divergent path was further approved by the attempt to interconvert the enzymatic activities between NAL and DHDPS. A conserved Arg residue in DHDPS was identified to be crucial for the DHDPS activity. Several DHDPS mutants with an enhanced NAL activity were also identified using combinatorial methods.