Investigation of the Structure, Function, and Regulation of Quinolinate Synthase: the Iron-Sulfur Cluster Enzyme Involved in Prokaryotic NAD+ Biosynthesis

Open Access
- Author:
- Saunders, Allison Hoover
- Graduate Program:
- Biochemistry, Microbiology, and Molecular Biology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 18, 2011
- Committee Members:
- Squire J Booker, Dissertation Advisor/Co-Advisor
Squire J Booker, Committee Chair/Co-Chair
Joseph M Bollinger Jr., Committee Member
Carsten Krebs, Committee Member
Ming Tien, Committee Member
Philip C. Bevilacqua, Committee Member - Keywords:
- NadA
iron-sulfur cluster
NAD
enzyme - Abstract:
- Nicotinamide adenine dinucleotide (NAD+) is an essential biological cofactor known primarily for its role in oxidation–reduction reactions, but is also involved in ADP ribosylation, adenylation and histone deacetylation. The key intermediate in NAD+ biosynthesis is quinolinic acid. The pathway to form quinolinic acid differs between most eukaryotes and prokaryotes, making the prokaryotic pathway a possible target for design of antibacterial agents. In eukaryotes, quinolinic acid is formed via the degradation of L-tryptophan by a series of enzymes, of which three require molecular oxygen for catalysis. In prokaryotes, quinolinic acid is synthesized via the action of two enzymes, L-aspartate oxidase (NadB) and quinolinate synthase (NadA). NadB oxidizes L-aspartate to form iminoaspartate by a two–electron oxidation utilizing the cofactor, flavin adenine dinucleotide (FAD). The iminoaspartate is then condensed with dihydroxyacetone phosphate by NadA to form quinolinic acid. In contrast to the eukaryotic pathway, the prokaryotic pathway can function under strictly anaerobic conditions. NadA from Escherichia coli was initially proposed to contain an iron–sulfur (Fe/S) cluster, due to its instability in the presence of molecular oxygen and the presence of a CysXXCysXXCys motif, which contains the cysteines that commonly ligate a [4Fe-4S] cluster. This enzyme was later confirmed to contain one [4Fe-4S] that is essential for catalysis. Through further investigation of the cysteine variants of NadA, we have demonstrated that the cluster is ligated by three cysteines that are conserved in all quinolinate synthases, of which only one lies in the CysXXCysXXCys motif. Interestingly, we found that the other two cysteines in the CysXXCys motif form a redox–active disulfide bond, which regulates the rate of formation of quinolinic acid. Disulfide bond formation is regulated by thioredoxin, a disulfide bond reductase, and we have determined the redox potential of NadA to be − 264 mV using the thioredoxin redox couple. Despite the lack of a CysXXCys motif, we have demonstrated that NadA from Mycobacterium tuberculosis is also regulated by disulfide bond formation. Our results suggest that the redox potential is higher than that of the E. coli enzyme because optimal activity is obtained with DsbA, a periplasmic oxidoreductase more oxidizing than thioredoxin. Site–directed mutagaenesis studies demonstrate that the disulfide bond is formed between two cysteines in a CysXCys motif. The role of the disulfide bond in vivo has also been investigated using an E. coli strain in which the nadA gene has been replaced with a kanamaycin cassette (ΔnadA). A variant with a single substitution of the disulfide bond forming cysteine is unable to complement the ΔnadA strain unless the level of expression of the variant is increased and the strain is grown in the absence of oxygen. These results suggest that the disulfide bond may play a role in protecting the Fe/S cluster from oxidative stress. The only published crystal structure of NadA is from Pyrococcus horikoshii and is lacking the essential Fe/S cluster. Utilizing the apo–structure bound to the substrate mimic, malate, and the knowledge of the three conserved cysteines ligating the Fe/S cluster, we have modeled the Fe/S cluster and quinolinic acid precursor into the structure. Using this model, we identified eight conserved active site residues positioned to interact with substrate and changed each of these residues by site–directed mutagenesis. We then characterized the resulting variant proteins for their ability to form both quinolinic acid and inorganic phosphate, a second product of the reaction. Through these studies we have identified three variants that are completely inactive for formation of both products and five others with varying reaction rates. In future experiments, these variants will be used to build up intermediates to determine the order of events during the condensation reaction.