Doctoral Dissertations

Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Biochemistry and Cellular and Molecular Biology

Major Professor

Elizabeth E. Howell

Committee Members

Barry Bruce, Hong Guo, Steve J. Kennel


Dihydrofolate reductases (DHFR) are important, ubiquitous enzymes catalyzing the hydride transfer from NADPH to dihydrofolate and producing the tetrahydrofolate intermediate that is essential for many metabolic processes, particularly for its role in DNA synthesis. R67 DHFR is a plasmid encoded enzyme that confers resistance to the antibiotic drug trimethoprim. This enzyme is active as a homotetramer. The active site pore possesses 222 symmetry as a result. This symmetry gives rise to a different mechanism by which it binds its ligands. There are potentially four symmetry related binding sites, but various studies have indicated only three possible combinations that include: two dihydrofolate molecules or two NADPH molecules or one substrate plus one cofactor. The latter is the productive ternary complex. Further mutagenesis and crystallography studies have not yet painted a clear picture of ligand binding. Thus, a thorough structure-activity relationship study for R67 DFHR has been hindered. Therefore, binding studies were pursued with the use of various analogs and fragments of cofactor and substrate to provide a basis for protein-ligand interactions as well as ligand-ligand interactions. Isothermal titration calorimetry was used as well as Ki studies. One of the goals was to determine the role of any substituents involved in formation of the binary and ternary complexes. In general, the tails of cofactor and substrate are important for binding as they each possess negative charge for interacting with K32 near the outer pore edge. The cofactor tail consists of the charged phosphate groups, adenine ring and ribosyl moiety. The 2' phosphate group was shown to be important for binding as well as the pyrophosphate bridge. The substrate tail consists of the charged pABA-Glu moiety where glutamate's carboxylate groups are important for binding. For both ligands, a large portion of the measured enthalpy of binding arose from these charged groups as their removal resulted in reduced binding affinities. Thus, connectivity effects are proposed to be important for ligand binding. The binding enthalpies arising from these charged groups as well as from other parts of each ligand's structure contributed to the preferred binding order between cofactor and substrate where enthalpy/entropy compensation plots showed an enthalpy-driven binding for the first site by the cofactor and for the second site by the substrate. In addition to the tail, another important group involved in the formation of both complexes was the carboxamide group off the nicotinamide ring head of cofactor and the analogous atom arrangement at the pteridine ring's N3 and O4 positions of substrate. The O4 atom was found particularly to be important for the formation of the productive ternary complex. From this study, a picture has emerged for the role of symmetry in R67 DHFR in binding of its ligands.

The binding of ligands to an enzyme or the protein folding mechanism involves a number of weak electrostatic types of forces. The cation-p interaction has recently been included as a number of theoretical, experimental and biological evidence was gained on this type of interaction between a cation and the p electron cloud face of aromatic residues. The resulting quadrupole moment of an aromatic ring, such as in benzene, gives rise to a density of partial negative charge above and below the ring plane. Arising from these calculations, an observation was noted on the occurrence of a partial positive charge around the aromatic ring edge. Thus, a question was asked if the ring edge of aromatic residues could interact with negatively charged anions such as with carboxylate groups off glutamate and aspartate residues in proteins. This type of interaction is termed an anion-quadrupole interaction and thus was investigated by a quantum mechanical approach not only to establish this effect but also to illustrate its importance in biology. The goal of the quantum mechanical calculations was to set up a self-sufficient theoretical basis for the anion-quadrupole effect. Calculations involved a simpler model for the aromatic-carboxylate group pairing in protein structures - the benzene-formate pair. Calculations were also done using benzene with point charges to provide a reference for the effect on the quadrupole moment of benzene by an approaching, charged anion. The study showed that the anion-quadrupole effect was established for a favorable edge on approach orientation and is the predominate, weak electrostatic interaction with a distance dependency of 1/r3. Thus the anion-quadrupole effect was similar for the electrostatic type of interaction as for cation-p, although not in magnitude, but was different for the polarizability effects with benzene's p electron cloud orbital.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."