Cholesterol oxidase is a water-soluble, interfacial enzyme. It is a flavoenzyme that catalyzes two reactions in one active site: the oxidation of cholesterol to cholest-5-en-3-one followed by isomerization to cholest-4-en-3-one. Cholesterol oxidase is produced by a variety of microorganisms to utilize cholesterol as their energy source. In addition, some pathogenic bacteria might require cholesterol oxidase for host macrophage infection. The enzyme has been widely used in serum cholesterol assays; it was also one of the early tools used to probe the presence of lipid rafts in cell membranes. In boll weevils, the midgut epithelial membrane is disrupted by cholesterol oxidase activity, which led to investigation of its use as a pesticidal agent.

Since cholesterol oxidase is active at the surface of cell membranes, a detailed understanding of the structure of the enzyme-membrane complex was required. This understanding would guide optimization of the use of the enzyme, as well as help investigate its role in pathogenicity. From a kinetic standpoint, cholesterol oxidase must associate with lipid bilayers and bind cholesterol directly from the membrane. This thesis work centered on determining the orientation of cholesterol oxidase from Streptomyces with respect to the membrane surface. The structure was probed using site-selective conjugation of fluorescent probes and fluorescence quenching measurements. In combination with high-resolution X-ray crystal structures of the enzyme in the absence of lipid, the experiments suggest that residues 80, 154, 274 and 333 are not in membrane contact upon binding to the lipid in the absence of the substrate. Under conditions in which the enzyme activity is negligible, these residues are not in membrane contact in the presence of cholesterol.

In addition, a mass spectrometry-based method, which involves an isotope tag, was proposed. L80C mutant ChoA was used to optimize protease digestion and thiol labeling with iodoacetamide (IAM) and N-ethylmaleimide (NEM). Peptide fragments containing IAM or NEM labeled residue 80 were detected by MALDI-TOF. A structure for anisotope-coded mass tag was proposed. This method will be usable to determine a membrane bound model for ChoA, and should be a general method for studying other interfacial proteins.

Atomic resolution structures of a double mutant of cholesterol oxidase in the presence and absence of an alcohol substrate analog, glycerol, were solved by Vrielink and her coworkers. Four distinct populations of glycerol in the active site were observed in the crystal structures of the enzyme-bound glycerol complex, including a structure with a covalent adduct between glycerol and the FAD cofactor. Two geometries of the FAD cofactor, planar and bent, were shown in the glycerol bound structures. Activity assays with glycerol indicate that in solution glycerol is not a substrate of ChoA. Results of enzyme kinetics suggest that glycerol has a very weak affinity for the enzyme and will only bind in the absence of competitive ligands. The distortion of FAD geometry is thought to be caused by the movement of the aromatic triad of tyrosine107, phenylalanine444 and tyrosine446, upon the binding of glycerol in the active site.