TWO APPROACHES TO THE STUDY OF PROTEIN INTERACTIONS WITH SMALL MOLECULES: (A) STRUCTURAL ANALYSIS OF PYRIDOXAL L-PHOSPHATE BINDING ENZYMES (B) PURIFICATION, RECONSTITUTION, AND DRUG-BINDING CAPABILITIES OF THE PLASMODIUM FALCIPARUM MULTIDRUG RESISTANCE PROTEIN (PfMDR1)

Abstract

Structural genomics initiatives are producing new protein structures at a rate that will soon

exceed the rate at which biochemical experiments can validate their properties. In the near future,

there will be a need for faster methods of accurate identification of proteins and prediction of their

functions. Attempts to address this problem by functional prediction methods based only on protein

sequences are often limited in scope, as protein sequences diverge at very fast rates. In cases of highly

diverged proteins, structural approaches have proven more promising, as it is a well-known fact that

structures retain over longer evolutionary timescales than sequences alone. Hence, a combined

approach using both sequence- and structure-based methods seems ideal for function characterizations.

An estimated 4% of all enzymes utilize pyridoxal- 5&rsquo-phosphate (PLP) as a cofactor (1). The

majority of PLP-dependent enzymes (PLP-DEs) are involved amino acid biosynthesis, but also catalyze a

variety of other biological reactions. Nevertheless, PLP is surprisingly unique in its interaction with each

enzyme. PLP forms a conserved covalent linkage from its aldehyde group with the epsilon-amino group of a

lysine residue in the enzyme's binding pocket. The imine linkage between cofactor and enzyme is the

only commonality among all these enzymes. In some enzymes, PLP-DEs have evolved to accommodate

two distinct substrates-- amino group donors and amino group acceptors-- within the same binding

pocket, a unique feature for enzymes which typically can only recognize a single substrate.

Furthermore, PLP-DE can bind and utilize their cofactor in a stereoselective way (2).