HETEROLOGOUS EXPRESSION OF HUMAN MEMBRANE PROTEIN DRUG TARGETS AND THE X-RAY CRYSTALLOGRAPHIC DETERMINATION OF THE HUMAN AQUAPORIN 4 STRUCTURE

Membrane proteins comprise 60% of all known drug targets. They perform essential processes in the cell by functioning as receptors, transporters, or channels, controlling the flow of information and materials between the cell and its environment. In any given organism, about 1/3 of the genome encodes membrane proteins, but even given the clinical importance, the three-dimension structures of membrane proteins are scarce at best. As of June 2009, there are a total of ~58000 structures in the Protein Data Bank, but only 473 are from membrane proteins, and out of those, only 193 are unique structures, and out of those, only 6 structures are human integral membrane proteins. Factors that contributed to the paucity of membrane protein structures include difficulty in large quantity heterologous expression, purification and stabilization of the protein molecules in detergent, and forming crystals that diffract to high resolution relevant for biochemical studies and structure-based drug design. We have tackled these problems on several fronts. First, the Pichia pastoris expression system was selected for heterologous expression of human membrane proteins. We have optimized the expression system for membrane protein expression, and from that system, we have obtained large quantity of human aquaporin 4 (hAQP4) as well as human GPCRs and acetylcholine receptors. In order to obtain the high resolution crystal structure of hAQP4, we have characterized its protein-detergent complex using the Tetra Detector Array (Viscotek Corp.) and found ways to remove the flexible termini of the protein to improve diffraction resolution. We have also developed a way for academic laboratories to economically codon optimize and synthesize genes "in-house." Codon optimization is an important part of the expression optimization of eukaryotic membrane proteins due to the incompatibility of the codon usage in the gene of interest and the tRNA population of the expression host. By intelligently selecting the codons that are favored by the expression host, one can improve the expression significantly