ABSTRACT
Of the more than 30,000 protein structures obtained by x-ray crystallography and deposited in the protein databank (PDB),1 less than 2% represent membrane proteins,2 (161 entries as of February 2005). The vast majority of these depict bacterial, mitochondrial, and chloroplast proteins. A closer look reveals that only a handful of those structures are derived from proteins produced in recombinant form.3 This is in sharp contrast to soluble protein structures, where recombinant production is the standard. In fact, not a single cell-surface protein structure of eukaryotic, let alone human, cells has been obtained through recombinant production. This shortage
of structures is not only a concern to structural biologists but even more to medicinal chemists who would acclaim experimental structures of human membrane proteins for rational drug design.
