Molecular characterization of a prokaryotic polypeptide sequence that catalyzes Au crystal formation

The gold crystal-forming E. coli polypeptide sequence, MHGKTQATSGTIQS, is one of several polypeptide sequences that interacts with gold interfaces and catalyzes the formation of Au crystals in solution, with nucleated Au crystals preferentially featuring the (111) interface. To date, there have been no experimental studies which explore the structure of E. coli-expressed gold binding proteins or the binding of Au(III) ions by these polypeptides. In this present report, multidisciplinary approaches were applied to the 42-AA gold binding protein-1 (GBP-1/42) and to a model polypeptide representing the 14-AA integral repeat of this protein (GBP-1/14). CD and NMR spectroscopy indicate that neither the integral repeat nor the GBP-1 protein adopt folded structures in the apo form or in the presence of Au(III) ions; the integral repeat adopts a random coil-extended structure conformation [i.e., (MHGKTQA)random coil-(TSGTIQS)extended] and the GBP-1 protein appears to be similarly structured. These features are inconsistent with a templating structure. Mass spectrometry experiments indicate that the integral repeat binds up to two Au(III) ions per polypeptide molecule, and ¹H NMR ROESY experiments pinpoint the interaction of Au(III) within two sites: the -QAT- region of the integral repeat MHGKTQATSGTIQS sequence, and, at the negatively charged C-terminus of this sequence. Collectively, our findings support the hypothesis that GBP-1 does not catalyze Au crystal formation via a templating mechanism; rather, the open, unfolded structure of this protein, combined with the presence of accessible proton donor/acceptor amino acids (Ser, Thr, Lys, Gln, His) most likely play a role in Au crystal formation in solution and may also explain the interactive nature of this polypeptide with Au interfaces.