— —   Bacteriorhodopsin -Helices in Lipid Settings: Insights for Structure Prediction THOMAS B. WOOLF Department of Physiology and of Biophysics and Biophysical Chemistry, Johns Hopkins University, School of Medicine, 725 N. Wolfe St., Baltimore, Maryland 21205 Received 29 September 1997; accepted 13 March 1998 ABSTRACT: The two-stage model of membrane protein folding predicts that isolated transmembrane -helices form stably in the bilayer before coming together to form the fully functional protein. Insight into the molecular implications of this model are possible with detailed molecular dynamics calculations. Thirty molecular dynamics simulations of both individual and pairs of -helices from bacteriorhodopsin were calculated with the CHARMm program. This data base will continue to grow and expand. Already, differences between identical helices in different media and different helices in the same media have been found. The current results are summarized in this contribution.  1998 John Wiley & Sons, Inc. Int J Quant Chem 69: 105116, 1998 Key words: bacteriorhodopsin; protein: lipid interactions; -helices; molecular dynamics; membrane protein tertiary structure prediction Introduction nderstanding of the environment for mem- U brane proteins is currently limited relative to the aqueous setting of globular proteins. This is unfortunate, since the membrane bilayer and the proteins embedded within it are essential for cellu- lar function. For example, many important cellular signals are first recognized through G-protein cou- Contract grant sponsors: Bard Foundation; Department of Physiology, Johns Hopkins University. pled receptors that are believed to be heptahelical  -helical membrane proteins 1 . It has been argued that -helical membrane protein tertiary structure prediction should be eas-  ier than the situation for globular proteins 2, because helical backbone hydrogen bonding stabi- lizes helices as folding elements in the bilayer. By contrast -strands, although also found in mem- brane proteins, must form some sort of closed structure, such as a barrel, before all backbone hydrogen bonds are satisfied. The relatively rigid -helices could then transfer as a structural unit into the bilayer and assemble into a functional protein. This is the two-stage model of membrane ( ) International Journal of Quantum Chemistry, Vol. 69, 105 116 1998  1998 John Wiley & Sons, Inc. CCC 0020-7608 / 98 / 010105-12