Knowledge Based Structure Prediction of the Light-Harvesting Complex II of Rhodospirillum molischianum

Abstract. We illustrate in this article how one proceeds to predict the struc-ture of integral membrane proteins using a combined approach in which mole-cular dynamics simulations and energy minimization are performed based on structural information from conventional structure prediction methods and ex-perimental constraints derived from biochemical and spectroscopical data. We focus here on the prediction of the structure of the light-harvesting complex II (LH{II) of Rhodospirillum molischianum, an integral membrane protein of 16 polypeptides aggregating and binding to 24 bacteriochlorophyll a’s and 12 lycopenes. Hydropathy analysis was performed to identify the putative trans-membrane segments. Multiple sequence alignment propensity analyses further pinpointed the exact sites of the 20 residue long transmembrane segment and the four residue long terminal sequence at both ends, which were independently veri ed and improved by homology modeling. A consensus assignment for sec-ondary structure was derived from a combination of all the prediction meth-ods used. The three-dimensional structures for the - and the -apoprotein were built by comparative modeling. The resulting tertiary structures were combined into an dimer pair with bacteriochlorophyll a’s attached under constraints provided by site directed mutagenesis and FT Resonance Raman spectra, as well as by conservation of residues. The dimer pairs were then aggregated into a quaternary structure through molecular dynamics simula-tions and energy minimization. The structure of LH{II, so determined, was an octamer of heterodimers forming a ring with a diameter of 70 A. We discuss how the resulting structure may be used to solve the phase problem in X-ray crystallography in a procedure called molecular replacement