The general research aims of my group are to use molecular modeling and bioinformatics to analyze structure, function, and molecular evolution of membrane proteins. Membrane proteins are one of the most important classes of proteins. They comprise about 30% of most genomes and are involved in many biological processes. They are especially important in biomedical research because most targets of current pharmaceutical projects are membrane proteins. Unfortunately, membrane protein structures are difficult to determine experimentally, and most that are determined come from prokaryotes. We fill some of this structural void by developing computational methods of analyzing sequences and developing structural models of membrane proteins. We use computational analyses to do the following:1)Address questions that are not answered by crystal structures. 2)Assist in understanding similarities and differences among homologous proteins.3)Relate structural and sequence information to functional properties.4)Assist in the design and interpretation of experimental studies.Our current projects can be classified into three areas: 1) models of the structure and gating mechanisms of potassium (K+) channels and their relatives; 2) models of the structure and gating mechanisms of the large mechanosensitive channel, MscL; and 3) development of methods to analyze sequences and construct structural models of membrane proteins. Project 1. Models of K+ Channels and Their RelativesPotassium channels and related protein comprise the third largest superfamily of human genes. These proteins are found in almost all cells from bacteria on up. This category of membrane proteins contains several diverse superfamilies of channels including Na+, Ca2+, cyclic nucleotide-gated, TRP and its homologs, glutamate-activated, and Ca2+ release channels plus some K+ symporters and transporters. The smallest of these proteins are 2TM K+ channels that have four identical subunits; each of which has only two transmembrane helices, M1 and M2. A 'P' hairpin segment that spans only the outer half of the transmembrane region is located between M1 and M2. The P segment determines the selectivity of the channel. 6TM K+ channels are more complex, with each alpha subunit having four additional transmembrane segments, S1-S4, that precede the pore-forming S5-P-S6 motif (analogous to the M1-P-M2 motif of 2TM channels) and that forms a voltage-sensing domain in voltage-gated channels. Voltage-gated Ca2+ and Na+ channels have only one alpha subunit; however, it contains four homologous 6TM motifs. One of our goals is to develop structural models of the transmembrane region of at least one member of each major family of K+ channel related proteins. The importance of understanding the structure and functional mechanisms of K+ channels was recognized last year by the awarding of the Nobel Prize in Chemistry to Roderick MacKinnon for his work in solving the crystal structures of two 2TM (KcsA and MthK) and one 6TM (KvAP) K+ channels. An additional 2TM channel structure, KirBac1.1, has also been determined recently. The effort to determine the KirBac1.1 structure was stimulation by our identification of this bacterial homolog of eukaryotic inward rectifying K+ channels. All of these structures are from prokaryotes. We are utilizing these crystallographic data in developing structural models of the gating mechanisms of both the crystallized prokaryotic proteins and of some of their eukaryotic homologs that have been studied extensively and that are important drug targets. The KvAP structure presents a particularly interesting molecular modeling challenge. It is difficult to reconcile the crystal structure of the complete KvAP channel protein and the paddle model of the voltage-dependent gating that MacKinnon's group developed based on this structure with many experimental results and with basic principles of membrane protein energetics. We suspect that the voltage-sensing domain (S1-S4) of this structure is grossly distorted, but that a second crystal structure of an isolated voltage-sensing domain has a native open conformation. We have developed alternative models of the open KvAP channel's structure by attaching the isolated voltage-sensing domain crystal structure to the pore domain (S5-P-S6) from the crystal structure of the complete protein. Using this model as a starting point, we have developed models of resting and intermediate conformations. Our models have the 'traditional' transmembrane topology in which each of the S1-S4 segments transverses the entire transmembrane region in all conformations. Much of the movement of S4 occurs via the helical screw mechanism. We were the first group to propose this topology and gating mechanism shortly after the first voltage-gated Na+ channel sequence was determined in the mid '80's. Molecular dynamics simulations that we have performed of the protein embedded in a lipid bilayer indicate that our models are substantially more stable than is the crystal structure of the complete KvAP protein. Our models of the KvAP channel were constrained to be consistent with experimental results from other Kv channels, primarily the Shaker channel. These constraints are complicated by the facts that these proteins are evolutionarily distant and substantial data have been obtained from Shaker residue positions that are deleted or that may be in a different conformation in the KvAP sequence. To better address these issues, we have developed models of the Shaker channel that are similar to those of our KvAP models, but with some important structural differences. These adjustments make the Shaker models consistent with many experimental observations that are inconsistent with MacKinnon's paddle model of gating. We are using the general structure of our models of the KvAP and Shaker channels in different conformations to model the general backbone folding and gating mechanism of another important class of K+ channels, the Herg channels. Our alignment of the Herg, KvAP, and Shaker sequences is based upon a very large multisequence alignment of all voltage-gated and cyclic nucleotide-gated 6TM channels and upon analyses of correlated mutations among different protein families. We use results of mutagenesis experiments on Herg and closely related EAG channels, results of NMR studies, plus basic modeling principles, to adjust features of these models and to model regions, such as the long S5-P loop in Herg channels for which analogous residues are deleted in KvAP. We are also modeling how the BeKm-1 scorpion toxin binds in the outer vestibule of Herg. The Herg project is being performed in collaboration with Dr. Gea-Ny Tseng, whose lab performs mutagenesis experiments that constrain and test our models. We are also developing models for the structure and gating mechanism of a prokaryotic sodium selective channel, NaChBac. NaChBac was selected because several labs are attempting to solve its structure experimentally. Also, its sequence is intermediate between those of the voltage-gated K+ channels we have modeled and eukaryotic Ca2+ and Na+ channels. Thus models of NaChBac are a logical first step in our efforts to model these important superfamilies of Na+ and Ca2+ channels.We are performing molecular dynamics simulations of all of our membrane protein structures embedded in a lipid bilayer with water on each side. These simulations contain many atoms and are computationally intense. They help us evaluate the feasibility of our structures by comparing the energetically stability of our models to the stability of experimentally determined crystal structures. Project 2: Models of the Mechanosensitive Channel, MscLThis project exemplifies our general approach to modeling the structures and functional mechanisms of membrane proteins. We have modeled the structure of the prokaryote mechanosensitive channel, MscL, as it undergoes a very large
