Protein Dynamics and Conformational Changes

Proteins perform a remarkably large number of functions in living species ranging from regulation and control of various compounds at the cellular level, to biochemical transformations by enzymes as well as forming structural tissue such as muscles, hair and the cytoskeleton. The biological function of a given protein is largely dependent on its 3D structure and especially the dynamical properties of this. Recent advances in bio-structural sciences as protein crystallography and NMR techniques have resulted in more than 45.000 known 3D structures of various proteins, stored at the Protein Data Bank. This enormous amount of information has paved an avenue for studying the dynamic behaviour of proteins where more than one stable structure is found. It has become more and more obvious that many proteins function by conformational changes between stable structures, and we are deeply involved in such studies using classical and advanced MD simulation and analysis techniques.

• Membrane Transporters: During the last couple of years we have worked on modeling important properties of selected members of the Neurotransmitter Transporter family. Specifically we have been involved in modeling the binding of the endogenous ligand, serotonin, and selected antidepressants to the serotonin transporter (hSERT) and the binding pathway for the substrate leucine in the Leucine Transporter. New projects related to membrane transporters involves: i) use of advanced MD techniques, as e.g. steered MD, coarse-grained force fields and accelerated MD, ii) use of advanced analytical techniques as principal component analysis (PCA) and normal-mode analysis (NMA) for identification of important collective motions and iii) construction of a putative inward conformation by combining various methods. Proteins to be included are LeuT, hSERT, hDAT and the aspartate transporter, Glt_^Ph.
• Antimicrobial Peptides: The last class of membrane proteins that we are currently studying concerns the antimicrobial peptides. We have successfully shown how the peptide alamethicin can self assemble in a membrane bilayer by using coarse-grained MD simulations. New studies focus on other similar peptides. 
• Estrogen Receptor: We have performed exhaustive MD simulations of the estrogen receptor (ER) to learn more about the dynamics of the large conformational change that is observed upon activation of the apo-protein to a transcriptional active form. From the studies we were able to identify stable intermediate structures which may be important for the observed mal-functioning of ER, e.g. in the presence of endocrine disrupting chemicals. Current studies focus on outlining the ligand binding pathway as well as computing changes in free energy related to the different conformational changes.
• Fibrillating Peptides/Proteins: Essentially all proteins are able to fold into two radically different types of structures: the biologically active folded state and the fibrillated state, in which the protein molecules align ?-strands to form long thin fibrils or dense plaques. For most proteins, fibrillation occurs under extreme conditions which are not found in vivo. However, a subset of proteins are able to fibrillate under physiological conditions, and this gives rise to deposition diseases such as Alzheimer's and Parkinson's Disease. We carry out model studies of the fibrillation process using MD techniques focusing on glucagon, transthyretin and small model systems.
• Development of New MD Methodologies: To combat the apparent time scale problem with classical MD techniques, being on the nanosecond scale, compared to the laboratory scale of micro- and milliseconds we are engaged in developing methods for speeding up simulations. We are exploring the use of coarse-grain methods, as well as advanced simulation protocols where the MD integration scheme is interfaced with “smart” stepping protocols to force a certain conformational/structural change to happen.