Protein folding and conformational search improvements Molecular simulations of protein folding at native conditions with atomic details can elucidate protein-folding events. To overcome the limitation in time scale, an enhanced simulation method, the self-guided molecular/Langevin dynamics (SGMD/SGLD) method was develope to boost systematic motion in molecular systems. This approach is capable of addressing slow events like crystallization, peptide folding, and molecular capturing. It allows us to directly access reversible protein folding events. Protein folding in the cell is not always a spontaneous process due to unproductive pathways of misfolding and aggregation. Chaperonin molecules prevent such off-pathway reactions and promote protein folding through spectacular ATP-driven cycles of binding and releasing substrate proteins. Protein folding in a confined environment. Coarse grained Langevin dynamics was used to examine the stability of different helix-forming sequences confined to a carbon nanotube. Several factors, including sequence, solvent conditions, strength (λ) of nanotube-peptide interactions, and the nanotube diameter (D), determine confinement-induced stability of helices. The results derived are in agreement with polymer theory. There is a strong sequence dependence as the strength of the λincreases. For an amphiphilic sequence, the helical stability increases with λ, whereas for polyalanine the diagram of states is a complex function of λand D. Decreasing the size of the hydrophobic patch lining the nanotube, which mimics the chemical heterogeneity of the ribosome tunnel, increases the helical stability of the sequence. Our results provide a framework for interpreting the structure formation of peptides in the ribosome tunnel and transport of biopolymers through nanotubes. Protein switches. Many proteins involved in cellular signal transduction switch between inactive and active conformations upon binding or release of ligands. The switching between these conformations of nitrogen regulatory protein C(NtrC) is one of the key steps in bacterial nitrogen metabolism. The structures of the active and inactive forms of NtrC have recently been determined through NMR spectroscopy. NtrC becomes active through phosphorylation of an active site aspartate. We have used multiple SGLD simulations to study the dynamics of the active and the inactive conformation of NtrC. Calculations of pKa values with the finite difference Poisson-Boltzmann method have suggested that phosphorylation can change the pKa value of a His residue that is close to the active site, while SGLD simulations have suggested that phosphorylation combined with charging of this His can stabilize the ensemble of the active form structures. Furthermore the simulations suggested that the regulatory helix may change its conformation through a partial unfolding mechanism. This mechanism is at the core of cellular regulatory mechanisms. Coupling between ionization of internal groups and protein dynamics. Ionization of internal groups in proteins is at the core of energy transduction in biological systems. The ionization can trigger conformational rearrangements, which in turn can change the pKa values of ionizable groups. To study how the protein responds to the ionization of internal groups, we have performed a series of SGLD simulations of variants of staphylococcal nuclease (SN) in which ionizable groups are buried in the protein core. The work was performed in collaboration with Prof. Bertrand Garcia-Moreno at Johns Hopkins University, who has experimentally characterized a large number of variants of SN. The simulations have shown that the protein can respond to charging of internal groups through large scale reorganization of the backbone. They also suggested that such charging events can trigger increased hydration of these internal groups. This study emphasizes the difficulties in calculations of pKa values of internal groups: those of a simultaneous description of changes in hydration patterns and possibly large scale conformational rearrangements. Exploring myosin II efficiency using the Langevin Network Model (LNM). The Langevin Network Model was developed and used to study the power-stroke efficiency of scallop myosin II. Previous normal mode studies of this protein did not consider the effect of solvent friction on behavior. The Langevin Network Model improves upon these by combining the Elastic Network Model (ENM) with Langevin Modes to create a method to calculate protein vibrations in simulated solvent using a relatively small amount of computer time. This new method was used to study pre- and post-power stroke structures of scallop myosin II, along with chicken lysozyme and a 4-bead test system. The Rotne-Prager tensor was used to effect solvent friction for all systems. By comparing the Langevin modes with the frictionless ENM modes, this study showed that the critical power-stroke modes are relatively unperturbed by friction when compared with other modes of the myosin structures. This result can be used as a first step toward examining the structural efficiency of myosin II. Structure and Reaction Mechanisms of Boronic Acids. In collaboration with Charles W. Bock (Philadelphia University), Tony D. James (Bath University, UK), and George D. Markham (FoxChase Cancer Center), the chemical structure and reaction mechanisms of various boronic, borinic, and orthoboric acids have been investigated. Bortezomib (formerly known as PS-341, and marketed as VELCADE) is a novel dipeptidyl boronic acid inhibitor of the S26 proteasome that was recently approved by the FDA for the treatment of patients with relapsed multiple myeloma where the disease is refractory to conventional therapies. Ab initio calculations were performed in several LCB studies on boronic acids (BAs): the structural characterization of BA monomers and dimers;the nature of boron bonding was described (specifically dative, hydrogen, and multiple bonding);and proto- and oxidative-deboronation mechanisms in the solution phase were elucidated. These results provide much needed insight into boron chemistry that will help guide future QM/MM investigations on boronic acid inhibition of proteasomes that show significant promise in cancer therapy. Investigating the effects of a six residue connecting peptide on insulin stability. Insulin is a 51 residue dimer (Chain A 21 residues and Chain B 30 residues) that regulates blood glucose level. Recently a study134 showed improved stability of insulin where a six residue peptide (GGGPRR) was used to connect the C-terminus of the B-chain to the N-terminus of the A-chain. NMR experiments showed that resulting single chain insulin analog (SCI-57) has improved stability over wild type insulin with similar binding affinity. A more refined structure with less fluctuations was observed but the extent of the effects of the connecting peptide is not known. We used molecular dynamics simulations in explicit solvent to investigate the effects of this connecting peptide on insulin using CHARMM. SCI-57 and its two chain analog (2CA) were simulated at various temperatures (300K, 325K, 340K and 350K). Both chains maintained their native conformations at 300K with no significant NOE violations. 2CA shows more fluctuations at the terminal residues compared to SCI-57. At elevated temperatures SCI-57 still maintains its conformation where 2CA starts losing some of its secondary structure elements. For enhanced sampling SGLD simulations are being also performed that show similar trends as seen in the high temperature simulations for both systems. Specific interactions between both chains and the connecting peptide are still being investigated. This project has implications for therapy design.
Showing the most recent 10 out of 40 publications
