On the Thermal Stability of O6-Methylguanine-DNA Methyltransferase from Archaeon Pyrococcus kodakaraensis by Molecular Dynamics Simulations.

We have employed molecular dynamics simulations to analyze the thermal stability of the O6-methylguanine-DNA methyltransferase (MGMT) protein, both hyperthermophilic archaeon Pyrococcus kodakaraensis (Pk-MGMT) and its mesophilic homologue pair, obtained from enterobacterium Escherichia coli (AdaC). This theoretical study was done at three different temperatures: 302, 371, and 450 K. The molecular dynamics has been performed in explicit aqueous solvent during a period of time of 95 ns, including periodic boundary conditions and constant pressure. The same procedure has been used for both proteins, and each simulation has been carried out by triplicate. Hence, we performed 18 simulations. In this way, we have done different analyses to explore the factors that may affect the thermal stability of Pk-MGMT. The structural behavior was analyzed using indicators such as root-mean-square deviation, radius of gyration, solvent-accessible surface area, hydrogen bonds, native contacts, secondary structure, and salt bridge formation. The results showed that when the temperature increases, the global atomic fluctuations increase too, which suggests that both proteins lose thermal stability, but as expected, this fact is highlighted in AdaC. Moreover, the contacts of the native state in AdaC are considerably lower than those found in Pk-MGMT at 450 K. Also, the structural studies showed that conserved and nonconserved salt bridges kept close contacts with the Pk-MGMT protein at high temperatures. These interaction types act as molecular staples and are mainly responsible to provide thermostability to the hyperthermophilic protein.

A Computational Approach to Studying Protein Folding Problems Considering the Crucial Role of the Intracellular Environment.

Intracellular protein folding (PF) is performed in a highly inhomogeneous, crowded, and correlated environment. Due to this inherent complexity, the study and understanding of PF phenomena is a fundamental issue in the field of computational systems biology. In particular, it is important to use a modeled medium that accurately reflects PF in natural systems. In the current study, we present a simulation wherein PF is carried out within an inhomogeneous modeled medium. Simulation resources included a two-dimensional hydrophobic-polar (HP) model, evolutionary algorithms, and the dual site-bond model. The dual site-bond model was used to develop an environment where HP beads could be folded. Our modeled medium included correlation lengths and fractal-like behavior, which were selected according to HP sequence lengths to induce folding in a crowded environment. Analysis of three benchmark HP sequences showed that the modeled inhomogeneous space played an important role in deeper energy folding and obtained better performance and convergence compared with homogeneous environments. Our computational approach also demonstrated that our correlated network provided a better space for PF. Thus, our approach represents a major advancement in PF simulations, not only for folding but also for understanding functional chemical structure and physicochemical properties of proteins in crowded molecular systems, which normally occur in nature.