• 2018-07
  • 2018-10
  • 2018-11
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • In this study using REMD with the


    In this study, using REMD with the crystal environment, we investigate the effect of cryo-cooling on the crystal structure of Escherichia coli DHFR. DHFR is an important model enzyme (for review, see, e.g., (30)), and has been studied for decades (31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43). Through structural transformation between two conformations—referred to as “occluded” and “closed” forms—during the catalytic Kobe0065 (33), the Met20 loop specifically assumes distinct forms. In the catalytic cycle, the loop conversion is involved with product release, which is a rate-limiting step (37). The x-ray crystallographic structures of E. coli DHFR were determined at both room and cryo-temperatures (Protein Data Bank (PDB): 4P3Q and 4P3R, respectively) (10) by the time-averaged ensemble method (44, 45). In the time-averaged ensemble method, MD simulation is performed under the restraint of the time-weighted average of structure factor to the experimental structure factor. In the protocol presented in (45), the timescale of MD simulations is around 10–100 ps to take into account the intramolecular dynamics, whereas rigid-body disorder is taken care of by the translation-libration-screw model. Resulting structures are reported as an ensemble. For the case of PDB: 4P3Q and 4P3R, 167 and 250 models were reported, respectively. In this article, based on MD simulation of the crystal environment of DHFR expedited by the REMD method, we present the temperature dependence of physical quantities such as volumes and solvent-accessible surface areas (SASAs) from 180 to 300 K and describe how the arrangements of proteins within crystal changes at lower temperatures. We then discuss the temperature-dependent fluctuation of the protein, showing that backbone structure can also be affected by cryo-cooling. We also describe thinly populated structures found in the simulations, thereby suggesting greater heterogeneity of structures than the structural ensemble obtained from the experimental data (PDB: 4P3Q and 4P3R) (10), using the time-averaged ensemble method (44, 45).
    Materials and Methods
    Results and Discussion
    Author Contributions
    Computation was performed with K computer provided by the RIKEN Center for Computational Science through the High Performance Computing Infrastructure System Research project (Project ID: hp160073). This work was supported by Japan Society for the Promotion of Science KAKENHI Grant Numbers 26119006, 15K21711, 26790083, and 25891031 and FOCUS for Establishing Supercomputing Center of Excellence. Input files can be obtained from the authros upon request.
    Introduction In order to predict binding affinity of small molecule inhibitors, a variety of post-docking methods have been established. These methods range from simple consensus scoring to free energy perturbation (FEP) [1], [2], [3], [4] among others. Undoubtedly, the post-docking methods can improve significantly the prediction of the energies of L-R binding, however they are still far from being able to predict with a high degree of accuracy the differences in L-R affinities for those ligands having similar binding energies. The situation is even more complex when we are in front of compounds with structural differences. In such cases, most of the times one must accept only if we can differentiate between very active compounds from compounds with low affinity for the receptor (very poor activity). It is clear that any progress or improvement that we can find to enhance these post-docking methods is of paramount importance for the structure based drug design and they are very welcome. In a recent paper we attempted to find a correlation that would allow us to differentiate between DHFR inhibitors with similar affinities, but we had no success [5]. In fact, we were only able to differentiate between highly active compounds of those who had very poor activity, but we were not able to differentiate between compounds with similar affinities. In that paper we also showed that if one has a good geometry, the QTAIM study provides an important insight into the molecular interactions between ligand and receptor. In this new work, we used the electron density obtained from QTAIM analysis as a descriptor of the molecular interactions of the L-R complex, which has allowed us to discriminate very well between compounds with similar binding affinities.