No abstract

YidC, a bacterial member of the YidC/Alb3/Oxa1 insertase family, mediates membrane protein assembly and insertion. Cytoplasmic loops are known to have functional significance in membrane proteins such as YidC. Employing microsecond-level molecular dynamics (MD) simulations, we show that the crystallographically unresolved C2 loop plays a crucial role in the structural dynamics of Bacillus halodurans YidC2. We have modeled the C2 loop and used all- atom MD simulations to investigate the structural dynamics of YidC2 in its apo form, both with and without the C2 loop. The C2 loop was found to stabilize the entire protein and particularly the C1 region. C2 was also found to stabilize the alpha-helical character of the C-terminal region. Interestingly, the highly polar or charged lipid head groups of the simulated membranes were found to interact with and stabilize the C2 loop. These findings demonstrate that the crystallographically unresolved loops of membrane proteins could be important for the stabilization of the protein despite the apparent lack of structure, which could be due to the absence of the relevant lipids to stabilize them in crystallographic conditions.

Profilin-1 (PFN1) is a 140-amino-acid protein with two distinct binding sites―one for actin and one for poly-L-proline (PLP). The best-described function of PFN1 is to catalyze actin elongation and polymerization. Thus far, eight DNA mutations in the PFN1 gene encoding the PFN1 protein are associated with human amyotrophic lateral sclerosis (ALS). We and others recently showed that two of these mutations (Gly118Val or G118V and Cys71Gly or C71G) cause ALS in rodents. In vitro studies suggested that Met114Thr and Thr109Met cause the protein to behave abnormally and cause neurotoxicity. The mechanism by which a single amino acid change in human PFN1 causes the degeneration of motor neurons is not known. In this study, we investigated the structural perturbations of PFN1 caused by each ALS-associated mutation. We used molecular dynamics simulations to assess how these mutations alter the secondary and tertiary structures of human PFN1. Herein, we present our in silico data and analysis on the effect of G118V and T109M mutations on PFN1 and its interactions with actin and PLP. The substitution of valine for glycine reduces the conformational flexibility of the loop region between the α-helix and β-strand and enhances the hydrophobicity of the region. Our in silico analysis of T109M indicates that this mutation alters the shape of the PLP-binding site and reduces the flexibility of this site. Simulation studies of PFN1 in its wild type (WT) and mutant forms (both G118V and T109M mutants) revealed differential fluctuation patterns and the formation of salt bridges and hydrogen bonds between critical residues that may shed light on differences between WT and mutant PFN1. In particular, we hypothesize that the flexibility of the actin- and PLP-binding sites in WT PFN1 may allow the protein to adopt slightly different conformations in its free and bound forms. These findings provide new insights into how each of these mutations in PFN1 might increase its propensity for misfolding and aggregation, leading to its dysfunction.

The protein–ligand binding affinity quantifies the binding strength between a protein and its ligand. Computer modeling and simulations can be used to estimate the binding affinity or binding free energy using data- or physics-driven methods or a combination thereof. Here we discuss a purely physics-based sampling approach based on biased molecular dynamics simulations. Our proposed method generalizes and simplifies previously suggested stratification strategies that use umbrella sampling or other enhanced sampling simulations with additional collective-variable-based restraints. The approach presented here uses a flexible scheme that can be easily tailored for any system of interest. We estimate the binding affinity of human fibroblast growth factor 1 to heparin hexasaccharide based on the available crystal structure of the complex as the initial model and four different variations of the proposed method to compare against the experimentally determined binding affinity obtained from isothermal titration calorimetry experiments.

Within the last two decades, severe acute respiratory syndrome (SARS) coronaviruses 1 and 2 (SARS-CoV-1 and SARS-CoV-2) have caused two major outbreaks. For reasons yet to be fully understood the COVID-19 outbreak caused by SARS-CoV-2 has been significantly more widespread than the 2003 SARS epidemic caused by SARS-CoV-1, despite striking similarities between the two viruses. One of the most variable genes differentiating SARS-CoV-1 and SARS-CoV-2 is the S gene that encodes the spike glycoprotein. This protein mediates a crucial step in the infection, i.e., host cell recognition and viral entry, which starts with binding to the host cell angiotensin converting enzyme 2 (ACE2) protein for both viruses. Recent structural and functional studies have shed light on the differential binding behavior of the SARS-CoV-1 and SARS-CoV-2 spike proteins. In particular, cryogenic electron microscopy (cryo-EM) studies show that ACE2 binding is preceded by a large-scale conformational change in the spike protein to expose the receptor binding domain (RBD) to its binding partner. Unfortunately, these studies do not provide detailed information on the dynamics of this activation process. Here, we have used an extensive set of unbiased and biased microsecond-timescale all-atom molecular dynamics (MD) simulations of SARS-CoV-1 and SARS-CoV-2 spike protein ectodomains in explicit solvent to determine the differential behavior of spike protein activation in the two viruses. Our results based on nearly 50 microseconds of equilibrium and nonequilibrium MD simulations indicate that the active form of the SARS-CoV-2 spike protein is considerably more stable than the active SARS-CoV-1 spike protein. Unlike the active SARS-CoV-2 spike, the active SARS-CoV-1 spike spontaneously undergoes a large-scale conformational transition to a pseudo-inactive state, which occurs in part due to interactions between the N-terminal domain (NTD) and RBD that are absent in the SARS-CoV-2 spike protein. Steered MD (SMD) simulations indicate that the energy barriers between active and inactive states are comparatively lower for the SARS-CoV-1 spike protein. Based on these results, we hypothesize that the greater propensity of the SARS-CoV-2 spike protein to remain in the active conformation contributes to the higher transmissibility of SARS-CoV-2 in comparison to SARS-CoV-1. These results strongly suggest that the differential binding behavior of the active SARS-CoV-1 and 2 spike proteins is not merely due to differences in their RBDs as other domains of the spike protein such as the NTD could play a crucial role in the effective binding process, which involves the prebinding activation. Therefore, our hypothesis predicts that mutations in regions such as the NTD, which is not directly involved in binding, may lead to a change in the effective binding behavior of the coronavirus.