Significant Refinement of Protein Structure Models Using a Residue-Specific Force Field

A New Combination: Anti Glypican-3 scFv and Diphtheria Toxin with the Best Flexible Linker

Along with all cancer treatments, including chemotherapy, radiotherapy, and surgery, targeting therapy is a new treatment manner. Immunotoxins are new recombinant structures that kill cancer cells by targeting specific antigens. Immunotoxins are composed of two parts: toxin moiety, which disrupts protein synthesis process, and antigen binding moiety that bind to antigens on the surface of cancer cells. Glypican 3 (GPC3) is an oncofetal antigen on the surface of Hepatocellular carcinoma (HCC) cells. In this study, truncated Diphtheria toxin (DT389) was fused to humanized scFv YP7 by one, two and three repeats of GGGGS linkers (DT389-(GGGGS)_1-3YP7). In-silico and experimental investigation were performed to find out how many repeats of linker between toxin and scFv moieties are sufficient. Results of in-silico investigations revealed that the difference in the number of linkers does not have a significant effect on the main structures of the immunotoxin; however, the three-dimensional structure of two repeats of linker had a more appropriate structure compared to others with one and three linker replications. In addition, with enhancing the number of linkers, the probability of protein solubility has increased. Generally, the bioinformatics results of DT389-(GGGGS)₂-YP7 structure showed that expression and folding is suitable; and YP7 scFv has appropriate orientation to bind GPC3. The experimental investigations indicated that the fusion protein was expressed as near to 50% soluble. Due to the high binding affinity of YP7 scFv and the proven potency of diphtheria in inhibiting protein synthesis, the proposed DT389-(GGGGS)₂-YP7 immunotoxin is expected to function well in inhibiting HCC.

A novel piperazine derivative that targets hepatitis B surface antigen effectively inhibits tenofovir resistant hepatitis B virus

Chronic hepatitis B virus (HBV) infection is a global problem. The loss of hepatitis B surface antigen (HBsAg) in serum is a therapeutic end point. Prolonged therapy with nucleoside/nucleotide analogues targeting the HBV-polymerase may lead to resistance and rarely results in the loss of HBsAg. Therefore, inhibitors targeting HBsAg may have potential therapeutic applications. Here, we used computational virtual screening, docking, and molecular dynamics simulations to identify potential small molecule inhibitors against HBsAg. After screening a million molecules from ZINC database, we identified small molecules with potential anti-HBV activity. Subsequently, cytotoxicity profiles and anti-HBV activities of these small molecules were tested using a widely used cell culture model for HBV. We identified a small molecule (ZINC20451377) which binds to HBsAg with high affinity, with a KD of 65.3 nM, as determined by Surface Plasmon Resonance spectroscopy. Notably, the small molecule inhibited HBsAg production and hepatitis B virion secretion (10 μM) at low micromolar concentrations and was also efficacious against a HBV quadruple mutant (CYEI mutant) resistant to tenofovir. We conclude that this small molecule exhibits strong anti-HBV properties and merits further testing.

Improvement, cloning, and expression of an in silico designed protein enriched with large neutral amino acids in Pichia pastoris for possible application in phenylketonuria

Phenylketonuria (PKU) is an inborn disease caused by defective phenylalanine hydroxylase, which consequently results in the accumulation of phenylalanine in the brain leading to further complications. One of the promising approaches in dietary treatment is the supplementation of large neutral amino acid (LNAA). The LNAA compete with phenylalanine for the common L-type LNAA transporter across the blood-brain barrier, and decrease phenylalanine levels in the brain. In this study, the earlier LNAA-enriched protein model was improved (Protein Model-66) and validated in silico. The reverse translated and codon-optimized synthetic LNAA66 gene was cloned into pPICZαC and expressed in Pichia pastoris. The expressed protein was purified by His Select affinity chromatography. SDS-PAGE and Western blotting analysis showed a band at an expected molecular weight of 12 kDa, confirming the expression of the modeled protein. To our knowledge, this is the first report showing the cloning and expression of an in silico designed LNAA-enriched protein. PRACTICAL APPLICATIONS: One of the promising dietary treatment of phenylketonuria (PKU) is the supplementation of large neutral amino acid (LNAA), wherein high levels of LNAA compete with phenylalanine for the same L-type LNAA transporter, and consequently decrease phenylalanine accumulation in the brain, thereby decreasing neurological complications. For the first time, here, we are showing that an in silico designed and validated Protein Model-66, rich in LNAA, can be successfully cloned and expressed in Pichia pastoris. The complete biochemical and structural characterization of this protein will give a clear insight into its potential application for PKU treatment. The protein can be potentially used as a supplement to treat PKU to those who are non-adherent to the restricted, non-palatable, and expensive diet. Furthermore, this novel and effective strategy of in silico designing, cloning and expression can be exploited to develop proteins for various applications of industrial, food, medical, and academic relevance.

Applications of Molecular Dynamics Simulation in Structure Prediction of Peptides and Proteins

Compared with rapid accumulation of protein sequences from high-throughput DNA sequencing, obtaining experimental 3D structures of proteins is still much more difficult, making protein structure prediction (PSP) potentially very useful. Currently, a vast majority of PSP efforts are based on data mining of known sequences, structures and their relationships (informatics-based). However, if closely related template is not available, these methods are usually much less reliable than experiments. They may also be problematic in predicting the structures of naturally occurring or designed peptides. On the other hand, physics-based methods including molecular dynamics (MD) can utilize our understanding of detailed atomic interactions determining biomolecular structures. In this mini-review, we show that all-atom MD can predict structures of cyclic peptides and other peptide foldamers with accuracy similar to experiments. Then, some notable successes in reproducing experimental 3D structures of small proteins through MD simulations (some with replica-exchange) of the folding were summarized. We also describe advancements of MD-based refinement of structure models, and the integration of limited experimental or bioinformatics data into MD-based structure modeling.

Experimental accuracy in protein structure refinement via molecular dynamics simulations

Refinement is the last step in protein structure prediction pipelines to convert approximate homology models to experimental accuracy. Protocols based on molecular dynamics (MD) simulations have shown promise, but current methods are limited to moderate levels of consistent refinement. To explore the energy landscape between homology models and native structures and analyze the challenges of MD-based refinement, eight test cases were studied via extensive simulations followed by Markov state modeling. In all cases, native states were found very close to the experimental structures and at the lowest free energies, but refinement was hindered by a rough energy landscape. Transitions from the homology model to the native states require the crossing of significant kinetic barriers on at least microsecond time scales. A significant energetic driving force toward the native state was lacking until its immediate vicinity, and there was significant sampling of off-pathway states competing for productive refinement. The role of recent force field improvements is discussed and transition paths are analyzed in detail to inform which key transitions have to be overcome to achieve successful refinement.

Significantly different contact patterns between Aβ40 and Aβ42 monomers involving the N-terminal region

Aβ42 peptide, with two additional residues at C-terminus, aggregates much faster than Aβ40. We performed equilibrium replica-exchange molecular dynamics simulations of their monomers using our residue-specific force field. Simulated ³ J_{HNH α} -coupling constants agree excellently with experimental data. Aβ40 and Aβ42 have very similar local conformational features, with considerable β-strand structures in the segments: A2-H6 (A), L17-A21 (B), A30-V36 (C) of both peptides and V39-I41 (D) of Aβ42. Both peptides have abundant A-B and B-C contacts, but Aβ40 has much more contacts between A and C than Aβ42, which may retard its aggregation. Only Aβ42 has considerable A-B-C-D topology. Decreased probability of A-C contact in Aβ42 relates to the competition from C-D contact. Increased A-C contact probability may also explain the slower aggregation of A2T and A2V mutants of Aβ42.

Protein homology model refinement by large-scale energy optimization

Proteins fold to their lowest free-energy structures, and hence the most straightforward way to increase the accuracy of a partially incorrect protein structure model is to search for the lowest-energy nearby structure. This direct approach has met with little success for two reasons: first, energy function inaccuracies can lead to false energy minima, resulting in model degradation rather than improvement; and second, even with an accurate energy function, the search problem is formidable because the energy only drops considerably in the immediate vicinity of the global minimum, and there are a very large number of degrees of freedom. Here we describe a large-scale energy optimization-based refinement method that incorporates advances in both search and energy function accuracy that can substantially improve the accuracy of low-resolution homology models. The method refined low-resolution homology models into correct folds for 50 of 84 diverse protein families and generated improved models in recent blind structure prediction experiments. Analyses of the basis for these improvements reveal contributions from both the improvements in conformational sampling techniques and the energy function.

Computational protein structure refinement: Almost there, yet still so far to go

Protein structures are essential in modern biology yet experimental methods are far from being able to catch up with the rapid increase in available genomic data. Computational protein structure prediction methods aim to fill the gap while the role of protein structure refinement is to take approximate initial template-based models and bring them closer to the true native structure. Current methods for computational structure refinement rely on molecular dynamics simulations, related sampling methods, or iterative structure optimization protocols. The best methods are able to achieve moderate degrees of refinement but consistent refinement that can reach near-experimental accuracy remains elusive. Key issues revolve around the accuracy of the energy function, the inability to reliably rank multiple models, and the use of restraints that keep sampling close to the native state but also limit the degree of possible refinement. A different aspect is the question of what exactly the target of high-resolution refinement should be as experimental structures are affected by experimental conditions and different biological questions require varying levels of accuracy. While improvement of the global protein structure is a difficult problem, high-resolution refinement methods that improves local structural quality such as favorable stereochemistry and the avoidance of atomic clashes are much more successful.

Mechanism of Phosphorylation-Induced Folding of 4E-BP2 Revealed by Molecular Dynamics Simulations

Site-specific phosphorylation of an intrinsically disordered protein, eIF4E-binding protein isoform 2 (4E-BP2), can suppress its native function by folding it into a four-stranded β-sheet, but the mechanism of this phosphorylation-induced folding is unclear. In this work, we use all-atom molecular dynamics simulations to investigate both the folded and unfolded states of 4E-BP2 under different phosphorylation states of T37 and T46. The results show that the phosphorylated forms of both T37 and T46 play important roles in stabilizing the folded structure, especially for the β-turns and the sequestered binding motif. The phosphorylated residues not only guide the folding of the protein through several intermediate states but also affect the conformational distribution of the unfolded ensemble. Significantly, the phosphorylated residues can function as nucleation sites for the folding of the protein by forming certain local structures that are stabilized by hydrogen bonding involving the phosphate group. The region around phosphorylated T46 appears to fold before that around phosphorylated T37. These findings provide new insight into the intricate effects of protein phosphorylation.

Nonlinear backbone torsional pair correlations in proteins

Protein allostery requires dynamical structural correlations. Physical origin of which, however, remain elusive despite intensive studies during last two and half decades. Based on analysis of molecular dynamics (MD) simulation trajectories for ten proteins with different sizes and folds, we found that nonlinear backbone torsional pair (BTP) correlations, which are mainly spatially long-ranged and are dominantly executed by loop residues, exist extensively in most analyzed proteins. Examination of torsional motion for correlated BTPs suggested that such nonlinear correlations are mainly associated aharmonic torsional state transitions and in some cases strongly anisotropic local torsional motion of participating torsions, and occur on widely different and relatively longer time scales. In contrast, correlations between backbone torsions in stable α helices and β strands are mainly linear and spatially short-ranged, and are more likely to associate with harmonic local torsional motion. Further analysis revealed that the direct cause of nonlinear contributions are heterogeneous linear correlations. These findings implicate a general search strategy for novel allosteric modulation sites of protein activities.

Accurate Structure Prediction and Conformational Analysis of Cyclic Peptides with Residue-Specific Force Fields

Cyclic peptides (CPs) are promising candidates for drugs, chemical biology tools, and self-assembling nanomaterials. However, the development of reliable and accurate computational methods for their structure prediction has been challenging. Here, 20 all-trans CPs of 5-12 residues selected from Cambridge Structure Database have been simulated using replica-exchange molecular dynamics with four different force fields. Our recently developed residue-specific force fields RSFF1 and RSFF2 can correctly identify the crystal-like conformations of more than half CPs as the most populated conformation. The RSFF2 performs the best, which consistently predicts the crystal structures of 17 out of 20 CPs with rmsd < 1.1 Å. We also compared the backbone (ϕ, ψ) sampling of residues in CPs with those in short linear peptides and in globular proteins. In general, unlike linear peptides, CPs have local conformational free energies and entropies quite similar to globular proteins.

Folding Simulations of an α-Helical Hairpin Motif αtα with Residue-Specific Force Fields

α-Helical hairpin (two-helix bundle) is a structure motif composed of two interacting helices connected by a turn or a short loop. It is an important model for protein folding studies, filling the gap between isolated α-helix and larger all-α domains. Here, we present, for the first time, successful folding simulations of an α-helical hairpin. Our RSFF1 and RSFF2 force fields give very similar predicted structures of this αtα peptide, which is in good agreement with its NMR structure. Our simulations also give site-specific stability of α-helix formation in good agreement with amide hydrogen exchange experiments. Combining the folding free energy landscapes and analyses of structures sampled in five different ranges of the fraction of native contacts (Q), a folding mechanism of αtα is proposed. The most stable sites of Q9-E15 in helix-1 and E24-A30 in helix-2 close to the loop region act as the folding initiation sites. The formation of interhelix side-chain contacts also initiates near the loop region, but some residues in the central parts of the two helices also form contacts quite early. The two termini fold at a final stage, and the loop region remains flexible during the whole folding process. This mechanism is similar to the "zipping out" pathway of β-hairpin folding.

Protein structure refinement via molecular-dynamics simulations: What works and what does not?

Protein structure refinement during CASP11 by the Feig group was described. Molecular dynamics simulations were used in combination with an improved selection and averaging protocol. On average, modest refinement was achieved with some targets improved significantly. Analysis of the CASP submission from our group focused on refinement success versus amount of sampling, refinement of different secondary structure elements and whether refinement varied as a function of which group provided initial models. The refinement of local stereochemical features was examined via the MolProbity score and an updated protocol was developed that can generate high-quality structures with very low MolProbity scores for most starting structures with modest computational effort. Proteins 2016; 84(Suppl 1):282-292. © 2015 Wiley Periodicals, Inc.