What I have learned by using chemical model systems to study biomolecular structure and interactions

Chemical model systems provide valuable insights into biomolecular structure and interactions by allowing researchers to simplify, isolate, and manipulate aspects of the complex molecular machinery of living systems. This perspective describes my laboratory's design, synthesis, and study of chemical model systems that fold and self-assemble like proteins and elucidates the insights that have come from studying these systems. Many of these studies have focused on protein beta-sheets, which exhibit fascinating intra- and intermolecular interactions and play important roles in protein folding, aggregation, and molecular recognition.     

Future in biomolecular computation

Summary Large-scale computations for biomolecules are dominated by three levels of theory: rigorous quantum mechanical calculations for molecules with up to about 30 atoms, semi-empirical quantum mechanical calculations for systems with up to several hundred atoms, and force-field molecular dynamics studies of biomacromolecules with 10,000 atoms and more including surrounding solvent molecules. It can be anticipated that increased computational power will allow the treatment of larger systems of ever growing complexity. Due to the scaling of the computational requirements with increasing number of atoms, the force-field approaches will benefit the most from increased computational power. On the other hand, progress in methodologies such as density functional theory will enable us to treat larger systems on a fully quantum mechanical level and a combination of molecular dynamics and quantum mechanics can be envisioned. One of the greatest challenges in biomolecular computation is the protein folding problem. It is unclear at this point, if an approach with current methodologies will lead to a satisfactory answer or if unconventional, new approaches will be necessary. In any event, due to the complexity of biomolecular systems, a hierarchy of approaches will have to be established and used in order to capture the wide ranges of length-scales and time-scales involved in biological processes. In terms of hardware development, speed and power of computers will increase while the price/performance ratio will become more and more favorable. Parallelism can be anticipated to become an integral architectural feature in a range of computers. It is unclear at this point, how fast massively parallel systems will become easy enough to use so that new methodological developments can be pursued on such computers. Current trends show that distributed processing such as the combination of convenient graphics workstations and powerful general-purpose supercomputers will lead to a new style of computing in which the calculations are monitored and manipulated as they proceed. The combination of a numeric approach with artificial-intelligence approaches can be expected to open up entirely new possibilities. Ultimately, the most exciding aspect of the future in biomolecular computing will be the unexpected discoveries.     

Time-dependent communication between multiple amino acids during protein folding

Cooperativity is considered to be a key organizing principle behind biomolecular assembly, recognition and folding. However, it has remained very challenging to quantitatively characterize how cooperative processes occur on a concerted, multiple-interaction basis. Here, we address how and when the folding process is cooperative on a molecular scale. To this end, we analyze multipoint time-correlation functions probing time-dependent communication between multiple amino acids, which were computed from long folding simulation trajectories. We find that the simultaneous multiple amino-acid contact formation, which is absent in the unfolded state, starts to develop only upon entering the folding transition path. Interestingly, the transition state, whose presence is connected to the macrostate cooperative behavior known as the two-state folding, can be identified as the state in which the amino-acid cooperativity is maximal. Thus, our work not only provides a new mechanistic view on how protein folding proceeds on a multiple-interaction basis, but also offers a conceptually novel characterization of the folding transition state and the molecular origin of the phenomenological cooperative folding behavior. Moreover, the multipoint correlation function approach adopted here is general and can be used to expand the understanding of cooperative processes in complex chemical and biomolecular systems.

Cooperativity in contact formation among multiple amino acids starts to develop upon entering the folding transition path and attains a maximum at the folding transition state, providing the molecular origin of the two-state folding behavior.     

Highly sensitive biomolecular fluorescence detection using nanoscale ZnO platforms

Fluorescence detection is currently one of the most widely used methods in the areas of basic biological research, biotechnology, cellular imaging, medical testing, and drug discovery. Using model protein and nucleic acid systems, we demonstrate that engineered nanoscale zinc oxide structures can significantly enhance the detection capability of biomolecular fluorescence. Without any chemical or biological amplification processes, nanoscale zinc oxide platforms enabled increased fluorescence detection of these biomolecules when compared to other commonly used substrates such as glass, quartz, polymer, and silicon. The use of zinc oxide nanorods as fluorescence enhancing substrates in our biomolecular detection permitted sub-picomolar and attomolar detection sensitivity of proteins and DNA, respectively, when using a conventional fluorescence microscope. This ultrasensitive detection was due to the presence of ZnO nanomaterials which contributed greatly to the increased signal-to-noise ratio of biomolecular fluorescence. We also demonstrate the easy integration potential of zinc oxide nanorods into periodically patterned nanoplatforms which, in turn, will promote the assembly and fabrication of these materials into multiplexed, high-throughput, optical sensor arrays. These zinc oxide nanoplatforms will be extremely beneficial in accomplishing highly sensitive and specific detection of biological samples involving nucleic acids, proteins and cells, particularly under detection environments involving extremely small sample volumes of ultratrace-level concentrations.     

Conformational selection or induced fit for Brinker and DNA recognition

Brinker is the key target protein of the Drosophila Decapentaplegic morphogen signalling pathway. Brinker is widely expressed and can bind with DNA. NMR spectra suggest that apo-Brinker is intrinsically unstructured and undergoes a folding transition upon DNA-binding. However, the coupled mechanism of binding and folding is poorly understood. Here, we performed molecular dynamics (MD) simulations for both bound and apo-Brinker to study the mechanism. Room-temperature MD simulations suggest that Brinker becomes more rigid and stable upon DNA-binding. Kinetic analysis of high-temperature MD simulations shows that both bound and apo-Brinker unfold via a two-state process. The time scale of tertiary unfolding is significantly different between bound and apo-Brinker. The predicted Φ-values suggest that there are more residues with native-like transition state ensembles (TSEs) for bound Brinker than for apo-Brinker. The average RMSD differences between bound and apo-Brinker and Kolmogorov-Smirnov (KS) test analysis illustrate that Brinker folding upon DNA-binding might obey induced-fit mechanism based on MD simulations. These methods can be used for the research of other biomolecular folding upon ligand-binding.     

Recent advances in molecular dynamics simulation towards the realistic representation of biomolecules in solution

Coupled advances in empirical force fields and classical molecular dynamics simulation methodologies, combined with the availability of faster computers, has lead to significant progress towards accurately representing the structure and dynamics of biomolecular systems, such as proteins, nucleic acids, and lipids in their native environments. Thanks to these advances, simulation results are moving beyond merely evaluating force fields, displaying expected structural fluctuations, or demonstrating low root-mean-squared deviations from experimental structures and now provide believable structural insight into a variety of processes such as the stabilization of A-DNA in mixed water and ethanol solution or reversible β-peptide folding in methanol. The purpose of this overview is to take stock of these recent advances in biomolecular simulation and point out some common deficiencies exposed in longer simulations. The most significant methodological advances relate to the development of fast methods to properly treat long-range electrostatic interactions, and in this regard the fast Ewald methods are becoming the de facto standard. Received: 9 April 1998 / Accepted: 21 May 1998 / Published online: 13 August 1998     

Convergence and sampling efficiency in replica exchange simulations of peptide folding in explicit solvent

Replica exchange methods (REMs) are increasingly used to improve sampling in molecular dynamics (MD) simulations of biomolecular systems. However, despite having been shown to be very effective on model systems, the application of REM in complex systems such as for the simulation of protein and peptide folding in explicit solvent has not been objectively tested in detail. Here we present a comparison of conventional MD and temperature replica exchange MD (T-REMD) simulations of a beta-heptapeptide in explicit solvent. This system has previously been shown to undergo reversible folding on the time scales accessible to MD simulation and thus allows a direct one-to-one comparison of efficiency. The primary properties compared are the free energy of folding and the relative populations of different conformers as a function of temperature. It is found that to achieve a similar degree of precision T-REMD simulations starting from a random set of initial configurations were approximately an order of magnitude more computationally efficient than a single 800 ns conventional MD simulation for this system at the lowest temperature investigated (275 K). However, whereas it was found that T-REMD simulations are more than four times more efficient than multiple independent MD simulations at one temperature (300 K) the actual increase in conformation sampling was only twofold. The overall gain in efficiency using REMD resulted primarily from the ordering of different conformational states over temperature, as opposed to a large increase of conformational sampling. It is also shown that in this system exchanges are accepted primarily based on (random) fluctuations within the solvent and are not strongly correlated with the instantaneous peptide conformation raising questions in regard to the efficiency of T-REMD in larger systems.     

Rapid Perturbation of Free‐Energy Landscapes: From In Vitro to In Vivo

Biological systems are often studied under the most “physiological” conditions possible. However, purposeful perturbation of biological systems can provide much information about their dynamics, robustness, and function. Such perturbations are particularly easy to apply at the interface of molecular biophysics and cellular biology, at which complex and highly regulated networks emerge from the behavior of individual biomolecules. Due to the size of diffusion coefficients and the length scale of biomolecules, the fastest timescales at this interface extend to below a microsecond. Thus perturbations must be induced and detected rapidly. We focus on examples of proteins and RNAs interacting with themselves (folding) or one another (binding, signaling). Beginning with general principles that have been learned from simple models and perturbation experiments in vitro, we progress to more complex environments that mimic aspects of the living cell, and finally rapid perturbation experiments in living cells. On the experimental side we highlight in particular two classes of rapid perturbation methods (nanoseconds to seconds) that have been traditionally employed in biophysical chemistry, but that will become increasingly important in cell biology and in vivo: fast relaxation techniques and phase‐sensitive modulation techniques. These techniques are now increasingly married with imaging to produce a spatiotemporal map of biomolecular stability, dynamics and, in the near future, interaction networks inside cells. Many important chemical processes occur on biologically fast timescales, and yet have important ramifications for slower biological networks.     

Rapid perturbation of free-energy landscapes: from in vitro to in vivo

Biological systems are often studied under the most "physiological" conditions possible. However, purposeful perturbation of biological systems can provide much information about their dynamics, robustness, and function. Such perturbations are particularly easy to apply at the interface of molecular biophysics and cellular biology, at which complex and highly regulated networks emerge from the behavior of individual biomolecules. Due to the size of diffusion coefficients and the length scale of biomolecules, the fastest timescales at this interface extend to below a microsecond. Thus perturbations must be induced and detected rapidly. We focus on examples of proteins and RNAs interacting with themselves (folding) or one another (binding, signaling). Beginning with general principles that have been learned from simple models and perturbation experiments in vitro, we progress to more complex environments that mimic aspects of the living cell, and finally rapid perturbation experiments in living cells. On the experimental side we highlight in particular two classes of rapid perturbation methods (nanoseconds to seconds) that have been traditionally employed in biophysical chemistry, but that will become increasingly important in cell biology and in vivo: fast relaxation techniques and phase-sensitive modulation techniques. These techniques are now increasingly married with imaging to produce a spatiotemporal map of biomolecular stability, dynamics and, in the near future, interaction networks inside cells. Many important chemical processes occur on biologically fast timescales, and yet have important ramifications for slower biological networks.     

On the Role of Flexibility in Protein–Ligand Interactions: the Example of p53 Tetramerization Domain

The recognition of protein surfaces by designed ligands has become an attractive approach in drug discovery. However, the variable nature and irregular behavior of protein surfaces defy this new area of research. The easy to understand “lock‐and‐key” model is far from being the ideal paradigm in biomolecular interactions and, hence, any new finding on how proteins and ligands behave in recognition events paves a step of the way. Herein, we illustrate a clear example on how an increase in flexibility of both protein and ligand can result in an increase in the stability of the macromolecular complex. The biophysical study of the interaction between a designed flexible tetraguanidinium‐calix[4]arene and the tetramerization domain of protein p53 (p53TD) and its natural mutant p53TD‐R337H shows how the floppy mutant domain interacts more tightly with the ligand than the well‐packed wild‐type protein. Moreover, the flexible calixarene ligand interacts with higher affinity to both wild‐type and mutated protein domains than a conformationally rigid calixarene analog previously reported. These findings underscore the crucial role of flexibility in molecular recognition processes, for both small ligands and large biomolecular surfaces.     

Monte Carlo vs molecular dynamics for all-atom polypeptide folding simulations

An efficient Monte Carlo (MC) algorithm including concerted rotations is directly compared to molecular dynamics (MD) in all-atom statistical mechanics folding simulations of small polypeptides. The previously reported algorithm "concerted rotations with flexible bond angles" (CRA) has been shown to successfully locate the native state of small polypeptides. In this study, the folding of three small polypeptides (trpzip2/H1/Trp-cage) is investigated using MC and MD, for a combined sampling time of approximately 10(11) MC configurations and 8 micros, respectively. Both methods successfully locate the experimentally determined native states of the three systems, but they do so at different speed, with 2-2.5 times faster folding of the MC runs. The comparison reveals that thermodynamic and dynamic properties can reliably be obtained by both and that results from folding simulations do not depend on the algorithm used. Similar to previous comparisons of MC and MD, it is found that one MD integration step of 2 fs corresponds to one MC scan, revealing the good sampling of MC. The simplicity and efficiency of the MC method will enable its future use in folding studies involving larger systems and the combination with replica exchange algorithms.     

Protein-based materials,toward a new level of structural control

Through billions of years of evolution nature has created and refined structural proteins for a wide variety of specific purposes. Amino acid sequences and their associated folding patterns combine to create elastic, rigid or tough materials. In many respects, nature's intricately designed products provide challenging examples for materials scientists, but translation of natural structural concepts into bio-inspired materials requires a level of control of macromolecular architecture far higher than that afforded by conventional polymerization processes. An increasingly important approach to this problem has been to use biological systems for production of materials. Through protein engineering, artificial genes can be developed that encode protein-based materials with desired features. Structural elements found in nature, such as beta-sheets and alpha-helices, can be combined with great flexibility, and can be outfitted with functional elements such as cell binding sites or enzymatic domains. The possibility of incorporating non-natural amino acids increases the versatility of protein engineering still further. It is expected that such methods will have large impact in the field of materials science, and especially in biomedical materials science, in the future.     

Dynamical signature of abasic damage in DNA

Time-dependent Stokes shift (TDSS) responses in proteins and DNA exhibit a broad range of long time scales (>10 ps) that are not present in bulk aqueous solution. The physical interpretation of the long TDSS time scales in biomolecular systems is a matter of considerable debate because of the many different components present in the sample (water, biomolecule, counterions), which have highly correlated motions and intrinsically different abilities to adapt to local perturbations. Here we use molecular dynamics (MD) simulations to show that the surprisingly slow (～10 ns) TDSS response of coumarin 102 (C102), a base pair replacement, reflects a distinct dynamical signature for DNA damage. When the C102 molecule is covalently incorporated into DNA, an abasic site is created on the strand opposite the C102 probe. The abasic sugar exhibits a reversible interchange between intra- and extrahelical conformations that are kinetically stable on a nanosecond time scale. This conformational change, only possible in damaged DNA, was found to be responsible for the long time scales in the measured TDSS response. For the first time, a TDSS measurement has been attributed to a specific biomolecular motion. This finding directly contradicts the prevailing notion that the TDSS response in biomolecular contexts is dominated by hydration dynamics. It also suggests that TDSS experiments can be used to study ultrafast biomolecular dynamics that are inaccessible to other techniques.     

准确快速计算含多肽酰胺和核酸碱基的氢键复合物的氢键强度和氢键作用势能曲线

N-H···O=C、C-H···O=C、N-H···N和C-H···N等氢键作用是蛋白质a-螺旋结构、b-折叠结构和DNA双螺旋结构形成的主要因素,在生物分子识别、蛋白质复制以及遗传信息传递等过程中起重要作用。准确快速计算生物体系中存在的N-H···O=C、C-H···O=C、N-H···N和C-H···N等氢键作用强度以及氢键强度随分子几何结构(距离和角度)变化的势能曲线对正确模拟(从而正确认识和理解)蛋白质折叠机制和DNA双螺旋结构形成机制等生物过程意义重大,对设计合成具有特殊功能的生物分子材料有重要指导价值。本文主要介绍了近年来建立的偶极-偶极氢键作用模型及其在快速预测多肽-多肽分子间和核酸碱基-多肽分子间氢键作用强度和氢键作用势能曲线方面的应用。     

A multistate empirical valence bond description of protonatable amino acids

The multistate empirical valence bond (MS-EVB) model, which was developed for molecular dynamics simulations of proton transport in water and biomolecular systems, is extended for the modeling of protonatable amino acid residues in aqueous environments, specifically histidine and glutamic acid. The parameters of the MS-EVB force field are first determined to reproduce the geometries and energetics of the gas phase amino acid-water clusters. These parameters are then optimized to reproduce experimental pK(a) values. The free energy profiles for acid ionization and the corresponding pK(a) values are calculated by MS-EVB molecular dynamics simulations utilizing the umbrella sampling technique, with the center of excess charge coordinate chosen as the dissociation reaction coordinate. A general procedure for fitting the MS-EVB parameters is formulated, which allows for the parametrization of other amino acid residues with protonatable groups and the subsequent use of the MS-EVB approach for molecular dynamics simulations of proton transfer processes in proteins involving protonation/deprotonation of the protonatable amino acid groups.     

Machine learning algorithms for predicting protein folding rates and stability of mutant proteins: comparison with statistical methods

Machine learning algorithms have wide range of applications in bioinformatics and computational biology such as prediction of protein secondary structures, solvent accessibility, binding site residues in protein complexes, protein folding rates, stability of mutant proteins, and discrimination of proteins based on their structure and function. In this work, we focus on two aspects of predictions: (i) protein folding rates and (ii) stability of proteins upon mutations. We briefly introduce the concepts of protein folding rates and stability along with available databases, features for prediction methods and measures for prediction performance. Subsequently, the development of structure based parameters and their relationship with protein folding rates will be outlined. The structure based parameters are helpful to understand the physical basis for protein folding and stability. Further, basic principles of major machine learning techniques will be mentioned and their applications for predicting protein folding rates and stability of mutant proteins will be illustrated. The machine learning techniques could achieve the highest accuracy of predicting protein folding rates and stability. In essence, statistical methods and machine learning algorithms are complimenting each other for understanding and predicting protein folding rates and the stability of protein mutants. The available online resources on protein folding rates and stability will be listed.     

A recent development in computational chemistry: chemical reactions from first principles molecular dynamics simulations

First-principles molecular dynamics simulations have recently been found an effective tool to study a large variety of chemical problems. Finite temperature simulations reveal unique information, including explicit dynamical effects and the evaluation of proper free energy differences. Moreover, dynamics simulations reveal information on the flexibility of molecular systems, and elucidate, often otherwise inaccessible, mechanistic details of chemical reactions. In addition this methodology allows the study of larger, periodic, systems, revealing computationally unique information which may be directly compared to experiments on realistic chemical systems. A variety of examples will be given, although most focus on the important field of catalysis.     

Common folding processes of mini‐proteins: Partial formations of secondary structures initiate the immediate protein folding

The folding processes of mini‐proteins (FSD‐EY/FBPWW28 domain) were computationally investigated by an enhanced conformational sampling method. Through the analyses of trajectories, these mini‐proteins had multiple folding pathways different from a simple two‐state folding, and the multiple folding processes were initiated by partial formations of secondary structures. These findings can be used to understand the folding of large proteins, that is, which secondary structures are partially folded in the early process, and how the remaining parts are sequentially folded. It is found that FSD‐EY (α/β topology) folds by a simple diffusion‐collision mechanism, while the folding process of the FBPWW28 domain (all‐β topology) requires a modification of the diffusion‐collision theory to adequately treat the coil‐sheet transition for the β sheet formation. © 2017 Wiley Periodicals, Inc.     

Synergistic DNA- and Protein-Based Recognition Promote an RNA-Templated Bio-orthogonal Reaction

Biomolecular assemblies composed of proteins and oligonucleotides play a central role in biological processes. While in nature, oligonucleotides and proteins usually assemble via non-covalent interactions, synthetic conjugates have been developed which covalently link both modalities. The resulting peptide-oligonucleotide conjugates have facilitated novel biological applications as well as the design of functional supramolecular systems and materials. However, despite the importance of concerted protein/oligonucleotide recognition in nature, conjugation approaches have barely utilized the synergistic recognition abilities of such complexes. Herein, the structure-based design of peptide-DNA conjugates that bind RNA through Watson-Crick base pairing combined with peptide-mediated major groove recognition is reported. Two distinct conjugate families with tunable binding characteristics have been designed to adjacently bind a particular RNA sequence. In the resulting ternary complex, their peptide elements are located in proximity, a feature that was used to enable an RNA-templated click reaction. The introduced structure-based design approach opens the door to novel functional biomolecular assemblies.     

Self-assembly of trehalose molecules on a lysozyme surface: the broken glass hypothesis

To help understand how sugar interactions with proteins stabilise biomolecular structures, we compare the three main hypotheses for the phenomenon with the results of long molecular dynamics simulations on lysozyme in aqueous trehalose solution (0.75 M). We show that the water replacement and water entrapment hypotheses need not be mutually exclusive, because the trehalose molecules assemble in distinctive clusters on the surface of the protein. The flexibility of the protein backbone is reduced under the sugar patches supporting earlier findings that link reduced flexibility of the protein with its higher stability. The results explain the apparent contradiction between different experimental and theoretical results for trehalose effects on proteins.