Computing infrared spectra of proteins using the exciton model

An electron density clustering based adaptive segmentation method for protein Raman spectrum calculation

Raman spectroscopy is a powerful technique for protein detection, but the calculation of Raman spectrum is a longstanding challenging problem due to the large sizes and complex structures of protein molecules. Dividing proteins into fragments can greatly accelerate the calculation, but this usually introduces large errors originating from ignored interactions between fragments into obtained spectra. In this paper, we proposed a new adaptive segmentation method based on the strength of interactions and molecular shapes and structures, i.e., electron density clustering, to divide proteins. It can reduce errors of obtained Raman spectra by about 20% compared to the uniform segmentation method without a significant increase in computational cost. This method can facilitate the validation and analysis of detected Raman spectra of proteins and promote the application of Raman spectroscopy in biological detection.

Artificial Intelligence-based Amide-II Infrared Spectroscopy Simulation for Monitoring Protein Hydrogen Bonding Dynamics

The structurally sensitive amide II infrared (IR) bands of proteins provide valuable information about the hydrogen bonding of protein secondary structures, which is crucial for understanding protein dynamics and associated functions. However, deciphering protein structures from experimental amide II spectra relies on time-consuming quantum chemical calculations on tens of thousands of representative configurations in solvent water. Currently, the accurate simulation of amide II spectra for whole proteins remains a challenge. Here, we present a machine learning (ML)-based protocol designed to efficiently simulate the amide II IR spectra of various proteins with an accuracy comparable to experimental results. This protocol stands out as a cost-effective and efficient alternative for studying protein dynamics, including the identification of secondary structures and monitoring the dynamics of protein hydrogen bonding under different pH conditions and during protein folding process. Our method provides a valuable tool in the field of protein research, focusing on the study of dynamic properties of proteins, especially those related to hydrogen bonding, using amide II IR spectroscopy.

Refining protein amide I spectrum simulations with simple yet effective electrostatic models for local wavenumbers and dipole derivative magnitudes

Analysis of the amide I band of proteins is probably the most wide-spread application of bioanalytical infrared spectroscopy. Although highly desirable for a more detailed structural interpretation, a quantitative description of this absorption band is still difficult. This work optimized several electrostatic models with the aim to reproduce the effect of the protein environment on the intrinsic wavenumber of a local amide I oscillator. We considered the main secondary structures - α-helices, parallel and antiparallel β-sheets - with a maximum of 21 amide groups. The models were based on the electric potential and/or the electric field component along the CO bond at up to four atoms in an amide group. They were bench-marked by comparison to Hessian matrices reconstructed from density functional theory calculations at the BPW91, 6-31G** level. The performance of the electrostatic models depended on the charge set used to calculate the electric field and potential. Gromos and DSSP charge sets, used in common force fields, were not optimal for the better performing models. A good compromise between performance and the stability of model parameters was achieved by a model that considered the electric field at the positions of the oxygen, nitrogen, and hydrogen atoms of the considered amide group. The model describes also some aspects of the local conformation effect and performs similar on its own as in combination with an explicit implementation of the local conformation effect. It is better than a combination of a local hydrogen bonding model with the local conformation effect. Even though the short-range hydrogen bonding model performs worse, it captures important aspects of the local wavenumber sensitivity to the molecular surroundings. We improved also the description of the coupling between local amide I oscillators by developing an electrostatic model for the dependency of the dipole derivative magnitude on the protein environment.

A Closed Loop Stimuli-Responsive Concanavalin A-Loaded Chitosan-Pluronic Hydrogel for Glucose-Responsive Delivery of Short-Acting Insulin Prototyped in RIN-5F Pancreatic Cells

The optimal treatment of diabetes (in particular, type 1 diabetes-T1D) remains a challenge. Closed-loop systems (implants/inserts) provide significant advantages for glucose responsivity and providing real-time sustained release of rapid-acting insulin. Concanavalin A (ConA), a glucose affinity agent, has been used to design closed-loop insulin delivery systems but not without significant risk of leakage of ConA from the matrices and poor mechanical strength of the hydrogels impacting longevity and control of insulin release. Therefore, this work focused on employing a thermoresponsive co-forming matrix between Pluronic F-127 (PL) and structurally robust chitosan (CHT) via EDC/NHS coupling (i.e., covalent linkage of -NH2 from CHT and ConA to the -COOH of PL). The system was characterized for its chemical structure stability and integrity (FTIR, XRD and TGA), injectability, rheological parameters and hydrogel morphology (Texture Analysis, Elastosens TM Bio2 and SEM). The prepared hydrogels demonstrated shear-thinning for injectability with a maximum force of 4.9 ± 8.3 N in a 26G needle with sol-gel transitioning from 25 to 38 °C. The apparent yield stress value of the hydrogel was determined to be 67.47 Pa. The insulin loading efficiency within the hydrogel matrix was calculated to be 46.8%. Insulin release studies revealed glucose responsiveness in simulated glycemic media (4 and 10 mg/mL) over 7 days (97%) (305 nm via fluorescence spectrophotometry). The MTT studies were performed over 72 h on RIN-5F pancreatic cells with viability results >80%. Results revealed that the thermoresponsive hydrogel is a promising alternative to current closed-loop insulin delivery systems.

Constructing high-accuracy theoretical Raman spectra of SARS-CoV-2 spike proteins based on a large fragment method

In order to control COVID-19, rapid and accurate detection of the pathogenic, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is an urgent task. The target spike proteins of SARS-CoV-2 have been detected experimentally via Raman spectroscopy. However, there lacks high-accuracy theoretical Raman spectra of the spike proteins to as a standard reference for the clinic diagnostic purpose. In this paper, we propose a large fragment method to construct the high-precision Raman spectra for the spike proteins. The large fragment method not only reduces the calculation error but also improves the accuracy of the protein Raman spectra by completely calculating the interactions within the large fragment. The Pearson correlation coefficient of theoretical Raman spectra is greater than 0.929 or more. Compared with the experimental spectra, the characteristic patterns are easily visible. This work provides a detection standard for the spike proteins which shall bring a step closer to the fast recognition of SARS-CoV-2 via Raman spectroscopy method.

Monitoring early-stage β-amyloid dimer aggregation by histidine site-specific two-dimensional infrared spectroscopy in a simulation study

Monitoring early-stage β-amyloid (Aβ) dimerization is a formidable challenge for understanding neurological diseases. We compared β-sheet formation and histidine site-specific two-dimensional infrared (2D IR) spectroscopic signatures of Aβ dimers with different histidine states (δ; N^δ1-H, ε; N^ε2-H, or π; both protonated). Molecular dynamics (MD) simulations revealed that β-sheet formation is favored for the δδδ:δδδ and πππ:πππ tautomeric isomers showing strong couplings and frequent contacts between the central hydrophobic core and C-terminus compared with the εεε:εεε isomer. Characteristic blue-shifts in the 2D IR central bands were observed upon monomer-dimer transformation. The εεε:εεε dimer exhibited larger frequency shifts than δδδ:δδδ and πππ:πππ implying that the red-shift may have a correlation with N^δ1-H(δ) protonation. Our results support the tautomerization/protonation hypothesis that attributes Aβ misfolding to histidine tautomers as a possible primary initiator for Aβ aggregation and facilitates the application of histidine site-specific 2D IR spectroscopy for studying early-stage Aβ self-assembly.

AI-based spectroscopic monitoring of real-time interactions between SARS-CoV-2 and human ACE2

The COVID-19 caused by SARS-CoV-2 virus has posed a tremendous threat to human health. The interactions between human angiotensin-converting enzyme 2 and the spike glycoprotein of SARS-CoV-2 hold the key to understanding the molecular mechanism to develop treatment and vaccines. However, the simulation of these interactions in fluctuating surroundings is challenging because it requires many electronic structure calculations at the quantum mechanics level for a large number of representative configurations. We report a machine learning protocol that can efficiently predict the IR spectra of SARS-CoV-2 with high efficiency and characterize fine changes in IR spectra associated with variations of protein secondary structures. Machine learning provides a cost-effective tool for monitoring of real-time interactions between the SARS-CoV-2 and human ACE2.

The novel coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), invades a human cell via human angiotensin-converting enzyme 2 (hACE2) as the entry, causing the severe coronavirus disease (COVID-19). The interactions between hACE2 and the spike glycoprotein (S protein) of SARS-CoV-2 hold the key to understanding the molecular mechanism to develop treatment and vaccines, yet the dynamic nature of these interactions in fluctuating surroundings is very challenging to probe by those structure determination techniques requiring the structures of samples to be fixed. Here we demonstrate, by a proof-of-concept simulation of infrared (IR) spectra of S protein and hACE2, that time-resolved spectroscopy may monitor the real-time structural information of the protein−protein complexes of interest, with the help of machine learning. Our machine learning protocol is able to identify fine changes in IR spectra associated with variation of the secondary structures of S protein of the coronavirus. Further, it is three to four orders of magnitude faster than conventional quantum chemistry calculations. We expect our machine learning protocol would accelerate the development of real-time spectroscopy study of protein dynamics.

Vibrational exciton delocalization precludes the use of infrared intensities as proxies for surfactant accumulation on aqueous surfaces

Surface-sensitive vibrational spectroscopy is a common tool for measuring molecular organization and intermolecular interactions at interfaces. Peak intensity ratios are typically used to extract molecular information from one-dimensional spectra but vibrational coupling between surfactant molecules can manifest as signal depletion in one-dimensional spectra. Through a combination of experiment and theory, we demonstrate the emergence of vibrational exciton delocalization in infrared reflection–absorption spectra of soluble and insoluble surfactants at the air/water interface. Vibrational coupling causes a significant decrease in peak intensities corresponding to C–F vibrational modes of perfluorooctanoic acid molecules. Vibrational excitons also form between arachidic acid surfactants within a compressed monolayer, manifesting as signal reduction of C–H stretching modes. Ionic composition of the aqueous phase impacts surfactant intermolecular distance, thereby modulating vibrational coupling strength between surfactants. Our results serve as a cautionary tale against employing alkyl and fluoroalkyl vibrational peak intensities as proxies for concentration, although such analysis is ubiquitous in interface science.

Coupling between surfactant molecules at the air/water interface bleeds intensity into a diffuse background, such that single-wavelength vibrational intensity is effectively depleted at high surface coverage.

Lessons from combined experimental and theoretical examination of the FTIR and 2D-IR spectroelectrochemistry of the amide I region of cytochrome c

Amide I difference spectroscopy is widely used to investigate protein function and structure changes. In this article, we show that the common approach of assigning features in amide I difference signals to distinct secondary structure elements in many cases may not be justified. Evidence comes from Fourier transform infrared (FTIR) and 2D-IR spectroelectrochemistry of the protein cytochrome c in the amide I range, in combination with computational spectroscopy based on molecular dynamics (MD) simulations. This combination reveals that each secondary structure unit, such as an alpha-helix or a beta-sheet, exhibits broad overlapping contributions, usually spanning a large part of the amide I region, which in the case of difference absorption experiments (such as in FTIR spectroelectrochemistry) may lead to intensity-compensating and even sign-changing contributions. We use cytochrome c as the test case, as this small electron-transferring redox-active protein contains different kinds of secondary structure units. Upon switching its redox-state, the protein exhibits a different charge distribution while largely retaining its structural scaffold. Our theoretical analysis suggests that the change in charge distribution contributes to the spectral changes and that structural changes are small. However, in order to confidently interpret FTIR amide I difference signals in cytochrome c and proteins in general, MD simulations in combination with additional experimental approaches such as isotope labeling, the insertion of infrared labels to selectively probe local structural elements will be required. In case these data are not available, a critical assessment of previous interpretations of protein amide I 1D- and 2D-IR difference spectroscopy data is warranted.

A Machine Learning Protocol for Predicting Protein Infrared Spectra

Infrared (IR) absorption provides important chemical fingerprints of biomolecules. Protein secondary structure determination from IR spectra is tedious since its theoretical interpretation requires repeated expensive quantum-mechanical calculations in a fluctuating environment. Herein we present a novel machine learning protocol that uses a few key structural descriptors to rapidly predict amide I IR spectra of various proteins and agrees well with experiment. Its transferability enabled us to distinguish protein secondary structures, probe atomic structure variations with temperature, and monitor protein folding. This approach offers a cost-effective tool to model the relationship between protein spectra and their biological/chemical properties.

Combining Experimental Data and Computational Methods for the Non-Computer Specialist

Experimental methods are indispensable for the study of the function of biological macromolecules, not just as static structures, but as dynamic systems that change conformation, bind partners, perform reactions, and respond to different stimulus. However, providing a detailed structural interpretation of the results is often a very challenging task. While experimental and computational methods are often considered as two different and separate approaches, the power and utility of combining both is undeniable. The integration of the experimental data with computational techniques can assist and enrich the interpretation, providing new detailed molecular understanding of the systems. Here, we briefly describe the basic principles of how experimental data can be combined with computational methods to obtain insights into the molecular mechanism and expand the interpretation through the generation of detailed models.

Mapping the amide-I vibrations of model dipeptides with secondary structure sensitivity and amino acid residue specificity, and its application to amyloid β-peptide in aqueous solution

Vibrational spectroscopy has been known as particularly well-suited for deciphering the polypeptide's structure. To decode structural information encoded in IR spectra, we developed amide-I frequency maps on the basis of model dipeptides to correlate the amide-I frequency of interest to the combination of the calculated secondary structure dependent amide-I frequency by using DFT method and the electrostatic potentials that projected onto the amide unit from the micro-environment within molecular mechanics force field. The constructed maps were applied to model dipeptides and amyloid β-peptide fragment (Aβ_25-35). The dipeptide specified map (DS map) and the hybrid map (HYB map) predicted amide-I bands of Aβ_25-35 in solution satisfactorily reproduce experimental observation, and indicate the preference of forming β-sheet and random coil structure for Aβ_25-35 in D₂O just as the results of cluster analysis suggested. These maps with secondary structural sensitivity and amino acid residue specificity open up a way for the interpretation of amide-I vibrations and show their potentials in the understanding of molecular structure of polypeptides in solution.

Site-Specific Peptide Probes Detect Buried Water in a Lipid Membrane

Transmembrane peptides contain polar residues in the interior of the membrane, which may alter the electrostatic environment and favor hydration in the otherwise nonpolar environment of the membrane core. Here, we demonstrate a general, nonperturbative strategy to probe hydration of the peptide backbone at specific depths within the bilayer using a combination of site-specific isotope labels, ultrafast two-dimensional infrared spectroscopy, and spectral modeling based on molecular dynamics simulations. Our results show that the amphiphilic pH-low insertion peptide supports a highly heterogeneous environment, with significant backbone hydration of nonpolar residues neighboring charged residues. For example, a leucine residue located as far as 1 nm into the hydrophobic bulk reports hydrogen-bonded populations as high as ∼20%. These findings indicate that the polar nature of these residues may facilitate the transport of water molecules into the hydrophobic core of the membrane.

On the choice of coordinates in anharmonic theoretical vibrational spectroscopy: Harmonic vs. anharmonic coupling in vibrational configuration interaction

By a suitable choice of coordinates, the computational effort required for calculations of anharmonic vibrational spectra can be reduced significantly. By using suitable localized-mode coordinates obtained from an orthogonal transformation of the conventionally used normal-mode coordinates, anharmonic couplings between modes can be significantly reduced. However, such a transformation introduces harmonic couplings between the localized modes. To elucidate the role of these harmonic couplings, we consider the vibrational self-consistent field (VSCF)/vibrational configuration interaction (VCI) calculations for both few-mode model systems and for ethene as a molecular test case. We show that large harmonic couplings can result in significant errors in localized-mode L-VSCF/L-VCI calculations and study the convergence with respect to the size of the VCI excitation space. To further elucidate the errors introduced by harmonic couplings, we discuss the connection between L-VSCF/L-VCI and vibrational exciton models. With the help of our results, we propose an algorithm for the localization of normal modes in suitable subsets that are chosen to strictly limit the errors introduced by the harmonic couplings while still leading to maximally localized modes.

A quantitative connection of experimental and simulated folding landscapes by vibrational spectroscopy

We break the barrier between simulation and experiment by comparing identical computed and experimental infrared observables.

For small molecule reaction kinetics, computed reaction coordinates often mimic experimentally measured observables quite accurately. Although nowadays simulated and measured biomolecule kinetics can be compared on the same time scale, a gap between computed and experimental observables remains. Here we directly compared temperature-jump experiments and molecular dynamics simulations of protein folding dynamics using the same observable: the time-dependent infrared spectrum. We first measured the stability and folding kinetics of the fastest-folding β-protein, the GTT35 WW domain, using its structurally specific infrared spectrum. The relaxation dynamics of the peptide backbone, β-sheets, turn, and random coil were measured independently by probing the amide I′ region at different frequencies. Next, the amide I′ spectra along folding/unfolding molecular dynamics trajectories were simulated by accurate mixed quantum/classical calculations. The simulated time dependence and spectral amplitudes at the exact experimental probe frequencies provided relaxation and folding rates in agreement with experimental observations. The calculations validated by experiment yield direct structural evidence for a rate-limiting reaction step where an intermediate state with either the first or second hairpin is formed. We show how folding switches from a more homogeneous (apparent two-state) process at high temperature to a more heterogeneous process at low temperature, where different parts of the WW domain fold at different rates.

Central-metal effect on intramolecular vibrational energy transfer of M(CO)5Br (M = Mn, Re) probed by two-dimensional infrared spectroscopy

Vibrational energy transfer in transition metal complexes with flexible structures in condensed phases is of central importance to catalytical chemistry processes. In this work, two molecules with different metal atoms, M(CO)₅Br (where M = Mn, Re), were used as model systems, and their axial and radial carbonyl stretching modes as infrared probes. The central-metal effect on intramolecular vibrational energy redistribution (IVR) in M(CO)₅Br was investigated in polar and nonpolar solvents. The linear infrared (IR) peak splitting between carbonyl vibrations increases as the metal atom changes from Mn to Re. The waiting-time dependent two-dimensional infrared diagonal- and off-diagonal peak amplitudes reveal a faster IVR process in Re(CO)₅Br than in Mn(CO)₅Br. With the aid of density functional theory (DFT) calculations, the central-metal effect on IVR time linearly correlates with the vibrational coupling strength between the two involved modes. In addition, the polar solvent is found to accelerate the IVR process by affecting the anharmonic vibrational potentials of a solute vibration mode.

Computational Exploration of Conformational Transitions in Protein Drug Targets

Protein drug targets vary from highly structured to completely disordered; either way dynamics governs function. Hence, understanding the dynamical aspects of how protein targets function can enable improved interventions with drug molecules. Computational approaches offer highly detailed structural models of protein dynamics which are becoming more predictive as model quality and sampling power improve. However, the most advanced and popular models still have errors owing to imperfect parameter sets and often cannot access longer timescales of many crucial biological processes. Experimental approaches offer more certainty but can struggle to detect and measure lightly populated conformations of target proteins and subtle allostery. An emerging solution is to integrate available experimental data into advanced molecular simulations. In the future, molecular simulation in combination with experimental data may be able to offer detailed models of important drug targets such that improved functional mechanisms or selectivity can be accessed.

Examining the role of protein structural dynamics in drug resistance in Mycobacterium tuberculosis

2D-IR spectroscopy reveals a role for protein structural dynamics in antimicrobial-resistance.

Antimicrobial resistance represents a growing global health problem. The emergence of novel resistance mechanisms necessitates the development of alternative approaches to investigate the molecular fundamentals of resistance, leading ultimately to new strategies for counteracting them. To gain deeper insight into antibiotic–target interactions, the binding of the frontline anti-tuberculosis drug isoniazid (INH) to a target enzyme, InhA, from Mycobacterium tuberculosis was studied using ultrafast two-dimensional infrared (2D-IR) spectroscopy and molecular simulations. Comparing wild-type InhA with a series of single point mutations, it was found that binding of the INH–NAD inhibitor to susceptible forms of the enzyme increased the vibrational coupling between residues located in the Rossmann fold co-factor binding site of InhA and suppressed dynamic fluctuations of the enzyme structure. The effect correlated with biochemical assay data, being reduced in the INH-resistant S94A mutant and absent in the biochemically-inactive P193A control. Molecular dynamics simulations and calculations of inter–residue couplings indicate that the changes in coupling and dynamics are not localised to the co-factor binding site, but permeate much of the protein. We thus propose that the resistant S94A mutation circumvents subtle changes in global structural dynamics caused by INH upon binding to the wild-type enzyme that may impact upon the formation of important protein–protein complexes in the fatty acid synthase pathway of M. tuberculosis.

Computing infrared spectra of proteins using the exciton model

The ability to compute from first principles the infrared spectrum of a protein in solution phase representing a biological system would provide a useful connection to atomistic models of protein structure and dynamics. Indeed, such calculations are a vital complement to 2DIR experimental measurements, allowing the observed signals to be interpreted in terms of detailed structural and dynamical information. In this article, we have studied nine structurally and spectroscopically well-characterized proteins, representing a range of structural types. We have simulated the equilibrium conformational dynamics in an explicit point charge water model. Using the resulting trajectories based on MD simulations, we have computed the one and two dimensional infrared spectra in the Amide I region, using an exciton approach, in which a local mode basis of carbonyl stretches is considered. The role of solvent in shifting the Amide I band (by 30 to 50 cm-1 ) is clearly evident. Similarly, the conformational dynamics contribute to the broadening of peaks in the spectrum. The inhomogeneous broadening in both the 1D and 2D spectra reflects the significant conformational diversity observed in the simulations. Through the computed 2D cross-peak spectra, we show how different pulse schemes can provide additional information on the coupled vibrations. © 2016 The Authors. Journal of Computational Chemistry Published by Wiley Periodicals, Inc.