Stabilization of Charge-Transfer Excited States in Biological Systems: A Computational Focus on the Special Pair in Photosystem II Reaction Centers.

Charge-transfer (CT) excited states play an important role in many biological processes. However, many computational approaches often inadequately address the equilibration effects of nuclear and environmental degrees of freedom on these states. One prominent example of systems in which CT states are of utmost importance is reaction centers (RC) in photosystems. Here we use a multiscale approach combined with time-dependent density functional theory to explore the lowest CT excited state of the special pair PD1-PD2 in the Photosystem II-RC of a cyanobacterium. We find that the nonequilibrium CT excited state resides near the Soret band, making an exciton the lowest-energy excited state. However, accounting for nuclear and state-specific dielectric equilibration along the CT potential energy surface (PES), the CT state PD1--PD2+ stabilizes energetically below the excitonic state. This underscores the crucial role of state-specific solvation in mapping the PES of CT states, as demonstrated in a simplified dimer model.

Electric Field Susceptibility of Chlorophyll c Leads to Unexpected Excitation Dynamics in the Major Light-Harvesting Complex of Diatoms.

Diatoms are one of the most abundant photosynthetic organisms on earth and contribute largely to atmospheric oxygen production. They contain fucoxanthin and chlorophyll-a/c binding proteins (FCPs) as light-harvesting complexes with a remarkable adaptation to the fluctuating light on ocean surfaces. To understand the basis of the photosynthetic process in diatoms, the excitation energy funneling within FCPs must be probed. A state-of-the-art multiscale analysis within a quantum mechanics/molecular mechanics framework has been employed. To this end, the chlorophyll (Chl) excitation energies within the FCP complex from the diatom Phaeodactylum tricornutum have been determined. The Chl-c excitation energies were found to be 5-fold more susceptible to electric fields than those of Chl-a pigments and thus are significantly lower in FCP than in organic solvents. This finding challenges the general belief that the excitation energy of Chl-c is always higher than that of Chl-a in FCP proteins and reveals that Chl-c molecules are much more sensitive to electric fields within protein scaffolds than in Chl-a pigments. The analysis of the linear absorption spectrum and the two-dimensional electronic spectra of the FCP complex strongly supports these findings and allows us to study the excitation transfer within the FCP complex.

Antibiotic Charge Profile Determines the Extent of L3 Dynamics in OmpF: An Expedited Passage for Molecules with a Positive Charge.

Efficient permeation into Gram-negative bacterial cells is a much-desired property in the design of antibacterial agents. The goal is to arrive at one or more chemical modifications of molecules that improve their uptake into the cell while maintaining a good binding affinity to the intracellular target. Previously, we proposed a mechanistic rationale for the fast permeation of bulky antibiotics that involves induced conformational dynamics in the constriction loop L3 of the OmpF channel. This flexibility is caused by the perturbation of a hydrogen bond network stabilizing the L3 loop due to the strong interactions of the positively charged moiety on the antibiotic with the residues of the L3 loop. In the present work, we examine how differences in the charge profile of antibiotic molecules can affect the permeation process, in particular, the L3 dynamics. To this end, we have performed all-atom molecular dynamics simulations to study the permeation process of molecules with differences in the net charge through the Escherichia coli OmpF channel. The results from these simulations suggest that a positively charged moiety on the antibiotic is responsible for strong interactions with the negatively charged residues of the L3 loop, promoting conformational dynamics in the L3 loop. In contrast, antibiotics without a positively charged moiety are unable to initiate such a dynamic response in the L3 loop. This distinct behavior of the L3 loop in the presence of molecules with different charge characteristics provides a plausible mechanism whereby large molecules with an appropriate charge distribution can leverage an L3 dynamic-dependent pathway to permeate efficiently. The results are relevant to the structure-based design of molecules with improved uptake properties achieved through systematic chemical modifications that effectively engage the L3 loop.

Multifidelity Machine Learning for Molecular Excitation Energies.

The accurate but fast calculation of molecular excited states is still a very challenging topic. For many applications, detailed knowledge of the energy funnel in larger molecular aggregates is of key importance, requiring highly accurate excitation energies. To this end, machine learning techniques can be a very useful tool, though the cost of generating highly accurate training data sets still remains a severe challenge. To overcome this hurdle, this work proposes the use of multifidelity machine learning where very little training data from high accuracies is combined with cheaper and less accurate data to achieve the accuracy of the costlier level. In the present study, the approach is employed to predict vertical excitation energies to the first excited state for three molecules of increasing size, namely, benzene, naphthalene, and anthracene. The energies are trained and tested for conformations stemming from classical molecular dynamics and density functional based tight-binding simulations. It can be shown that the multifidelity machine learning model can achieve the same accuracy as a machine learning model built only on high-cost training data while expending a much lower computational effort to generate the data. The numerical gain observed in these benchmark test calculations was over a factor of 30 but certainly can be much higher for high-accuracy data.

Incoherent Energy Transfer between the Baseplate and FMO Protein Explored at Ideal Geometries.

The Förster resonance energy transfer (FRET) between the Fenna-Matthews-Olson (FMO) protein complex and the chlorosomal baseplate (CBP) is investigated by using an idealized model. This simplified model is based on crystal structure and molecular dynamics conformations. Some of the further input, such as the transition dipole moments, was extracted from earlier molecular-level simulations. The resulting model mimics the effects of the relative position between the CBP and the FMO complex on the corresponding FRET efficiency under ideal conditions, involving about 1.3 billion FRET calculations per investigated model. In this idealized model and employing some approximations, it is found that FRET efficiency is almost completely independent of the FMO trimer orientation (displacement, distance, and rotation), despite FMO and CBP being highly structured complexes. Even removing individual FMO BChl triples will only reduce the FRET efficiency by up to 8.6%. An FMO containing only the least efficient BChl triple will retain about 25% of the FRET efficiency of a full FMO complex. In addition to its proposed function as an energetic funnel, FMO is thus identified to act as a highly robust spatial funnel for CBP excitation harvesting, independent of the mutual CBP-FMO orientation.

Machine-learned correction to ensemble-averaged wave packet dynamics.

For a detailed understanding of many processes in nature involving, for example, energy or electron transfer, the theory of open quantum systems is of key importance. For larger systems, an accurate description of the underlying quantum dynamics is still a formidable task, and, hence, approaches employing machine learning techniques have been developed to reduce the computational effort of accurate dissipative quantum dynamics. A downside of many previous machine learning methods is that they require expensive numerical training datasets for systems of the same size as the ones they will be employed on, making them unfeasible to use for larger systems where those calculations are still too expensive. In this work, we will introduce a new method that is implemented as a machine-learned correction term to the so-called Numerical Integration of Schrödinger Equation (NISE) approach. It is shown that this term can be trained on data from small systems where accurate quantum methods are still numerically feasible. Subsequently, the NISE scheme, together with the new machine-learned correction, can be used to determine the dissipative quantum dynamics for larger systems. Furthermore, we show that the newly proposed machine-learned correction outperforms a previously handcrafted one, which, however, improves the results already considerably.

BtuB TonB-dependent transporters and BtuG surface lipoproteins form stable complexes for vitamin B12 uptake in gut Bacteroides.

Vitamin B12 (cobalamin) is required for most human gut microbes, many of which are dependent on scavenging to obtain this vitamin. Since bacterial densities in the gut are extremely high, competition for this keystone micronutrient is severe. Contrasting with Enterobacteria, members of the dominant genus Bacteroides often encode several BtuB vitamin B12 outer membrane transporters together with a conserved array of surface-exposed B12-binding lipoproteins. Here we show that the BtuB transporters from Bacteroides thetaiotaomicron form stable, pedal bin-like complexes with surface-exposed BtuG lipoprotein lids, which bind B12 with high affinities. Closing of the BtuG lid following B12 capture causes destabilisation of the bound B12 by a conserved BtuB extracellular loop, causing translocation of the vitamin to BtuB and subsequent transport. We propose that TonB-dependent, lipoprotein-assisted small molecule uptake is a general feature of Bacteroides spp. that is important for the success of this genus in colonising the human gut.

Correction: Benchmark and performance of long-range corrected time-dependent density functional tight binding (LC-TD-DFTB) on rhodopsins and light-harvesting complexes.

Correction for 'Benchmark and performance of long-range corrected time-dependent density functional tight binding (LC-TD-DFTB) on rhodopsins and light-harvesting complexes' by Beatrix M. Bold et al., Phys. Chem. Chem. Phys., 2020, 22, 10500-10518, https://doi.org/10.1039/C9CP05753F.

Fast prediction of antibiotic permeability through membrane channels using Brownian dynamics.

The efficient permeation across the Gram-negative bacterial membrane is an important step in the overall process of antibacterial action of a molecule and the one that has posed a significant hurdle on the way toward approved antibiotics. Predicting the permeability for a large library of molecules and assessing the effect of different molecular transformations on permeation rates of a given molecule is critical to the development of effective antibiotics. We present a computational approach for obtaining estimates of molecular permeability through a porin channel in a matter of hours using a Brownian dynamics approach. The fast sampling using a temperature acceleration scheme enables the approximate estimation of permeability using the inhomogeneous solubility diffusion model. Although the method is a significant approximation to similar all-atom approaches tested previously, we show that the present approach predicts permeabilities that correlate fairly well with the respective experimental permeation rates from liposome swelling experiments and accumulation rates from antibiotic accumulation assays, and is significantly, i.e., about 14 times, faster compared with a previously reported approach. The possible applications of the scheme in high-throughput screening for fast permeators are discussed.

Conformational Dynamics of Loop L3 in OmpF: Implications toward Antibiotic Translocation and Voltage Gating.

In the present work, we delineate the molecular mechanism of a bulky antibiotic permeating through a bacterial channel and uncover the role of conformational dynamics of the constriction loop in this process. Using the temperature accelerated sliced sampling approach, we shed light onto the dynamics of the L3 loop, in particular the F118 to S125 segment, at the constriction regions of the OmpF porin. We complement the findings with single channel electrophysiology experiments and applied-field simulations, and we demonstrate the role of hydrogen-bond stabilization in the conformational dynamics of the L3 loop. A molecular mechanism of permeation is put forward wherein charged antibiotics perturb the network of stabilizing hydrogen-bond interactions and induce conformational changes in the L3 segment, thereby aiding the accommodation and permeation of bulky antibiotic molecules across the constriction region. We complement the findings with single channel electrophysiology experiments and demonstrate the importance of the hydrogen-bond stabilization in the conformational dynamics of the L3 loop. The generality of the present observations and experimental results regarding the L3 dynamics enables us to identify this L3 segment as the source of gating. We propose a mechanism of OmpF gating that is in agreement with previous experimental data that showed the noninfluence of cysteine double mutants that tethered the L3 tip to the barrel wall on the OmpF gating behavior. The presence of similar loop stabilization networks in porins of other clinically relevant pathogens suggests that the conformational dynamics of the constriction loop is possibly of general importance in the context of antibiotic permeation through porins.

Importance of the lysine cluster in the translocation of anions through the pyrophosphate specific channel OprO.

Pseudomonas aeruginosa is a Gram-negative bacterium with an intrinsic resistance towards antibiotics due to the lack of a large diffusion pores. Exchange of substances with the environment is done mainly through a set of narrow and substrate-specific porins in its outer membrane that filter molecules according to their size and chemical composition. Among these proteins are OprP and OprO involved in the selective uptake of mono- and pyrophosphates, respectively. Both proteins are homotrimers and each monomer features an hourglass-shaped channel structure including a periplasmic cavity with a lysine cluster. In this study, we focus on the characterization of this lysine cluster in OprO. The importance of these lysine residues was shown with alanine substitutions in single channel conductance experiments, by titration of mono- and pyrophosphate in multi-channel analysis and by molecular dynamics simulations. All obtained data demonstrated that the closer the mutated lysine residues are to arginine 133, the lower gets the single channel conductance. It was found that the ion flow through each monomer can follow two different lysine paths indicating that phosphate ions have a larger freedom on the periplasmic side of the constriction region. Our results emphasize the important role of the lysine residue 121 in the binding site together with arginine 133 and aspartic acid 94. An improved understanding of the ion mobility across these channels can potentially lead to an optimized permeation of (phosphonic acid containing) antibiotics through the outer membrane of P. aeruginosa and the development of new drug molecules.

Recent progress in atomistic modeling of light-harvesting complexes: a mini review.

In this mini review, we focus on recent advances in the atomistic modeling of biological light-harvesting (LH) complexes. Because of their size and sophisticated electronic structures, multiscale methods are required to investigate the dynamical and spectroscopic properties of such complexes. The excitation energies, in this context also known as site energies, excitonic couplings, and spectral densities are key quantities which usually need to be extracted to be able to determine the exciton dynamics and spectroscopic properties. The recently developed multiscale approach based on the numerically efficient density functional tight-binding framework followed by excited state calculations has been shown to be superior to the scheme based on pure classical molecular dynamics simulations. The enhanced approach, which improves the description of the internal vibrational dynamics of the pigment molecules, yields spectral densities in good agreement with the experimental counterparts for various bacterial and plant LH systems. Here, we provide a brief overview of those results and described the theoretical foundation of the multiscale protocol.

Assembly of transmembrane pores from mirror-image peptides.

Tailored transmembrane alpha-helical pores with desired structural and functional versatility have promising applications in nanobiotechnology. Herein, we present a transmembrane pore DpPorA, based on the natural pore PorACj, built from D-amino acid α-helical peptides. Using single-channel current recordings, we show that DpPorA peptides self-assemble into uniform cation-selective pores in lipid membranes and exhibit properties distinct from their L-amino acid counterparts. DpPorA shows resistance to protease and acts as a functional nanopore sensor to detect cyclic sugars, polypeptides, and polymers. Fluorescence imaging reveals that DpPorA forms well-defined pores in giant unilamellar vesicles facilitating the transport of hydrophilic molecules. A second D-amino acid peptide based on the polysaccharide transporter Wza forms transient pores confirming sequence specificity in stable, functional pore formation. Finally, molecular dynamics simulations reveal the specific alpha-helical packing and surface charge conformation of the D-pores consistent with experimental observations. Our findings will aid the design of sophisticated pores for single-molecule sensing related technologies.

Identification of Single Amino Acid Chiral and Positional Isomers Using an Electrostatically Asymmetric Nanopore.

Chirality is essential in nearly all biological organizations and chemical reactions but is rarely considered due to technical limitations in identifying L/D isomerization. Using OmpF, a membrane channel from Escherichia coli with an electrostatically asymmetric constriction zone, allows discriminating chiral amino acids in a single peptide. The heterogeneous distribution of charged residues in OmpF causes a strong lateral electrostatic field at the constriction. This laterally asymmetric constriction zone forces the sidechains of the peptides to specific orientations within OmpF, causing distinct ionic current fluctuations. Using statistical analysis of the respective ionic current variations allows distinguishing the presence and position of a single amino acid with different chiralities. To explore potential applications, the disease-related peptide ß-Amyloid and its d-Asp1 isoform and a mixture of the icatibant peptide drug (HOE 140) and its d-Ser7 mutant have been discriminated. Both chiral isomers were not applicable to be distinguished by mass spectroscopy approaches. These findings highlight a novel sensing mechanism for identifying single amino acids in single peptides and even for achieving single-molecule protein sequencing.

Spectral densities and absorption spectra of the core antenna complex CP43 from photosystem II.

Besides absorbing light, the core antenna complex CP43 of photosystem II is of great importance in transferring excitation energy from the antenna complexes to the reaction center. Excitation energies, spectral densities, and linear absorption spectra of the complex have been evaluated by a multiscale approach. In this scheme, quantum mechanics/molecular mechanics molecular dynamics simulations are performed employing the parameterized density functional tight binding (DFTB) while the time-dependent long-range-corrected DFTB scheme is applied for the excited state calculations. The obtained average spectral density of the CP43 complex shows a very good agreement with experimental results. Moreover, the excitonic Hamiltonian of the system along with the computed site-dependent spectral densities was used to determine the linear absorption. While a Redfield-like approximation has severe shortcomings in dealing with the CP43 complex due to quasi-degenerate states, the non-Markovian full second-order cumulant expansion formalism is able to overcome the drawbacks. Linear absorption spectra were obtained, which show a good agreement with the experimental counterparts at different temperatures. This study once more emphasizes that by combining diverse techniques from the areas of molecular dynamics simulations, quantum chemistry, and open quantum systems, it is possible to obtain first-principle results for photosynthetic complexes, which are in accord with experimental findings.

Atomistic Simulation of Molecules Interacting with Biological Nanopores: From Current Understanding to Future Directions.

Biological nanopores have been at the focus of numerous studies due to their role in many biological processes as well as their (prospective) technological applications. Among many other topics, recent studies on nanopores have addressed two key areas: antibiotic permeation through bacterial channels and sensing of analytes. Although the two areas are quite far apart in terms of their objectives, in both cases atomistic simulations attempt to understand the solute dynamics and the solute-protein interactions within the channel lumen. While decades of studies on various channels have culminated in an improved understanding of the key molecular factors and led to practical applications in some cases, successful utilization is limited. In this Perspective we summarize recent progress in understanding key issues in molecular simulations of antibiotic translocation and in the development of nanopore sensors. Moreover, we comment on possible advancements in computational algorithms that can potentially resolve some of the issues.

Changes in Salt Concentration Modify the Translocation of Neutral Molecules through a ΔCymA Nanopore in a Non-monotonic Manner.

The voltage-dependent transport through biological and artificial nanopores is being used in many applications such as DNA or protein sequencing and sensing. The primary approach to determine the transport has been to measure the temporal ion current fluctuations caused by solutes when applying external voltages. Crossing the nanoscale confinement in the presence of an applied electric field primarily relies on two factors, i.e., the electrophoretic drag and electroosmosis. The electroosmotic flow (EOF) is a voltage-dependent ion-associated flow of solvent molecules, i.e., usually water, and depends on many factors, such as pH, temperature, pore diameter, and also the concentration of ions. The exact interplay between these factors is so far poorly understood. In this joint experimental and computational study, we have investigated the dependence of the EOF on the concentration of the buffer salt by probing the transport of α-cyclodextrin molecules through the ΔCymA channel. For five different KCl concentrations in the range between 0.125 and 2 M, we performed applied-field molecular dynamics simulations and analyzed the ionic flow and the EOF across the ΔCymA pore. To our surprise, the concentration-dependent net ionic flux changes non-monotonically and nonlinearly and the EOF is seen to follow the same pattern. On the basis of these findings, we were able to correlate the concentration-dependent EOF with experimental kinetic constants for the translocation of α-cyclodextrin through the ΔCymA nanopore. Overall, the results further improve our understanding of the EOF-mediated transport through nanopores and show that the EOF needs to seriously be taken into consideration when analyzing the permeation of (neutral) substrates through nanopores.

Coherent and incoherent contributions to molecular electron transport.

We numerically isolate the limits of validity of the Landauer approximation to describe charge transport along molecular junctions in condensed phase environments. To do so, we contrast Landauer with exact time-dependent non-equilibrium Green's function quantum transport computations in a two-site molecular junction subject to exponentially correlated noise. Under resonant transport conditions, we find Landauer accuracy to critically depend on intramolecular interactions. By contrast, under nonresonant conditions, the emergence of incoherent transport routes that go beyond Landauer depends on charging and discharging processes at the electrode-molecule interface. In both cases, decreasing the rate of charge exchange between the electrodes and molecule and increasing the interaction strength with the thermal environment cause Landauer to become less accurate. The results are interpreted from a time-dependent perspective where the noise prevents the junction from achieving steady-state and from a fully quantum perspective where the environment introduces dephasing in the dynamics. Using these results, we analyze why the Landauer approach is so useful to understand experiments, isolate regimes where it fails, and propose schemes to chemically manipulate the degree of transport coherence.

Permeation of Fosfomycin through the Phosphate-Specific Channels OprP and OprO of Pseudomonas aeruginosa.

Pseudomonas aeruginosa is a Gram-negative opportunistic pathogen responsible for many nosocomial infections. It is quite resistant to various antibiotics, caused by the absence of general diffusion pores in the outer membrane. Instead, it contains many substrate-specific channels. Among them are the two phosphate- and pyrophosphate-specific porins OprP and OprO. Phosphonic acid antibiotics such as fosfomycin and fosmidomycin seem to be good candidates for using these channels to enter P. aeruginosa bacteria. Here, we investigated the permeation of fosfomycin through OprP and OprO using electrophysiology and molecular dynamics (MD) simulations. The results were compared to those of the fosmidomycin translocation, for which additional MD simulations were performed. In the electrophysiological approach, we noticed a higher binding affinity of fosfomycin than of fosmidomycin to OprP and OprO. In MD simulations, the ladder of arginine residues and the cluster of lysine residues play an important role in the permeation of fosfomycin through the OprP and OprO channels. Molecular details on the permeation of fosfomycin through OprP and OprO channels were derived from MD simulations and compared to those of fosmidomycin translocation. In summary, this study demonstrates that the selectivity of membrane channels can be employed to improve the permeation of antibiotics into Gram-negative bacteria and especially into resistant P. aeruginosa strains.

Robust Strategy for Photoprotection in the Light-Harvesting Antenna of Diatoms: A Molecular Dynamics Study.

Diatoms generate a large portion of the oxygen produced on earth due to their exceptional light-harvesting properties involving fucoxanthin and chlorophyll-binding proteins (FCP). At the same time, an efficient adaptation of these complexes to fluctuating light conditions is necessary to protect the diatoms against photodamage. So far, structural and dynamic data for the interaction between FCP and the photoprotective LHCX family of proteins in diatoms are lacking. In this computational study, we provide a structural basis for a remarkable pH-dependent adaptation at the molecular level. Upon binding of the LHCX1 protein to the FCP complex together with a change in pH, conformational changes within the FCP protein result in a variation of the electronic coupling in a specific chlorophyll-fucoxanthin pair, leading to a change in the exciton transfer rate by almost an order of magnitude. A common strategy for photoprotection between diatoms and higher plants is identified and discussed.