Three-Photon Infrared Stimulation of Endogenous Neuroreceptors in Vivo.

To interrogate neural circuits and crack their codes, in vivo brain activity imaging must be combined with spatiotemporally precise stimulation in three dimensions using genetic or pharmacological specificity. This challenge requires deep penetration and focusing as provided by infrared light and multiphoton excitation, and has promoted two-photon photopharmacology and optogenetics. However, three-photon brain stimulation in vivo remains to be demonstrated. We report the regulation of neuronal activity in zebrafish larvae by three-photon excitation of a photoswitchable muscarinic agonist at 50 pM, a billion-fold lower concentration than used for uncaging, and with mid-infrared light of 1560 nm, the longest reported photoswitch wavelength. Robust, physiologically relevant photoresponses allow modulating brain activity in wild-type animals with spatiotemporal and pharmacological precision. Computational calculations predict that azobenzene-based ligands have high three-photon absorption cross-section and can be used directly with pulsed infrared light. The expansion of three-photon pharmacology will deeply impact basic neurobiology and neuromodulation phototherapies.

On the Computational Design of Azobenzene-Based Multi-State Photoswitches.

In order to theoretically design multi-state photoswitches with specific properties, an exhaustive computational study is first carried out for an azobenzene dimer that has been recently synthesized and experimentally studied. This study allows for a full comprehension of the factors that govern the photoactivated isomerization processes of these molecules so to provide a conceptual/computational protocol that can be applied to generic multi-state photoswitches. From this knowledge a new dimer with a similar chemical design is designed and also fully characterized. Our theoretical calculations predict that the new dimer proposed is one step further in the quest for a double photoswitch, where the four metastable isomers could be selectively interconverted through the use of different irradiation sequences.

Predicting the Electronic Absorption Band Shape of Azobenzene Photoswitches.

Simulations based on molecular dynamics coupled to excitation energy calculations were used to generate simulated absorption spectra for a family of halide derivatives of azobenzene, a family of photoswitch molecules with a weak absorption band around 400-600 nm and potential uses in living tissue. This is a case where using the conventional approach in theoretical spectroscopy (estimation of absorption maxima based on the vertical transition from the potential energy minimum on the ground electronic state) does not provide valid results that explain how the observed band shape extends towards the low energy region of the spectrum. The method affords a reasonable description of the main features of the low-energy UV-Vis spectra of these compounds. A bathochromic trend was detected linked to the size of the halide atom. Analysis of the excitation reveals a correlation between the energy of the molecular orbital where excitation starts and the energy of the highest occupied atomic orbital of the free halide atom. This was put to the test with a new brominated compound with good results. The energy level of the highest occupied orbital on the free halide was identified as a key factor that strongly affects the energy gap in the photoswitch. This opens the way for the design of bathochromically shifted variants of the photoswitch with possible applications.

A high-throughput computational approach to UV-Vis spectra in protein mutants.

In this work we present a high-throughput approach to the computation of absorption UV-Vis spectra tailored to mutagenesis studies. The scheme makes use of a single molecular dynamics trajectory of a reference (non-mutated) species. The shifts in absorption energy caused by a residue mutation are evaluated by building an effective potential of the environment and computing a correction term based on perturbation theory. The sampling is only performed in the phase space of the initial protein. We analyze the robustness of the method by comparing different approximations for the effective potential, the sampling of mutant residue geometries and observing the impact in the prediction of both bathocromic and hypsochromic shifts. As a test subject, we consider a red fluorescent protein variant with potential biotechnological applications.

Deciphering the grounds of the suitability of acylhydrazones as efficient photoswitches.

Many efforts are currently being devoted to designing molecular photoswitches with specific properties. In this sense, a recent publication [D. J. van Dijken et al., J. Am. Chem. Soc., 2015, 137, 14982-14991] has synthesized and analyzed the photochromic properties of a large set of acylhydrazones (ACHs), a relatively unexploited class of potential photoswitches with two stable E and Z isomers. This study has revealed a very diverse and complex pattern of the absorption/emission properties of ACHs depending on the substituents attached to the ACH motif. In this work, high level theoretical calculations are performed on a representative set of the experimentally studied ACHs in order to analyze, at the molecular level, the reasons behind the different photochemistries experimentally observed. This systematic study allows for the classification of the full set of ACHs into just four categories. The two more common groups display a small, either positive or negative, shift of the maximum wavelength of absorption between the E and Z isomers. Less common, but far more interesting from a practical point of view, are the compounds that show a large (>100 nm) Stokes shift. This behavior may arise from two different situations. The most common one implies the possibility of an intramolecular proton transfer in the excited electronic state of the less stable Z isomer. The less likely scenario would also involve a loss of the azidic proton through an intermolecular proton transfer that would take place with the aid of the solvent.

Rationally designed azobenzene photoswitches for efficient two-photon neuronal excitation.

Manipulation of neuronal activity using two-photon excitation of azobenzene photoswitches with near-infrared light has been recently demonstrated, but their practical use in neuronal tissue to photostimulate individual neurons with three-dimensional precision has been hampered by firstly, the low efficacy and reliability of NIR-induced azobenzene photoisomerization compared to one-photon excitation, and secondly, the short cis state lifetime of the two-photon responsive azo switches. Here we report the rational design based on theoretical calculations and the synthesis of azobenzene photoswitches endowed with both high two-photon absorption cross section and slow thermal back-isomerization. These compounds provide optimized and sustained two-photon neuronal stimulation both in light-scattering brain tissue and in Caenorhabditis elegans nematodes, displaying photoresponse intensities that are comparable to those achieved under one-photon excitation. This finding opens the way to use both genetically targeted and pharmacologically selective azobenzene photoswitches to dissect intact neuronal circuits in three dimensions.

Ultrafast action chemistry in slow motion: atomistic description of the excitation and fluorescence processes in an archetypal fluorescent protein.

We report quantum mechanical/molecular mechanical non-adiabatic molecular dynamics simulations on the electronically excited state of green fluorescent protein mutant S65T/H148D. We examine the driving force of the ultrafast (τ < 50 fs) excited-state proton transfer unleashed by absorption in the A band at 415 nm and propose an atomistic description of the two dynamical regimes experimentally observed [Stoner Ma et al., J. Am. Chem. Soc., 2008, 130, 1227]. These regimes are explained in terms of two sets of successive dynamical events: first the proton transfers quickly from the chromophore to the acceptor Asp148. Thereafter, on a slower time scale, there are geometrical changes in the cavity of the chromophore that involve the distance between the chromophore and Asp148, the planarity of the excited-state chromophore, and the distance between the chromophore and Tyr145. We find two different non-radiative relaxation channels that are operative for structures in the reactant region and that can explain the mismatch between the decay of the emission of A* and the rise of the emission of I*, as well as the temperature dependence of the non-radiative decay rate.

Chromophore interactions leading to different absorption spectra in mNeptune1 and mCardinal red fluorescent proteins.

Extensive MD simulations combined with QM/MM calculations have been performed on mNeptune1 and mCardinal red fluorescent proteins to establish the reasons behind the red shift of the excitation wavelength of mCardinal with respect to mNeptune1. In both cases, it is seen that Arg197 stabilizes the chromophore but cannot be described as stabilizing preferentially the excited state because of the anchor point of the interaction. The interactions of the linking bonds to the α-helix of both proteins to the chromophore have been analyzed. It has been found that, besides the presence of a strategically placed residue Gln41 in mCardinal, solvation water molecules play an active role in the energetics of the stabilization of the excited state, which is preferentially stabilized in the case of mCardinal in contrast to mNeptune1.

The Quest for Photoswitches Activated by Near-Infrared Light: A Theoretical Study of the Photochemistry of BF2 -Coordinated Azo Derivatives.

Recently synthesized BF2 -coordinated azo derivatives have been proposed as photoswitches that operate in the optical window (λ=600-1200 nm) for use in bioimaging applications. Herein, we have theoretically analyzed these compounds and modified some substituents to analyze which properties of the molecule govern its photochemistry. Our results compare rather well with the available experimental data, so our methodology, based on density functional theory (DFT) calculations for the ground electronic state and time-dependent-DFT for the first excited electronic state, is validated. Through systematic modification of different substituents of the parent system, we designed compounds that are predicted to operate fully within the optical window. We also analyzed several molecules for which the cis isomer is the more stable isomer, a quite unusual result for azobenzene derivatives that is a much coveted property for some applications of these photoactive molecules in pharmacology. Our results also provide insight into other properties relevant for photoswitches, such as the thermal stability of the less stable isomer and the magnitude of the gap between the wavelengths of the radiation that activates each isomerization process, which must be as large as possible to improve the yield of each photoisomerization. From a more general perspective, our results may provide a step towards the rational design of new photoswitches that fulfill a set of desired characteristics.

Molecular modelling of the pH influence in the geometry and the absorbance spectrum of near-infrared TagRFP675 fluorescent protein.

Classical molecular dynamics (MD) simulations are carried out for the recently developed TagRFP675 fluorescent protein (FP), which is specifically designed to fully absorb and emit in the near infrared (NIR) region of the electromagnetic spectrum. Since the X-ray data of TagRFP675 reveal that the chromophore exists in both the cis and trans configuration and it can also be neutral (protonated) or anionic (deprotonated) depending on the pH of the media, a total of 8 molecular dynamic simulations have been run to simulate all the possible states of the chromophore. Time-dependent DFT (TDDFT) single point calculations are performed at selected points along the simulation to theoretically mimic the absorption spectrum of the protein. Our simulations compare well (within the expected error of the computational method) with the experimental results. Our theoretical procedure allows for an analysis of the molecular orbitals involved in the lowest energy electronic excitations of the chromophore and, more interestingly, for a full analysis of the H-bond interactions between the chromophore and its surrounding residues and solvent (water) molecules. This study does not support the hypothesis, exclusively based on the analysis of X-ray data, that the isomerization of nearby residues provokes the rearrangement of the hydrogen bonds in the chromophore's immediate environment leading to the observed red shift of the absorption bands at higher pHs. Instead, we attribute this shift mainly to the superposition of bands of the neutral and anionic chromophores that are expected to coexist at almost the full range of pHs experimentally analyzed. An additional factor that could contribute to this shift is the experimentally observed increase of the cis configuration of the chromophore at higher pHs.

Transient low-barrier hydrogen bond in the photoactive state of green fluorescent protein.

In this paper, we have analyzed the feasibility of spontaneous proton transfer in GFP at the Franck-Condon region directly after photoexcitation. Computation of a sizeable portion of the potential energy surface at the Franck-Condon region of A the structure shows the process of proton transfer to be unfavorable by 3 kcal mol(-1) in S1 if no further structural relaxation is permitted. The ground vibrational state is found to lie above the potential energy barrier of the proton transfer in both S0 and S1. Expectation values of the geometry reveal that the proton shared between the chromophore and W22, and the proton shared between Ser205 and Glu222 are very close to the center of the respective hydrogen bonds, giving support to the claim that the first transient intermediate detected after photoexcitation (I0*) has characteristics similar to those of a low-barrier hydrogen bond [Di Donato et al., Phys. Chem. Chem. Phys., 2012, 13, 16295]. A quantum dynamical calculation of the evolution in the excited state shows an even larger probability of finding those two protons close to the center compared to in the ground state, but no formation of the proton-transferred product is observed. A QM/MM photoactive state geometry optimization, initiated using a configuration obtained by taking the A minimum and moving the protons to the product side, yields a minimum energy structure with the protons transferred and in which the His148 residue is substantially closer to the now anionic chromophore. These results indicate that: (1) proton transfer is not possible if structural relaxation of the surroundings of the chromophore is prevented; (2) protons H1 and H3 especially are found very close to the point halfway between the donor and acceptor after photoexcitation when the zero-point energy is considered; (3) a geometrical parameter exists (the His148-Cro distance) under which the structure with the protons transferred is not a minimum, and that, if included, should lead to the fluorescing I* structure. The existence of an oscillating stationary state between the reactants and products of the triple proton transfer reaction can explain the dual emission reported for the I0* intermediate of wtGFP.

Theoretical Computer-Aided Mutagenic Study on the Triple Green Fluorescent Protein Mutant S65T/H148D/Y145F.

Green fluorescent protein (GFP) mutant S65T/H148D has been proposed to host a photocycle that involves an excited-state proton transfer between the chromophore (Cro) and the Asp148 residue and takes place in less than 50 fs without a measurable kinetic isotope effect. It has been suggested that the interaction between the unsuspected Tyr145 residue and the chromophore is needed for the ultrafast sub-50 fs rise in fluorescence. To verify this, we have performed a computer-aided mutagenic study to introduce the additional mutation Y145F, which eliminates this interaction. By means of QM/MM molecular dynamics simulations and time-dependent density functional theory studies, we have assessed the importance of the Cro-Tyr145 interaction and the solvation of Asp148 and shown that in the triple mutant S65T/H148D/Y145F a significant loss in the ultrafast rise of the Stokes-shifted fluorescence should be expected.

Unveiling how an archetypal fluorescent protein operates: theoretical perspective on the ultrafast excited state dynamics of GFP variant S65T/H148D.

Green fluorescent protein variant S65T/H148D has been reported to host a photocycle involving the photoinduced proton transfer reaction between the chromophore and residue Asp148 under 50 fs and without a measurable kinetic isotope effect, and experimental evidence is suggestive of the existence of a highly delocalized proton between these residues. The blinding speed at which this biological system undergoes proton transfer has been ascribed to the extreme increase of acidity of the GFP chromophore in the electronic excited state where proton transfer takes place. This work strives to present a coherent, complete, and balanced description of the dynamics of this specific variant of GFP in which it will be shown that this increase of acidity is insufficient to explain the behavior observed. This study tracks the behavior of this photosystem to the delicate interplay between structure and dynamics shown in the presence of solvent. In this way, it has been found that the dynamics of this protein intertwines its structure with the intervening solvent to give rise to effectively degenerate situations in what concerns the reactants and products of the proton transfer reaction in ground and, most importantly, photoexcited state, in terms of potential energy profiles associated with the proton migration. Under these conditions, proton transfer can occur in accordance with the experimental data available. This set of characteristics is possibly common to a host of other proton transfer based fluorescent proteins, and helps promoting GFP S65T/H148D to a case of archetypal significance. Thus, our results can be useful to understand the way many fluorescent proteins work and, more generally, the molecular basis for proton transfer reactions in proteins.

New insights into the structure-spectrum relationship in S65T/H148D and E222Q/H148D green fluorescent protein mutants: a theoretical assessment.

The green fluorescent protein (GFP) variant S65T/H148D recovers the A-band fluorescence lost in the single mutant S65T, and it has been established that Asp148 is the alternate proton acceptor for the excited state proton transfer (ESPT). This mutant has been widely studied and presents unique spectroscopic properties, such as an ultrafast rise in the fluorescence (<50 fs). Also it exhibits a red-shift of the A absorption band of 20 nm with respect to wt-GFP's. The double mutant E222Q/H148D presents a very similar behaviour, at least within the experimental data available (which is scarcer than those of S65T/H148D). By means of dynamic theoretical studies we have been able to (1) reproduce and thoroughly analyse the red-shifted absorption spectra of both mutants and (2) predict the structure that the variant E222Q/H148D (for which there is no X-ray-resolved structure available) most probably adopts in water at room temperature. Our results deepen the understanding of the way GFP variants work and give some new insights into the rational design of fluorescent proteins and biological photosystems in general.

Are there really low-barrier hydrogen bonds in proteins? The case of photoactive yellow protein.

For a long time, low-barrier hydrogen bonds (LBHBs) have been proposed to exist in many enzymes and to play an important role in their catalytic function, but the proof of their existence has been elusive. The transient formation of an LBHB in a protein system has been detected for the first time using neutron diffraction techniques on a photoactive yellow protein (PYP) crystal in a study published in 2009 (Yamaguchi, S.; et al. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 440-444). However, very recent theoretical studies based on electronic structure calculations and NMR resonance experiments on PYP in solution (Saito, K.; et al. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 167-172) strongly indicate that there is not such an LBHB. By means of electronic structure calculations combined with the solution of the nuclear Schrödinger equation, we analyze here under which conditions an LBHB can exist in PYP, thus leading to a more reasonable and conciliating understanding of the above-mentioned studies.

A theoretical study of the photochemistry of indigo in its neutral and dianionic (leucoindigo) forms.

A thorough analysis of the single and double proton transfer and the internal rotations of neutral indigo and its dianionic leucoindigo form has been performed for the ground and first singlet excited electronic states using, respectively, DFT and TDDFT state-of-the-art methods. Our theoretical analysis discloses that the diketo isomer is the most stable one in the ground state of indigo but not in leucoindigo where the dienol minimum is more stable. Single and double proton transfer processes are not energetically favored in the ground electronic state but a single proton transfer gives a more stable tautomer in the excited electronic state of indigo whereas a double proton transfer is energetically favorable in the excited state of leucoindigo. The internal rotations are not thermodynamically allowed except for the keto-enol tautomer where a full rotation of the inter-ring C-C bond leads to another stable keto-enol structure. A preliminary analysis of the plausible conical intersections for both indigo and leucoindigo allows the discussion of the likely deactivation paths that will follow light irradiation. Our results point to a very different photochemistry of the two molecules. For indigo the proton transfer can only take place through tunneling so the main deactivation path would involve a conical intersection accessed directly upon internal rotation of the keto-keto tautomer. For leucoindigo a richer photochemistry is expected as the single and double proton transfer processes are energetically open. The more favorable path involves single proton transfer followed by a trans to cis isomerization and a second proton transfer. This process competes on equal grounds with several non-radiative decays through conical intersections. All these results are in agreement with the experimental facts known to date.

How Does the Environment Affect the Absorption Spectrum of the Fluorescent Protein mKeima?

The absorption spectrum of a fluorescent protein is determined by its chromophore, but the residues that surround it also have a remarkable role, leading to noticeable spectral shifts. We have theoretically analyzed the monomeric protein Keima (mKeima), a red fluorescent protein most remarkable for an outstanding difference between the absorption and emission frequencies, and potentially suited for multicolor imaging applications. In the present work, we have performed excited state electronic calculations on the chromophore with an increasing number of atoms surrounding it, and we have compared these results with the excited states calculations on an ensemble of structures obtained from a molecular dynamics simulation of the complete protein. The importance of the inclusion of the effects of the whole protein in the electronic calculations has been proved, and it is concluded that only with the consideration of the thermal effects can the absorption spectra of the protein be properly characterized.

Peek at the potential energy surfaces of the LSSmKate1 and LSSmKate2 proteins.

To determine the energetic feasibility of the mechanisms involved in the generation of the fluorescent species in red fluorescent proteins LSSmKate1 and LSSmKate2 developed by Piatkevich et al. (Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 5369-5374 and J. Am. Chem. Soc. 2010, 132, 10762-10770), a potential energy scan for the respective reaction coordinates was performed in large cluster models including the surroundings of the chromophores, based on the respective crystallographic structures, using DFT and TDDFT. The predicted absorption wavelengths agree to within 5 nm with experiment, thus confirming the accuracy of the calculational level and modeling done. In both proteins, it was found that the adiabatic electronic state with the largest oscillator strength at the Franck-Condon region was not the one from which fluorescence could occur in the products. A diabatization procedure was used to determine an approximate photoactive state, based on selecting the state with the largest oscillator strength throughout. For LSSmKate1, this led to a rather flat potential energy profile but still did not predict a minimum in the product side. It is suggested that relaxation processes, absent from the model, could bring about such a minimum. LSSmKate2, on the other hand, clearly displays a favorable exoergic process in the photoactive state, and its double-proton transfer can be described as concerted but highly asynchronous, involving a barrier in the transfer of the first proton. In this way, the model provides strong support for the mechanism proposed for LSSmKate2.

Photo-deactivation pathways of a double H-bonded photochromic Schiff base investigated by combined theoretical calculations and experimental time-resolved studies.

The photophysics of N,N'-bis(salicylidene)-p-phenylenediamine (BSP) is analyzed both theoretically and experimentally. The alternative intramolecular proton-transfer reactions lead to three different tautomers. We performed DFT and TDDFT calculations to analyze the topography of the reactions connecting the three tautomers. Deactivation paths through a Conical Intersection (CI) region are also analyzed to explain the low fluorescence quantum yield of the phototautomers. The complex molecular structure of BSP provides a large number of deactivation paths, almost all of them energetically available following the initial photoexcitation. Femtosecond (fs) time-resolved emission studies in solution and flash photolysis experiments (nano to millisecond regime) were performed to get detailed information on the time domain of the full photocycle. The picture that emerges by combining theoretical and experimental results shows a very fast (less than 100 fs) photoinduced single proton transfer process leading to a phototautomer where a single proton has moved. This species may deactivate through a low-energy CI leading in about 20 ps to a rotameric form in the ground state that has a lifetime of several tens of microseconds in solution. This process competes with another deactivation path taking place prior to the proton-transfer reaction which involves a low-energy CI leading to a rotamer of the enol structure. In the flash photolysis studies, the rotamer of the enol structure was directly identified by the positive transient absorption band in the 250-260 nm and its lifetime in n-hexane (10 ms) is almost 3 orders of magnitude longer than the lifetime of the photochrome (around 40 µs). Our findings do not exclude a double proton transfer reaction in the excited enol form to give a tautomer in less than 100 fs during the first (impulsive) phase of the reaction which reverts back to the photoproducts of the simple proton transfer in 1-3 ps.

A method to compute probability current in generic coordinates.

A method to compute probability current and its surface integral, the total flux, for systems of many particles of different masses is presented, based on transforming the wave function and its gradient onto a mass-weighted coordinate system. As a test for this methodology, it has been applied to a nontrivial 6-dimensional quantum dynamics study of a model of the operation of the proton-wire in Green Fluorescent Protein [O. Vendrell, R. Gelabert, M. Moreno, and J. M. Lluch, J. Phys. Chem. B, 112, 5500-5511 (2008)]. An adaptive Monte Carlo method is proposed, with favorable scaling properties for future applications, to solve the flux integral. Comparison of total reactive flux with the time derivative of the survival probability is satisfactory, corroborating the adequacy of the derivation. Using the new method the flux can quantitatively be divided into its positive and negative contributions, or more relevantly, into tunneling and classical parts.