Unraveling the role of thermal fluctuations on the exciton structure of the cryptophyte PC612 and PC645 photosynthetic antenna complexes.

Protein scaffolds play a crucial role in tuning the light harvesting properties of photosynthetic pigment-protein complexes, influencing pigment-protein and pigment-pigment excitonic interactions. Here, we investigate the influence of thermal dynamic effects on the protein tuning mechanisms of phycocyanin PC645 and PC612 antenna complexes of cryptophyte algae, featuring closed or open quaternary structures. We employ a dual molecular dynamics (MD) strategy that combines extensive classical MD simulations with multiple short Born-Oppenheimer quantum/molecular mechanical (QM/MM) simulations to accurately account for both static and dynamic disorder effects. Additionally, we compare the results with an alternative protocol based on multiple QM/MM geometry optimizations of the pigments. Subsequently, we employ polarizable QM/MM calculations using time-dependent density functional theory (TD-DFT) to compute the excited states, and we adopt the full cumulant expansion (FCE) formalism to describe the absorption and circular dichroism spectra. Our findings indicate that thermal effects have only minor impacts on the energy ladder in PC612, despite its remarkable flexibility owing to an open quaternary structure. In striking contrast, thermal effects significantly influence the properties of PC645 due to the absence of a hydrogen bond controlling the twist of ring D in PCB ß82 bilins, as well as the larger impact of fluctuations on the excited states of MBV pigments, which possess a higher conjugation length compared to other bilin types. Overall, the dual MD protocol combined with the FCE formalism yields excellent spectral properties for PC612 and PC645, and the resultant excitonic Hamiltonians pave the way for future investigations concerning the implications of open and closed quaternary structures on phycocyanin light harvesting properties.

Prediction of toluene/water partition coefficients in the SAMPL9 blind challenge: assessment of machine learning and IEF-PCM/MST continuum solvation models.

In recent years the use of partition systems other than the widely used biphasic n-octanol/water has received increased attention to gain insight into the molecular features that dictate the lipophilicity of compounds. Thus, the difference between n-octanol/water and toluene/water partition coefficients has proven to be a valuable descriptor to study the propensity of molecules to form intramolecular hydrogen bonds and exhibit chameleon-like properties that modulate solubility and permeability. In this context, this study reports the experimental toluene/water partition coefficients (log Ptol/w) for a series of 16 drugs that were selected as an external test set in the framework of the Statistical Assessment of the Modeling of Proteins and Ligands (SAMPL) blind challenge. This external set has been used by the computational community to calibrate their methods in the current edition (SAMPL9) of this contest. Furthermore, the study also investigates the performance of two computational strategies for the prediction of log Ptol/w. The first relies on the development of two machine learning (ML) models, which are built up by combining the selection of 11 molecular descriptors in conjunction with either the multiple linear regression (MLR) or the random forest regression (RFR) model to target a dataset of 252 experimental log Ptol/w values. The second consists of the parametrization of the IEF-PCM/MST continuum solvation model from B3LYP/6-31G(d) calculations to predict the solvation free energies of 163 compounds in toluene and benzene. The performance of the ML and IEF-PCM/MST models has been calibrated against external test sets, including the compounds that define the SAMPL9 log Ptol/w challenge. The results are used to discuss the merits and weaknesses of the two computational approaches.

Benchmarking DFT functionals for photophysics of pyranoflavylium cations.

An intense absorption, phosphorescence, a long triplet excited state lifetime and singlet oxygen generation capabilities are characteristics of pyranoflavylium cations, analogues to pyranoanthocyanidins originated in the maturation process of red wine. Such properties make these compounds potential photosensitizers to be applied in photodynamic therapy. In this context, the photophysical processes underlying that treatment critically depend on the electronic structure of the pyranoflavylium molecules. When employing density functional theory to describe the electronic structure of molecules, the choice of the most suitable functional is not trivial, and benchmark studies are needed to orient practitioners in the field. In this work, a benchmark of seven of the most commonly used density functionals in addressing the photophysical properties of a set of eight pyranoflavylium cations is reported. Ground and excited state geometries, molecular orbitals, and absorption, fluorescence and phosphorescence transition energies were calculated using density functional theory approaches, and evaluated and compared to experimental data and monoreferential wave function-based methodologies. Statistical analysis of the results indicates that global-hybrid functionals allow an excellent description of absorption and emission energies, with errors around 0.05 eV, while range-separated variants led to somewhat larger errors in the range 0.1-0.2 eV. In contrast, range-separated functionals display excellent phosphorescence energies with errors close to 0.05 eV, in this case global-hybrids showing increased discrepancies around 0.5-0.1 eV.

Engineering excitonically coupled dimers in an artificial protein for light harvesting via computational modeling.

In photosynthesis, pigment-protein complexes achieve outstanding photoinduced charge separation efficiencies through a set of strategies in which excited states delocalization over multiple pigments ("excitons") and charge-transfer states play key roles. These concepts, and their implementation in bioinspired artificial systems, are attracting increasing attention due to the vast potential that could be tapped by realizing efficient photochemical reactions. In particular, de novo designed proteins provide a diverse structural toolbox that can be used to manipulate the geometric and electronic properties of bound chromophore molecules. However, achieving excitonic and charge-transfer states requires closely spaced chromophores, a non-trivial aspect since a strong binding with the protein matrix needs to be maintained. Here, we show how a general-purpose artificial protein can be optimized via molecular dynamics simulations to improve its binding capacity of a chlorophyll derivative, achieving complexes in which chromophores form two closely spaced and strongly interacting dimers. Based on spectroscopy results and computational modeling, we demonstrate each dimer is excitonically coupled, and propose they display signatures of charge-transfer state mixing. This work could open new avenues for the rational design of chromophore-protein complexes with advanced functionalities.

Prediction of n-octanol/water partition coefficients and acidity constants (pKa) in the SAMPL7 blind challenge with the IEFPCM-MST model.

Within the scope of SAMPL7 challenge for predicting physical properties, the Integral Equation Formalism of the Miertus-Scrocco-Tomasi (IEFPCM/MST) continuum solvation model has been used for the blind prediction of n-octanol/water partition coefficients and acidity constants of a set of 22 and 20 sulfonamide-containing compounds, respectively. The log P and pKa were computed using the B3LPYP/6-31G(d) parametrized version of the IEFPCM/MST model. The performance of our method for partition coefficients yielded a root-mean square error of 1.03 (log P units), placing this method among the most accurate theoretical approaches in the comparison with both globally (rank 8th) and physical (rank 2nd) methods. On the other hand, the deviation between predicted and experimental pKa values was 1.32 log units, obtaining the second best-ranked submission. Though this highlights the reliability of the IEFPCM/MST model for predicting the partitioning and the acid dissociation constant of drug-like compounds compound, the results are discussed to identify potential weaknesses and improve the performance of the method.

On the Binding of Congo Red to Amyloid Fibrils.

Amyloids are characterized by their capacity to bind Congo red (CR), one of the most used amyloid-specific dyes. The structural features of CR binding were unknown for years, mainly because of the lack of amyloid structures solved at high resolution. In the last few years, solid-state NMR spectroscopy enabled the determination of the structural features of amyloids, such as the HET-s prion forming domain (HET-s PFD), which also has recently been used to determine the amyloid-CR interface at atomic resolution. Herein, we combine spectroscopic data with molecular docking, molecular dynamics, and excitonic quantum/molecular mechanics calculations to examine and rationalize CR binding to amyloids. In contrast to a previous assumption on the binding mode, our results suggest that CR binding to the HET-s PFD involves a cooperative process entailing the formation of a complex with 1:1 stoichiometry. This provides a molecular basis to explain the bathochromic shift in the maximal absorbance wavelength when CR is bound to amyloids.

Prediction of the n-octanol/water partition coefficients in the SAMPL6 blind challenge from MST continuum solvation calculations.

The IEFPCM/MST continuum solvation model is used for the blind prediction of n-octanol/water partition of a set of 11 fragment-like small molecules within the SAMPL6 Part II Partition Coefficient Challenge. The partition coefficient of the neutral species (log P) was determined using an extended parametrization of the B3LYP/6-31G(d) version of the Miertus-Scrocco-Tomasi continuum solvation model in n-octanol. Comparison with the experimental data provided for partition coefficients yielded a root-mean square error (rmse) of 0.78 (log P units), which agrees with the accuracy reported for our method (rmse = 0.80) for nitrogen-containing heterocyclic compounds. Out of the 91 sets of log P values submitted by the participants, our submission is within those with an rmse < 1 and among the four best ranked physical methods. The largest errors involve three compounds: two with the largest positive deviations (SM13 and SM08), and one with the largest negative deviations (SM15). Here we report the potentiometric determination of the log P for SM13, leading to a value of 3.62 ± 0.02, which is in better agreement with most empirical predictions than the experimental value reported in SAMPL6. In addition, further inclusion of several conformations for SM08 significantly improved our results. Inclusion of these refinements led to an overall error of 0.51 (log P units), which supports the reliability of the IEFPCM/MST model for predicting the partitioning of neutral compounds.

How abasic sites impact hole transfer dynamics in GC-rich DNA sequences.

Changes in DNA charge transfer properties upon the creation of apurinic and apyrimidinic sites have been used to monitor DNA repair processes, given that such lesions generally reduce charge transfer yields. However, because these lesions translate into distinct intra and extrahelical conformations depending on the nature of the unpaired base and its DNA context, it is unclear the actual impact of such diverse conformations on charge transfer. Here we combine classical molecular dynamics, quantum/molecular mechanics (QM/MM) calculations, and kinetic Monte Carlo simulations to investigate the impact of abasic sites on the structure and hole transfer (HT) properties of DNA. We consider both apurinic and apyrimidinic sites in polyG and polyGC sequences and find that most situations lead to intrahelical conformations where HT rates are significantly slowed down due to the energetic disorder induced by the abasic void. In contrast, the presence of an unpaired C flanked by C bases leads to an extrahelical conformation where stacking among G sites is reduced, leading to an attenuation of electronic couplings and a destabilization of hole states. Interestingly, this leads to an asymmetric HT behavior, given that the 5' to 3' transfer along the G strand is slowed down by one order of magnitude while the opposite 3' to 5' transfer remains similar to that estimated for the reference polyG sequence. Our simulations thus suggest that electrochemical monitoring of the DNA repair process following changes in charge transfer properties can miss repair events linked to abasic sites adopting extrahelical conformations.

Determination of the protonation preferences of bilin pigments in cryptophyte antenna complexes.

The light-harvesting mechanisms of cryptophyte antenna complexes have attracted considerable attention due to their ability to exhibit maximal photosynthetic activity under very low-light conditions and to display several colors, as well as the observation of vibronic coherent features in their two-dimensional electronic spectra. However, detailed investigations on the interplay between the protein environment and their light-harvesting properties are hampered by the uncertainty related to the protonation state of the underlying bilin pigments. Here we study the protonation preferences of four types of bilin pigments including 15,16-dihydrobiliverdin (DBV), phycoerythrobilin (PEB), phycocyanobilin (PCB) and mesobiliverdin (MBV), which are found in phycoerythrin PE545 and phycocyanin PC577, PC612, PC630 and PC645 complexes. We apply quantum chemical calculations coupled to continuum solvation calculations to predict the intrinsic acidity of bilins in aqueous solution, and then combine molecular dynamics simulations with empirical pKa estimates to investigate the impact of the local protein environment on the acidity of the pigments. We also report measurements of the absorption spectra of the five complexes in a wide range of pH in order to validate our simulations and investigate possible changes in the light harvesting properties of the complexes in the range of physiological pH found in the lumen (pH ∼ 5-7). The results suggest a pKa > 7 for DBV and MBV pigments in the α polypeptide chains of PE545 and PC630/PC645 complexes, which are not coordinated to a negatively charged amino acid. For the other PEB, DBV and PCB pigments, which interact with a Glu or Asp side chain, higher pKa values (pKa > 8) are estimated. Overall, the results support a preferential population of the fully protonated state for bilins in cryptophyte complexes under physiological conditions regardless of the specific type of pigment and local protein environment.

Density-Dependent Formulation of Dispersion-Repulsion Interactions in Hybrid Multiscale Quantum/Molecular Mechanics (QM/MM) Models.

Mixed multiscale quantum/molecular mechanics (QM/MM) models are widely used to explore the structure, reactivity, and electronic properties of complex chemical systems. Whereas such models typically include electrostatics and potentially polarization in so-called electrostatic and polarizable embedding approaches, respectively, nonelectrostatic dispersion and repulsion interactions are instead commonly described through classical potentials despite their quantum mechanical origin. Here we present an extension of the Tkatchenko-Scheffler semiempirical van der Waals (vdWTS) scheme aimed at describing dispersion and repulsion interactions between quantum and classical regions within a QM/MM polarizable embedding framework. Starting from the vdWTS expression, we define a dispersion and a repulsion term, both of them density-dependent and consistently based on a Lennard-Jones-like potential. We explore transferable atom type-based parametrization strategies for the MM parameters, based on either vdWTS calculations performed on isolated fragments or on a direct estimation of the parameters from atomic polarizabilities taken from a polarizable force field. We investigate the performance of the implementation by computing self-consistent interaction energies for the S22 benchmark set, designed to represent typical noncovalent interactions in biological systems, in both equilibrium and out-of-equilibrium geometries. Overall, our results suggest that the present implementation is a promising strategy to include dispersion and repulsion in multiscale QM/MM models incorporating their explicit dependence on the electronic density.

Prediction of pH-Dependent Hydrophobic Profiles of Small Molecules from Miertus-Scrocco-Tomasi Continuum Solvation Calculations.

Hydrophobicity is a key physicochemical descriptor used to understand the biological profile of (bio)organic compounds as well as a broad variety of biochemical, pharmacological, and toxicological processes. This property is estimated from the partition coefficient between aqueous and nonaqueous environments for neutral compounds (PN) and corrected for the pH-dependence of ionizable compounds as the distribution coefficient (D). Here, we have extended the parametrization of the Miertus-Scrocco-Tomasi continuum solvation model in n-octanol to nitrogen-containing heterocyclic compounds, as they are present in many biologically relevant molecules (e.g., purines and pyrimidines bases, amino acids, and drugs), to obtain accurate log PN values for these molecules. This refinement also includes solvation calculations for ionic species in n-octanol with the aim of reproducing the experimental partition of ionic compounds (PI). Finally, the suitability of different formalisms to estimate the distribution coefficient for a wide range of pH values has been examined for a set of small acidic and basic compounds. The results indicate that in general the simple pH-dependence model of the ionizable compound in water suffices to predict the partitioning at or around physiological pH. However, at extreme pH values, where ionic species are predominant, more elaborate models provide a better prediction of the n-octanol/water distribution coefficient, especially for amino acid analogues. Finally, the results also show that these formalisms are better suited to reproduce the experimental pH-dependent distribution curves of log D for both acidic and basic compounds as well as for amino acid analogues.

Structural and energetic study of cation-π-cation interactions in proteins.

Cation-π interactions of aromatic rings and positively charged groups are among the most important interactions in structural biology. The role and energetic characteristics of these interactions are well established. However, the occurrence of cation-π-cation interactions is an unexpected motif, which raises intriguing questions about its functional role in proteins. We present a statistical analysis of the occurrence, composition and geometrical preferences of cation-π-cation interactions identified in a set of non-redundant protein structures taken from the Protein Data Bank. Our results demonstrate that this structural motif is observed at a small, albeit non-negligible frequency in proteins, and suggest a preference to establish cation-π-cation motifs with Trp, followed by Tyr and Phe. Furthermore, we have found that cation-π-cation interactions tend to be highly conserved, which supports their structural or functional role. Finally, we have performed an energetic analysis of a representative subset of cation-π-cation complexes combining quantum-chemical and continuum solvation calculations. Our results point out that the protein environment can strongly screen the cation-cation repulsion, leading to an attractive interaction in 64% of the complexes analyzed. Together with the high degree of conservation observed, these results suggest a potential stabilizing role in the protein fold, as demonstrated recently for a miniature protein (Craven et al., J. Am. Chem. Soc. 2016, 138, 1543). From a computational point of view, the significant contribution of non-additive three-body terms challenges the suitability of standard additive force fields for describing cation-π-cation motifs in molecular simulations.

Can Förster Theory Describe Stereoselective Energy Transfer Dynamics in a Protein-Ligand Complex?

Förster resonance energy transfer (FRET) reactions involving ligands and aromatic amino acids can substantially impact the fluorescence properties of a protein-ligand complex, an impact intimately related to the corresponding binding mode. Structural characterization of such binding events in terms of intermolecular distances can be done through the well-known R-6 distance-dependent Förster rate expression. However, such an interpretation suffers from uncertainties underlying Förster theory in the description of the electronic coupling that promotes FRET, mostly related to the dipole-dipole orientation factor, dielectric screening effects, and deviations from the ideal dipole approximation. Here, we investigate how Förster approximations impact the prediction of energy transfer dynamics in the complex between flurbiprofen (FBP) and human serum albumin (HSA), as well as a model FBP-Trp dyad, in which recent observation of enantioselective fluorescence quenching has been ascribed to energy transfer from FBP to Trp. To this end, we combine classical molecular dynamics simulations with polarizable quantum mechanics/molecular mechanics calculations that allow overcoming Förster approximations. On the basis of our results, we discuss the potential of structure-based simulations in the characterization of drug-binding events through fluorescence techniques. Overall, we find an excellent agreement between theory and experiment both in terms of enantioselectivity and FRET times, thus strongly supporting the reliability of the binding modes proposed for the (S) and (R) enantiomers of FBP. In particular, we show that the dynamic quenching arises from a small fraction of drug bound to the secondary site of HSA at the interface between subdomains IIA and IIB, whereas the enantioselectivity arises from the larger flexibility of the (S)-FBP enantiomer in the binding pocket.

Impact of Electronic Fluctuations and Their Description on the Exciton Dynamics in the Light-Harvesting Complex PE545.

Temperature-dependent fluctuations of both site energies and electronic couplings are known to affect the excitation energy transfer in light-harvesting complexes. Environment effects on such fluctuations as well as possible spatial correlations among them are investigated here in the PE545 complex from cryptophyte algae using ensemble-averaged wave packet dynamics to extract the exciton dynamics. This strategy directly uses the time-dependent fluctuations of the system Hamiltonian, as described by quantum mechanics/molecular mechanics calculations performed along a classical MD trajectory. Neither the fluctuations in the couplings nor spatial correlations including cross-correlations between site energies and couplings are found to be important in the exciton dynamics of the complex. This finding does not change if a polarizable embedding is used instead of its electrostatic counterpart. The impact of variations in spectral densities and screening of excitonic couplings based on the electrostatic and polarizable embeddings are discussed as well.

Quantum Chemical Studies of Light Harvesting.

The design of optimal light-harvesting (supra)molecular systems and materials is one of the most challenging frontiers of science. Theoretical methods and computational models play a fundamental role in this difficult task, as they allow the establishment of structural blueprints inspired by natural photosynthetic organisms that can be applied to the design of novel artificial light-harvesting devices. Among theoretical strategies, the application of quantum chemical tools represents an important reality that has already reached an evident degree of maturity, although it still has to show its real potentials. This Review presents an overview of the state of the art of this strategy, showing the actual fields of applicability but also indicating its current limitations, which need to be solved in future developments.

Electrostatic versus Resonance Interactions in Photoreceptor Proteins: The Case of Rhodopsin.

Light sensing in photoreceptor proteins is subtly modulated by the multiple interactions between the chromophoric unit and its binding pocket. Many theoretical and experimental studies have tried to uncover the fundamental origin of these interactions but reached contradictory conclusions as to whether electrostatics, polarization, or intrinsically quantum effects prevail. Here, we select rhodopsin as a prototypical photoreceptor system to reveal the molecular mechanism underlying these interactions and regulating the spectral tuning. Combining a multireference perturbation method and density functional theory with a classical but atomistic and polarizable embedding scheme, we show that accounting for electrostatics only leads to a qualitatively wrong picture, while a responsive environment can successfully capture both the classical and quantum dominant effects. Several residues are found to tune the excitation by both differentially stabilizing ground and excited states and through nonclassical "inductive resonance" interactions. The results obtained with such a quantum-in-classical model are validated against both experimental data and fully quantum calculations.

Chromophore-protein coupling beyond nonpolarizable models: understanding absorption in green fluorescent protein.

The nature of the coupling of the photoexcited chromophore with the environment in a prototypical system like green fluorescent protein (GFP) is to date not understood, and its description still defies state-of-the-art multiscale approaches. To identify which theoretical framework of the chromophore-protein complex can realistically capture its essence, we employ here a variety of electronic-structure methods, namely, time-dependent density functional theory (TD-DFT), multireference perturbation theory (NEVPT2 and CASPT2), and quantum Monte Carlo (QMC) in combination with static point charges (QM/MM), DFT embedding (QM/DFT), and classical polarizable embedding through induced dipoles (QM/MMpol). Since structural modifications can significantly affect the photophysics of GFP, we also account for thermal fluctuations through extensive molecular dynamics simulations. We find that a treatment of the protein through static point charges leads to significantly blue-shifted excitation energies and that including thermal fluctuations does not cure the coarseness of the MM description. While TDDFT calculations on large cluster models indicate the need of a responsive protein, this response is not simply electrostatic: An improved description of the protein in the ground state or in response to the excitation of the chromophore via ground-state or state-specific DFT and MMpol embedding does not significantly modify the results obtained with static point charges. Through the use of QM/MMpol in a linear response formulation, a different picture in fact emerges in which the main environment response to the chromophore excitation is the one coupling the transition density and the corresponding induced dipoles. Such interaction leads to significant red-shifts and a satisfactory agreement with full QM cluster calculations at the same level of theory. Our findings demonstrate that, ultimately, faithfully capturing the effects of the environment in GFP requires a quantum treatment of large photoexcited regions but that a QM/classical model can be a useful approximation when extended beyond the electrostatic-only formulation.

Theoretical Characterization of the Spectral Density of the Water-Soluble Chlorophyll-Binding Protein from Combined Quantum Mechanics/Molecular Mechanics Molecular Dynamics Simulations.

Over the past decade, both experimentalists and theorists have worked to develop methods to describe pigment-protein coupling in photosynthetic light-harvesting complexes in order to understand the molecular basis of quantum coherence effects observed in photosynthesis. Here we present an improved strategy based on the combination of quantum mechanics/molecular mechanics (QM/MM) molecular dynamics (MD) simulations and excited-state calculations to predict the spectral density of electronic-vibrational coupling. We study the water-soluble chlorophyll-binding protein (WSCP) reconstituted with Chl a or Chl b pigments as the system of interest and compare our work with data obtained by Pieper and co-workers from differential fluorescence line-narrowing spectra (Pieper et al. J. Phys. Chem. B 2011, 115 (14), 4042-4052). Our results demonstrate that the use of QM/MM MD simulations where the nuclear positions are still propagated at the classical level leads to a striking improvement of the predicted spectral densities in the middle- and high-frequency regions, where they nearly reach quantitative accuracy. This demonstrates that the so-called "geometry mismatch" problem related to the use of low-quality structures in QM calculations, not the quantum features of pigments high-frequency motions, causes the failure of previous studies relying on similar protocols. Thus, this work paves the way toward quantitative predictions of pigment-protein coupling and the comprehension of quantum coherence effects in photosynthesis.

Electronic Couplings for Resonance Energy Transfer from CCSD Calculations: From Isolated to Solvated Systems.

Quantum mechanical (QM) calculations of electronic couplings provide great insights for the study of resonance energy transfer (RET). However, most of these calculations rely on approximate QM methods due to the computational limitations imposed by the size of typical donor-acceptor systems. In this work, we present a novel implementation that allows computing electronic couplings at the coupled cluster singles and doubles (CCSD) level of theory. Solvent effects are also taken into account through the polarizable continuum model (PCM). As a test case, we use a dimer of indole, a common model system for tryptophan, which is routinely used as an intrinsic fluorophore in Förster resonance energy transfer studies. We consider two bright π → π* states, one of which has charge transfer character. Lastly, the results are compared with those obtained by applying TD-DFT in combination with one of the most popular density functionals, B3LYP.

Polarizable QM/MM Multiconfiguration Self-Consistent Field Approach with State-Specific Corrections: Environment Effects on Cytosine Absorption Spectrum.

We present the formulation and implementation of a polarizable quantum mechanics/molecular mechanics (QM/MM) strategy to describe environment effects in multiconfiguration self-consistent field calculations. The strategy is applied to the calculation of the vertical absorption spectrum of cytosine in water. In our approach, mutual polarization of the solute and the solvent is solved self-consistently at the complete-active-space self-consistent-field (CASSCF) level, and the resulting set of charges and dipoles is used to calculate vertical excitation energies using the complete-active-space second-order perturbative (CASPT2) approach and its multistate (MS-CASPT2) variant. In order to treat multiple excited states, we converge the solvent polarization with respect to the state-averaged density of the solute. In order to obtain the final energies, however, we introduce a state-specific correction, where the solvent polarization is recomputed with the density of each state, and demonstrate that this correction brings the excitation energies closer to the values obtained with state-optimized orbitals. Comparison with PCM and nonpolarizable QM/MM calculations shows the importance of specific solute-solvent interactions and environment polarization in describing experiments. Overall, the calculated excitations for the π → π* states in water show good agreement with the experimental spectrum, whereas the n → π* appear at energies above 6 eV, approximately 1 eV higher than in the gas phase. Beyond solvents, the new method will allow studying the impact of heterogeneous biological environments in multiple excited states, as well as the treatment of multichromophoric systems where charge transfer and exciton states play important roles.