Sequential Proton-Coupled Electron Transfer Mediates Excited-State

Feb 14, 2017 - Skin photoprotection is commonly believed to rely on the photochemistry of 5,6-dihydroxyindole (DHI)- and 5,6-dihydroxyindole-2-carboxy...
0 downloads 0 Views 2MB Size
Download PDF
Letter pubs.acs.org/JPCL

Sequential Proton-Coupled Electron Transfer Mediates Excited-State Deactivation of a Eumelanin Building Block Juan J. Nogueira,† Alice Corani,‡ Amal El Nahhas,‡ Alessandro Pezzella,§ Marco d’Ischia,§ Leticia González,*,† and Villy Sundström*,‡ †

Institute of Theoretical Chemistry, Faculty of Chemistry, University of Vienna, Währinger Str. 17, A-1090 Wien, Austria Department of Chemical Physics, Lund University, Box 124, 22100 Lund, Sweden § Department of Chemistry and Sciences, University of Naples Federico II, Via Cintia, 80126 Naples, Italy ‡

S Supporting Information *

ABSTRACT: Skin photoprotection is commonly believed to rely on the photochemistry of 5,6-dihydroxyindole (DHI)- and 5,6-dihydroxyindole-2-carboxylic acid (DHICA)based eumelanin building blocks. Attempts to elucidate the underlying excited-state relaxation mechanisms have been partly unsuccessful due to the marked instability to oxidation. We report a study of the excited-state deactivation of DHI using steady-state and time-resolved fluorescence accompanied by high-level quantum-chemistry calculations including solvent effects. Spectroscopic data show that deactivation of the lowest excited state of DHI in aqueous buffer proceeds on the 100 ps time scale and is 20 times faster than in methanol. Quantum-chemical calculations reveal that the excited-state decay mechanism is a sequential proton-coupled electron transfer, which involves the initial formation of a solvated electron from DHI, followed by the transfer of a proton to the solvent. This unexpected finding would prompt a revision of current notions about eumelanin photophysics and photobiology.

E

the photophysics of the monomer is crucial. DHI encompasses a most intriguing and peculiar molecular system consisting of a catechol moiety incorporated into a bicyclic indole ring in such a fashion that access to o-quinone or quinonoid π-electron arrangements is neither facile nor obvious due to concomitant disruption of the aromaticity of the pyrrole ring. In consequence, the investigation of excited states provides a unique model for testing concepts and probing mechanisms. UV pump−probe experiments on DHI in aqueous buffer solution have suggested deactivation of photoexcited DHI through two separate channels: nanosecond fluorescence (from S1) and faster cation radical formation (from higher lying excited states).8 The nanosecond fluorescence decay was suggested to reflect ESPT, but direct evidence of this mechanism was lacking. The formation of solvated electrons after UV excitation has been also postulated,8 but an active role of these species during the excited-state decay of DHI has never been demonstrated. Time-resolved fluorescence results on dimers and polymers of DHI have also been published, suggesting mainly nanosecond excited-state lifetimes.9 Calculations in the gas phase on DHI have evinced that UV excitation leads to hydrogen atom transfer from a hydroxyl group to a nearby C atom of the aromatic ring.10 This hydrogen attachment to the ring would result in a rapid tautomerization

umelanin, the major determinant of the dark-brown and black coloration of skin, hair, and eyes, is a heterogeneous polymer built from the small molecular units 5,6-dihydroxyindole (DHI) and 5,6-dihydroxyindole-2-carboxylic acid (DHICA).1,2 It is commonly believed that eumelanins play an important photoprotective function against the harmful effects of UV light through efficient dissipation of electronic excited-state energy.3 However, the photoprotection mechanism is still poorly understood because of the lack of detailed knowledge about the highly disordered eumelanin structure. In addition, spectroscopic studies on the eumelanin building block DHI are difficult to perform due to its tendency to polymerize in solution, especially in the presence of light. It has been shown that DHICA monomers dissipate the UV radiation energy through excited-state proton transfer (ESPT) at intra- and intermolecular levels, depending on the pH.4 The excited-state deactivation by intra- and intermolecular ESPT is much more efficient for DHICA oligomers, very similar to what occurs in polymeric DHICA.4−7 This efficient dissipation of the excitation energy suggests that DHICA may be the main responsible compound for the photoprotection in eumelanin. However, the biological role played by DHI is still unclear. The short excited-state lifetime of the monomer indicates that DHI could also be involved in the photoprotective function.8 Nonetheless, the formation of radicals and triplet excited states after UV excitation could generate reactive species with phototoxic effects.8 Thus further research is necessary to obtain a complete understanding of the biological function of polymeric DHI. As a first step, an accurate characterization of © XXXX American Chemical Society

Received: December 22, 2016 Accepted: February 14, 2017 Published: February 14, 2017 1004

DOI: 10.1021/acs.jpclett.6b03012 J. Phys. Chem. Lett. 2017, 8, 1004−1008

Letter

The Journal of Physical Chemistry Letters of the molecule to form 6-hydroxy-1,4-dihydroindol-5-one.10 Despite previous experimental and theoretical efforts, a clear picture of the excited-state deactivation mechanism of DHI is far from being complete. Herein we use time-resolved and steady-state fluorescence spectroscopy in combination with quantum-chemical calculations including the explicit effect of the solvent to investigate the excited-state dynamics of DHI. The spectroscopic measurements show that DHI dissipates its excited-state energy on a 100 ps time scale through a process involving both OH groups and requiring an aqueous solvent. The theoretical calculations predict that the excited-state decay occurs through a sequential proton-coupled electron transfer (PCET) process. The measured absorption spectra of DHI in solution are shown in Figure 1a. In aqueous solution, regardless of pH, DHI

CASPT2), and the explicit water and methanol molecules are described by a force field14,15 (see Section S1 of the Supporting Information for further details). The two computed bands peaking at 271 and 311 nm in water and at 273 and 309 nm in methanol compare well with the measured values. These bands correspond to excitations into the two lowest electronic excited states, S0 → S1 and S0 → S2, respectively, and have ππ* character in both solvents. The measured fluorescence spectra, upon 280 nm excitation, are also shown in Figure 1a. A main emission band at ∼340 nm is observed in water and methanol, and an additional band at ∼380 nm, with a wing extending all the way up to ∼500 nm, is present in aqueous solution. This red emission is attributed to DHI dimers and oligomers, which were formed through oxidation of DHI under the experimental irradiation conditions (see Section S2 of the Supporting Information for details). The excitation wavelength of 280 nm for steady-state and time-resolved fluorescence measurements was chosen as a compromise between signal-to-noise optimization and available wavelengths and with the knowledge that excitation wavelengths in the range 250−320 nm generate the same fluorescence bands at ∼340 and 380 nm.7 Fluorescence decay kinetics of DHI in aqueous buffer solution, measured in the range 320−460 nm, are also identical within experimental error for 266 and 280 nm excitation.7 Previous experiments7 have shown that DHI has two excitedstate decays in aqueous buffer solution: a fast one at ∼110 ps after excitation in the short wavelength part of the 300 nm absorption band and a long-lived with ∼1.6 ns decay on the red side of the emission after red excitation. The present timeresolved fluorescence measurements in water/methanol mixtures show that the emission at 340 nm is solventdependent and the lifetime increases from 103 ± 10 ps in aqueous buffer solution to 2.2 ± 0.1 ns in neat methanol (Figure 2).

Figure 2. 350 nm emission decays of DHI in water/methanol mixtures from 100% methanol (dark blue) to 0% methanol (light blue) with 20% of water content increment upon 280 nm excitation. Inset a shows DHI emission decays in 10 mM phosphate buffer pH 7 and pD 7. Inset b presents the decay rate constant of DHI as a function of the water content.

Figure 1. (a) Steady-state absorption (solid lines) and emission spectra (dashed line) of DHI, (b) calculated absorption spectra of DHI in water and methanol at MS-CASPT2/MM level of theory, and (c) steady-state absorption (solid lines) and emission spectra (dashed line) of 5M6HI in water (black lines) and methanol (red lines). Excitation wavelength for emission spectra was 280 nm.

Previous experimental work4 on DHICA has demonstrated efficient excited-state deactivation through ESPT to solvent. An intriguing question is to investigate whether ESPT is also a deactivation pathway in DHI and why the deactivation is faster in aqueous solution than in methanol. To this aim, two cluster models formed by DHI solvated by 10 water and 6 methanol molecules, respectively, were built based on QM/MM molecular dynamics simulations (see Section S3 of the Supporting Information). The electronic excitation energies

exhibits two bands peaking at 274 and 299 nm, as previously reported.11 In methanol, the absorption spectrum undergoes a bathochromic shift of 3 nm. The calculated absorption spectrum employing a quantum mechanics/molecular mechanics (QM/MM) electrostatic scheme12 and 100 geometries sampled by molecular dynamics is displayed in Figure 1b. In the QM/MM scheme, DHI is described by accurate multistate complete active-space second-order perturbation theory13 (MS1005

DOI: 10.1021/acs.jpclett.6b03012 J. Phys. Chem. Lett. 2017, 8, 1004−1008

Letter

The Journal of Physical Chemistry Letters

solvation shell and subsequent electron transfer from DHI to a cluster of solvent molecules generating a solvated electron (πe−sol)p,min (the subscript p indicates that DHI is still protonated). The formation of the solvated electron is energetically favorable in both solvents, and it is stabilized by hydrogen bonding with the OH groups of three solvent molecules and one C−H bond of DHI. This tetrahedral configuration has recently been suggested as the smallest model that correctly reproduces the experimental properties of the hydrated electron.17 The formation of a solvated electron induces a proton transfer from an OH group of DHI to a nearby solvent molecule. The proton acceptor water molecule is not one of the waters involved in the electron acceptance, which is not the most common situation. However, because proton transfer takes place in the same direction as the previous electron transfer, there is a strong attractive electrostatic interaction between the solvated electron and the transferring proton that drives the reaction. The proton transfer takes place adiabatically in the πe−sol electronic state, giving rise to the (πe−sol)d,min structures of Figure 3a,b (the subscript d indicates that DHI is deprotonated). When the proton transfer is completed, the energy gap between the πe−sol electronic state and the ground state is only 0.37 and 0.50 eV in water and methanol, respectively. Therefore, it is very likely that the system undergoes internal conversion to the ground state, where the electron will come back to the deprotonated DHI molecule. However, because of the attractive electrostatic interaction between the transferred proton and the solvated electron, the occurrence of multiple proton transfers along solvent wires is likely before internal conversion to the ground state. Such multiple proton transfers were predicted by surfacehopping simulations for 2-aminooxazole solvated by a cluster of water molecules.18 PCET processes can take place by concerted or sequential mechanisms.19 The former does not involve stable intermediates because both electron transfer and proton transfer occur in one step. On the contrary, in sequential reactions, electron transfer is followed by proton transfer or vice versa in a twostep process. The excited-state deactivation of aromatic molecules with OH and NH groups (e.g., phenol, pyrrole and indole) through concerted PCET mechanisms has been well-established in the last two decades.18,20−24 In this mechanism, the chromophore is excited to the bright ππ* electronic state, in which the predissociation of the NH or OH bond induces an energy decrease of a dissociative πσ* state. The πσ* state is populated through a conical intersection with the ππ* state; then, the proton transfer from the chromophore to the solvent is completed. Thus both electron and proton are transferred in a concerted way. The transferred electron located in a rather diffuse σ* orbital of a solvent molecule is the precursor of a solvated electron. The deactivation mechanism observed here for DHI is different. It takes place in a sequential manner because the (πe−sol)p,min intermediate structures are local minima of the potential energy surfaces. Additionally, intermolecular electron transfer is induced by reorganization of the solvation shell and not by the predissociation of the OH bond. The rate-limiting step of the reaction is the proton transfer with calculated energy barriers of 0.04 and 0.15 eV in water and methanol, respectively. The small energy barrier in water supports the faster decay observed in water than that in methanol. An accurate estimation of the individual rate constants is very hard considering the small value of the energy barriers and the approximations introduced by our

of the two cluster models computed at density functional theory level using the CAM-B3LYP functional16 are displayed in Figure 3a,b. The excitation energies of the two brightest

Figure 3. Relevant potential energy surface regions of the adiabatic S1 electronic excited state, computed by CAM-B3LYP, along the protoncoupled electron transfer process (PCET) in (a) water and (b) methanol.

states (S1 and S2) at the Franck−Condon geometry are 4.59 eV (270 nm) and 4.81 eV (258 nm) in water and 4.53 eV (274 nm) and 4.79 eV (259 nm) in methanol. The reasonable agreement of these energy values with the measured absorption spectra and the MS-CASPT2 spectra validates the model and level of theory employed. Thus the cluster models and the CAM-B3LYP functional have been used to explore the relevant regions of the potential energy surface of the S1 electronic state. A detailed discussion about the accuracy of our model can be found in Section 3 of the Supporting Information. The energetic schemes shown in Figure 3a,b reveal that the electronic excited-state deactivation of DHI follows a PCET mechanism. After excitation to the bright S1(ππ*) electronic excited state, DHI vibrationally relaxes to the minimum energy geometry (ππ*)min. Then, there is a reorganization of the 1006

DOI: 10.1021/acs.jpclett.6b03012 J. Phys. Chem. Lett. 2017, 8, 1004−1008

Letter

The Journal of Physical Chemistry Letters

of 5M6HI (Figure 4) is similar to that of DHI, with a 160 ± 10 ps monoexponential fluorescence decay in aqueous buffer

cluster model. However, the calculation of the ratio between both rate constants provides a more accurate result because it relies on error cancelation. Using the Arrhenius expression for the proton-transfer rate constants, as an approximation, one obtains a ratio k(H2O)/k(MeOH) = 50; that is, the reaction is 50 times faster in water than in methanol. This result is in qualitative agreement with the ratio k(H2O)/k(MeOH) = 20 extracted from the time-resolved fluorescence measurements (Figure 2). The presence of a solvated electron during the excited-state deactivation of DHI is evident from the orbital picture depicted in Figure 3a,b. However, a more quantitative characterization of the solvated electron features is possible by analyzing the oneelectron transition density matrix of the πe−sol electronic state.25,26 This electronic state in aqueous solution has 97% charge-transfer character with a hole−electron separation (exciton size) of 5.5 Å. The contribution of the different molecules to the molecular orbital where the solvated electron is located can also be quantified. Specifically, each of the three water molecules solvating the electron contributes 43, 34, and 18%, and the C−H bond of DHI contributes only 3%. The remaining 2% comes from other solvent molecules. A similar characterization of the solvated electron is obtained in methanol (see Section 3 of the Supporting Information). Thus the πe−sol electronic state has a strong long-range chargetransfer character. It is well known that DFT can underestimate the energy of charge-transfer states, even with a range-separated functional, as it has been used here. However, the calculation of the excitation energies and the electronic characterization of the (πe−sol)d,min structure at MS-CASPT2 level corroborates that the charge-transfer state is the lowest electronic excited state, validating the conclusions extracted from the DFT calculations (see Section 3 of the Supporting Information). The fast electron transfer/slow proton transfer mechanism suggested by the quantum-chemical calculations is indirectly corroborated by the spectroscopic measurements. The emission decay of DHI in deuterated buffer solution is three times slower than in hydrogenated buffer solution; that is, the kinetic isotope effect (KIE) is around 3 (Figure 2 inset a). The presence of a significant KIE indicates that the rate-limiting step is the proton transfer because deuteration is not expected to influence the rate constant of the electron-transfer process.19 Previous quantum-mechanical calculations for DHI in the gas phase suggested that excitation with UV light promotes hydrogen atom transfer from a hydroxyl group to a nearby C atom of the aromatic ring.10 However, the solvent dependence of the rate constant and the KIE observed in our experiments points toward an active participation of the solvent in the deactivation mechanism, and thus an intramolecular hydrogen atom transfer without solvent mediation can be ruled out in bulk solvation. To assess whether the two OH groups of DHI have an active role in the process, we investigate the excited-state behavior of 5-methoxy-6-hydroxyindole (5M6HI), in which the OH group at C5 is methylated. Inspection of the absorption spectrum at pH 7 (Figure 1c) reveals a minor blue shift with respect to DHI, while in methanol the ∼300 nm absorption band displayed two sub-bands, probably representing a vibronic structure. The fluorescence spectrum, upon 280 nm excitation, presents a main emission band at 340 nm in both solvents as for DHI. However, the band at ∼380 nm, observed for DHI in aqueous solution, is only present for 5M6HI in water when the excitation wavelength is above 320 nm (see Figure S4 of the Supporting Information). Interestingly, the fluorescence decay

Figure 4. Fluorescence decays at 340 nm of 5M6HI in H2O buffer, D2O buffer and methanol after excitation at 280 nm.

solution, which slows down to 2.8 ± 0.15 ns in methanol. In deuterated buffer the fluorescence decay time increases by the same factor of 3 as for DHI. The very similar fluorescence decay characteristics of DHI and 5M6HI led us to conclude that 5M6HI undergoes the same sequential PCET as the parent DHI, and the OH group in position C6 is the proton donor. In addition, the similar pKa values of the two hydroxyl groups of DHI and the decrease in the decay rate when proton transfer from the 5-OH group is inhibited in 5M6HI suggest that both OH groups are involved in the relaxation process of DHI. If we assume that the rate of proton/electron transfer from the 6-OH group of DHI is similar to that measured for 5M6HI, then transfer from the 5-OH group in DHI would be characterized by a rate constant of (3.0 ± 0.5) × 109 s−1, only half that for transfer from the 6-OH group (5.9 ± 0.4) × 109 s−1, assuming that the 2.8 ± 0.15 ns fluorescence decay time of 5M6HI measured in MeOH represents the excited-state lifetime in the absence of proton/electron transfer. In conclusion, time-resolved fluorescence measurements show an excited-state decay of DHI at 103 ps in aqueous solution and at 2.2 ns in methanol. For the first time, a sequential proton-coupled electron transfer was identified as a deactivation pathway for the eumelanin building block DHI by means of quantum-chemical calculations including explicit solvent molecules. The process is initiated by a reorganization of the solvent, which favors the fast formation of a solvated electron. Then, a proton is transferred from DHI to a nearby solvent molecule. Both OH groups of DHI are involved in the process. It is likely that multiple proton transfers along solvent wires, driven by the electrostatic interaction between the proton and the solvated electron, take place. Finally, DHI likely decays to the ground state by internal conversion, where the solvated electron recombines with the deprotonated DHI species. However, other channels might also be operative, and additional calculations are required to clarify this point. Further theoretical and experimental work is necessary to investigate whether the mechanism observed here is also relevant for other eumelanin building blocks and their oligomeric and polymeric forms. Moreover, the unexpected mechanistic scenario emerging from this study would reinforce the notion of a different role of DHI versus DHICA in eumelanin photoprotection. While absorption of UV light in DHICA results in reversible migration of a proton between two different locations (the COOH and NH functionalities), DHI is now shown to undergo excited-state relaxation with the formation of a reactive species, a solvated electron. The actual scope and photo1007

DOI: 10.1021/acs.jpclett.6b03012 J. Phys. Chem. Lett. 2017, 8, 1004−1008

Letter

The Journal of Physical Chemistry Letters biological implications of this finding are manifold and likely to spur more important advancement in melanin photoprotection than the direct investigation of the polymer itself as a mimic of natural pigments. Elucidating DHI photophysics may allow the disclosure and characterization of pathways of potential relevance to indoles, catechols, and other biologically relevant systems and provides the necessary groundwork to ultimately elucidate energy dissipation channels in eumelanins. The reverse top-down approach, from the overall photophysical behavior of the polymer to specific mechanisms at the molecular level, remains a less rewarding approach in the present stage of knowledge in the field.

■

dihydroxyindole, a key eumelanin building block: nonradiative decay mechanism. J. Phys. Chem. B 2009, 113, 12575−12580. (9) Corani, A.; Huijser, A.; Iadonisi, A.; Pezzella, A.; Sundström, V.; d’Ischia, M. Bottom-Up approach to eumelanin photoprotection: Emission dynamics in parallel sets of water-soluble 5,6-dihydroxyindole-based model systems. J. Phys. Chem. B 2012, 116, 13151− 13158. (10) Sobolewski, A. L.; Domcke, W. Photophysics of eumelanin: Ab initio studies on the electronic spectroscopy and photochemistry of 5,6-dihydroxyindole. ChemPhysChem 2007, 8, 756−762. (11) d’Ischia, M.; Napolitano, A.; Pezzella, A. 5,6-dihydroxyindole chemistry: Unexplored opportunities beyond eumelanin. Eur. J. Org. Chem. 2011, 2011, 5501−5516. (12) Melaccio, F.; Olivucci, M.; Lindh, R.; Ferré, N. Unique QM/ MM potential energy surface exploration using microiterations. Int. J. Quantum Chem. 2011, 111, 3339−3346. (13) Finley, J.; Malmqvist, P. A.; Roos, B. O.; Serrano-Andrés, L. The multi-state CASPT2 method. Chem. Phys. Lett. 1998, 288, 299−306. (14) Caldwell, J. W.; Kollman, P. A. Structure and properties of neat liquids using nonadditive molecular dynamics: Water, methanol, and N-methylacetamide. J. Phys. Chem. 1995, 99, 6208−6219. (15) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. Comparison of simple potential functions for simulating liquid water. J. Chem. Phys. 1983, 79, 926−935. (16) Yanai, T.; Tew, D. P.; Handy, N. C. A new hybrid exchangecorrelation functional using the Coulomb-attenuating method (CAMB3LYP). Chem. Phys. Lett. 2004, 393, 51−57. (17) Kumar, A.; Walker, J. A.; Bartels, D. M.; Sevilla, M. D. A Simple ab Initio Model for the Hydrated Electron That Matches Experiment. J. Phys. Chem. A 2015, 119, 9148−9159. (18) Szabla, R.; Šponer, J.; Góra, R. W. Electron-driven proton transfer along H2O wires enables photorelaxation of πσ* states in chromophore - Water clusters. J. Phys. Chem. Lett. 2015, 6, 1467− 1471. (19) Hammes-Schiffer, S.; Stuchebrukhov, A. A. Theory of coupled electron and proton transfer reactions. Chem. Rev. 2010, 110, 6939− 6960. (20) Sobolewski, A. L.; Domcke, W.; Dedonder-Lardeux, C.; Jouvet, C. Excited-state hydrogen detachment and hydrogen transfer driven by repulsive πσ* states: A new paradigm for nonradiative decay in aromatic biomolecules. Phys. Chem. Chem. Phys. 2002, 4, 1093−1100. (21) Sobolewski, A. L.; Domcke, W. Computational studies of the photophysics of hydrogen-bonded molecular systems. J. Phys. Chem. A 2007, 111, 11725−11735. (22) Pino, G. A.; Oldani, A. N.; Marceca, E.; Fujii, M.; Ishiuchi, S. I.; Miyazaki, M.; Broquier, M.; Dedonder, C.; Jouvet, C. Excited state hydrogen transfer dynamics in substituted phenols and their complexes with ammonia: ππ*-πσ* energy gap propensity and ortho-substitution effect. J. Chem. Phys. 2010, 133, 124313. (23) Sobolewski, A. L.; Domcke, W. Photoinduced electron and proton transfer in phenol and its clusters with water and ammonia. J. Phys. Chem. A 2001, 105, 9275−9283. (24) Nagashima, K.; Takatsuka, K. Early-stage dynamics in coupled proton-electron transfer from the π-π* State of phenol to solvent ammonia clusters: A nonadiabatic electron dynamics study. J. Phys. Chem. A 2012, 116, 11167−11179. (25) Plasser, F.; Lischka, H. Analysis of excitonic and charge transfer interactions from quantum chemical calculations. J. Chem. Theory Comput. 2012, 8, 2777−2789. (26) Plasser, F.; Wormit, M.; Dreuw, A. New tools for the systematic analysis and visualization of electronic excitations. I. Formalism. J. Chem. Phys. 2014, 141, 024106.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.6b03012. Computational details, discussion about the origin of red ∼380−500 nm emission of DHI, and details of sample preparation and additional spectroscopic measurements (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*L.G.: E-mail: [email protected]. *V.S.: E-mail: [email protected]. ORCID

Leticia González: 0000-0001-5112-794X Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Erik Ekengard for help with the preparation of the deuterated buffer solution. V.S. acknowledges funding from the Swedish Research Council and the Knut&Alice Wallenberg Foundation. The calculations have been in part performed in the Vienna Scientific Cluster (VSC).

■

REFERENCES

(1) d’Ischia, M.; Napolitano, A.; Pezzella, A.; Meredith, P.; Sarna, T. Chemical and structural diversity in eumelanins: Unexplored biooptoelectronic materials. Angew. Chem., Int. Ed. 2009, 48, 3914−3921. (2) Prota, G. Melanins and Melanogenesis; Academic Press: San Diego, CA, 1992. (3) Meredith, P.; Sarna, T. The physical and chemical properties of eumelanin. Pigm. Cell Res. 2006, 19, 572−594. (4) Corani, A.; Pezzella, A.; Pascher, T.; Gustavsson, T.; Markovitsi, D.; Huijser, A.; d’Ischia, M.; Sundström, V. Excited-state protontransfer processes of DHICA resolved: From sub-picoseconds to nanoseconds. J. Phys. Chem. Lett. 2013, 4, 1383−1388. (5) Corani, A.; Huijser, A.; Gustavsson, T.; Markovitsi, D.; Malmqvist, P. A.; Pezzella, A.; d’Ischia, M.; Sundström, V. Superior photoprotective motifs and mechanisms in eumelanins uncovered. J. Am. Chem. Soc. 2014, 136, 11626−11635. (6) Huijser, A.; Pezzella, A.; Hannestad, J. K.; Panzella, L.; Napolitano, A.; d’Ischia, M.; Sundström, V. UV-dissipation mechanisms in the eumelanin building block DHICA. ChemPhysChem 2010, 11, 2424−2431. (7) Huijser, A.; Pezzella, A.; Sundströ m, V. Functionality of epidermal melanin pigments: Current knowledge on UV-dissipative mechanisms and research perspectives. Phys. Chem. Chem. Phys. 2011, 13, 9119−9127. (8) Gauden, M.; Pezzella, A.; Panzella, L.; Napolitano, A.; d’Ischia, M.; Sundströ m, V. Ultrafast excited state dynamics of 5,61008

DOI: 10.1021/acs.jpclett.6b03012 J. Phys. Chem. Lett. 2017, 8, 1004−1008