Ignacio Vayá

Electronic excited states responsible for dimer formation upon UV absorption directly by thymine strands: joint experimental and theoretical study.

The study addresses interconnected issues related to two major types of cycloadditions between adjacent thymines in DNA leading to cyclobutane dimers (T<>Ts) and (6-4) adducts. Experimental results are obtained for the single strand (dT)(20) by steady-state and time-resolved optical spectroscopy, as well as by HPLC coupled to mass spectrometry. Calculations are carried out for the dinucleoside monophosphate in water using the TD-M052X method and including the polarizable continuum model; the reliability of TD-M052X is checked against CASPT2 calculations regarding the behavior of two stacked thymines in the gas phase. It is shown that irradiation at the main absorption band leads to cyclobutane dimers (T<>Ts) and (6-4) adducts via different electronic excited states. T<>Ts are formed via (1)ππ* excitons; [2 + 2] dimerization proceeds along a barrierless path, in line with the constant quantum yield (0.05) with the irradiation wavelength, the contribution of the (3)ππ* state to this reaction being less than 10%. The formation of oxetane, the reaction intermediate leading to (6-4) adducts, occurs via charge transfer excited states involving two stacked thymines, whose fingerprint is detected in the fluorescence spectra; it involves an energy barrier explaining the important decrease in the quantum yield of (6-4) adducts with the irradiation wavelength.

Base Pairing Enhances Fluorescence and Favors Cyclobutane Dimer Formation Induced upon Absorption of UVA Radiation by DNA

The photochemical properties of the DNA duplex (dA)(20)·(dT)(20) are compared with those of the parent single strands. It is shown that base pairing increases the probability of absorbing UVA photons, probably due to the formation of charge-transfer states. UVA excitation induces fluorescence peaking at ∼420 nm and decaying on the nanosecond time scale. The fluorescence quantum yield, the fluorescence lifetime, and the quantum yield for cyclobutane dimer formation increase upon base pairing. Such behavior contrasts with that of the UVC-induced processes.

Fluorescence of natural DNA: from the femtosecond to the nanosecond time-scales

The fluorescence of calf thymus DNA is studied by steady-state and time-resolved spectroscopy combining fluorescence upconversion and time-correlated single photon counting. The fluorescence spectrum is very similar to that of a stoichiometric mixture of monomeric chromophores, arising from bright pi pi* states, and contrasts with the existing picture of exciplex emission in natural DNA. Yet, the DNA fluorescence decays span over five decades of time, with 98% of the photons being emitted at times longer than 10 ps. These findings, in association with recent studies on model duplexes, are explained by the involvement of dark states, possibly related to charge separation, serving as a reservoir for the repopulation of the bright pi pi* states.

Electronic Excitation Energy Transfer between Nucleobases of Natural DNA

Transfer of the electronic excitation energy in calf thymus DNA is studied by time-resolved fluorescence spectroscopy. The fluorescence anisotropy, after an initial decay starting on the femtosecond time scale, dwindles down to ca. 0.1. The in-plane depolarized fluorescence decays are described by a stretched exponential law. Our observations are consistent with one-dimensional transfer mediated by charge-transfer excited states.

High‐Energy Long‐Lived Excited States in DNA Double Strands

The knowledge that UV photons absorbed by DNA induce photoreactions leading to carcinogenic mutations has triggered numerous studies. During the past few years spectroscopic investigations performed with femtosecond resolution brought new insight in the electronic excited states of DNA duplexes. The detailed nature of these states is still under debate. However, based on results from transient absorption measurements performed for short helices at selected wavelengths, the following picture has been proposed. The initially populated pp* states relax within 100 fs to energetically low-lying excimers or exciplexes decaying with time constants up to ca. 100 ps. Such partial dissipation of the excitation energy, considered to contribute to the stability of the genetic code, was inferred from the steady-state fluorescence spectra of some DNA helices presenting an emission band at longer wavelengths compared to those the monomeric chromophores. 9, 10] However, while monitoring non-emitting transient species, transient absorption measurements do not provide any information about their energy relative to the ground state. In parallel, theoretical studies pointed out the importance of conformational dynamics on the electronic states of double helices. 12] Conformational changes affect both the electronic coupling, which determines the properties of exciton states, and the polarity of the local environment, which determines the energy of highly polarizable states. Consequently, the ordering of the various excited states of a duplex may vary with the conformation leading to intricate relaxation pathways. Herein we present novel and unexpected fluorescence properties evidenced for duplexes with alternating GC sequence in their B form (GC duplexes) which could be explained by such conformational variations. Our observations relate to excited states which emit at energies ca. 4000 cm 1 higher than the pp* states and decay on the nanosecond timescale indicating trapping of the excitation energy by high-energy channels. Previous fluorescence upconversion (FU) measurements, which detect only emission from bright excited states, showed that the pp* states of GC duplexes have a lifetime of 0.2 ps. 13] Herein, we study the polymer poly(dGdC).poly(dGdC) [ca. 1000 bp] and the oligomer (dGdC)10.(dGdC)10 by time-correlated single-photon counting (TCSPC) using the same laser (100 fs, 267 nm) as for the FU experiments. TCSPC also probe emission from “dark” states which may borrow oscillator strength from close-lying bright states. The time-resolved signals are correlated with the steady-state fluorescence spectra. Such a parallel study, in connection with the femtosecond investigation, is necessary in order to disentangle complex processes. Probably for this reason, the fluorescence spectrum of poly(dGdC).poly(dGdC), reported about two decades ago, did not reveal the full picture. A key point in our experiments was the use of low excitation intensities (<3 kW cm 2 peak power) in order to avoid photodamage of the duplexes. The fluorescence spectra of GC duplexes are presented in Figure 1 a. The polymer spectrum is characterized by a sharp peak close to the Raman line. The oligomer spectrum is about five times less intense than that of the polymer and has a rather flat profile. Despite the difference in spectral shape, the decays of both duplexes at 305 nm (Figure 1 b) contain a common feature, namely a long component decaying on the nanosecond timescale. An ultrafast component, which cannot be resolved by TCSPC, is also present in the decays, its contribution being more important in the case of the oligomer. The amplitude of the fast component becomes larger when the emission wavelength increases from 305 to 330 nm (Figure 2). The latter wavelength corresponds to the maximum of the spectrum recorded at 0.2 ps. The details of the fits with multi-exponential functions are given in the Supporting Information. The most important outcome is that the major contribution to the fluorescence decays of both the polymer and the oligomer is related to a time constant of 1.31 0.05 ns. The associated fluorescence anisotropy (insets in Figure 2) is also independent of the duplex size. After 100 ps, when the instrumental response function becomes negligible,

A joint experimental/theoretical study of the ultrafast excited state deactivation of deoxyadenosine and 9-methyladenine in water and acetonitrile.

The excited states of deoxyadenosine (dA) and 9-methyladenine (9Me-Ade) were studied in water and acetonitrile by a combination of steady-state and time-resolved spectroscopy and quantum chemical calculations. Femtosecond fluorescence upconversion experiments show that the decays of dA and 9Me-Ade after excitation at 267 nm are very similar, confirming that 9Me-Ade is a valid model for the calculations. The fluorescence decays can be described by an ultrafast component (<100 fs) and a slower one (≈ 300-500 fs); they are slightly slower in acetonitrile than in water. Time-dependent DFT calculations on 9Me-Ade, using PBE0 and M052X functionals and including both bulk and specific solvent effects, provide absorption and emission spectra in good agreement with experiments, giving a comprehensive description of the decay mechanism. It is shown that, in the Franck-Condon region, the lowest in energy state is the optically bright La state, with the Lb state situated about 2000 cm(-1) higher. Both states are populated when excited at 267 nm, but the Lb state undergoes an ultrafast Lb → La decay, too fast for our time-resolution (≈ 80 fs). This is confirmed by the experimentally observed fluorescence anisotropies, attaining values lower than 0.4 already at time zero. Consequently, the ensuing excited state relaxation mechanism can be described as the evolution along an almost barrierless path from the Franck-Condon region of the La potential energy surface towards a conical intersection with the ground state. This internal conversion mechanism proceeds without any significant involvement of any near-lying nπ* state.

Characterisation of the lowest singlet and triplet excited states of S-flurbiprofen.

The photophysical properties of S-flurbiprofen [S-2-fluoro-alpha-methyl-4-biphenylacetic acid], a nonsteroidal anti-inflammatory drug, have been examined using steady-state and time-resolved spectroscopic techniques. The energy of its first singlet excited state is 99 kcal mol(-1). The fluorescence quantum yields and lifetimes (at 300 nm) have been determined in acetonitrile, methanol, hexane and PBS; they are in the range 0.15<phi(F)< 0.33 and 0.7<tau(F)<2.0 ns. The intersystem crossing quantum yields are between 0.45 and 0.71; the lambda(max) of the T-T absorption is 360 nm, and the triplets live from 15 to 106 micros. Steady state photolysis in aqueous medium leads to S-2-hydroxy-alpha-methyl-4-biphenylacetic acid via photonucleophilic aromatic substitution, in addition to the photodecarboxylation products observed in organic solvents.

Use of triplet excited States for the study of drug binding to human and bovine serum albumins.

The triplet excited states of (S)‐ and (R)‐flurbiprofen (FBP) have been used as reporters for the microenvironments experienced within the binding sites of human and bovine serum albumins. Regression analysis of triplet decay provides valuable information on the degree of protection that these excited states are afforded from attack by a second FBP molecule, oxygen, or other reagents. The multiexponential fitting of these decays can be satisfactorily correlated with the distribution of the drug among the two binding sites and its presence as the noncomplexed form in the bulk solution. This assignment has been confirmed by using (S)‐ibuprofen or capric acid as selective site II replacement probes. Triplet lifetimes and site occupancy are sensitive to the type of serum albumin employed (human versus bovine). Finally, the binding behaviour of (S)‐ and (R)‐FBP exhibits little stereoselectivity.

Determination of Enantiomeric Compositions by Transient Absorption Spectroscopy using Proteins as Chiral Selectors

Determination of the enantiomeric composition is a relevant issue in different areas, including the synthesis of enantiopure chemicals and biologically active substances. In pharmacology, the enantiomers exhibit different behavior in terms of activity, side-effects, toxicity, metabolism, or transport mechanism. Therefore, the development of analytical methods to assess the enantiomeric composition of chiral drugs is of great interest. Direct determination of the enantiomeric excess (ee) can be achieved by polarimetry; however, this technique presents some practical limitations, mainly related to sensitivity and low tolerance to impurities. Other commonly used analytical methods are based on GC and HPLC on chiral stationary phases; their main disadvantages are that they are time-consuming and require serial analysis, which limits the number of samples that can be studied. Other methods have been developed in the last decades, based on the determination of different properties; they include MS, UV/ Vis absorption spectroscopy, IR thermography, circular dichroism, capillary electrophoresis, NMR spectroscopy, fluorescence spectroscopy, biochemical assays, and so forth. In spite of this effort, quantification of stereoisomer levels continues to be an important problem, and therefore further research is still required to develop new analytical methodologies. In principle, discrimination between enantiomers is possible by making use of supramolecular host–guest interactions. In this context, useful information has been obtained by chemometric analysis of the UV/Vis absorption or fluorescence spectra in the presence of cyclodextrins. Proteins are another important class of chiral selectors. Particularly, serum albumins (SA) have been widely used as stationary phases for the chromatographic resolution of enantiomeric mixtures, a concept based on the possible stereoselectivity of the binding process. Recently, we have explored the suitability of triplet excited states as reporters for the binding of drugs to transport proteins, as the properties of these states are very sensitive to the experienced microenvironment. Thus, laser flash photolysis (LFP) measurements have been performed on (S)and (R)-flurbiprofen methyl esters (FBPMe) in the presence of human serum albumin (HSA). Actually, FBPMe is a prodrug of the nonsteroidal antiinflammatory drug flurbiprofen (FBP). In the absence of protein, after LFP at 266 nm (2.5 10 m, in phosphate-buffered saline (PBS), air), (S)-FBPMe exhibits a characteristic transient triplet– triplet absorption spectrum centered at 360 nm, with a lifetime (tT) of 1.5 ms. However, in the presence of 2.5 10 m HSA, two tT values are observed (31.5 and 4.1 ms); [15a] they have been assigned to (S)-FBPMe inside the two known HSA binding sites (named site I and site II by Sudlow). From the relative contributions of the different tT values at various (S)-FBPMe/HSA ratios, it has been possible to determine the drug distribution among the bulk solution and the protein binding sites. The same trend, but with remarkably different triplet lifetimes (157.6 and 16.6 ms), has been observed for (R)-FBPMe/HSA systems. For both (S)and (R)-FBPMe, the major (longer-lived) component under non-saturating conditions was assigned to FBPMe within site I, and the minor (shorter-lived) component corresponded to site II-bound FBPMe. [a] Dr. M. C. Jim nez, Prof. M. A. Miranda Departamento de Qu mica Instituto de Tecnolog a Qu mica UPV-CSIC Universidad Polit cnica de Valencia, Camino de Vera s/n Apdo 46022, Valencia (Spain) Fax: (+34)963877344 E-mail : mcjimene@qim.upv.es mmiranda@qim.upv.es [b] Dr. I. Vay , C. J. Bueno, Dr. M. C. Jim nez, Prof. M. A. Miranda Departamento de Qu mica Instituto de Tecnolog a Qu mica UPV-CSIC Universidad Polit cnica de Valencia, Camino de Vera s/n Apdo 46022, Valencia (Spain) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.200801657.

Transient absorption spectroscopy for determining multiple site occupancy in drug-protein conjugates. A comparison between human and bovine serum albumins using flurbiprofen methyl ester as a probe.

Laser flash photolysis (LFP) has been used to determine the degree of binding of (S)- or (R)-flurbiprofen methyl ester (FBPMe) to human and bovine serum albumins (HSA and BSA, respectively). Regression analysis of the triplet decay of the drug (lambda = 360 nm) in the presence of the proteins led to a satisfactory fitting when considering a set of three lifetimes; the corresponding Afree, AI and AII preexponential coefficients can be correlated with the presence of FBPMe in the bulk solution and within the two known binding sites. The most remarkable differences between HSA and BSA were found under nonsaturating conditions; thus, when the [FBPMe]/[SA] ratio was 1:1, all the drug was bound to HSA, whereas 20-30% of it remained free in the bulk solution in the presence of BSA. The LFP approach was also applicable to the study of more complex FBPMe/HSA/BSA mixtures; the obtained results were in good agreement with the previous findings in FBPMe/HSA and FBPMe/BSA systems. This suggests the possibility of making use of the transient triplet-triplet absorption for investigating the distribution of a drug between several compartments in different host biomolecules.