Photosensitized Splitting of Thymine Dimer or Oxetane Unit by a

Jul 8, 2010 - Although many similarities exist between the two classes of enzymes, cyclobutane photolyases and (6−4) photolyases have certain import...
1 downloads 0 Views 392KB Size
Download PDF
J. Phys. Chem. B 2010, 114, 9827–9832

9827

Photosensitized Splitting of Thymine Dimer or Oxetane Unit by a Covalently N-Linked Carbazole via Electron Transfer in Different Marcus Regions Qing-Qing Wu and Qin-Hua Song* Department of Chemistry, UniVersity of Science and Technology of China, Hefei 230026, China ReceiVed: April 20, 2010; ReVised Manuscript ReceiVed: June 22, 2010

Although many similarities exist between the two classes of enzymes, cyclobutane photolyases and (6-4) photolyases have certain important differences. The most significant difference is in their repair quantum yields, cyclobutane photolyases with a uniformly high efficiency (0.7-0.98) and very low repair efficiency for (6-4) photolyases (0.05-0.1). To understand the significant difference, we prepared two classes of model compounds, covalently N-linked dimer- (1) or oxetane-carbazole (2) compounds with a dimethylene or trimethylene group as a linker. Under light irradiation, the dimer or oxetane unit of model compounds can be sensitized to split by the excited carbazole via an intramolecular electron transfer. The splitting reaction of dimer or oxetane unit in model compounds is strongly solvent dependent. In nonpolar solvents, such as cyclohexane or THF, no fluorescence quenching of the carbazole moiety of model compounds relative to a free carbazole, N-methylcarbazole, was observed and thus no splitting occurred. In polar solvents, two classes of model compounds reveal two reverse solvent effects on the splitting quantum yield. One is an inverse relation between the quantum yield and the polarity of the solvent for dimer-model systems, and another is a normal relation for oxetane-model systems. This phenomenon was also observed with another two classes of model compounds, covalently linked dimer- or oxetane-indole. Based on Marcus theory and thermodynamic data, it has been rationalized that the two reverse solvent effects derive from back electron transfer in the splitting process lying in the different Marcus regions. Back electron transfer lies in the Marcus inverted region for dimer-model systems and the normal region for oxetane-model systems. From repair solvent behavior of the two classes of model compounds, we gained some insights into the major difference in the repair efficiency for the two classes of photolyases. Introduction Two major lesions in DNA induced by the UV component in solar light (280-320 nm), the cyclobutane pyrimidine dimers (CPDs) and the pyrimidine (6-4) pyrimidone adducts ((6-4) photoproducts) (Figure 1), are responsible for the harmful effects of the UV on organisms, such as growth delay, mutagenesis, and killing, and constitute 70-80% and 20-30% of the total photoproducts, respectively.1 The two photolesions can be repaired through DNA photoreactivation catalyzed by CPD photolyases and (6-4) photolyases, respectively.2 CPD photolyases are flavin-containing repair enzymes which catalyze the efficient repair of cis-syn cyclobutane pyrimidine dimers by utilizing the energy of visible light to break the cyclobutane ring of the dimer. The mechanisms of CPD photolyases have been well investigated.2,3 The enzymes contain a redox active cofactor, flavin adenine dinucleotide (FAD) that operates in the full-reduced and deprotonated form (FADH-), and an auxiliary antenna chromophore. The second cofactor activates the process of repair by absorbing near UV-vis radiation (300-500 nm) and transferring the energy to the flavin. Subsequently, the excited FADH- transfers one electron to the cyclobutane pyrimidine dimer to form the dimer radical anion, which cleaves spontaneously and then back electron transfer restores the dipyrimidine and the functional form of flavin ready for a new cycle of catalysis. The discovery of (6-4) photolyase and the subsequent identification of structural and cofactor similarities * Corresponding author. Telephone: +86-551-3607524. Fax: +86-5513601592. E-mail: [email protected].

to CPD photolyase led to a proposal of a reaction scheme very similar to that of the cyclobutane photolyase.4 Cyclobutane dimers can be restored to their canonical forms by simply breaking the C5-C5′ and C6-C6′ bonds, but the breaking of the C5-OH and C6-C4′ bonds of (6-4) photoproducts by any means would not result in repair. Hence, in the model for (6-4) photolyase, a critical step, in which (6-4) photolyase differs from CPD photolyase, is that upon binding to the substrate; the enzyme converts the open form of the (6-4)

Figure 1. Formation of the two major photoproducts in DNA under UV light, CPDs and (6-4) photoproducts using thymine as an example.

10.1021/jp1035579  2010 American Chemical Society Published on Web 07/08/2010

9828

J. Phys. Chem. B, Vol. 114, No. 30, 2010

CHART 1: Two Classes of Model Compounds with a Short Linker

photoproduct to the oxetane intermediate by a thermal reaction.4 However, recent crystal structural data of the (6-4) photolyase from Drosophila melanogaster bound to a DNA duplex containing a T(6-4)T photoproduct5a or a T(6-4)C lesion5b support a nonoxetane repair mechanism that a direct electron injection into the lesion forming a (6-4) lesion radical anion can undergo repair via hydroxyl group transfer. Several model compounds6-10 that mimic the action of photolyase have been designed, such as a chromophore attached to a pyrimidine dimer/oxetane unit. Studies performed with these compounds have been helpful to unravel the mechanisms sketched above in detail. Additionally, these model systems are also good substrates for the study of intramolecular electron transfer, such as solvent and7c,d,8a,10a,b,d,e conformation effects.7c,10e However, there are still many unsolved problems on the physical chemical mechanisms involved in the whole process. Although the two classes of enzymes have similar structures, the same chromophores and the same basic mechanism, certain important differences exist between CPD and (6-4) photolyases. The most significant difference is that, while cyclobutane photolyases repair the photodimers with a uniformly high quantum yield (0.7-0.98),3 the quantum yield of (6-4) photolyases is very low (0.05-0.11).4 Up to now, mechanistic details of this significant difference are unclear.3,4,10b,d In this work, we have prepared two kinds of model compounds, covalently linked carbazole-thymine dimer (1a and 1b) and carbazole/indole-thymine oxetane compounds (2a, 2b and 3, shown in Chart 1), and investigated photosensitized splitting of thymine dimer/oxetane by the attached chromophore in various solvents. From the splitting reactions in two classes of model systems, two reverse solvent effects were observed and rationalized based on Marcus theory. Experimental Details General. Melting points were uncorrected. All materials were obtained from commercial suppliers and used as received. Solvents of technical quality were distilled prior to use. Dimethylformamide (DMF) was dried overnight with K2CO3 and distilled. Acetonitrile, cyclohexane, and tetrahydrofuran (THF) were dried with CaH2 or metal sodium, and methanol was dried with metal magnesium and distilled before use, for the photosplitting measurements of model compounds. Measurement of Steady-State Fluorescence Emission. Fluorescence emission spectra were measured at room temperature on a luminescence spectrometer. The extent of fluorescence quenching, Q, was determined by comparing fluorescence intensities (Fmodel) of model compounds with that (Fcarbazole) of a free carbazole, N-methylcarbazole, that is Q ) 1 - Fmodel/ Fcarbazole. The concentrations of the carbazole moiety of model compounds and N-methylcarbazole were controlled within 0.05

Wu and Song for absorbance at the excitation wavelength of 328 nm, and fluorescence intensities were normalized with the absorbances. The measurement error is less than 10%. Measurements of Splitting Quantum Yields of Model Compounds. To obtain the quantum yields for dimer splitting of model compounds, Φ ) rate of dimer splitting / rate of photons absorbed, 3 mL (V0) solutions containing model compounds were prepared in a cuvette with a Teflon stopper and irradiated with monochromatic light (328 and 295 nm for the carbazole and indole systems, respectively) from a fluorescence spectrometer with a 20 nm slit. The extent of dimer or oxetane splitting was measured by monitoring the increase in the absorbance (A273) at 273 nm due to the regeneration of the pyrimidine bases and benzophenone (for oxetane models). The A273 change (∆A273) of the solution depends on the splitting extent of model compounds. The plot of ∆A273 against the irradiation time (t, /min) is well fitted as a straight line, where the slope B reflected the splitting rate of dimer or oxetane unit in the model compound. The intensity of the incident light I0 (unit: einstein min-1) was measured using ferrioxalate actinometry.11 The intensity of light absorbed (Ia) by solution was calculated in terms of Beer’s law (Ia ) I0(1-10-A328(295))). ∆ε273 was the difference of mole extinction coefficient of splitting products against their starting materials. These values above allow the calculation of the quantum yield (Φ ) BV0/∆ε273Ia), the experimental error within 2%. No significant difference in the splitting quantum yields was observed for deaerated and nondeaerated solutions. Hence, the nondeaerated solution was employed in all measurements. To limit competition of absorption of the irradiated light between model compounds and photoproducts, the splitting extent of model compounds was controlled within 10% in all the measurements of the quantum yield. Results and Discussion Photosplitting Properties of Model Compounds. Model compounds 1 and 2 in methanol were irradiated with a Xe lamp, and photosplitting products 4 and 5 were isolated from the irradiation solutions, respectively. The photosplitting reactions for model compounds 1 and 2, shown in Scheme 1, were further analyzed by 1H NMR spectra. Utilizing 1H NMR spectra, the splitting reaction of model compounds could be tracked. The deuterated CD3OD/CD3CN (v/v 1:1) solvent mixture containing a model compound 1a or 2a was prepared in a NMR tube and irradiated with monochromatic light (328 nm). After irradiation for a certain time interval, the 1H NMR spectrum was recorded. With increasing irradiation time, new NMR peaks appeared gradually and accorded completely with the expected photoproducts (4a in Figure 2 (left), 5a and benzophenone in Figure 2 (right)). Hence, the photosplitting of model compounds was a clean conversion generating no other detectable products. Fluorescence Quenching of Model Compounds. Fluorescence emission spectra of model compounds 1 and 2 in various solvents were measured on a fluorescence spectrometer. Figure 3 showed the fluorescence emission spectra of compounds 1 and the free carbazole in acetonitrile and methanol. The fluorescence intensity of the carbazole moiety in these compounds is weaker than that of the free carbazole, especially in methanol. Similar phenomena were observed from model compound 2, namely, fluorescence of the carbazole moiety is quenched by a covalently linked dimer or oxetane unit. The fluorescence quenching of the carbazole moiety in model compounds is not a result of absorption of the excitation light

Photosensitized Repair of CPD and Oxetane

J. Phys. Chem. B, Vol. 114, No. 30, 2010 9829

SCHEME 1: Photosensitized Splitting Reactions of Dimer or Oxetane Unit of Model Compounds Under Irradiation with 328 nm Light

at 328 nm by the oxetane or dimer unit because they have no absorption above 300 nm. Additionally, there is no overlap between the emission spectra of N-methylcarbazole and the absorption spectra of the dimer and the oxetane. Thus singlet-singlet energy transfer is an improbable pathway for the fluorescence quenching. Therefore, an intramolecular electron transfer from the excited carbazole moiety to the dimer or oxetane unit should be responsible for the fluorescence quenching, and the degree of fluorescence quenching would reflect the efficiency (Q) of the electron transfer reaction.7e,i,10a In cyclohexane or THF solution, however, the fluorescence quenching was not observed, and fluorescence intensity of model compounds is even stronger than that of the free carbazole. Through comparing with fluorescence intensity of the free carbazole in a corresponding solvent, fluorescence quenching extents (Q) of the carbazole moiety in model compounds were obtained in various solvents, listed in Tables 1 and 2. Except for compound 3, no fluorescence quenching was observed in cyclohexane or THF. Model compounds 1 and 2 have low fluorescence quenching extents in acetonitrile and high values of Q in methanol or water/acetonitrile mixture. The model compounds with a short linker (a dimethylene), 1a and 2a, have higher Q values than the corresponding compounds with a relative long linker (a trimethylene), 1b and 2b. Using the Rehm-Weller equation (eq 1),12 free energy changes (∆Gfet) for the proposed electron-transfer reactions from the excited carbazole moiety to the dimer or oxetane unit can be estimated.

∆Gfet(eV) ) Eox(D) - Ered(A) + ∆Ecoul - ∆E0,0

(1) ∆Ecoul(eV) )

e 1 2 4πε0a ε 37.5

(

)

(2)

Where Ered of the dimer and the oxetane in acetonitrile was -2.213 and -1.8 ∼ -2.0 V14 vs saturated calomel electrode (SCE), respectively, and Eox of N-ethylcarbazole is 1.12 V

(SCE)15 and 1.20 V (normal hydrogen electrode, NHE)16 for N-methylindole. ∆Ecoul and ∆E0,0 are the coulomb term and the energy level of the excited state, respectively. The energy level of the excited state can be obtained from fluorescence peaks of model compounds in the corresponding solvents. While ε and a are the static dielectric constant of a solvent and the centerto-center distance between a donor and an acceptor, respectively. If a ) 5 Å, values of free energy changes (∆Gfet) can be calculated, as listed in Table 3. Calculation results show that the forward electron transfer is thermodynamically possible in polar solvents for all four model systems and is unfavorable for nonpolar solvents (cyclohexane, THF) for carbazole systems. This can well explain the fluorescence quenching phenomenon in various solvents. The photophysical and photochemical processes of 1 and 2 (represented as Cz-D(O)) are illuminated with a simple mechanistic scheme (Figure 4). Upon irradiation with light, the carbazole moiety absorbs a photon to produce the excited state (1Cz*-D(O)). The excited state has the following relaxation pathways: fluorescence (kf), internal conversion (kic), and electron transfer to a covalently linked dimer or oxetane (kfet). The charge-separated species (Cz•+-D(O)•-), formed by the electron transfer, undergoes two competitive processes: splitting (kspl) to produce M′ and Cz•+-M•-(it then becomes Cz-M by charge combination) and back electron transfer (kbet) to return to the starting substrate. In these processes, kfet and kspl contribute to the observed splitting quantum yield (Φ) of dimer or oxetane, while kbet, kf, and kic reduce the quantum efficiency. Quantum Yields for Splitting of Model Compounds. To measure the observed quantum yields (Φ) of the model compounds, all sample solutions were prepared in six solvents, cyclohexane, THF, acetonitrile, methanol, water/acetonitrile (3: 7), and water/acetonitrile (6:4) mixture, placed in cuvettes with a Teflon stopper, and then irradiated with 328 nm light for carbazole-model compounds and 295 nm light for the indolemodel compound from a fluorescence spectrometer. After certain time intervals, the absorption spectra of the irradiated solution were recorded by a UV-vis spectrometer. The intensity of irradiation light was measured for three times during one sample measurement, and the average of three measurements was employed. Based on these data, the quantum yields of dimer or oxetane splitting of model compounds were obtained, as listed in Tables 1 and 2. In nonpolar solvents, cyclohexane and THF, no fluorescence quenching relative to the free carbazole and no detectable splitting on UV-vis absorption spectra were observed for all four compounds containing carbazole. It is indicative of the fact that no forward electron transfer occurs in cyclohexane or THF, thus no subsequent splitting reaction takes place. Photosensitized splitting of the dimer or the oxetane unit of model compounds was observed in polar solvents except compound 1 in acetonitrile. The low fluorescence quenching extents of compound 1 in acetonitrile, reflecting low efficiencies of forward electron transfer, would lead to an undetectable level of splitting or even none at all. It is more important that splitting reactions of two kinds of model compounds (dimer and oxetane) seem to reveal two reverse solvent effects on the quantum yields. The splitting quantum yields decrease with increasing solvent polarity for carbazole-dimer systems (Table 1). For carbazole-oxetane models, however, the solvent effect is just opposite, an increase in the quantum yield with solvent polarity being observed (Table 2).

9830

J. Phys. Chem. B, Vol. 114, No. 30, 2010

Wu and Song

Figure 2. 1H NMR spectra of 1a (left) and 2a (right) in CD3OD/CD3CN (v/v 1:1) recorded after irradiation for (a) 0 min, (b) 15 min, (c) 45 min, and (d) 75 min and (e) neat 4a (left); neat 5a (right).

Figure 3. Fluorescence emission spectra of model compounds 1a, 1b, and N-methylcarbazole in acetonitrile (left) and methanol (right), normalized with absorbance of their solutions at 328 nm.

TABLE 1: Fluorescence Quenching Extent (Q) and the Splitting Quantum Yield (Φ) of Model Compounds 1a and 1b in Various Solvents 1a solvent

Q

cyclohexane/THF acetonitrile methanol CH3CN/H2O(7:3) CH3CN/H2O(4:6)

a

0.18 0.88 0.90 0.98

a

TABLE 3: Free Energy Change (∆Gfet, eV) of Forward Electron Transfer ReactionS for Four Model Systems in Three Solvents

1b Φ

Q

b

a

b

b

b

0.101 0.089 0.074

0.12 0.58 0.54 0.71

Φ

0.089 0.088 0.084

No fluorescence quenching. b No splitting detected.

Solvent

carbazoledimera

carbazoleoxetanea

indoledimerb

indoleoxetaneb

THF MeOH water

+0.33 -0.13 -0.15

-0.07∼ +0.13 -0.53∼ -0.33 -0.55∼ -0.35

-0.24 -0.29 -0.31

-0.64∼ -0.44 -0.70∼ -0.50 -0.71∼ -0.51

a Calculation with ∆E0,0 in various solvents and Eox ) 1.12 V (SCE)15 of N-ethylcarbazole. b Calculation with ∆E0,0 of N-ethylindole in various solvents and Eox ) 1.20 V (NHE)16 of N-methylindole.

TABLE 2: Fluorescence Quenching Extent (Q) and the Splitting Quantum Yield (Φ) of Model Compounds 2a, 2b, and 3 in Various Solvents 2a Solvent cyclohexane THF acetonitrile methanol CH3CN/H2O(7:3) CH3CN/H2O(4:6) a

2b

Q

3

Q

Φ

Q

Φ

0.13 0.86 0.94 0.98

0.011 0.122 0.174 0.246

Φ

a

b

a

b

a

b

a

b

0.11 0.70 0.76 0.92

0.014 0.110 0.098 0.124

0.086 0.58 0.50 0.59

0.010 0.089 0.089 0.111

c

0.99

c

0.279

see Table 1. b see Table 1. c No measurement.

In order to verify the reverse solvent behaviors, the splitting efficiency of the dimer or the oxetane unit of model compounds 1a and 2a was measured in a series of CH3CN/H2O binary solvents, and data are listed in Table 4. As the proportion of water in solvent mixture increases, quantum yields of carbazole-dimer model 1a decrease ranging from 0.089 in CH3CN/H2O (7:3) to 0.072 in CH3CN/H2O (1:9), and the quantum yields of 1a in high proportion of acetonitrile (9:1 and 10:0) are low due to low efficiencies of forward electron transfer (Q). However, a reverse solvent behavior for carbazole-oxetane model 2a was observed, and the quantum yield is ranged from 0.014 in neat CH3CN to 0.146 in CH3CN/

Figure 4. Photophysical and photochemical processes of model compounds.

H2O (3:7). This shows that two kinds of model compounds, dimer-carbazole and oxetane-carbazole, reveal the two reverse solvent effects on the splitting efficiency. In indole-dimer model systems with the same and short linker,10e we have observed the same solvent behavior with carbazole-dimer systems. To further confirm the solvent behavior of the oxetane model system with a short linker, a covalently linked indole-oxetane model compound 3 was prepared, and the corresponding data were measured, as listed

Photosensitized Repair of CPD and Oxetane

J. Phys. Chem. B, Vol. 114, No. 30, 2010 9831

TABLE 4: Fluorescence Quenching Extent (Q) and Splitting Quantum Yield (Φ) of Model Compounds 1a and 2a in CH3CN/H2O Binary Solvents 1a solvent CH3CN/H2O CH3CN/H2O CH3CN/H2O CH3CN/H2O CH3CN/H2O CH3CN/H2O

2a

Q (10:0) (9:1) (7:3) (5:5) (3:7) (1:9)

Q

Φ

-∆Gbet ) Eox - Ered + ∆Ecoul

0.11 0.53 0.76 0.88 0.95

0.014 0.067 0.098 0.123 0.146

The solvent reorganization energy of back electron transfer (λs) can be estimated using the equation as follows:18

b

b

Φ a

0.18 0.79 0.90 0.95 0.98 0.99

be estimated by using thermodynamic redox potentials. The free energy difference between the charge-transfer state and the ground state is given by eq 5:

0.070 0.089 0.088 0.081 0.072

a

No splitting detected. b No measurement due to poor solubility of 2a in CH3CN/H2O (1:9).

in Table 2. The solvent effect on the splitting quantum efficiency agreed well with that of the carbazole-oxetane systems, i.e., values of Φ increase with polarity of solvents. Hence, two reverse solvent effects were also observed from two indole model systems (dimer and oxetane). It can be concluded that the two model systems (dimer and oxetane) with a short linker reveal two reverse solvent effects on the splitting efficiencies. Interception of Two Reverse Solvent Effects Based on Marcus Theory. In our previous paper,10e two reverse solvent effects on the splitting efficiency have been observed from covalently linked indole-dimer compounds with different length linkers. One is that the splitting quantum yield is a normal relation with solvent polarity for the compounds with long linkers. Another is an inverse relation for the compounds with short linkers. We have demonstrated that the two reverse solvent effects derive from the difference in the conformational change between short- and long-linker compounds. In this work, all model systems are compounds with a short linker. The dimer-containing systems reveal a contrary solvent effect to the oxetane-containing systems. For all reported model systems,7d,i,10a,b,e fluorescence quenching extents increase with solvent polarity, namely, forward electron transfer is accelerated with increasing polarity. This implies that forward electron transfer lies in the Marcus normal region (vide infra). Hence, it can be excluded that the forward electron transfer is responsible for two reverse solvent effects, and back electron transfer, which is a key leading to low splitting efficiencies of model compounds, may be the factor controlling the two reverse solvent effects. To rationalize this observation, back electron transfer is discussed according to Marcus’ theory17(eqs 3 and 4):

k ) A′ exp(-G+ /kBT)

(3)

G+ ) (∆G + λs)2 /4λs

(4)

The free energy of activation (G+) can be obtained from the change of free energy (∆G) and the solvent reorganization energy (λs) for back electron transfer. The former is the energy level of the charge-separated state (Cz•+-D(O)•-), which can

(5)

-1 λs ) (e/4πε0)[(2rD)-1 + (2rA)-1 - (RDA)-1](ε-1 op - εs ) (6)

where rD and rA are the ionic radii of the donor and the acceptor, respectively, and RDA is the distance between a donor and an acceptor, that is a in eq 2. While εop and εs are the optical and static dielectric constant of the solvent, respectively, with εop ≈ n2, n being the solvent refractive index. According to Marcus’ theory (eqs 3 and 4), if λs value is less than -∆Gbet, then the back electron transfer would enter the so-called Marcus inverted region, and back electron transfer would be slowed when the value of -∆Gbet increases (eq 5), e.g., with decreasing solvent polarity. Back electron transfer slowing would cause an increase in the splitting efficiency, i.e., the splitting efficiency increases with decreasing solvent polarity. In contrast, if λs > -∆Gbet, then back electron transfer is in the Marcus normal region, and it is a reverse situation that more efficient splitting occurs in solvents of higher polarity. The terms influencing the values of -∆Gbet and λs are ∆Ecoul, RDA, and (εop-1 - εs-1) in different solvents. Among them, two key factors are dielectric constant of solvents and the centerto-center distance between a donor and an acceptor, RDA (a). The conformation change of model compounds in different solvents would give different values of RDA (a). To evaluate the solvent effect on the conformational change of model compounds, 1H NMR spectra and fluorescence emission spectra of model compounds were measured in various solvents and compared with the free dimer (10) and the oxetane (16) for 1H NMR measurements. The changes in spectra for different solvents are small, and data show that the physical properties of solvents, but not conformational changes, are responsible for these spectral changes, e.g., the changes in both 1H NMR and fluorescence spectra are related to the dielectric constant of solvents. If the small conformational change is neglected, namely, RDA (a) is taken as a constant, we can estimate solvent effects on ∆Ecoul and (εop-1 - εs-1) as well as -∆Gbet and λs, see Table 5. With increasing solvent polarity, values of ∆Ecoul and -∆Gbet would decrease and (εop-1 - εs-1) or λs increases. If back electron transfer lies in the Marcus normal region, then the free energy of activation (G+) would increase with solvent polarity, leading to slow back electron transfer kbet, and enhance the splitting quantum yield (Table 5). If back electron transfer lies in the Marcus inverted region, then a contrary varying trend to the situation above is expected. Therefore, back electron transfer

TABLE 5: Solvent Effects on Back Electron Transfer solvents polarity +

∆Ecoul (-∆Gbet)

(εop-1 - εs-1) (λs)

-b

+

a

(εs +, εop -) a

b

Symbol “+” indicates increasing. Symbol “-” indicates decreasing.

Marcus region

G+

kbet

Φ

normal, -∆Gbet < λs

+

-

+

inverted, -∆Gbet > λs

-

+

-

9832

J. Phys. Chem. B, Vol. 114, No. 30, 2010

lying in two Marcus regions would reveal two reverse solvent effects on the splitting quantum yield. If back electron transfer lies in the Marcus inverted region for dimer-model systems and the Marcus normal region for oxetane-model systems, then the above theoretical expectation would explain the two reverse solvent effects completely. Based on thermodynamic data, the probability of back electron transfer lying in different Marcus regions can be discussed. First, due to the difference in values of Ered, -2.2 V13 for dimer-model systems and -1.8 ∼ -2.0 V14 for oxetane-model systems, the former is higher by 0.2-0.4 V than the latter. Second, oxetane systems (2 and 3) with a larger driving force of forward electron transfer (∆Gfet, lie in Marcus normal region) have a lower fluorescence quenching extent than the corresponding dimermodel systems (1 and 8 in ref 10e), see Tables 1 and 2. This indicates that the oxetane systems have a longer donor-acceptor distance than the corresponding dimer system. If the change of donor-acceptor distance in the back electron transfer process is neglected, then a larger a (RDA) would give a lower -∆Gbet and a higher λs for oxetane systems than dimer systems. Hence, combining 0.2-0.4 V lower value of -∆Gbet for oxetane systems than dimer systems, it is possible that back electron transfer would lie in the Marcus normal region for oxetane systems with higher λs and lower -∆Gbet, and dimer systems lie in the Marcus inverted region. Comparing quantum yields of two model systems with the same linker and electron donor, oxetane systems have higher Φ over dimer systems. A reasonable inference is that the repair of the oxetane intermediate via electron transfer would be also highly efficient, that is, back electron transfer is suppressed well in the charge separated state for oxetane systems over dimer systems. This step should not be responsible for low repair efficiency of (6-4) photolyase. Thus, the dark reaction converting (6-4) product to the oxetane intermediate may cause a low repair efficiency. For example, there may be an equilibrium between (6-4) product-enzyme and oxetane-enzyme complexes after the enzyme binds to (6-4) product. When the excited cofactor transfers an electron to the substrate in these complexes, the latter complex can undergo an efficient repair, and the former may be inefficient, thus leading to a low repair efficiency. However, in the crystal structure of the (6-4) photolyase from D. melanogaster bound to the (6-4) lesion, no oxetane formation was observed, and the determined positions of the two conserved histidines are incompatible with the proposed oxetane formation.5a Hence, it has been proposed that a new nonoxetane mechanism that avoids the dark reaction converting (6-4) product to oxetane and suggests a direct electron injection into the lesion, followed by the formed (6-4) radical anion undergoing repair via hydroxyl-group transfer. Recently, quantum chemical calculations19 also showed that it is probable for a nonadiabatic repair of the excited (6-4) radical anion, which is formed by a direct electron transfer to (6-4) product, via hydroxyl transfer. However, many problems in the nonoxetane mechanism remain to be clarified. Acknowledgment. This work was supported by the National Natural Science Foundation of China (Grant no. 30870581, 20972149). Supporting Information Available: Synthesis procedures, spectral characterization data, and copies of the 1H (300 MHz) and 13C NMR (75 or 100 MHz) spectra, for new compounds 1-5, 7-10, and 14. This information is available free of charge via the Internet at http://pubs.acs.org.

Wu and Song References and Notes (1) Cadet, J.; Vigny, P. The Photochemistry of Nucleic Acids. In Bioorganic Photochemistry; Morrison, H., Ed.; Wiley Interscience: New York, 1990; Vol. 1, pp 1-272; Friedberg, E. C.; Walker, G. C.; Siede, W. DNA Repair and Mutagenesis, ASM Press: Washington, DC, 1995; Taylor, J.-S. Pure Appl. Chem. 1995, 67, 183–190. Cadet, J.; Berger, M.; Douki, T.; Morin, B.; Raoul, S.; Ravanat, J.-L.; Spinelli, S. Biol. Chem. 1997, 378, 1275–1286. (2) Sancar, A. Biochemistry 1994, 33, 2–9. Sancar, A. Chem. ReV. 2003, 103, 2203–2237. Sancar, A. J. Biol. Chem. 2008, 283, 32153–32157. (3) Kao, Y.-T.; Saxena, C.; Wang, L. J.; Sancar, A.; Zhong, D. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 16128–16132. Chang, C.-W.; Guo, L.; Kao, Y.-T.; Li, J.; Tan, C.; Li, T.; Saxena, C.; Liu, Z.; Wang, L.; Sancar, A.; Zhong, D. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 2914–2919. (4) Kim, S. T.; Malhotra, K.; Smith, C. A.; Taylor, J.-S.; Sancar, A. J. Biol. Chem. 1994, 269, 8535–8540. Zhao, X.; Liu, J.; Hsu, D. S.; Zhao, S.; Taylor, J. S.; Sancar, A. J. Biol. Chem. 1997, 272, 32580–32590. Hitomi, K.; Nakamura, H.; Kim, S. H.; Mizukoshi, T.; Ishikawa, T.; Iwai, S.; Todo, T. J. Biol. Chem. 2001, 276, 10103–10109. Wang, Y. S.; Gaspar, P. P.; Taylor, J.-S. J. Am. Chem. Soc. 2000, 122, 5510–5519. Asgatay, S.; Petermann, C.; Harakat, D.; Guillaume, D.; Taylor, J.-S.; Clivio, P. J. Am. Chem. Soc. 2008, 130, 12618–12619. (5) (a) Maul, M. J.; Barends, T. R. M.; Glas, A. F.; Cryle, M. J.; Domratcheva, T.; Schneider, S.; Schlichting, I.; Carell, T. Angew. Chem., Int. Ed. 2008, 47, 10076–10080. (b) Glas, A. F.; Schneider, S.; Maul, M. J.; Hennecke, U.; Carell, T. Chem.sEur. J. 2009, 15, 10387–10396. (6) For reviews: Begley, T. P. Acc. Chem. Res. 1994, 27, 394–401. Heelis, P. F.; Hartman, R. F.; Rose, S. D. Chem. Soc. ReV. 1995, 24, 289– 297. Carell, T.; Burgdorf, L. T.; Kundu, L. M.; Cichon, M. Curr. Opin. Chem. Biol. 2001, 5, 491–498. (7) (a) Hartman, R. F.; Van Camp, J. R.; Rose, S. D. J. Org. Chem. 1987, 52, 2684–2689. (b) Kim, S.-T.; Rose, S. D. Photochem. Photobiol. 1988, 47, 725–729. (c) Van Camp, J. R.; Hartman, R. F.; Rose, S. D. Photochem. Photobiol. 1988, 48, 635–641. (d) Kim, S.-T.; Hartman, R. F.; Rose, S. D. Photochem. Photobiol. 1990, 52, 789–794. (e) Kim, S.-T.; Rose, S. D. J. Phys. Org. Chem. 1990, 3, 581–586. (f) Hartman, R. F.; Rose, S. D. J. Am. Chem. Soc. 1992, 114, 3559–3560. (g) Hartman, R. F. S.; Rose, D. S. J. Org. Chem. 1992, 57, 2302–2306. (h) Kim, S.-T.; Rose, S. D. J. Photochem. Photobiol., B 1992, 12, 179–191. (i) Hartzfeld, D. G.; Rose, S. D. J. Am. Chem. Soc. 1993, 115, 850–854. (j) Young, T.; Kim, S.-T.; Rose, S. D. In CRC Handbook of Organic Photochemistry and Photobiology; Horspool, W., Song, P.-S., Ed.; CRC Press: New York, 1995, pp 1332-1346. (8) (a) Epple, R.; Wallenborn, E.-U.; Carell, T. J. Am. Chem. Soc. 1997, 119, 7440–7451. (b) Butenandt, J.; Eker, A. P. M.; Carell, T. Chem.sEur. J. 1998, 4, 642–654. (c) Carell, T.; Epple, R. Eur. J. Org. Chem. 1998, 1245–1258. (d) Epple, R.; Carell, T. Angew. Chem., Int. Ed. Engl. 1998, 37, 938–941. (e) Epple, R.; Carell, T. J. Am. Chem. Soc. 1999, 121, 7318– 7329. (f) Butenandt, J.; Burgdorf, L. T.; Carell, T. Angew. Chem., Int. Ed. Engl. 1999, 38, 708–711. (g) Butenandt, J.; Epple, R.; Wallenborn, E.-U.; Eker, A. P. M.; Gramlich, E.-U.; Carell, T. Chem.sEur. J. 2000, 6, 62–72. (h) Cichon, M. K.; Arnold, S.; Carell, T. Angew. Chem., Int. Ed. 2002, 41, 767–770. (i) Breeger, S.; von Meltzer, M.; Hennecke, U.; Carell, T. Chem.sEur. J. 2006, 12, 6469–6477. (9) Stafforst, T.; Diederichsen, U. Chem. Commun. 2005, 3430–3432. (10) (a) Song, Q. H.; Tang, W. J.; Hei, X. M.; Wang, H. B.; Guo, Q. X.; Yu, S. Q. Eur. J. Org. Chem. 2005, 1097–1106. (b) Song, Q. H.; Wang, H. B.; Tang, W. J.; Guo, Q. X.; Yu, S. Q. Org. Biomol. Chem. 2006, 4, 291–298. (c) Tang, W. J.; Song, Q. H.; Wang, H. B.; Yu, J. Y.; Guo, Q. X. Org. Biomol. Chem. 2006, 4, 2575–2580. (d) Song, Q. H.; Tang, W. J.; Ji, X. B.; Wang, H. B.; Guo, Q. X. Chem.sEur. J. 2007, 13, 7762–7770. (e) Tang, W. J.; Guo, Q. X.; Song, Q. H. J. Phys. Chem. B 2009, 113, 7205– 7210. (11) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochemistry, 2nd ed.; Marcel Dekker Inc.: New York, 1993. (12) Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 259–271. Weller, A. Z. Phys. Chem. (Muenchen, Ger.) 1982, 133, 93–98. (13) Scannell, M. P.; Prakash, G.; Falvey, D. E. J. Phys. Chem. A 1997, 101, 4332–4337. (14) Joseph, A.; Prakash, G.; Falvey, D. E. J. Am. Chem. Soc. 2000, 122, 11219–11225. (15) Ambrose, J. F.; Nelson, R. F. J. Electrochem. Soc. 1968, 16, 1159– 1164. (16) Lind, J.; Shen, X.; Eriksen, T. E.; Mere´nyi, G. J. Am. Chem. Soc. 1990, 112, 419–482. (17) Marcus, R. A. J. Chem. Phys. 1956, 24, 966–978. (18) Marcus, R. A. Can. J. Chem. 1959, 37, 155–163. Marcus, R. A. Annu. ReV. Phys. Chem. 1964, 15, 155–196. (19) Domratcheva, T.; Schlichting, I. J. Am. Chem. Soc. 2009, 131, 17793–17799.

JP1035579