Theoretical study on the repair mechanism of the (6-4) photolesion by the (6-4) photolyase

UV irradiation of DNA can lead to the formation of mutagenic (6-4) pyrimidine-pyrimidone photolesions. The (6-4) photolyases are the enzymes responsible for the photoinduced repair of such lesions. On the basis of the recently published crystal structure of the (6-4) photolyase bound to DNA [Maul et al. 2008] and employing quantum mechanics/molecular mechanics techniques, a repair mechanism is proposed, which involves two photoexcitations. The flavin chromophore, initially being in its reduced anionic form, is photoexcited and donates an electron to the (6-4) form of the photolesion. The photolesion is then protonated by the neighboring histidine residue and forms a radical intermediate. The latter undergoes a series of energy stabilizing hydrogen-bonding rearrangements before the electron back transfer to the flavin semiquinone. The resulting structure corresponds to the oxetane intermediate, long thought to be formed upon DNA-enzyme binding. A second photoexcitation of the flavin promotes another electron transfer to the oxetane. Proton donation from the same histidine residue allows for the splitting of the four-membered ring, hence opening an efficient pathway to the final repaired form. The repair of the lesion by a single photoexcitation was shown not to be feasible.     

Excess electron transfer in flavin-capped,thymine dimer-containing DNA hairpins

The transfer of an excess electron through DNA was investigated with DNA hairpins, which contain a flavin cap functioning as an electron donor. A thymine dimer with an open backbone acts as the electron acceptor. The dimer translates the electron capture into a strand break, which is readily detectable by HPLC. Analysis of four hairpins, in which the distance between the flavin donor and the dimer acceptor was systematically increased, revealed a flat distance dependence of the repair efficiency supporting the view that excess electrons hop through DNA using intermediate A-T base pairs as temporary charge carriers.     

Photoinduced reductive repair of thymine glycol: implications for excess electron transfer through DNA containing modified bases

Photoinduced reduction of thymine glycol in oligodeoxynucleotides was investigated using either a reduced form of flavin adenine dinucleotide (FADH(-)) as an intermolecular electron donor or covalently linked phenothiazine (PTZ) as an intramolecular electron donor. Intermolecular electron donation from photoexcited flavin (FADH(-)) to free thymidine glycol generated thymidine in high yield, along with a small amount of 6-hydroxy-5,6-dihydrothymidine. In the case of photoreduction of 4-mer long single-stranded oligodeoxynucleotides containing thymine glycol by *FADH(-), the restoration yield of thymine was varied depending on the sequence of oligodeoxynucleotides. Time-resolved spectroscopic study on the photoreduction by laser-excited N,N-dimethylaniline (DMA) suggested elimination of a hydroxyl ion from the radical anion of thymidine glycol with a rate constant of approximately 10(4) s(-1) generates 6-hydroxy-5,6-dihydrothymidine (6-HOT(*)) as a key intermediate, followed by further reduction of 6-HOT(*) to thymidine or 6-hydroxy-5,6-dihydrothymdine (6-HOT). On the other hand, an excess electron injected into double-stranded DNA containing thymine glycol was not trapped at the lesion but was further transported along the duplex. Considering redox properties of the nucleobases and PTZ, competitive excess electron trapping at pyrimidine bases (thymine, T and cytosine, C) which leads to protonation of the radical anion (T(-)(*), C(-)(*)) or rapid back electron transfer to the radical cation of PTZ (PTZ(+)(*)), is presumably faster than elimination of the hydroxyl ion from the radical anion of thymine glycol in DNA.     

Insights into Light‐driven DNA Repair by Photolyases: Challenges and Opportunities for Electronic Structure Theory

Ultraviolet radiation causes two of the most abundant mutagenic and cytotoxic DNA lesions: cyclobutane pyrimidine dimers and 6‐4 photoproducts. (6‐4) Photolyases are light‐activated enzymes that selectively bind to DNA and trigger repair of mutagenic 6‐4 photoproducts via photoinduced electron transfer from flavin adenine dinucleotide anion (FADH^?) to the lesion triggering repair. This review provides an overview of the sequential steps of the repair process, that is light absorption and resonance energy transfer, photoinduced electron transfer and electron‐induced splitting mechanisms, with an emphasis on the role of theory and computation. In addition, theoretical calculations and physical properties that can be used to classify specific mechanism are discussed in an effort to trace the fundamental aspects of each individual step and assist the interpretation of experimental data. The current challenges and suggested future directions are outlined for each step, concluding with a view on the future.     

Investigation of excess-electron transfer in DNA double-duplex systems allows estimation of absolute excess-electron transfer and CPD cleavage rates

To investigate the parameters and rates that determine excess-electron transfer processes in DNA duplexes, we developed a DNA double-duplex system containing a reduced and deprotonated flavin donor at the junction of two duplexes with either the same or different electron acceptors in the individual duplex substructures. This model system allows us to bring the two electron acceptors in the duplex substructures into direct competition for injected electrons and this enables us to decipher how the kind of acceptor influences the transfer data. Measurements with the electron acceptors 8-bromo-dA (BrdA), 8-bromo-dG (BrdG), 5-bromo-dU (BrdU), and a cyclobutane pyrimidine dimer, which is a UV-induced DNA lesion, allowed us to obtain directly the maximum overall reaction rates of these acceptors and especially of the T=T dimer with the injected electrons in the duplex. In line with previous observations, we detected that the overall dimer cleavage rate is about one order of magnitude slower than the debromination of BrdU. Furthermore, we present a more detailed explanation of why sequence dependence cannot be observed when a T=T dimer is used as the acceptor and we estimate the absolute excess-electron hopping rates.     

Context-dependent photodimerization in isolated thymine-thymine steps in DNA

The effect of 280 nm irradiation on a family of synthetic DNA hairpins possessing an alkane linker connecting a six-base pair stem having a single T-T step located at different positions within the hairpin has been investigated. A single adduct assigned to the product of 2+2 dimerization is obtained except in the case of a T-T step located adjacent to the linker, in which case both 2+2 and 6-4 adducts are obtained. The efficiency of dimerization is similar for three hairpins having a T-T step located within the duplex interior. Lower efficiency is observed for a T-T step located at the open end of the hairpin and in T overhangs, whereas higher efficiency is observed for the T-T step adjacent to the linker and in a single T bulge. The context-dependence of dimerization efficiency is discussed.     

Neutral histidine and photoinduced electron transfer in DNA photolyases

The two major UV-induced DNA lesions, the cyclobutane pyrimidine dimers (CPD) and (6-4) pyrimidine-pyrimidone photoproducts, can be repaired by the light-activated enzymes CPD and (6-4) photolyases, respectively. It is a long-standing question how the two classes of photolyases with alike molecular structure are capable of reversing the two chemically different DNA photoproducts. In both photolyases the repair reaction is initiated by photoinduced electron transfer from the hydroquinone-anion part of the flavin adenine dinucleotide (FADH(-)) cofactor to the photoproduct. Here, the state-of-the-art XMCQDPT2-CASSCF approach was employed to compute the excitation spectra of the respective active site models. It is found that protonation of His365 in the presence of the hydroquinone-anion electron donor causes spontaneous, as opposed to photoinduced, coupled proton and electron transfer to the (6-4) photoproduct. The resulting neutralized biradical, containing the neutral semiquinone and the N3'-protonated (6-4) photoproduct neutral radical, corresponds to the lowest energy electronic ground-state minimum. The high electron affinity of the N3'-protonated (6-4) photoproduct underlines this finding. Thus, it is anticipated that the (6-4) photoproduct repair is assisted by His365 in its neutral form, which is in contrast to the repair mechanisms proposed in the literature. The repair via hydroxyl group transfer assisted by neutral His365 is considered. The repair involves the 5'base radical anion of the (6-4) photoproduct which in terms of electronic structure is similar to the CPD radical anion. A unified model of the CPD and (6-4) photoproduct repair is proposed.     

Evaluation of dsDNA as a wire for redox-active proteins

The redox-active multiligand-binding flavoprotein dodecin binds flavins with high affinity when they are oxidized, whereas flavin reduction induces the dissociation of the holoprotein complex in apododecin and free flavin ligands. Dodecin could be reconstituted at flavin-terminated dsDNA monolayers. The binding and release of apododecin triggered by the redox state of the flavins can be monitored by surface-sensitive techniques such as surface plasmon resonance and quartz crystal microbalance measurements with dissipation monitoring. It has been shown that flavin reduction followed by the release of apododecin can be achieved by mediated electron transfer in the presence of the redox mediator amino ethyl viologen and by chemical flavin reduction, whereas flavin reduction by direct electron transfer via the dsDNA tethers is not possible. The combination of electrochemistry with surface-sensitive techniques such as surface plasmon resonance or quartz crystal microbalance measurements with dissipation monitoring could be highly beneficial to confirm or disprove the mechanism, which has been postulated for the action of primases, which contain a [4Fe4S] cluster and are involved in DNA replication. It has been postulated that these enzymes bind the DNA template when the cluster is in the [4Fe4S]³⁺ state, whereas they are released when the cluster is reduced via electron transfer through DNA and the protein environment.     

Influence of DNA structure on adjacent site cooperative binding

Previous NMR studies of Hoechst 33258 with the d(CTTTTGCAAAAG)2 sequence have shown very strong (K2 > K1) cooperativity between two adjacent binding sites (Searle, M. S.; Embrey, K. J. Nucleic Acids Res. 1990, 18 (13), 3753- 3762). In contrast, surface plasmon resonance (SPR) results with the hairpin analog of the same sequence show significantly reduced cooperativity. In an effort to explain the difference, two-dimensional (2-D) NMR experiments were done on both duplex and hairpin. Hoechst 33258 and an amidine analog, DB183, show very strong cooperativity with the duplex DNA but much weaker cooperativity with the hairpin. The significantly lower thermal melting temperature (Tm) of the duplex (34.8 degrees C) in comparison to its hairpin analog (62.3 degrees C) supports the idea of a dynamic difference between the two DNA structures that can influence cooperativity in binding. These results confirm the role of conformational entropy in positive cooperativity in some DNA interactions.     

Light-induced conformational change and product release in DNA repair by (6-4) photolyase

Proteins of the cryptochrome/photolyase family share high sequence similarities, common folds, and the flavin adenine dinucleotide (FAD) cofactor, but exhibit diverse physiological functions. Mammalian cryptochromes are essential regulatory components of the 24 h circadian clock, whereas (6-4) photolyases recognize and repair UV-induced DNA damage by using light energy absorbed by FAD. Despite increasing knowledge about physiological functions from genetic analyses, the molecular mechanisms and conformational dynamics involved in clock signaling and DNA repair remain poorly understood. The (6-4) photolyase, which has strikingly high similarity to human clock cryptochromes, is a prototypic biological system to study conformational dynamics of cryptochrome/photolyase family proteins. The entire light-dependent DNA repair process for (6-4) photolyase can be reproduced in a simple in vitro system. To decipher pivotal reactions of the common FAD cofactor, we accomplished time-resolved measurements of radical formation, diffusion, and protein conformational changes during light-dependent repair by full-length (6-4) photolyase on DNA carrying a single UV-induced damage. The (6-4) photolyase by itself showed significant volume changes after blue-light activation, indicating protein conformational changes distant from the flavin cofactor. A drastic diffusion change was observed only in the presence of both (6-4) photolyase and damaged DNA, and not for (6-4) photolyase alone or with undamaged DNA. Thus, we propose that this diffusion change reflects the rapid (50 μs time constant) dissociation of the protein from the repaired DNA product. Conformational changes with such fast turnover would likely enable DNA repair photolyases to access the entire genome in cells.     

Theoretical study of excitation energy transfer in DNA photolyase

Photolyase (PL) is a DNA repair enzyme which splits UV light-induced thymine dimers on DNA by an electron transfer reaction occurring between the photoactivated FADH(-) cofactor and the DNA dimer in the DNA/PL complex. The crystal structure of the DNA/photolyase complex from Anacystis nidulans has been solved. Here, using the experimental crystal structure, we re-examine the details of the repair electron transfer reaction and address the question of energy transfer from the antenna HDF to the redox active FADH(-) cofactor. The photoactivation of FADH(-) immediately preceding the electron transfer is a key step in the repair mechanism that is largely left unexamined theoretically. An important butterfly thermal motion of flavin is identified in ab initio calculations; we propose its role in the back electron transfer from DNA to photolyase. Molecular dynamics simulation of the whole protein/DNA complex is carried out to obtain relevant cofactor conformations for ZINDO/S spectroscopic absorption and fluorescence calculations. We find that significant thermal broadening of the spectral lines, due to protein dynamics, as well as the alignment of the donor HDF and the acceptor FADH(-) transition dipole moments both contribute to the efficiency of energy transfer. The geometric factor of F?rster's dipolar coupling is calculated to be 1.82, a large increase from the experimentally estimated 0.67. Using F?rster's mechanism, we find that the energy transfer occurs with remarkable efficiency, comparable with the experimentally determined value of 98%.     

Quantum Chemical Study of the Enzymatic Repair of T(6‐4)C/C(6‐4)T UV‐Photolesions by DNA Photolyases

Several strategies have evolved to repair one of the abundant UV radiation‐induced damages caused to DNA, namely the mutagenic pyrimidine (6‐4) pyrimidone photolesions. DNA (6‐4)‐photolyases are enzymes repairing these lesions by a photoinitiated electron transfer. An important aspect of a possible repair mechanism is its generality and transferability to different (6‐4) lesions. Therefore, previously suggested mechanisms for the repair of the T(6‐4)T lesion are here transferred to the T(6‐4)C and C(6‐4)T lesions and investigated theoretically using quantum chemical methods. Despite the different functional groups of the pyrimidine bases involved, a general valid molecular mechanism was identified, in which the initial step is an electron transfer coupled to a proton transfer from the protonated HIS365 to the N3^′ nitrogen of the 3^′ pyrimidine, followed by an intramolecular OH/NH₂ transfer in one concerted step, which does not require an oxetane/azetidine or isolated water/ammonia intermediate.     

Chemistry and biology of DNA repair

Numerous agents of endogenous and exogenous origin damage DNA in our genome. There are several DNA-repair pathways that recognize lesions in DNA and remove them through a number of diverse reaction sequences. Defects in DNA-repair proteins are associated with several human hereditary syndromes, which show a marked predisposition to cancer. Although DNA repair is essential for a healthy cell, DNA-repair enzymes counteract the efficiency of a number of important antitumor agents that exert their cytotoxic effects by damaging DNA. DNA-repair enzymes are therefore also targets for drug design. DNA-repair processes differ greatly in their nature and complexity. Whereas some pathways only require a single enzyme to restore the original DNA sequence, others operate through the coordinated action of 30 or more proteins. Our understanding of the genetic, biochemical, and structural basis of DNA repair and related processes has increased dramatically over the past decade. This review summarizes the latest developments in this field.     

Crystal Structure of the T(6‐4)C Lesion in Complex with a (6‐4) DNA Photolyase and Repair of UV‐Induced (6‐4) and Dewar Photolesions

UV‐light irradiation induces the formation of highly mutagenic lesions in DNA, such as cis‐syn cyclobutane pyrimidine dimers (CPD photoproducts), pyrimidine(6‐4)pyrimidone photoproducts ((6‐4) photoproducts) and their Dewar valence isomers ((Dew) photoproducts). Here we describe the synthesis of defined DNA strands containing these lesions by direct irradiation. We show that all lesions are efficiently repaired except for the T(Dew)T lesion, which cannot be cleaved by the repair enzyme under our conditions. A crystal structure of a T(6‐4)C lesion containing DNA duplex in complex with the (6‐4) photolyase from Drosophila melanogaster provides insight into the molecular recognition event of a cytosine derived photolesion for the first time. In light of the previously postulated repair mechanism, which involves rearrangement of the (6‐4) lesions into strained four‐membered ring repair intermediates, it is surprising that the not rearranged T(6‐4)C lesion is observed in the active site. The structure, therefore, provides additional support for the newly postulated repair mechanism that avoids this rearrangement step and argues for a direct electron injection into the lesion as the first step of the repair reaction performed by (6‐4) DNA photolyases.     

Flavin- and Deazaflavin-Containing Model Compounds Mimic the Energy Transfer Step in Type-II DNA-Photolyases

A distinct energy transfer between the deazaflavin and the flavin is observed in compounds that mimic the DNA repair function of DNA-photolase enzymes (see sketch). Studies on these models prove that the deazaflavin acts solely as photoantenna and suggest that the two cofactors have to be separated within the enzyme to suppress a competitive intercofactor electron transfer reaction.     

GA and AG sequences of DNA react with cisplatin at comparable rates

The sequence selectivity of the antitumor drug cisplatin (cis-[PtCl(2)(NH(3))(2)] (1)) between the 5'-AG-3' and 5'-GA-3' sites of DNA has been a matter of discussion for more than twenty years. In this work, we compared the reactivity of GA and AG sequences of DNA towards the aquated forms of cisplatin (cis-[PtCl(NH(3))(2)(H(2)O)](+) (2), cis-[Pt(NH(3))(2)(H(2)O)(2)](2+) (3), and cis-[Pt(OH)(NH(3))(2)(H(2)O)](+) (4)) using two sets of experiments. In the first, we investigated a DNA hairpin, whose duplex stem contained a TGAT sequence as the single reactive site, and determined the individual rate constants of platination with 2 and 3 for G and A in acidic solution. The rate constants at 20 degrees C in 0.1M NaClO(4) at pH 4.5+/-0.1 were 0.09(4) M(-1)s(-1) (G) and 0.11(3) M(-1)s(-1) (A) for 2, and 9.6(1) M(-1)s(-1) (G) and 1.7(1) M(-1)s(-1) (A) for 3. These values are similar to those obtained previously for an analogous hairpin that contained a TAGT sequence. The monoadducts formed with 2 by both GA purines are extremely long-lived, partly as a result of the slow hydrolysis of the chloro monoadduct at A, and partly because of the very low chelation rate (1.4 x 10(-5)s(-1) at 20 degrees C) of the aqua monoadduct on the guanine. In the second set of experiments, we incubated pure or enriched samples of 1, 2, 3, or 4 for 18-64 h at 25 degrees C with a 19 base pair (bp) DNA duplex, whose radiolabeled top strand contained one GA and one AG sequence as the only reactive sites. Quantification of the number of GA and AG cross-links afforded a ratio of about two in favor of AG, irrespective of the nature of the leaving ligands. These results disagree with a previous NMR spectroscopy study, and indicate that GA sequences of DNA are substantially more susceptible to attack by cisplatin than previously thought.     

Sequence-specific fluorescence detection of DNA by polyamide-thiazole orange conjugates

Fluorescent methods to detect specific double-stranded DNA sequences without the need for denaturation may be useful in the field of genetics. Three hairpin pyrrole-imidazole polyamides 2-4 that target their respective sequences 5'-WGGGWW-3', 5'-WGGCCW-3', and 5'-WGWWCW-3' (W = A or T) were conjugated to thiazole orange dye at the C-termini to examine their fluorescence properties in the presence and absence of match duplex DNA. The conjugates fluoresce weakly in the absence of DNA but showed significant enhancement (>1000-fold) upon the addition of 1 equiv of match DNA and only slight enhancement with the addition of mismatch DNA. The polyamide-dye conjugates bound specific DNA sequences with high affinity (Ka > 10(8) M(-1)) and unwound the DNA duplex through intercalation (unwinding angle, phi, approximately 8 degrees). This new class of polyamides provides a method to specifically detect DNA sequences without denaturation.     

DNA Microenvironment Monitored by Controlling Redox Blinking

The rate of a bimolecular reaction between a fluorophore and a freely diffusing molecule in the solvent depends on the accessibility of the fluorophore for collision with the molecule. We previously reported that the observation of blinking, caused by the formation of R6G in the excited triplet state (³R6G*) and its quenching reaction with O₂, allowed us to monitor the DNA conformational changes between a duplex and a hairpin. However, the small molecular size of O₂ hampered sensitive monitoring of the microenvironment changes around R6G. In this study, we control redox blinking by adding a reductant ascorbic acid 2‐phosphate (VcP), which converts ³R6G* into the radical anion form R6G^.?, and by adding a bulky oxidant FeDTPA. The bimolecular electron‐transfer rate between R6G^.? and bulky FeDTPA was more strongly affected by microenvironment changes around R6G, compared with that between ³R6G* and the smaller O₂. This allowed us to monitor subtle DNA conformational changes caused by a single different nucleotide.     

A Simple,High‐Yield Synthesis of DNA Duplexes Containing a Covalent,Thermally Cleavable Interstrand Cross‐Link at a Defined Location

Interstrand DNA–DNA cross‐links are highly toxic to cells because these lesions block the extraction of information from the genetic material. The pathways by which cells repair cross‐links are important, but not well understood. The preparation of chemically well‐defined cross‐linked DNA substrates represents a significant challenge in the study of cross‐link repair. Here a simple method is reported that employs “post‐synthetic” modifications of commercially available 2′‐deoxyoligonucleotides to install a single cross‐link in high yield at a specified location within a DNA duplex. The cross‐linking process exploits the formation of a hydrazone between a non‐natural N⁴‐amino‐2′‐deoxycytidine nucleobase and the aldehyde residue of an abasic site in duplex DNA. The resulting cross‐link is stable under physiological conditions, but can be readily dissociated and re‐formed through heating–cooling cycles.     

Dynamics and mechanism of DNA repair in a biomimetic system: flavin-thymine dimer adduct

To mimic photolyase for efficient repair of UV-damaged DNA, numerous biomimetic systems have been synthesized, but all show low repair efficiency. The molecular mechanism of this low-efficiency process is still poorly understood. Here we report our direct mapping of the repair processes of a flavin-thymine dimer adduct with femtosecond resolution. We followed the entire dynamic evolution and observed direct electron transfer (ET) from the excited flavin to the thymine dimer in 79 ps. We further observed two competitive pathways, productive dimer ring splitting within 435 ps and futile back-ET in 95 ps. Our observations reveal that the underlying mechanism for the low repair quantum yield of flavin-thymine dimer adducts is the short-lived excited flavin moiety and the fast dynamics of futile back-ET without repair.