Article pubs.acs.org/biochemistry
Structural Changes of the Active Center during the Photoactivation of Xenopus (6−4) Photolyase Daichi Yamada,† Junpei Yamamoto,‡ Yu Zhang,† Tatsuya Iwata,† Kenichi Hitomi,§,∥ Elizabeth D. Getzoff,§ Shigenori Iwai,‡ and Hideki Kandori*,† †
Department of Frontier Materials, Nagoya Institute of Technology, Showa-ku, Nagoya 466-8555, Japan Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka 560-8531, Japan § Department of Integrative Structural and Computational Biology and The Skaggs Institute for Chemical Biology, The Scripps Research Institute, La Jolla, California 92037, United States ∥ Life Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States ‡
S Supporting Information *
ABSTRACT: Photolyases (PHRs) repair the UV-induced photoproducts, cyclobutane pyrimidine dimer (CPD) or pyrimidinepyrimidone (6−4) photoproduct [(6−4) PP], restoring normal bases to maintain genetic integrity. CPD and (6−4) PP are repaired by substrate-specific PHRs, CPD PHR and (6−4) PHR, respectively. Flavin adenine dinucleotide (FAD) is the chromophore of both PHRs, and the resting oxidized form (FADox), at least under in vitro purified conditions, is first photoconverted to the neutral semiquinoid radical (FADH•) form, followed by photoconversion into the enzymatically active fully reduced (FADH−) form. Previously, we reported light-induced difference Fourier transform infrared (FTIR) spectra corresponding to the photoactivation process of Xenopus (6−4) PHR. Spectral differences between the absence and presence of (6−4) PP were observed in the photoactivation process. To identify the FTIR signals where these differences appeared, we compared the FTIR spectra of photoactivation (i) in the presence and absence of (6−4) PP, (ii) of 13C labeling, 15N labeling, and [14N]His/15N labeling, and (iii) of H354A and H358A mutants. We successfully assigned the vibrational bands for (6−4) PP, the α-helix and neutral His residue(s). In particular, we assigned three bands to the CO groups of (6−4) PP in the three different redox states of FAD. Furthermore, the changed hydrogen bonding environments of CO groups of (6−4) PP suggested restructuring of the binding pocket of the DNA lesion in the process of photoactivation.
U
(6−4) PHR provides insight into product recognition by the enzyme.4,5 The overall structure of (6−4) PHR was similar to that of CPD PHR, consisting of α/β and α barrel domains connected with a unique long loop.4,5 Flavin adenine dinucleotide (FAD) is the common chromophore for both PHRs. The chromophore is buried in the α barrel domain with the redox-active isoalloxazine rings sequestered from the solvent. Reduced FAD (FADH−) is the active form for PHR catalysis. The oxidized
ltraviolet (UV) light is harmful to life because it triggers various chemical reactions inside cells, including DNA, and organisms have developed diverse defense systems to deal with this. Photolyases (PHRs) are unique DNA repair enzymes that maintain genetic integrity by reverting UV-induced photoproducts on DNA strands into normal bases with nearUV/blue light.1,2 Most prokaryotes have a single PHR that specifically repairs the cyclobutane pyrimidine dimer (CPD), CPD PHR. On the other hand, some higher eukaryotes possess an additional PHR that restores the pyrimidine-pyrimidone (6−4) photoproduct [(6−4) PP] to parental bases (Figure 1a). The discovery of (6−4) PHR occurred 40 years after a CPD PHR gene was first isolated.3 The structural determination of © 2015 American Chemical Society
Received: October 13, 2015 Revised: December 2, 2015 Published: December 31, 2015 715
DOI: 10.1021/acs.biochem.5b01111 Biochemistry 2016, 55, 715−723
Article
Biochemistry
of (6−4) PHR, and (iii) of H354A and H358A mutants. Additional insight into the molecular mechanism of (6−4) PHR in the process of photoactivation is discussed on the basis of these FTIR observations.
■
MATERIALS AND METHODS
Sample Preparation. Xenopus (6−4) PHR was expressed in E. coli as a fusion protein with a His6 tag at the N-terminus.14 The H354A and H358A mutants of Xenopus (6−4) PHR were constructed by polymerase chain reaction using the QuikChange site-directed mutagenesis method (Stratagene). Nucleotide substitutions were confirmed by DNA sequencing. For isotope labeling, E. coli was grown in M9 minimal medium containing either 0.5 g/L 15NH4Cl (Cambridge Isotope Laboratories) or 4 g/L [13C]glucose (Chlorella Industry). For [14N]His reverse labeling in the 15N background, E. coli was grown in M9 minimal medium containing 0.5 g/L 15NH4Cl and 0.1 g/L [14N]His (ICN Biomedicals Inc.). (6−4) PP was synthesized and incorporated into doublestranded DNA according to a previously described method.15 The damaged DNA substrate had the following sequence:
Figure 1. (a) Molecular structures of two thymines (right) and UVinduced photoproducts, (6−4) PP (left). (b) (6−4) PP binding site of (6−4) PHR (structure from Protein Data Bank entry 3cvu). FAD chromophore, (6−4) PP, and residues are shown as yellow, orange, and green sticks, respectively. The number of amino acid residues corresponds to that of Xenopus (6−4) PHR.
We used redissolved samples for FTIR spectroscopy, which was established in a previous study.13 We prepared the (6−4) PHR sample with or without (6−4) PP as follows. First, we placed 1 μL of the sample solution containing 2 mM (6−4) PHR, which was estimated from the absorbance at 450 nm (ε450 = 11200 M−1 cm−1)16 before concentration, in 50 mM Tris-HCl and 200 mM NaCl (pH 8.0) on a BaF2 window, and dried. To measure (6−4) PHR in complex with (6−4) PP, 1 μL of 2 mM (6−4) PP in an aqueous solution was gently mixed with the sample solution of (6−4) PHR on the BaF2 window and then dried. We then added 0.4 μL of H2O directly onto the dried film that was sandwiched by another BaF2 window. (6−4) PHR in D2O was prepared by diluting (6−4) PHR with the same buffer prepared in D2O and concentrating it three times with an Amicon YM-30 device (Millipore). For the measurement of (6−4) PHR in D2O, (6−4) PP dissolved in D2O was used. FTIR Spectroscopy. FTIR spectra of the redissolved samples were measured using an FTS-7000 (DIGILAB) spectrophotometer, as reported previously.13,14,17−20 Samples were placed in an Optistat-DN cryostat (Oxford) mounted in the spectrophotometer, which was also equipped with a temperature controller (ITC-4, Oxford). The illumination source was a high-power xenon lamp (MAX-302, ASAHI SPECTRA), providing three illuminations: 450 nm (MZ0450, ASAHI SPECTRA), >550 nm (LVX550, ASAHI SPECTRA), or >450 nm (VY-45, Toshiba). FTIR spectra were constructed from 128 interferograms with a spectral resolution of 2 cm−1. The difference spectrum was calculated by subtracting the spectrum recorded before illumination from the spectrum recorded after illumination. Six to eight difference spectra obtained in this way were averaged for each difference spectrum.
(FADox) and semiquinoid (FADH•) forms of FAD are resting states in purified enzymes for both CPD PHR and (6−4) PHR.6 PHRs purified from organisms, however, are artificially subject to oxidation, becoming inactive. Remarkably, PHRs have a system for regaining activity. In the presence of reducing agents, buried FAD inside PHRs can be reduced in a lightdependent manner. PHRs have the tryptophan triad chain, which is considered to be important for electron transfer and links FAD to the protein surface. In Escherichia coli CPD PHR, substitution of the outside tryptophan of the triad chain disturbs this process.7 (6−4) PHRs structurally conserve the tryptophan triad chain with some modification. Two photons are needed for this photoactivation: FADox is first converted to FADH• by one light-induced electron and one proton transfer and then to FADH− by one light-induced electron transfer.6 This indicates the presence of FADH• as an intermediate. In addition, if the electron and proton transfer reaction is separated, an anion radical intermediate (FAD•−) will be captured. When the reduced enzyme absorbs another photon in the presence of (6−4) PP, electron transfer from FADH− to the photoproduct initiates the repair process.8,9 Light-induced difference Fourier transform infrared (FTIR) spectroscopy is a powerful, sensitive, and informative method for studying structure−function relationships in photoreceptive proteins.10−12 Recently, we obtained difference FTIR spectra of the photoactivation (among four redox states of FAD) process and light-dependent DNA repair reactions of (6−4) PHR.13,14 This initiated a new understanding of the molecular mechanisms of (6−4) PHR in atomic detail, particularly of the enzymatic catalysis. In this study, our aim was to elucidate the FTIR signals of (6−4) PP and identify specific amino acids of (6−4) PHR known to possess photoactivation ability, especially those in the DNA binding pocket (Figure 1b). To achieve this, we studied the detailed activation mechanism by FTIR measurements (i) in the presence and absence of (6−4) PP, (ii) of 13C labeling, 15N labeling, and [14N]His/15N labeling
■
RESULTS Light-Induced Difference FTIR Spectra of (6−4) PHR at 277 K. Figure 2a shows FADH• minus FADox spectra in the presence (red lines) and absence (black lines) of (6−4) PP, in 716
DOI: 10.1021/acs.biochem.5b01111 Biochemistry 2016, 55, 715−723
Article
Biochemistry
are likely to have originated from an amide I vibration, and the negative band at 1653 cm−1 suggests an α-helical perturbation.31 With regard to the band at 1738 (+)/ 1728 (−) cm−1, we considered the origin to be a CO stretch of a protonated carboxylic acid because these peaks are sensitive to D2O [the band at 1738 (+) cm−1 showed a downshift to 1734 (+) cm−1 (Figure S2d of the Supporting Information)].13 However, another possibility is that the signal originates from the CO stretch of (6−4) PP because the substrate was perturbed upon photoactivation of CPD PHR.32,33 Comparing parts a and b of Figure 2 indicates that the substrate does not influence the structural changes of the peptide backbone accompanying the change from FADox to FADH• but does influence the change from FADox to FADH−. Furthermore, the band at 1653 and 1738 (+)/1728 (−) cm−1 appeared only in the change from FADH• to FADH− (Figure S3 of the Supporting Information). Comparison of 13C-Labeled (6−4) PHR in the 1780− 1660 cm−1 Region. To identify the origin of the band at 1738 (+)/1728 (−) cm−1 in the difference FTIR spectrum of FADH− minus FADox with (6−4) PP, uniformly 13C-labeled (6−4) PHR with unlabeled (6−4) PP was measured. Figure 3
Figure 2. Light-induced difference FTIR spectra of Xenopus (6−4) PHR without (black lines) and with (red lines) the (6−4) PP measured at 277 K. (a) FADH• minus FADox spectra calculated from spectra in parts a and c of Figure S1 of the Supporting Information (black lines) to remove the contribution of FADH− (see ref 14). (b) FADH− minus FADox difference spectrum obtained by illumination of FADox at >450 nm for 4 min. One division of the y-axis corresponds to 0.008 absorbance unit.
which the black line was reproduced form our previous study.14 Because illumination with 450 nm light of the FADox form of (6−4) PHR produces both FADH• and FADH− forms, the contribution of FADH− was removed by a calculation as we reported previously (Figure S1 of the Supporting Information).14 The overall spectral shape was similar between the spectra in the presence and absence of (6−4) PP. Negative signals at 1715, 1690, 1578, and 1545 cm−1 can be identified as C4O, C2O, C4aN5, and C10aN1 stretches of oxidized FAD, respectively.21−26 The positive bands at 1535 cm−1 are characteristic of FADH•, which was suggested as a C10aN1 stretch from flavin by theoretical calculations.27−29 Thus, most of the peaks originate from the FAD chromophore. On the other hand, small but reproducible differences were observed for the negative band at 1227 cm−1 and the positive band at 1084 cm−1, which disappeared and appeared in the presence of (6−4) PP, respectively. As these bands are affected by the substrate, it is likely that they originate from the apoprotein and/or substrate. From the frequencies, the former and latter candidates can be attributed to Tyr and His, respectively.30 Figure 2b shows FADH− minus FADox spectra in the presence (red line) and absence (black line) of (6−4) PP. The obtained spectra for His6-tagged (6−4) PHR are quite identical with those for (6−4) PHR in a previous study, where the fusion protein with the GST tag that had been treated with thrombin to remove the GST tag was used.13 The prominent characteristics are as follows: (i) the negative band appeared only in the presence of (6−4) PP; (ii) the bands are affected by the presence of (6−4) PP; and (iii) paired peaks were newly observed at 1738 (+)/ 1728 (−) cm−1. The band at 1102 (−) cm−1 may originate from a C−N stretch of His; this is also the case for FADH• (Figure 2a).30 The bands at 1670−1630 cm−1
Figure 3. Comparison of difference FTIR spectra of unlabeled (black lines) and 13C-labeled (red solid lines) Xenopus (6−4) PHR in the 1780−1660 cm−1 region. (a) FADH• minus FADox difference FTIR spectra with (6−4) PP. (b) FADH− minus FADox difference FTIR spectra with (6−4) PP. Red dotted lines in parts a and b show the FADH• minus FADox and FADH− minus FADox difference FTIR spectra without (6−4) PP, respectively. One division of the y-axis corresponds to 0.005 absorbance unit.
shows FADH• minus FADox (a) and FADH− minus FADox (b) difference FTIR spectra for unlabeled (black lines) and 13Clabeled (6−4) PHR (red solid lines) in the presence of (6−4) PP. The spectra in the 1800−1000 cm−1 region are presented in Figure S4 of the Supporting Information. CO stretches appear in the 1780−1660 cm−1 region, and they show a downshift by ∼40 cm−1 after 13C labeling.34,35 In this measurement, the isotope shift takes place in the signals from (6−4) PHR. 717
DOI: 10.1021/acs.biochem.5b01111 Biochemistry 2016, 55, 715−723
Article
Biochemistry We observed a negative peak at 1715 cm−1, which shifted to 1671 cm−1 by 13C labeling (Figure 3). The band originated from (6−4) PHR, and the most likely candidate is a C4O stretching vibration of the isoalloxazine ring from oxidized FAD.22 On the other hand, the bands at 1738 (+)/1728 (−) cm−1 in the FADH− minus FADox spectrum were not shifted by 13 C labeling (Figure 3b, red line). This clearly shows that these bands originate from the substrate, and not from (6−4) PHR. Furthermore, the bands at 1698 (+)/1687 (−) and 1693 (+)/1686 (−) cm−1 appeared in the 13C-labeled (6−4) PHR with (6−4) PP (red solid lines in parts a and b of Figure 3, respectively). These features are in contrast to those of the spectrum of 13C-labeled (6−4) PHR without (6−4) PP (red dotted lines in Figure 3), where these bands were not observed. Even if we change the subtracting ratio, the 1698(+)/1687 (−) cm−1 band cannot be completely canceled. Therefore, there should be two kinds of paired bands. If the bands had originated from (6−4) PHR, they should have appeared in the 1740−1725 cm−1 zone in the spectra from unlabeled (6−4) PHR. Because there are no such signals except for the bands at 1738 (+)/1728 (−) cm−1 in Figure 3b, we concluded that the bands originated from (6−4) PP, not a carboxylic acid. The peak pair at ∼1695 (+)/∼1687 (−) cm−1 was observed in both the FADH• minus FADox and FADH− minus FADox difference FTIR spectra, but the amplitude of the former was half that of the latter. Therefore, two CO groups from (6−4) PP may be involved in the photoactivation reaction. From the X-ray structure, only CO groups of (6−4) PP interacted with (6−4) PHR. (6−4) PP has three CO groups at the C4 and C2 positions of 5′-thymine and at the C2 position of 3′-thymine (Figure 1a). The number of paired bands corresponds to the number of CO groups of (6−4) PP. We identified these three bands originating from (6−4) PP of three CO stretches by isotope labeling, suggesting structural coupling between the chromophore and substrate. Isotope Effects of 15N- and [14N]His/15N-Labeled (6−4) PHR at 1150−1050 cm−1. The bands at 1084 (+)/1102 (−) cm−1 appeared in the photoactivation of (6−4) PHR with (6− 4) PP (Figure 2). The His residue(s) is a good candidate for these bands because C−N stretching vibrations of a His residue appear at around 1100 cm−1.36,37 Actually the bands showed a downshift by uniformly 15N-labeled (6−4) PHR with unlabeled (6−4) PP (Figure S5b,d of the Supporting Information). We prepared [14N]His/15N-labeled (6−4) PHR, whose nitrogen atoms of His residues are unlabeled (14N-labeled), but all other nitrogen atoms are labeled with 15N. If His residues contributed to these signals, then the spectrum from [14N]His/15N-labeled would be upshifted compared to uniformly 15N-labeled spectra. Figure 4 compares the difference spectra of 15N-labeled (black lines) and [14N]His/15N-labeled (red lines) (6−4) PHR in the 1150−1050 cm−1 region. An overview of the resulting set of spectra is presented in Figures S5 and S6, respectively, of the Supporting Information. The shifted peak pair was observed at ∼1100 (−)/1097 (+) cm−1 for the [14N]His/15N-labeled spectra of FADH• minus FADox (Figure 4a) and FADH− minus FADox (Figure 4c) with (6−4) PP, which was downshifted by ∼4 cm−1 for 15N-labeled spectra [1097 (−)/1093 (+) cm−1]. On the other hand, in the spectra of FADH• minus FADox (Figure 4b) and FADH− minus FADox (Figure 4d) without (6−4) PP, no isotope effect was observed. Thus, the band at 1101 (−)/1097 (+) cm−1 originated from His residues. However, we could not find the 1084 cm−1 (+) band in FADH• minus FADox spectra of [14N]His/15N-labeled (red
Figure 4. Comparison of difference FTIR spectra of 15N-labeled (black line) and [14N]His/15N-labeled (red line) Xenopus (6−4) PHR in the 1150−1050 cm−1 region. (a) FADH• minus FADox difference FTIR spectra with (6−4) PP. (b) FADH• minus FADox difference FTIR spectra without (6−4) PP. (c) FADH− minus FADox difference FTIR spectra with (6−4) PP. (d) FADH− minus FADox difference FTIR spectra without (6−4) PP. One division of the y-axis corresponds to 0.002 absorbance unit.
line) (6−4) PHR. The frequency at 1084 cm−1 (+) might be shifted to 1081 cm−1 (+) by an 15N labeling effect (Figure 4a, black line). Therefore, the 1084 cm−1 (+) band originated from the N atom of (6−4) PHR that is not from His. This band is a good candidate for Trp,38 the adenine moiety,39 or the isoalloxazine ring40 of FAD vibrations, which will be investigated by isotopic labeling in the near future. Two nitrogen atoms in the imidazole ring (Nτ and Nπ) serve as protonation sites (Figure S7 of the Supporting Information). C−N stretches of Nτ-protonated and Nπ-protonated imidazole appear at 1100 cm−1, respectively.41 The bands around 1100 cm−1 can mainly be attributed to the C−Nτ stretching vibration.42 From the frequency region, the paired bands at 1101 (−)/1097 (+) cm−1 are likely to have originated from the C−Nτ stretch of Nπ-protonated His. The peak pair at 1101 (−)/1097 (+) cm−1 is similarly observed in the FADH• minus FADox difference FTIR spectrum (Figure 4a, red line), but the amplitude is smaller (0.4 times the spectrum of Figure 4c, red line). If the calculation were conducted so that the bands at 1100/1097 cm−1 disappear in the [14N]His/[15N]protein spectrum (Figure 4a), the bands specific to FADH− [1386 cm−1 for the 15N-labeled sample (Figure S5 of the Supporting Information)] appear at negative sides, in the calculated FADH• minus FADox difference spectrum. In this case, the subtraction factor must be overestimated. Therefore, the calculated spectrum of Figure 4a is reasonable. From these results, signals from two His residues may overlap in the photoactivation reaction from FADox to FADH−. While the structure of one His is preserved, another His changed upon formation of FADH•. Thus, upon the reduction of FADH• to FADH−, structural changes of another His were involved. These environmental changes from His appeared only in the 718
DOI: 10.1021/acs.biochem.5b01111 Biochemistry 2016, 55, 715−723
Article
Biochemistry
Supporting Information. The spectra were similar regardless of the presence of (6−4) PP in the FADH• minus FADox spectrum process of WT (Figure 5a). In addition, in the spectra of H354A and H358A mutants (parts b and c of Figure 5, respectively), overall light-induced difference FTIR spectra were very similar regardless of the existence of (6−4) PP. This indicates that the substrate has little influence on the structural changes accompanying the change from FADox to FADH•. However, as described above, the amide-I vibrations at 1670− 1630 cm−1 are clearly affected by the presence of (6−4) PP in the FADH− minus FADox spectrum of WT (Figure 5d). These protein conformational changes also appeared in CPD PHR with the substrate in the FADH− minus FADH• spectrum32,33 and might have been triggered by FAD bending movement.43,44 Similarly, the structural change of the α-helical perturbation was observed in the H354A mutant but was not the same as that of WT (Figure 5e). In contrast, in the spectra of the H358A mutant (Figure 5f), the spectral shape in the amide-I region was very similar regardless of the existence of (6−4) PP. These results clearly demonstrate that the replacement of H358 with Ala eliminates the effect of the light-induced structural changes in the peptide backbone in the presence of (6−4) PP. These results suggest that H358 participates in the hydrogen bond network to perturb the α-helix. The X-ray structure of the Drosophila (6−4) PHR with (6−4) PP indicates that the H365 and Y423 side chains have the potential to form hydrogen bonds with H369 [the corresponding amino acid residues in Xenopus (6−4) PHR are H354, Y412, and H358, respectively]. These mutants at H354 and H358 can bind (6−4) PP but are poor at repairing (6−4) PP.45,46 This hydrogen bonding network might be important for perturbing the α-helix to establish an enzymatically active state. The 1738 (+)/1728 (−) cm−1 paired band in the FADH− minus FADox difference FTIR spectra with (6−4) PP (Figure 5d, red line) originated from (6−4) PP. The CO vibrations at 1738 (+)/1728 (−) cm−1 of (6−4) PP were clearly affected by the H354A mutant (Figure 5e). However, the peak pair at 1739 (+)/1729 (−) cm−1 was similarly observed in the spectra of the H358A mutant (Figure 5f). The 1738 (+)/1728 (−) cm−1 paired band might interact with H354. According to an enzyme−substrate complex X-ray structure of Drosophila (6− 4) PHR,4 the C4O group of 5′-(6−4) PP is located within 2.7 Å of H365 [H354 for Xenopus (6−4) PHR]. The paired bands at 1738 (+)/1728 (−) cm−1 originate form C4O groups in 5′-(6−4) PP. The negative peak in the H354A mutant (1745 cm−1) appeared in a higher-frequency region compared to that in WT (1728 cm−1). On the other hand, the positive peak in H354A (tentative assignment at 1703 cm−1) was lower than that of WT (1738 cm−1) (Figure 5e). These results show that the C4O group from (6−4) PP forms a hydrogen bond in the FADox state of H354A weaker than that of WT, and the reverse case takes place in the FADH− state. The assignment of FTIR spectra of (6−4) PHR photoactivation upon (6−4) PP binding is summarized in Tables 1 and 2.
presence of (6−4) PP. Thus, His residues that are located near the substrate binding site are the most likely candidates for these bands. On the basis of amino acid sequence similarity and X-ray structures,4,5 the possible candidates for the His residues are H353, H354, H358, H406, and H472. In particular, H354 and H358 are the most likely candidate His residues because they are located in the binding site of (6−4) PP (Figure 1b). To investigate the involvement of these His residues, photoactivation spectra of H354A and H358A mutants were measured. Secondary Structure Perturbations in H354A and H358A Mutants of (6−4) PHR. Figure S8 of the Supporting Information shows the difference FTIR spectra of H354A and H358A mutants with (red line) and without (black line) (6−4) PP in the 1125−1050 cm−1 region. An overview of the resulting set of spectra is presented in Figures S9 and S10 of the Supporting Information. Quite different spectra were obtained for the H354A mutation in the presence and absence of (6−4) PP in the 1125−1050 cm−1 region compared to the WT spectra (Figure S8b,e of the Supporting Information), indicating that the structural changes between them are different. On the other hand, overall light-induced difference FTIR spectra are very similar between the presence and absence of (6−4) PP in the H358A mutant (Figure S8c,f of the Supporting Information), and the spectra of the H358A mutant are very similar to the spectrum of WT without (6−4) PP. However, these changes induced by mutations in individual bands cannot be clearly identified. Consequently, from the experiment using mutants, we were unable to clarify whether the band of His residues that appeared originated from H354 and H358. In the 1700−1600 cm−1 region, CO stretches appeared from the peptide backbone, known as the amide-I vibration.31 In Figure 5, these spectra in the 1770−1550 cm−1 region are shown and were expanded from Figures S9 and S10 of the
■
DISCUSSION Origin of CO Stretches of (6−4) PP. We found three paired bands originating from CO stretches of (6−4) PP, 1738 (+)/1728 (−), 1698 (+)/1687 (−), and 1693 (+)/1684 (−) cm−1, by 13C labeling of (6−4) PHR (Figure 3 and Tables 1 and 2). The candidates of the origin of these bands are three CO groups, C4O and C2O groups in 5′-(6−4) PP and
Figure 5. Influence of mutations on difference FTIR spectra with (red line) and without (black line) (6−4) PP in the 1770−1550 cm−1 region. FADH• minus FADox spectra of (a) WT, (b) the H354A mutant, and (c) the H358A mutant of (6−4) PHR and FADH− minus FADox spectra of (d) WT, (e) the H354A mutant, and (f) the H358A mutant of (6−4) PHR are shown. One division of the y-axis corresponds to 0.01 absorbance unit. 719
DOI: 10.1021/acs.biochem.5b01111 Biochemistry 2016, 55, 715−723
Article
Biochemistry Table 1. FTIR Spectral Assignment for the Process from FADox to FADH• upon (6−4) PP Binding wavenumber (cm−1) unlabel
D2O
13
C
15
N
[14N] His/15N
assignment
1100 (−) 1097 (+) 1081 (+)
C2O C2O C−N C−N C−N
1698 (+) 1687 (−) 1102 (−) 1096 (+) 1084 (+)
1097 (−) 1093 (+) 1081 (+)
1086 (+)
description C2O group of 3′-(6−4) PP C−Nτ group of Nπ-protonated His Trp or FAD
Table 2. FTIR Spectral Assignment for the Process from FADox to FADH− upon (6−4) PP Binding wavenumber (cm−1) unlabeled
D2O
1738 (+) 1728 (−)
1734 (+) 1728 (−)
1653 1639 1102 1097
(−) (+) (−) (+)
1651 1637 1102 1097
(−) (+) (−) (+)
13
C
1738 1728 1693 1686 1610 1603 1086 1081
(+) (−) (+) (−) (−) (+) (−) (+)
15
N
[14N] His/15N
assignment
1738 (+) 1728 (−)
C4O C4O C2O C2O protein protein C−N C−N
1738 (+) 1728 (−)
1650 1639 1097 1094
(−) (+) (−) (+)
1650 1639 1101 1097
the C2O group in 3′-(6−4) PP. We propose a tentative assignment of the observed vibrational signals in the CO groups of (6−4) PP (Figure 1a). First, we postulate that the band at 1738 (+)/1728 (−) cm−1 originates form the CO stretch located near H354. The reason is that the CO stretch at 1738 (+)/1728 (−) cm−1 was clearly perturbed by the H354A mutation (Figure 5e), whereas the peak pair remained at 1739 (+)/1729 (−) cm−1 in the spectra of the H358A mutant (Figure 5f). According to the enzyme−substrate complex X-ray structure,4 the C4O group in 5′-(6−4) PP is located within 2.7 Å of H365 [H354 in Xenopus (6−4) PHR]. Thus, it is reasonable that the band at 1738 (+)/1728 (−) cm−1 originates form C4O groups in 5′-(6−4) PP. Next, when a structural change of the C4O group in 5′-(6−4) PP takes place, the C2O group in the same thymine would be influenced simultaneously. Therefore, the paired band at 1693 (+)/1684 (−) cm−1 that appeared in the FADH− minus FADox spectrum might have originated from C2O groups of 5′-(6− 4) PP. Furthermore, the remaining paired band at 1698 (+)/1687 (−) cm−1 that appeared in the FADH• minus FADox spectrum might have originated from the C2O stretch of 3′(6−4) PP. The paired band was upshifted in the change of FADox to FADH• and also to FADH−. This result suggests that the hydrogen bond of CO groups of (6−4) PP became weaker. The simplest interpretation is that the CO groups of (6−4) PP moved away from the hydrogen bonding donor. It seems apparently disadvantageous for photorepair because the hydrogen bonding distance is too far between (6−4) PP and the electron donor, FAD (FADH−). However, an ultrafast spectroscopy study observed (6−4) PP repair on a time scale of at least tens of nanoseconds, even a back electron transfer in 50 ps, without repair.46 Therefore, although it seems disadvantageous to forward electron transfer, such a structural change in (6−4) PHR with (6−4) PP might take place to prevent back electron transfer by maintaining a distance for repairing (6−4) PP. Previous time-resolved study under photon-regulated conditions indicates the two-photon repair processes and the existence of a long-lived intermediate.47 Considering this, a structural change in (6−4) PHR with (6−4) PP might be advantageous to forming the intermediate. These results
(−) (+) (−) (+)
description C4O group of 5′-(6−4) PP C2O group of 5′-(6−4) PP α-helices C−Nτ group of Nπ-protonated His
demonstrate how (6−4) PHR not only utilizes photons to change the redox state of its FAD cofactor but also utilizes the subsequent change to FAD’s shape to structurally preconfigure its conformation to optimally facilitate DNA repair in general. C−N Stretches of the His Residue. We found two pairs of bands originating from Nπ-protonated His residues (Figure 4 and Tables 1 and 2). These paired bands suggest a change to the hydrogen bonding environment of the Nπ-protonated His residues. However, we were unable to determine how the environments of the His residues changed because there are no reports about how the His C−N stretches change by hydrogen bonding environments. The candidate of the His residue should be near the substrate and chromophore, because these bands appeared only in the presence of (6−4) PP (Figure 4a,c). Therefore, reasonable candidates of these bands are H354 and H358, because they are located in the (6−4) PP binding site (Figure 1b). Both bands at 1101 (−)/1097 (+) cm−1 indicate a C−Nτ stretch of Nπ-protonated His. Thus, two His residues are protonated only in the Nπ position. We considered one possibility in which both H354 and H358 are neutralized in the presence of (6−4) PP. This hypothesis does not correspond to the results of the ENDOR experiment 48 or to MD calculations.49,50 These previous results showed that one of the two His residues was indeed in an Nτ- and Nπ-protonated state (Figure S7b of the Supporting Information). Thus, one of the two Nπ-protonated His residues detected in this study might originate from H354 or H358. In addition, a measurement was not observed for the Nτ-protonated His signal. This suggests that Nτ- and Nπ-protonated His residues were the same in the process of photoactivation, independent of the presence or absence of (6−4) PP. By measurement of the spectra of the H354A and H358A mutants, most of the bands in the 1125−1050 cm−1 region were changed by the H354A mutation (Figure S8), indicating that the structural changes for these bands were influenced by this mutation. We should note that we were unable to identify changes in individual bands induced by these mutations. As a result, we believe that H354 interacts with the adenine moiety of FAD and C5−OH of (6−4) PP, but not with the C4O of (6−4) PP. The replacement of H354 with Ala influenced light720
DOI: 10.1021/acs.biochem.5b01111 Biochemistry 2016, 55, 715−723
Article
Biochemistry
Figure 6. Structural model of the active site of (6−4) PHR in the process of photoactivation. In the FADox state, CO groups of 5′-(6−4) PP form a hydrogen bridge between adenine moiety of FAD and Q288 (a). The C2O group of 3′-(6−4) PP forms hydrogen bonding structures with the H−X group (for example, a water molecule). In the FADH• form, the hydrogen bonding structures of the C2O group of 3′-(6−4) PP and the H− X group changed and became weaker. In the FADH− form, perturbation of the α-helix was triggered by bending of the isoalloxazine ring of FAD (blue helix). The CO groups of 5′-(6−4) PP changed the hydrogen bonding structures between the adenine moiety of FAD and Q288.
■
CONCLUSION We studied the detailed photoactivation by FTIR measurements (i) in the presence and absence of (6−4) PP, (ii) of 13C labeling, 15N labeling, and [14N]His/15N labeling, and (iii) of H354A and H358A mutants. We found three paired bands originating from the CO group of (6−4) PP (Figure 3) and two paired bands originating from Nπ-protonated His residues (Figure 4) by isotope labeling in the change from FADox to FADH• and from FADH• to FADH−. Furthermore, the structural change of the α-helical perturbation is important for the enzymatically active state indicated by mutants in the process from FADH• to FADH− (Figure 5). Additionally, the CO groups of (6−4) PP distanced themselves from the partner of each hydrogen bonding donor (Figure 6). These structural changes may be important for creation of a structure allowing for an enzymatically active state.
induced structural changes in the peptide backbone and substrate, suggesting that H354 participates in the hydrogen bonding network with FAD and (6−4) PP (Figure 1b). Structural Model of the Active Site of (6−4) PHR during Photoactivation. On the basis of these findings and previously published data, including the crystal structure of the enzyme−substrate complex,4 we propose a structural model of the active site of (6−4) PHR upon photoactivation (Figure 6). In the FADox state, the CO groups of 5′-(6−4) PP form a hydrogen bond with the N−H group of the adenine moiety of FAD and with the N−H group of Q288 (Figure 6a). However, in our FTIR measurement, we were not able to detect Q288. Normally, the CO stretch, the NH2 bend, and the C−N stretch from the Gln moiety are expected to appear at 1687− 1668, 1610−1586, and 1410 cm−1, respectively.30 These frequency regions appeared in the amide-I vibrations of the enzyme,31 and in the CO and CN stretch from the isoalloxazine ring of the FAD moiety,21−26 which might have masked the Q288 signal. To clearly elucidate the role of this amino acid residue, a future mutational study of Q288 needs to be performed. The C2O group of 3′-(6−4) PP forms a hydrogen bond with a certain X−H group (for example, a water molecule). When FADH• forms by light illumination, the C2 O group of 3′-(6−4) PP and H−X change the hydrogen bonding structures (Figure 6b), weakening them. When an additional electron produces FADH− from FADH• by light illumination, the isoalloxazine ring of FAD is negatively charged and bends.44 These reactions might trigger the following reaction. In FADH−, the change in FAD’s structure triggered the perturbation of the α-helix (Figure 6c, blue arrow). Then, the C4O and C2O groups of 5′-(6−4) PP change the hydrogen bonding structures. These groups move away from the partner of each hydrogen bonding donor. These structural changes might result in the formation of this important structure to create an enzymatically active state.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.5b01111. Raw data of light-induced difference FTIR spectra of Xenopus (6−4) PHR with (6−4) PP at 277 K (Figure S1), light-induced difference FTIR spectra of Xenopus (6−4) PHR at 277 K showing photoactivation in H2O and D2O at 1800−1000 cm−1 (Figure S2), comparison with or without (6−4) PP in the change from FADH• to FADH− (Figure S3), light-induced difference FTIR spectra of Xenopus (6−4) PHR at 277 K showing photoactivation of unlabeled samples, 13C-labeled samples (Figure S4), 15N-labeled samples (Figure S5), and comparison of difference FTIR spectra of 15Nlabeled and [14N]His/15N-labeled samples (Figure S6) in the regions of 1800−1000 cm−1, structural diagram of a 721
DOI: 10.1021/acs.biochem.5b01111 Biochemistry 2016, 55, 715−723
Article
Biochemistry
■
(11) Kötting, C., and Gerwert, K. (2005) Proteins in action monitored by time-resolved FTIR spectroscopy. ChemPhysChem 6, 881−888. (12) Noguchi, T. (2007) Light-induced FTIR difference spectroscopy as a powerful tool toward understanding the molecular mechanism of photosynthetic oxygen evolution. Photosynth. Res. 91, 59−69. (13) Zhang, Y., Iwata, T., Yamamoto, J., Hitomi, K., Iwai, S., Todo, T., Getzoff, E. D., and Kandori, H. (2011) FTIR study of lightdependent activation and DNA repair processes of (6−4) photolyase. Biochemistry 50, 3591−3598. (14) Yamada, D., Zhang, Y., Iwata, T., Hitomi, K., Getzoff, E. D., and Kandori, H. (2012) Fourier-transform infrared study of the photoactivation process of Xenopus (6−4) photolyase. Biochemistry 51, 5774−5783. (15) Iwai, S., Shimizu, M., Kamiya, H., and Ohtsuka, E. (1996) Synthesis of a phosphoramidite coupling unit of the pyrimidine (6−4) pyrimidone photoproduct and its incorporation into oligodeoxynucleotides. J. Am. Chem. Soc. 118, 7642−7643. (16) Hitomi, K., Okamoto, K., Daiyasu, H., Miyashita, H., Iwai, S., Toh, H., Ishiura, M., and Todo, T. (2000) Bacterial cryptochrome and photolyase: characterization of two photolyase-like genes of Synechocystis sp. PCC6803. Nucleic Acids Res. 28, 2353−2362. (17) Iwata, T., Zhang, Y., Hitomi, K., Getzoff, E. D., and Kandori, H. (2010) Key dynamics of conserved asparagine in a cryptochrome/ photolyase family protein by Fourier transform infrared spectroscopy. Biochemistry 49, 8882−8891. (18) Zhang, Y., Yamamoto, J., Yamada, D., Iwata, T., Hitomi, K., Todo, T., Getzoff, E. D., Iwai, S., and Kandori, H. (2011) Substrate assignment of the (6−4) photolyase reaction by FTIR spectroscopy. J. Phys. Chem. Lett. 2, 2774−2777. (19) Iwata, T., Yamamoto, A., Tokutomi, S., and Kandori, H. (2007) Hydration and temperature similarly affects light-induced protein structural changes in the chromophore domain of phototropin. Biochemistry 46, 7016−7021. (20) Yamamoto, A., Iwata, T., Tokutomi, S., and Kandori, H. (2008) Role of Phe1010 in light-induced structural changes of the neo1-LOV2 domain of Adiantum. Biochemistry 47, 922−928. (21) Wille, G., Ritter, M., Friedemann, R., Mäntele, W., and Hübner, G. (2003) Redox-triggered FTIR difference spectra of FAD in aqueous solution and bound to flavoproteins. Biochemistry 42, 14814−14821. (22) Bowman, W. D., and Spiro, T. G. (1981) Normal mode analysis of lumiflavin and interpretation of resonance Raman spectra of flavoproteins. Biochemistry 20, 3313−3318. (23) Schmidt, J., Coudron, P., Thompson, A. W., Watters, K. L., and McFarland, J. T. (1983) Assignment and the effect of hydrogen bonding on the vibrational normal modes of flavins and flavoproteins. Biochemistry 22, 76−84. (24) Abe, M., and Kyogoku, Y. (1987) Vibrational analysis of flavin derivatives: normal coordinate treatments of lumiflavin. Spectrochim. Acta 43, 1027−1037. (25) Lively, C. R., and McFarland, J. T. (1990) Assignment and the effect of hydrogen bonding on the vibrational normal modes of flavins and flavoproteins. J. Phys. Chem. 94, 3980−3994. (26) Zhang, W., Vivoni, A., Lombardi, J. R., and Birke, R. L. (1995) Time-resolved SERS study of direct photochemical charge transfer between FMN and a Ag electrode. J. Phys. Chem. 99, 12846−12857. (27) Rieff, B., Bauer, S., Mathias, G., and Tavan, P. (2011) IR spectra of flavins in solution: DFT/MM description of redox effects. J. Phys. Chem. B 115, 2117−2123. (28) Li, J., Uchida, T., Ohta, T., Todo, T., and Kitagawa, T. (2006) Characteristic structure and environment in FAD cofactor of (6−4) photolyase along function revealed by resonance Raman spectroscopy. J. Phys. Chem. B 110, 16724−16732. (29) Zheng, Y., Carey, P. R., and Palfey, B. A. (2004) Raman spectrum of fully reduced flavin. J. Raman Spectrosc. 35, 521−524. (30) Barth, A. (2007) Infrared spectroscopy of proteins. Biochim. Biophys. Acta, Bioenerg. 1767, 1073−1101.
His side chain (Figure S7), spectra of WT, H354A, and H358A at 1125−1050 cm−1 (Figure S8), and lightinduced difference FTIR spectra of Xenopus (6−4) PHR at 277 K showing photoactivation of WT, H354A (Figure S9), and H358A (Figure S10) samples in the 1800−1000 cm−1 region (PDF)
AUTHOR INFORMATION
Corresponding Author
*Department of Frontier Materials, Nagoya Institute of Technology, Showa-ku, Nagoya 466-8555, Japan. Phone and fax: 81-52-735-5207. E-mail:
[email protected]. Funding
This work was supported by grants from the Japan Society for the Promotion of Science to D.Y. (25·170) and J.Y. (25870400) and the Japanese Ministry of Education, Culture, Sports, Science and Technology to H.K. (25104009 and 15H02391). E.D.G. was supported by National Institutes of Health Grant GM37684. Notes
The authors declare no competing financial interest.
■
ABBREVIATIONS PHR, photolyase; CPD, cyclobutane pyrimidine dimer; (6−4) PP, (6−4) photoproduct; FAD, flavin adenine dinucleotide; FTIR, Fourier transform infrared; FADox, fully oxidized form of FAD; FADH−, fully reduced form of FAD; FADH•, neutral semiquinoid form of FAD; FAD•−, anion radical form of FAD.
■
REFERENCES
(1) Sancar, A. (2003) Structure and function of DNA photolyase and cryptochrome blue-light photoreceptors. Chem. Rev. 103, 2203−2237. (2) Weber, S. (2005) Light-driven enzymatic catalysis of DNA repair: a review of recent biophysical studies on photolyase. Biochim. Biophys. Acta, Bioenerg. 1707, 1−23. (3) Todo, T., Takemori, H., Ryo, H., Ihara, M., Matsunaga, T., Nikaido, O., Sato, K., and Nomura, T. (1993) A new photoreactivating enzyme that specifically repairs ultraviolet light-induced (6−4) photoproducts. Nature 361, 371−374. (4) Maul, M. J., Barends, T. R., Glas, A. F., Cryle, M. J., Domratcheva, T., Schneider, S., Schlichting, I., and Carell, T. (2008) Crystal structure and mechanism of a DNA (6−4) photolyase. Angew. Chem., Int. Ed. 47, 10076−10080. (5) Hitomi, K., DiTacchio, L., Arvai, A. S., Yamamoto, J., Kim, S. T., Todo, T., Tainer, J. A., Iwai, S., Panda, S., and Getzoff, E. D. (2009) Functional motifs in the (6−4) photolyase crystal structure make a comparative framework for DNA repair photolyases and clock cryptochromes. Proc. Natl. Acad. Sci. U. S. A. 106, 6962−6967. (6) Hitomi, K., Kim, S. T., Iwai, S., Harima, N., Otoshi, E., Ikenaga, M., and Todo, T. (1997) Binding and catalytic properties of Xenopus (6−4) photolyase. J. Biol. Chem. 272, 32591−32598. (7) Aubert, C., Vos, M. H., Mathis, P., Eker, A. P., and Brettel, K. (2000) Intraprotein radical transfer during photoactivation of DNA photolyase. Nature 405, 586−590. (8) Kim, S. T., Li, Y. F., and Sancar, A. (1992) The third chromophore of DNA photolyase: Trp-277 of Escherichia coli DNA photolyase repairs thymine dimers by direct electron transfer. Proc. Natl. Acad. Sci. U. S. A. 89, 900−904. (9) Todo, T., Kim, S. T., Hitomi, K., Otoshi, E., Inui, T., Morioka, H., Kobayashi, H., Ohtsuka, E., Toh, H., and Ikenaga, M. (1997) Flavin adenine dinucleotide as a chromophore of the Xenopus (6−4) photolyase. Nucleic Acids Res. 25, 764−768. (10) Kandori, H. (2000) Role of internal water molecules in bacteriorhodopsin. Biochim. Biophys. Acta, Bioenerg. 1460, 177−191. 722
DOI: 10.1021/acs.biochem.5b01111 Biochemistry 2016, 55, 715−723
Article
Biochemistry (31) Krimm, S., and Bandekar, J. (1986) Vibrational spectroscopy and conformation of peptides, polypeptides, and proteins. Adv. Protein Chem. 38, 181−364. (32) Wijaya, I. M., Zhang, Y., Iwata, T., Yamamoto, J., Hitomi, K., Iwai, S., Getzoff, E. D., and Kandori, H. (2013) Detection of distinct αhelical rearrangements of cyclobutane pyrimidine dimer photolyase upon substrate binding by Fourier transform infrared spectroscopy. Biochemistry 52, 1019−1027. (33) Wijaya, I. M., Iwata, T., Yamamoto, J., Hitomi, K., Iwai, S., Getzoff, E. D., Kennis, J. T., Mathes, T., and Kandori, H. (2014) Flavin adenine dinucleotide chromophore charge controls the conformation of cyclobutane pyrimidine dimer photolyase α-helices. Biochemistry 53, 5864−5875. (34) Noguchi, T., and Sugiura, M. (2003) Analysis of flash-induced FTIR difference spectra of the S-state cycle in the photosynthetic water-oxidizing complex by uniform 15N and 13C isotope labeling. Biochemistry 42, 6035−6042. (35) Masuda, S., Hasegawa, K., and Ono, T. (2005) Light-induced structural changes of apoprotein and chromophore in the sensor of blue light using FAD (BLUF) domain of AppA for a signaling state. Biochemistry 44, 1215−1224. (36) Pflüger, F., Hernández, B., and Ghomi, M. (2010) Vibrational analysis of amino acids and short peptides in hydrated media. VII. Energy landscapes, energetic and geometrical features of L-histidine with protonated and neutral side chains. J. Phys. Chem. B 114, 9072− 9083. (37) Hasegawa, K., Ono, T.-A., and Noguchi, T. (2000) Vibrational spectra and ab initio DFT calculations of 4-methylimidazole and its different protonation forms: Infrared and Raman markers of the protonation state of a histidine side chain. J. Phys. Chem. B 104, 4253− 4265. (38) Hernández, B., Pflüger, F., Adenier, A., Kruglik, S. G., and Ghomi, M. (2010) Vibrational analysis of amino acids and short peptides in hydrated media. VIII. Amino acids with aromatic side chains: L-phenylalanine, L-tyrosine, and L-tryptophan. J. Phys. Chem. B 114, 15319−15330. (39) Nowak, M. J., Lapinski, L., Kwiatkowski, J. S., and Leszczyński, J. (1996) Molecular structure and infrared spectra of adenine. Experimental matrix isolation and density functional theory study of adenine 15N isotopomers. J. Phys. Chem. 100, 3527−3534. (40) Zheng, Y., Dong, J., Palfey, B. A., and Carey, P. R. (1999) Using Raman spectroscopy to monitor the solvent-exposed and ″buried″ forms of flavin in p-hydroxybenzoate hydroxylase. Biochemistry 38, 16727−16732. (41) Noguchi, T., Inoue, Y., and Tang, X. S. (1999) Structure of a histidine ligand in the photosynthetic oxygen-evolving complex as studied by light-induced Fourier transform infrared difference spectroscopy. Biochemistry 38, 10187−10195. (42) Majoube, M., Millié, Ph., and Vergoten, G. (1995) Vibrational spectra of 4-methylimidazole: assignment of modes and calculation of Raman and resonance Raman intensities at the ab initio 6-31G level. J. Mol. Struct. 344, 21−36. (43) Kao, Y. T., Saxena, C., He, T. F., Guo, L., Wang, L., Sancar, A., and Zhong, D. (2008) Ultrafast dynamics of flavin in five redox states. J. Am. Chem. Soc. 130, 13132−13139. (44) Mees, A., Klar, T., Gnau, P., Hennecke, U., Eker, A. P. M., Carell, T., and Essen, L.-O. (2004) Crystal structure of a photolyase bound to a CPD-like DNA lesion after in situ repair. Science 306, 1789−1793. (45) Hitomi, K., Nakamura, H., Kim, S. T., Mizukoshi, T., Ishikawa, T., Iwai, S., and Todo, T. (2001) Role of two histidines in the (6−4) photolyase reaction. J. Biol. Chem. 276, 10103−10109. (46) Li, J., Liu, Z., Tan, X., Guo, X., Wang, L., Sancar, A., and Zhong, D. (2010) Dynamics and mechanism of repair of ultraviolet-induced (6−4) photoproduct by photolyase. Nature 466, 887−890. (47) Yamamoto, J., Martin, R., Iwai, S., Plaza, P., and Brettel, K. (2013) Repair of the (6−4) photoproduct by DNA photolyase requires two photons. Angew. Chem., Int. Ed. 52, 7432−7436.
(48) Schleicher, E., Hitomi, K., Kay, C. W. M., Getzoff, E. D., Todo, T., and Weber, S. (2007) Electron nuclear double resonance differentiates complementary roles for active site histidines in (6−4) photolyase. J. Biol. Chem. 282, 4738−4747. (49) Condic-Jurkic, K., Smith, A.-S., Zipse, H., and Smith, D. M. (2012) The protonation states of the active-site histidines in (6−4) photolyase. J. Chem. Theory Comput. 8, 1078−1091. (50) Sadeghian, K., Bocola, M., Merz, T., and Schütz, M. (2010) Theoretical study on the repair mechanism of the (6−4) photolesion by the (6−4) photolyase. J. Am. Chem. Soc. 132, 16285−16295.
723
DOI: 10.1021/acs.biochem.5b01111 Biochemistry 2016, 55, 715−723
