Damage recognition in nucleotide excision repair
Malta, E.
Citation
Malta, E. (2008, December 9). Damage recognition in nucleotide excision repair. Retrieved from https://hdl.handle.net/1887/13327
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/13327
Chapter 5
Fluorescent DNA modifications to study nucleotide excision repair
Erik Malta, Carlo P. Verhagen, Geri F. Moolenaar, Dmitri V. Filippov, Gijs
A. van der Marel and Nora Goosen
Abstract
Nucleotide excision repair is a highly versatile DNA repair system that is capable of recognizing various structurally unrelated types of damage. To do so, the proteins of this system must recognize a common DNA distortion induced by all these damages or a propensity of the DNA structure to adopt a certain conformation due to the presence of a lesion.
To monitor these protein-induced DNA alterations several fluorescent DNA modifications were synthesized that alter their fluorescent properties depending on the position in the DNA helix. Here we demonstrate using a fluorescent 2-aminopurine menthol modification that the position of this damage with respect to its neighboring nucleotides does not change upon binding of either UvrA or UvrB to this lesion. With substrates containing a fluorescent pyrene attached to the ribose as a damage it was demonstrated that a conformational change at the site of damage can occur with UvrA and UvrB. In the case of UvrA the enhanced fluorescence most likely represents the formation of a stretch of ssDNA in order to allow UvrB to insert its highly conserved β-hairpin structure between the DNA strands. In case of UvrB the observed fluorescence could at least partially be ascribed to dynamic complexes in search of damage. The fluorescent properties of these modifications could very well be used to study conformational changes induced by other DNA repair proteins as well. In this study it is also shown that small planar DNA modifications that are readily detected by UvrA can escape detection by UvrB most likely because they can be translocated behind the β-hairpin structure of the protein, just like an undamaged nucleotide. This underlines that the two proteins of bacterial nucleotide excision repair use different strategies for damage detection.
Introduction
DNA is under constant threat of being damaged by various exogenous and endogenous agents. If left unrepaired the resulting DNA lesions can lead to mutations and/or cell death. To counteract these detrimental effects several DNA repair mechanisms have arisen, like direct reversal repair, base excision repair (BER), and nucleotide excision repair (NER) (Friedberg et al. 2005). These DNA repair systems all use different strategies to obtain a common goal namely to maintain genomic integrity. Whereas proteins involved in direct reversal repair and base excision repair recognize only one or at most a small subset of DNA lesions, NER uses the same subset of proteins to remove a large variety of structurally unrelated types of damage (Shuck et al. 2008; Truglio et al. 2006a). Most DNA modifications have in common that they alter the structure of the DNA helix, leading to a reduction in base pairing and/or base stacking interactions (Lukin and de los Santos, 2006). These altered properties of the DNA helical structure are likely to be the primary recognition targets of most DNA repair systems. After initial recognition of this common structural feature, however, different DNA repair pathways use diverse verification mechanisms to ensure correct lesion recognition and removal.
In BER and direct reversal repair lesion specificity is obtained by the presence of a highly specific binding pocket that only allows incorporation of one or a small group of damaged nucleotides (Bruner et al. 2000; Daniels et al. 2004). Non-damaged nucleotides and non-cognate DNA lesions are either not efficiently bound in the pocket or are prevented from entering the pocket by sterical hindrance. In general these repair systems mainly recognize DNA lesions that cause little distortion to the DNA helical structure. Lesions that do cause significant DNA distortion are substrates for the NER system, which predominantly recognizes bulky lesions and inter- and intra-strand crosslink damages (reviewed in Truglio et al. 2006a). The identification of such lesions without a similar chemical structure requires multiple proteins, each one probably probing a different parameter of the helical structure of the DNA for abnormalities. In bacteria both the UvrA (Seeberg and Steinum, 1982; Mazur and Grossman, 1991) and UvrB (Moolenaar et al. 2000) proteins can distinguish damaged from undamaged DNA and in humans this is accomplished by UV-DDB (Moser et al. 2005;
Wittschieben et al. 2005), the XPC-hHR23A complex (Sugasawa et al. 1998), XPA (Robins et al. 1991), and RPA (Clugston et al. 1992). Previous studies have also implicated the Saccharomyces cerevisiae XPD homolog Rad3 as a damage-recognizing protein. Its helicase activity was shown to specifically stall at sites of DNA lesions, thereby resulting in stable binding (Naegeli et al. 1992, 1993; Sung et al. 1994).
To probe for the presence of DNA modifications repair proteins can use a variety of activities to distinguish damaged from undamaged DNA. Most likely these activities involve conformational alterations in the DNA that are facilitated at the site of the lesion. Structural changes that might be induced are DNA bending (Chen et al. 2002), strand separation (Oh and Grossman, 1989; Evans et al. 1997) and base flipping (Mees et al. 2004). To monitor these protein-induced DNA alterations we synthesized several fluorescent DNA modifications that on one hand can be recognized as damage and on the other hand alter their fluorescent properties depending on the position in the DNA helix. These damages include fluorescent compounds replacing a normal DNA base (phenanthrene and pyrene modifications) and an adduct connected to a fluorescent base (a menthol derivative of 2-aminopurine (AP-M)).
Fluorescence emission in the free DNA as compared to that in the protein bound state can provide important information about conformational changes induced by the repair proteins.
To show the usefulness of the fluorescent damages we applied the E. coli NER proteins UvrA and UvrB as a model system. These proteins are expected to probe different parameters of the DNA for damage. The UvrB protein scans for damage by inserting its β-hairpin between the two DNA strands and translocating along the DNA (Truglio et al.
2006b). It was proposed that the presence of a lesion prevents it from passing behind the hairpin, thereby stalling translocation and allowing incision by UvrC. Upon UvrB binding to a damaged site on the DNA it was shown that the damaged nucleotide remains inserted in the DNA helix (Malta et al. 2008), but that its 3’ neighbor becomes extrahelical (Malta
et al. 2006). The mechanism that UvrA uses to scout for the presence of a lesion is however currently unknown.
In this study it is shown that all three DNA modifications are efficiently recognized by the UvrA protein. The AP-M adduct is also properly recognized by UvrB, but the phenanthrene and, to a lesser extent, the pyrene lesions escape detection by UvrB most likely because these planar residues can be translocated behind the β-hairpin. Upon binding of UvrA and/or UvrB to the DNA, AP-M fluorescence did not change. However, pyrene fluorescence was increased in both cases indicating a conformational change reducing either base stacking or base pairing interactions at the site of damage induced by the two proteins. The pyrene modification therefore proved to be a more sensitive probe for conformational alterations and might therefore be very useful to study other DNA repair systems.
Materials and methods
DNA Synthesis and modification
Appropriately protected 3’-O-phosphoramidites of 2’-deoxy-1’-phenantryl- and 2’-deoxy-1’-pyrenylnucleoside were prepared as individual α- and β-anomers as described (Ren et al. 1996). The detailed synthesis of a protected 3’-O-phosphoramidite of 6-(3-O-L- menthyl-propenyl)-9-(2’-deoxy-β-D-ribofuranosyl)purine (AP-M), designed for the present study, will be reported elsewhere. The commercially available reagents for DNA synthesis were all obtained from Proligo and the solvents were from Biosolve.
Oligonucleotide synthesis.
The solid-phase synthesis of oligonucleotides (ODN’s) was performed on a fully automated Expedite instrument(PerSeptive Biosystems) starting from controlled pore glass functionalized with an appropriate nucleoside. The synthesis was performed ona 1-μmol scale via phosphoramidite methodology (de Kort et al. 1999, 2001), but using mildly removable N- t-butylphenoxyacetyl (Tac) protection for nucleobases (Sinha et al. 1993). For the modified fluorescent nucleobases N-phenoxyacetyl was used as a substitute. Elongation was performed by coupling of the 3’-phosphoramidite derivativesof DMT-protected nucleosides (5’-DMT- dATac, 5’-DMT-dCTac,5’-DMT-dGTac, and 5’-DMT-T, 10 eq., 10 μmol, 0.1 M stock) with 4,5- dicyanoimidazole (Vargeese et al. 1998) as the activator (50 eq., 50 μmol 0.25 M stock), for 3 minutes. The phosphoramidites of the fluorescent nucleoside analogues (15 eq.) were coupled for 5 min with 60 eq. of 4,5-dicyanoimidazole (60 μmol).The 5’- DMT group was removed using a 3% TCA solution.After each coupling, remaining free 5’-hydroxyls were blocked using a mixture of cap A (t-butylphenoxyacetic anhydride, 0.2 M in THF) and cap B (1- methylimidazolein tetrahydrofuran/pyridine) followed by oxidation of thephosphite linkage to the phosphate using 0.02 M of I2 in pyridine/water(1 min).After final DMT removal the
modified oligonucleotide was cleaved from theresin by 25% ammonium hydroxide solution at room temperature (2 h). The resulting ODN was purified on a Q-Sepharosecolumn at pH 12 applying a gradientof buffer B (0.01 M NaOH + 2 M NaCl) in buffer A (0.01 M NaOH).
Fractions containing the pure productwere combined, desalted on Sephadex G-25 column (0.15 M ammonium bicarbonate) and lyophilized. The identity and purity of the product was confirmed by MALDI-TOF mass-spectrometry (Voyager DE PRO, PerSeptive Biosystems, positive ionization mode) and IE HPLC (DNA-Pac PA200 analytical column, Dionex).
The DNA modifications so obtained are shown in Fig.1. The 50 bp DNA substrates were obtained by hybridizing the 50-mer (un)damaged top strand (5’-GGGATTACTTAC- GGGCACATTACAAAXAAACCTCAGAACGACCTCACACG-3’) to its complementary DNA strand (5’-CGTGTGAGGTCGTTCTGAGGTTTYTTTGTAATGTGCCCGTAAG- TAATCCC-3’). In these DNA substrates X denotes the used DNA modification (phenan- threne, pyrene, menthol-derivative of 2AP) or thymine. The nature of the complementary nucleotide (Y) depends on X with Y = T for AP-M and Y = A for phenanthrene/pyrene/thy- mine.
Proteins and chemicals
Purification of the UvrA (Visse et al. 1992), UvrB (Moolenaar et al. 2001) and UvrC (Visse et al. 1992) proteins occurred as described. Creatine kinase (CK), creatine phosphate (CP) and ATP were obtained from Roche. 1-Pyrenebutyric acid was obtained from Acros.
Filter-binding assay
The DNA substrates were 5’ labelled as described (Verhoeven et al. 2002). The DNA substrates (2 nM) were incubated with different concentrations of UvrA (either 2 or 25 nM) and UvrB (either 0 or 100 nM) as indicated in 20 μl Uvr-endo buffer (50mM Tris–HCl pH 7.5, 10mM MgCl2, 100mM KCl, 0.1μg/μl BSA and 1mM ATP). After an incubation of 10 min at 37oC, 300 μl of pre-heated (37oC) Uvr-endo buffer (without BSA) was added and the mixture was applied to a nitrocellulose filter. The incubation vial was rinsed two times with 300 μl portions of the same buffer at 37oC. To study UvrA binding under fluorescence conditions the DNA substrates were incubated with 2 μM of UvrA in 60 μl Uvr-endo buffer.
After 10 minutes of incubation at 37 oC the samples were directly applied to a nitrocellulose filter. The incubation vial was rinsed once with 100 μl of the same buffer (37oC) and the filter was subsequently rinsed twice with 300 μl Uvr-endo buffer (37oC). Each sample was corrected for the amount of DNA bound to the filter in the absence of UvrA. DNA binding is indicated as the percentage of total input DNA that is retained on the filter.
Gel retardation assay
The samples for the gel-retardation assay (10 μl) were incubated under the same conditions as for the filter binding assay. Where indicated, 1 μl of preimmune (Pre)- or anti-
serum was added after incubation as described (Visse et al. 1992). The resulting mixtures were analyzed on a cooled 3.5% native polyacrylamide gel containing 1 mM of ATP and 10 mM MgCl2 in 1x Tris-borate as described. To determine complex formation under the conditions used for the fluorescence measurements 0.5 μM DNA was mixed with a small tracer of 5’ radioactively labelled DNA of the same kind. The DNA was incubated in 10 μl Uvr-endo buffer (without BSA) for 10 minutes at 37oC in the presence of UvrA (0.45 μM), UvrB (3.75 μM), CK (0.17 μg/μl) and CP (20 mM). After incubation the samples were loaded on a 3.5% native polyacrylamide gel as described above.
Incision assay
5’-Radioactively labelled DNA substrates were incubated with 2.5 nM UvrA, 100 nM UvrB and 25 nM UvrC in 20 μl Uvr-endo buffer. After an incubation of 30 minutes at 37oC the reactions were terminated by addition of 3 μl EDTA/SDS (0.33 M EDTA, 3.3%
SDS) and 2.4 μl glycogen (4 μg/μl). DNA was purified by ethanol precipitation and the incision products were analyzed on a 15% denaturing polyacrylamide gel.
Fluorescence measurements
60 μl samples containing 0.5 μM of DNA substrates carrying different nucleotide modifications were incubated in Uvr-endo buffer for 10 min at 37oC in the presence of 2 μM UvrA (UvrA alone) or 0.45 μM UvrA and 3.75 μM UvrB (UvrA and UvrB). All samples additionally contained 0.17 μg/μl CK and 20 mM CP. After incubation samples were handled as described (Malta et al. 2006). Emission spectra were obtained by setting the excitation wavelength at 310 nm for the 2-aminopurine menthol (Ap-M), at 349 nm for the pyrene modification and at 254 nm for the phenanthrene modification. In order to obtain excitation spectra emission was fixed at 370 nm for AP-M, at 398 nm for pyrene and at 374 nm for phenanthrene. The same wavelengths were used to quantitatively determine the emission signal. The emission signal of pyrene free in solution was obtained by measuring 0.5 μM of 1-pyrenebutyric acid (Acros) in 60 μl Uvr-endo buffer at 37oC.
Results and discussion
To study conformational changes in the DNA at the site of damage upon binding of DNA repair protein(s), we designed and synthesized several fluorescent DNA modifications that are expected to alter fluorescence as a consequence of a change in their molecular environment (Fig. 1). The fluorescent moieties include a menthol derivative of 2-aminopurine (AP-M), with the 2-aminopurine unit as its fluorescent center, and a fluorescent phenanthrene and pyrene modification. These groups are directly attached to the C1 position of a ribose unit, resulting in two different anomeric configurations (α and β).
Fig. 1 Chemical structure of the DNA modifications used in this study. All modifications were directly attached to the C1 position of the deoxyribose. The phenanthrene and pyrene modifications were attached in two different conformations, α and β with respect to the ribose ring.
Undamaged DNA
To determine the efficiency of recognition of these lesions, binding of UvrA and UvrB to these substrates was compared to binding to undamaged DNA. Quantitative determination by filter-binding analysis showed minor association of UvrA to undamaged DNA under equimolar concentrations (2 nM) of UvrA and DNA (4.6%, Fig. 2A). Increasing the UvrA concentration to 25 nM did result in significant UvrA-DNA complex formation (18.3%). Addition of an excess of UvrB (100 nM) to these samples hardly altered the amount of complexes (7.5% and 20.1% at 2 and 25 nM resp.) indicating that UvrB does not contribute to the affinity for undamaged DNA. Since filter-binding studies do not distinguish the nature of the formed complexes we also performed a more qualitative gel-retardation analysis using the same protein and DNA concentrations (Fig. 2B). At 2 nM UvrA, no UvrA- DNA complexes can be detected (lane 1) and also in the presence of UvrB (lane 3) hardly any binding is observed. At 25 nM however UvrA-DNA complexes are observed (lane 2).
Addition of an excess of UvrB showed the same level of binding (lane 4) but it is unclear whether the obtained complexes contain UvrA or UvrAB. To distinguish between these two possibilities serum containing UvrB-specific antibodies (αB) was post-incubated with the samples just prior to loading on the gel. After αB addition, only part of the complex band (~70%) reacted to the antibodies resulting in a shift upwards (lane 6). Apparently not only UvrAB (~70%) complexes were present on undamaged DNA but also complexes containing only UvrA (~30%). With an excess of UvrB (100 nM) compared to UvrA (25 nM) most of the UvrA protein is expected to be associated with UvrB. The observation of UvrA-DNA complexes without UvrB on gel can therefore only be explained by frequent dissociative loading of UvrB on the DNA. Most likely, the DNA-handoff from UvrA to UvrB at a potential damaged site results in dissociation of the complex when no damage is detected. Either UvrB alone dissociates from the DNA or the entire complex might fall apart leading to free UvrA and UvrB proteins in solution. Re-association of the free UvrA with the DNA might then
also explain the presence of the UvrA-DNA complexes free from UvrB. These results are in agreement with previous fluorescence resonance energy transfer (FRET) experiments where it was shown that the addition of undamaged DNA to UvrAB complexes formed in solution reduced FRET from a GFP- to a YFP-UvrB fusion (Malta et al. 2007). Since this study also showed that FRET can only be observed when UvrB is associated to UvrA this indicated that undamaged DNA enhances the turnover of the UvrAB complex.
In the presence of αB less free DNA is visible on gel (Fig. 2B lane 6) compared to the incubation without antibodies (lane 4). This is due to a stabilizing effect of the serum on the formed complexes since a control sample containing pre-immune serum (Pre) also yielded more complexes (lane 5).
A
DNA substrate (2 nM)
UvrA (nM) UvrB (nM) DNA binding (%)
Undamaged 2 --- 4.6 ± 0.2
25 --- 18.3 ± 1.4
2 100 7.5 ± 0.1
25 100 20.1 ± 0.7
B
Fig. 2 Analysis of a 50 bp undamaged DNA substrate. (A) Filter-binding analysis. 2nM of 50-mer undamaged DNA substrate was incubated for 10 min at 37oC with the indicated amounts of UvrA and UvrB. After incubation the samples were applied to a nitrocellulose filter. DNA binding is indicated as the percentage of total input DNA that is retained on the filter. (B) Gel- retardation analysis. DNA (2 nM) was incubated with UvrA and UvrB as specified. After incubation, where indicated, 1μl of serum with (αB) or without (Pre) UvrB-specific antibodies was added. Samples were analyzed on a 3.5% native acrylamide gel containing 1 mM ATP. Positions of the co-migrating UvrA-DNA and UvrAB-DNA complexes (A(B)-DNA), of the UvrB-DNA complex and of the αB-bound complex are indicated.
2-Aminopurine menthol modification (AP-M)
The AP-M modification consists of the fluorescent adenine analog 2-aminopurine to which a non-planar menthol modification is attached at the C6-position (Fig. 1). Since in the unmodified adenine residue there is an amino-group at this position that is engaged in base pairing, this compound is expected to alter the basepairing properties of the DNA and to induce a considerable DNA deformation.
To study recognition of the AP-M damage by UvrA a DNA substrate containing the AP-M modification (2 nM) was incubated with either 2 or 25 nM UvrA. At 2 nM UvrA significantly more UvrA-DNA complexes were obtained (15.0%; Fig. 3A) than with undamaged DNA (4.6%) indicating that these complexes are mostly damage-specific (Fig.
2A). Increasing the UvrA concentration elevated damage-specific UvrA-DNA complex
formation (60.2%) to a level again higher than on undamaged DNA (18.3%). Similar results were obtained from the gel-retardation assay (Fig. 3B lanes 1 and 2) in which at 25 nM significantly more UvrA-DNA complexes were detected compared to undamaged DNA (Fig.
2B).
A
DNA substrate (2 nM)
UvrA (nM) UvrB (nM) DNA binding (%)
AP-M 2 --- 15.0 ± 0.5
25 --- 60.2 ± 0.5
2 100 55.2 ± 5.7
25 100 77.6 ± 0.5
B C E
D
DNA substrate
Free 2AP ssDNA dsDNA DNA +
UvrA
DNA + UvrAB
AP-M 240 ± 4 41 ± 5 68 ± 2 65 ± 2 69 ± 1
Fig. 3 Analysis of a 50 bp DNA substrate containing a menthol-derivative of the fluorescent base 2-aminopurine (AP-M). (A) Filter-binding analysis. Incubations with indicated protein concentrations were as described for Fig.
2A. DNA binding is indicated as the percentage of total input DNA that is retained on the filter. (B) Gel-retardation analysis. Incubations with indicated protein concentrations were as described for Fig. 2B. Positions of the co- migrating UvrA-DNA and UvrAB-DNA complexes (A(B)-DNA), of the UvrB-DNA complex and of the αB-bound complex are indicated. (C) Incision analysis. The 5’ terminally labelled DNA substrate was incubated at 37oC for 30 min in the presence of UvrA (2.5 nM), UvrB (100 nM), and UvrC (25 nM). Incision products were analyzed on a 15% denaturing polyacrylamide gel and are indicated. (D) 2-Aminopurine fluorescence emission. DNA (0.5 μM) was incubated in the presence of the CP/CK system for 10 min at 37oC with or without UvrA (0.45 μM) and UvrB (3.75 μM), as indicated. After incubation the samples were transferred to a cuvette and emission spectra were obtained by excitation at 310 nm. Values represent fluorescence emission at 370 nm. (E) Gel-retardation analysis.
0.5 μM DNA substrate mixed with tracer of 5’ radioactively labelled DNA of the same substrate was incubated with UvrA (0.45 μM) and UvrB (3.75 μM) in the presence of the CP/CK system. After incubation (10 min 37oC), samples were analyzed on a 3.5% native acrylamide gel containing 1 mM ATP. Positions of the co-migrating UvrA-DNA and UvrAB-DNA complexes (A(B)-DNA) and of the dimeric UvrB complex (B2-DNA) are indicated.
Addition of UvrB (100 nM) to a sample containing 2 nM of UvrA resulted in an increase in complex formation on filter (55.2%; Fig. 3A) which mainly consisted of UvrB- DNA complexes as shown by the gel (Fig. 3B lane 3). Additional UvrA (25 nM) further increased DNA binding (77.6%, compared to 60.2% with UvrA alone; Fig. 3A) but the resulting complex migrated at a higher position on gel indicating the presence of UvrA (Fig.
3B lane 4). The addition of αB revealed that all these complexes also contain UvrB (lane 6).
This is most likely due to re-association of UvrA to the UvrB-DNA complexes at this higher UvrA concentration. This re-association of UvrA has been shown before in the FRET studies of UvrB-GFP and UvrB-YFP. In these experiments FRET of the two UvrB subunits bound to a damaged site still increased at UvrA concentrations far above the concentration of DNA damage when UvrA levels are more than sufficient to load UvrB onto the DNA (Malta et al.
2007). In line with efficient damage recognition by both UvrA and UvrB incision was shown to be efficient as well (~90%; Fig. 3C).
Taken together, the results clearly indicate that the AP-M damage is very well recognized by both UvrA and UvrB and can therefore be used to study potential conformational changes of the lesion induced by these proteins.
First we measured the fluorescence of the DNA in the absence of proteins. The signal of ssDNA containing an AP-M (~41 AU; Fig. 3D) is significantly decreased compared to that of free 2-aminopurine (~240 AU). Base stacking interactions of the damaged nucleotide with its neighboring bases apparently quench the fluorescence signal. The emission signal of a sample containing dsDNA (~68 AU) is elevated compared to that of ssDNA indicating a decrease in base stacking. Probably in dsDNA the nucleotides in the complementary strand force the lesion in a different conformation where the stacking interactions of the 2- aminopurine residue are slightly altered. In the presence of an excess of UvrA (2 μM) with respect to DNA (0.5 μM) the fluorescence signal remains unaltered (~65 AU), even though damage-specific binding of UvrA does occur under these conditions as shown by filter- binding analysis (28.7 ± 7.1%; Table 1). The addition of both UvrA (0.45 μM) and UvrB (37.5 μM) resulted in the formation of a similar amount of UvrB-DNA (~30%; Fig. 3F).
Under these conditions the complexes exist in the dimeric UvrB2-DNA form as shown before (Malta et al. 2006). Also in this UvrB-DNA complex, however, fluorescence does not change (~69 AU; Fig. 3D). Apparently, neither UvrA nor UvrB induce a detectable conformational change of the DNA damage.
Table 1. UvrA (2 μM) complex formation on DNA (0.5 μM) on various DNA substrates as determined by filter- binding analysis.
DNA substrate UvrA-DNA complex formation (%)
AP-M 28.7 ± 7.1
α-pyrene 30.7 ± 3.5
β-pyrene 30.7 ± 3.4
A
DNA substrate (2 nM)
UvrA (nM) UvrB (nM) DNA binding (%)
α-phen 2 --- 9.4 ± 0.5
25 --- 51.4 ± 5.5
2 100 16.5 ± 0.1
25 100 52.0 ± 2.0
β-phen 2 --- 10.1 ± 0.6
25 --- 50.4 ± 1.6
2 100 13.5 ± 1.8
25 100 43.3 ± 0.7
B C
D
Emission
Excitation Fig. 4 Analysis of 50 bp DNA substrates
containing a phenanthrene modification directly connected to the deoxyribose unit in two different configurations (α and β with respect to the deoxyribose). (A) Filter- binding analysis. Incubations with indicated protein concentrations were as described for Fig. 2A. DNA binding is indicated as the percentage of total input DNA that is retained on the filter. (B) Gel-retardation analysis. Incubations with indicated protein concentration were as described for Fig. 2B. Positions of the co-migrating UvrA-DNA and UvrAB-DNA complexes (A(B)-DNA), of the UvrB-DNA complex and of the αB-bound complex are indicated. (C) Incision analysis. Incubations were as described in Fig. 3C.
Incision products are indicated. Note that both DNA fragments contain minor contaminations of shorter oligonucleotides. (D) Fluorescence characterization of the DNA substrate containing the α-configuration of phenanthrene. The emission spectrum (right panel) was obtained by exciting a sample containing 0.5 μM DNA at 254 nm. The excitation spectrum (left panel) was obtained by analyzing the emission at 374 nm.
Phenanthrene modification
The phenanthrene modification (Fig. 1) consists of a planar phenanthrene molecule that is directly attached to the C1-position of a deoxyribose subunit. This attachment resulted in two different anomeric conformations: α and β with respect to the ribose, where the β-
anomeric form is the natural configuration for normal DNA nucleobases. The two anomeric forms were separately incorporated into the DNA.
Filter-binding experiments (Fig. 4A) revealed that for both anomeric forms binding of UvrA is significantly higher than on undamaged DNA and only slightly reduced compared to AP-M. The same results were obtained using gel-retardation analysis (Fig. 4B lanes 1,2,7 and 8). UvrB binding in the filter-binding assay using equimolar UvrA compared to DNA (2 nM) is however very low on these substrates (Fig. 4A) although still slightly higher than on undamaged DNA (Fig. 2A). No UvrB-DNA complexes can be detected on gel either under these conditions (Fig. 4B lanes 3 and 9), showing that UvrB does not efficiently recognize the damage. Also at higher UvrA concentrations (25 nM) the presence or absence of UvrB does not alter DNA binding (Fig. 4A). Gel-retardation analysis using these protein concentrations (Fig. 4B lanes 4 and 10) shows that almost all DNA is now bound by protein (especially in the presence of the stabilizing serum (Fig. 4B lanes 5 and 11)) but incubation with αB (Fig. 4B lane 6 and 12) revealed that not all complexes contain UvrB. Apparently, a significant amount of complexes exists in the UvrA-DNA complex form. A similar effect was observed on undamaged DNA (Fig. 2B lane 6), where the absence of a damage triggered dissociation of the UvrAB complex. Apparently, after initial recognition of the phenanthrene damage by UvrA this protein tries to load UvrB onto the damaged site and the inability of UvrB to recognize this adduct results in complex dissociation. Comparing the α and β anomeric configurations shows that the β-configuration is less efficiently recognized by the UvrB protein than the α-form since with this substrate the presence of UvrB even results in a decrease in DNA binding (43.3%) compared to UvrA alone (50.4%). Apparently, on the β-phenanthrene, the UvrB protein induces dissociation of the complex from the DNA.
In addition, also the amount of UvrA-DNA complex free from UvrB, as visualized by αB addition, is slightly higher for the β-configuration (Fig. 4B). As a result from the poor recognition of the phenanthrene modification by UvrB the incision efficiencies on the two DNA substrates are very low as well (Fig. 4C). Incision is slightly more efficient on the α- phenanthrene (~10% incision; Fig. 5C lane 2) than on the β-phenanthrene (<5% incision, lane 4), again reflecting UvrB’s preference for the α-configured DNA substrate.
The combined results strongly suggest that UvrA efficiently recognizes the damage but when UvrB tries to verify the presence of a lesion it fails to do so. Therefore, poor recognition of the phenanthrene modification by the UvrB protein is the main cause of the repair defect.
In previous studies by our lab and by others (Malta et al. 2008; Waters et al. 2006) it was postulated that UvrB distinguishes damaged from undamaged DNA by preventing the adducted nucleotide from passing behind the β-hairpin motif. In addition, Truglio et al (2006b) postulated that nucleotides that are translocated behind this hairpin need to pass through a small planar hydrophobic pocket. In light of this model, the poor recognition of the phenanthrene moiety can be explained by its planar structure. Apparently, especially
in its ‘natural’ β-configuration it evades detection by UvrB because it can pass behind the β-hairpin structure and through the pocket of the protein. In the α-substituted modification the phenanthrene is differently orientated with respect to the ribose which might slow down its translocation behind the hairpin thereby explaining a somewhat higher incision of this anomer.
Initial fluorescence characterization of the synthesized dsDNA substrates containing a phenanthrene modification revealed that emission at 374 nm was optimal when excitation was set at 254 nm (Fig. 4D). Unfortunately, the localization of the excitation optimum far below 300 nm makes this construct impractical to study biological systems. The range of wavelengths between 200 and 300 nm is known to harbor excitation signals of several amino acid side chains (e.g. tyrosine and tryptophane). Even more importantly this region of wavelengths also contains the absorption band of ATP which is required by both UvrA and UvrB for damage-specific binding. The fluorescence signal in the protein-DNA complexes could therefore not be determined. The compound, however, might be a very good tool to study protein-induced changes by other repair proteins that do not require ATP i.e. XPA and XPC.
Pyrene modification
Although the pyrene modification is also planar it contains an additional ring as compared to the phenanthrene moiety and due to its larger size it might be better recognized by the UvrB protein. Moreover, the extra ring will increase the conjugated double bond system and is thereby expected to increase the excitation maximum of the compound. As for phenanthrene, the pyrene compound (Fig. 1) was incorporated in two different configurations (α and β with respect to the sugar ring).
Filter-binding (Fig. 5A) and gel-retardation analysis (Fig. 5B) at both UvrA concentrations (2 and 25 nM) indicate that UvrA binding to these DNA substrates is in the same order as binding of the phenanthrene modified DNA substrates (Fig. 4A and B).
In the presence of 2 nM UvrA and UvrB, binding to the pyrene-modified DNA substrates (Fig. 5A) is significantly higher than on the phenanthrene damage (Fig. 4A). In addition, the UvrB-DNA complexes can now be seen in the gel-retardation assay (Fig. 5B lanes 3 and 9).
These results are in agreement with the better incision observed on these substrates (Fig. 5C) compared to incision on the phenanthrene-modified DNA substrates (Fig. 4C) and show that indeed the larger pyrene is a better substrate for UvrB than phenanthrene.
In the β-configuration however the pyrene forms only very little UvrB-DNA complexes (Fig. 5B lane 9) and the incision efficiency (~40%; Fig. 5C) on this substrate is significantly lower than on AP-M (~90%; Fig. 3C). This suggests that even though the planar pyrene modification is larger than the phenanthrene it can still avoid detection by UvrB, probably again by its ability to pass behind the β-hairpin. However the rate at which this translocation occurs seems lower than for the β-phenanthrene since it does not result in
a high turn-over of the UvrAB complex at 25 nM UvrA with UvrB. For the phenanthrene constructs this turnover could be visualized by the appearance of UvrA-DNA complexes without UvrB (as shown with αB, Fig. 4B). For the β-pyrene the amount of UvrB-free complexes is significantly lower (Fig. 5B lane 12) suggesting a lower turn-over rate. Addition of serum however increased the amount of observed protein-DNA complexes showing that dissociation of complexes does still occur for the β-pyrene.
A
DNA substrate (2 nM)
UvrA (nM) UvrB (nM) DNA binding (%)
α-pyr 2 --- 11.5 ± 3.0
25 --- 62.0 ± 2.3
2 100 36.6 ± 0.4
25 100 69.5 ± 2.9
β-pyr 2 --- 16.2 ± 1.0
25 --- 55.3 ± 3.2
2 100 36.9 ± 4.6
25 100 69.4 ± 3.4
B C
D
E
DNA substrate
Free pyrene
ssDNA dsDNA DNA +
UvrA
DNA + UvrAB
α-pyr 495 ± 20 126 ± 1 11 ± 2 43 ± 1 141 ± 15
β-pyr 495 ± 20 127 ± 2 22 ± 1 44 ± 1 54 ± 2
Emission Excitation
Fig. 5 Analysis of 50 bp DNA substrates containing a pyrene modification directly connected to the deoxyribose unit in two different configurations (α and β with respect to the deoxyribose). (A) Filter-binding analysis. Incubations with indicated protein concentration were as described for Fig. 2A. DNA binding is indicated as the percentage of total input DNA that is retained on the filter. (B) Gel-retardation analysis. Incubations with indicated protein concentrations were as described for Fig. 2B. Positions of the co-migrating UvrA-DNA and UvrAB- DNA complexes (A(B)-DNA), of the UvrB-DNA complex and of the αB-bound complex are indicated. Note that in the DNA preparation of the β-pyrene there is a minor unknown contamination (lane 7) that migrates at a higher position and that also reacts with UvrA (lane 8). (C) Incision analysis. Incubations were as described in Fig. 3C. Incision products are indicated. (D) Fluorescence characterization of the DNA substrate containing the β-configuration of pyrene.
The emission spectrum (right panel) was obtained by exciting a sample containing 0.5 μM DNA at 349 nm. The excitation spectrum (left panel) was obtained by analyzing the emission at 398 nm. (E) Fluorescence emission data on substrates containing a pyrene modification in both the α and β form. Incubations were as described in Fig. 3D.
Emission spectra were obtained by excitation at 349 nm. Values represent fluorescence emission at 398 nm. (F) Gel-retardation analysis. Incubations were as described for Fig. 3E. Positions of the co-migrating UvrA-DNA and UvrAB-DNA complexes (A(B)-DNA) and of the dimeric UvrB complex (B2-DNA) are indicated.
In the α-configuration of pyrene more UvrB-DNA complexes are formed in the presence of 2 nM UvrA (lane 3), than on the β-pyrene. At 25 nM UvrA in the presence of UvrB a large amount of UvrA(B)-DNA complexes is observed (lane 4). Comparable to what was found for AP-M all these complexes exist in the UvrAB-DNA form (lane 6). Apparently, in the α-configuration the pyrene group forms a better block for translocation of UvrB than the β-pyrene. In line with this, incision of the α-pyrene is also more efficient (~90% after 30 min) and seems comparable to that on the AP-M substrate (Fig. 3C). However, incision analysis after 5 minutes of incubation did reveal a difference between the two lesions, 40%
for α-pyrene and 70% for AP-M (results not shown), reflecting a difference in UvrB-specific binding as can be observed at low UvrA concentrations (compare Fig.3B and 5B lanes 3).
Although UvrA binding to AP-M and α-pyrene is comparable, the α-pyrene is apparently less well recognized by UvrB, suggesting that the α-anomer can also escape detection and pass behind the hairpin structure. Just as was observed for the phenanthrene however this occurs less readily for the non-natural α-configuration compared to the β-anomer resulting in a longer retention time on the DNA and increased incision efficiency.
Taken together, both the phenanthrene and pyrene damage cause a significant DNA distortion resulting in proper recognition by UvrA. However the planar nature of these compounds enables them to escape detection by UvrB by passing behind the hairpin of the protein. Increasing the size of the adduct group or substituting it to the ribose in the non- natural α-configuration enhances the recognition by UvrB, probably by slowing down this translocation behind the hairpin.
As expected, substrates containing the pyrene modification exhibit a significantly red shifted excitation maximum (349 nm) compared to phenanthrene (Fig. 5D and 4D) and can
therefore be used in biological systems requiring ATP. Quantitative measurements showed that incorporation of the pyrene moiety in ssDNA and thereby stacking it to its neighboring adenine residues decreases the fluorescence signal of ~520 AU for pyrene free in solution to a level still exhibiting considerable fluorescence (α: ~126 AU β: ~127 AU; Fig. 6E). When incorporated in ssDNA consisting of a stretch of thymines however fluorescence is quenched completely (α: ~5 AU β: ~6 AU results not shown). This is consistent with a previous study where quenching was determined to be the result of photo-induced electron transfer with a neighboring G, C or T, but not with A (Manoharan et al. 1995). Since in our construct pyrene is flanked by two adenines the remaining signal can be explained by inefficient electron transfer with these adenines. Upon addition of equimolar amounts of the complementary DNA strand fluorescence strongly decreases for both substrates (α: ~11 AU β: ~22 AU).
Possibly the large pyrene group pushes the opposing nucleotide out of the DNA helical stack and as a result it is now quenched by the thymines in the opposite strand. An NMR structure of a pyrene adduct incorporated into the DNA opposite an abasic site indeed revealed the compound to occupy the position of the absent nucleobase stacking on the neighbors of the abasic site (Smirnov et al. 2002). Pyrene fluorescence in our constructs can therefore not only be used to study alterations in stacking interactions with its neighboring nucleotides but also to monitor DNA strand separation.
Addition of an excess of UvrA (2 μM) to the α- and β-pyrene containing dsDNA substrates (0.5 μM) resulted in an increase in the fluorescence signal for both constructs (α:
~43 AU β: ~44 AU; Fig. 5E). This increase in fluorescence is significant since under these conditions filter-binding analysis showed that about 30% of the DNA is bound by UvrA (Table 1). Since the signal obtained by UvrA on dsDNA is also approximately 30% of the emission of ssDNA it is suggestive to conclude that in the UvrA-DNA complex the DNA is unwound at the site of damage. Possibly, strand separation by UvrA is required to allow UvrB to insert its β-hairpin between the two DNA strands.
In the presence of both UvrA (0.45 μM) and UvrB (3.75 μM) a considerable amount of UvrB2-DNA complexes are formed on the α-pyrene substrate whereas only little UvrB2- DNA complexes are formed on the β-pyrene (Fig. 5F). This again shows poor recognition of the β-anomer by the UvrB protein. Under these conditions the α-anomer gave a signal of ~141 AU although less than 50% of the DNA was bound to the proteins. Since this fluorescence signal exceeds the signal of ssDNA it cannot be solely explained by strand separation by UvrA(B) at the site of the lesion. Fluorescence of the β-pyrene upon addition of UvrA and UvrB is increased as well (~54 AU). In light of the poor complex formation on this construct (<20%) fluorescence on this DNA substrate is substantial and again too high to be only due to ssDNA formation. This indicates that base stacking interactions with neighboring nucleotides must be disturbed (as well), which is in apparent contradiction to the results obtained on the AP-M which did not alter its fluorescence upon binding of a similar amount of UvrB (compare Fig. 3D and 5E).
Several explanations for the UvrB-induced signal of the two pyrene anomers can be considered. It has been shown (Malta et al. 2006) that upon UvrB binding to a cholesterol or menthol lesion the 3’ flanking base is flipped into an extrahelical conformation. Flipping of its neighbor might be responsible for enhanced fluorescence of the pyrene damage by reducing the quenching by this neighbor. On the other hand not such a high signal would be expected, since quenching by the 5’ neighbor would still occur. Moreover, for the AP-M UvrB-mediated flipping of the 3’ neighbor does not enhance the signal. Probably, energy transfer from the fluorescent group to its 5’ neighboring base is efficient enough to quench fluorescence.
Another explanation for the enhanced fluorescence might be that the pyrene damage now takes up the extrahelical position in the UvrB-DNA complex that for the cholesterol and menthol lesions is occupied by the 3’ neighbor. This, however, seems also less likely since it would be expected that shifting of the UvrB contacts by one nucleotide in the 5’ direction would also result in a shift in incision position of UvrC by one nucleotide. Incision positions turned out to be the same as on the other constructs (results not shown).
Alternatively upon recognition of the pyrene by UvrB the flat nature of the compound and the lack of basepairing might allow it to be partially extruded from the base stack without being fully rotated behind the hairpin.
If, as described above, the obtained fluorescence signal would only originate from complexes where UvrB is bound to the damaged site, the signal is unexpectedly high. Taken into account the low complex formation, especially in the case of the β-pyrene where hardly any UvrB-DNA complexes can be detected (Fig. 5F lane 2) it would mean that in each individual UvrB-DNA complex the fluorescence of the pyrene would equal or even exceed the fluorescence of the free pyrene. It is therefore very likely that the obtained fluorescence signal not only derives from the UvrB-DNA preincision complexes but at least also from complexes formed during the search for DNA damage which escape detection by UvrB.
If during this dynamic process a pyrene molecule is passed behind the β-hairpin it will be temporarily in a fully extrahelical position thereby emitting a high fluorescence signal. In the case of the α-configuration the passage of the pyrene might be a slower process because of its non-natural configuration, which could cause an enhanced fluorescence compared to the β-configuration.
Taken together it is not clear what exactly causes the fluorescence signal of the pyrene anomers in the presence of UvrB. These studies however do show that the pyrene modifications enable visualization of conformational changes that remain hidden for the AP- M. This makes a pyrene-modified DNA substrate a more sensitive tool to study other DNA repair systems.
Conclusion
In this study we have synthesized several fluorescent DNA modifications that can be used to study conformational changes in the DNA: the 2-aminopurine menthol modification (AP-M), phenanthrene and pyrene. All of these constructs are expected to change their fluorescence upon a conformational change at the site of damage. However, we have shown that pyrene is the most sensitive probe since its fluorescence is responsive to alterations in both base stacking and base pairing interactions. AP-M fluorescence on the other hand is only responsive to a reduction in base stacking. The phenanthrene compound could not be studied with the UvrA and UvrB proteins since its excitation maximum overlaps with the absorption spectrum of ATP, an important cofactor for the action of these proteins. Taken together however, it would be interesting to use these fluorescent probes to monitor conformational changes at the site of damage induced by other DNA repair proteins. Examples of proteins or protein complexes that might be studied in this way include UV-DDB, XPC, XPA and XPD.
References
Bruner, S.D., Norman, D.P. and Verdine, G.L. (2000) Structural basis for recognition and repair of the endogenous mutagen 8-oxoguanine in DNA. Nature, 403, 859-866.
Chen, L., Haushalter, K.A., Lieber, C.M. and Verdine, G.L. (2002) Direct visualization of a DNA glycosylase searching for damage. Chem. Biol., 9, 345-350.
Clugston, C.K., McLaughlin, K., Kenny, M.K. and Brown, R. (1992) Binding of human single-stranded DNA binding protein to DNA damaged by the anticancer drug cis-diamminedichloroplatinum (II). Cancer Res., 52, 6375-6379.
Daniels, D.S., Woo, T.T., Luu, K.X., Noll, D.M., Clarke, N.D., Pegg, A.E. and Tainer, J.A. (2004) DNA binding and nucleotide flipping by the human DNA repair protein AGT. Nat. Struct. Mol. Biol.,Struct. Mol. Biol., 11, 714-720.
de Kort, M., Ebrahimi, E., Wijsman, E.R., van der Marel, G.A. and van Boom, J.H. (1999) Synthesis of oligodeoxynucleotides containing 5-(beta-D-glycopyranosyloxymethyl)-2 ‘-deoxyuridine, a modified nucleoside in the DNA of Trypanosoma Brucei. Eur. J. Org. Chem. 2337-2344.
de Kort, M., de Visser, P.C., Kurzeck, J., Meeuwenoord, N.J., van der Marel, G.A., Ruger, W. and van Boom, J.H. (2001) Chemical and enzymatic synthesis of DNA fragments containing 5-(beta-D- glucopyranosyloxymethyl)-2 ‘-deoxycytidine - a modified nucleoside in T4 phage DNA. Eur. J. Org. Chem., 2075-2082.
Evans, E., Moggs, J.G., Hwang, J.R., Egly, J.M. and Wood, R.D. (1997) Mechanism of open complex and dual incision formation by human nucleotide excision repair factors. EMBO J., 16, 6559-6573.
Friedberg, E.C., Walker, G.C., Siede, W., Wood, R.D., Shchultz, R.A. and Ellenberger, T. (2005) Friedberg EC, ed. DNA repair and mutagenesis. Washington, DC: ASM Press.
Lukin, M. and de los Santos, C. (2006) NMR structures of damaged DNA. Chem. Rev., 106, 607-686.
Malta, E., Moolenaar, G.F. and Goosen, N. (2006) Base flipping in nucleotide excision repair. J. Biol. Chem., 281, 2184-2194.
Malta, E., Moolenaar, G.F. and Goosen, N. (2007) Dynamics of the UvrABC nucleotide excision repair proteins analyzed by fluorescence resonance energy transfer. Biochemistry, 46, 9080-9088.
Malta, E., Verhagen, C.P., Moolenaar, G.F., Filippov, D.V., van der Marel, G.A. and Goosen, N. (2008) Functions of base flipping in E. coli nucleotide excision repair. DNA repair, 7, 1647-1658.
Manoharan, M., Tivel, K.L., Zhao, M., Nafisi, K. and Netzel, T.L. (1995) Base-Sequence Dependence of Emission Lifetimes for DNA Oligomers and Duplexes Covalently Labeled with Pyrene – Relative Electron- Transfer Quenching Efficiencies of A-Nucleoside, G-Nucleoside, C-Nucleoside, and T-Nucleoside toward Pyrene.
J. Phys. Chem., 99, 17461–17472.
Mazur, S.J. and Grossman, L. (1991) Dimerization of Escherichia coli UvrA and its binding to undamaged and ultraviolet light damaged DNA. Biochemistry, 30, 4432-4443.
Mees, A., Klar, T., Gnau, P., Hennecke, U., Eker, A.P., 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.
Moolenaar, G.F., Monaco, V., van der Marel, G.A., van Boom, J.H., Visse, R. and Goosen, N. (2000) The effect of the DNA flanking the lesion on formation of the UvrB-DNA preincision complex. Mechanism for the UvrA-mediated loading of UvrB onto a DNA damaged site. J. Biol. Chem., 275, 8038-8043.
Moolenaar, G.F., Höglund, L. and Goosen, N. (2001) Clue to damage recognition by UvrB: residues in the beta- hairpin structure prevent binding to non-damaged DNA. EMBO J., 20, 6140-6149.
Moser, J., Volker, M., Kool, H., Alekseev, S., Vrieling, H., Yasui, A., van Zeeland, A.A. and Mullenders, L.H. (2005) The UV-damaged DNA binding protein mediates efficient targeting of the nucleotide excision repair complex to UV-induced photo lesions. DNA repair, 4, 571-582.
Naegeli, H., Bardwell, L. and Friedberg, E.C. (1992) The DNA helicase and adenosine triphosphatase activities of yeast Rad3 protein are inhibited by DNA damage. J. Biol. Chem., 267, 392-398.
Naegeli, H., Modrich, P. and Friedberg, E.C. (1993) The DNA helicase activities of Rad3 protein of Saccharomyces cerevisiae and Helicase II of Escherichia coli are differentially inhibited by covalent and noncovalent DNA modifications. J. Biol. Chem., 268, 10386-10392.
Oh, E.Y. and Grossman, L. (1989) Characterization of the helicase activity of the Escherichia coli UvrAB protein complex. J. Biol. Chem., 264, 1336-1343.
Ren, R.X.F., Chaudhuri, N.C., Paris, P.L., Rumney, S. and Kool, E.T. (1996) Naphthalene, phenanthrene, and pyrene as DNA base analogues: Synthesis, structure, and fluorescence in DNA. Am. Chem. Soc., 118, 7671-7678.
Robins, P., Jones, C.J., Biggerstaff, M., Lindahl, T. and Wood, R.D. (1991) Complementation of DNA repair in xeroderma pigmentosum group A cell extracts by a protein with affinity for damaged DNA. EMBO J., 10, 3913-3921.
Seeberg, E. and Steinum, A.L. (1982) Purification and properties of the uvrA protein from Escherichia coli.
Proc. Natl. Acad. Sci. USA, 79, 988-992.
Shuck, S.C., Short, E.A. and Turchi, J.J. (2008) Eukaryotic nucleotide excision repair: from understanding mechanisms to influencing biology. Cell Res., 18, 64-72.
Sinha, N.D., Davis, P. Usman, N., Perez, J., Hodge, R., Kremsky, J. and Casale, R. (1993) Labile exocyclic amine protection of nucleosides in DNA, RNA and oligonucleotide analog synthesis facilitating N- deacylation,minimizing depurination and chain degradation. Biochimie, 75, 13-23.
Smirnov, S., Matray, T.J., Kool, E.T. and de los Santos, C. (2002) Integrity of duplex structures without hydrogen bonding: DNA with pyrene paired at abasic sites. Nucleic Acids Res., 30, 5561-5569.
Sugasawa, K., Ng, J.M., Masutani, C., Iwai, S., van der Spek, P.J., Eker, A.P., Hanaoka, F., Bootsma, D. and Hoeijmakers, J.H. (1998) Xeroderma pigmentosum group C protein complex is the initiator of global genome nucleotide excision repair. Mol. Cell, 2, 223-232.
Sung, P., Watkins, J.F., Prakash, L. and Prakash, S. (1994) Negative superhelicity promotes ATP-dependent binding of yeast Rad3 protein to ultraviolet-damaged DNA. J. Biol. Chem., 269, 8303-8308.
Truglio, J.J., Croteau, D.L., Van Houten, B. and Kisker, C. (2006a) Prokaryotic nucleotide excision repair: the UvrABC system. Chem. Rev., 106, 233-252.
Truglio, J.J., Karakas, E., Rhau, B., Wang, H., DellaVecchia, M.J., Van Houten, B. and Kisker, C. (2006b) Structural basis for DNA recognition and processing by UvrB. Nat. Struct. Mol. Biol., 13, 360-364.
Vargeese, C., Carter, J., Yegge, J., Krivjansky, S., Settle, A., Kropp, E., Peterson, K. and Pieken, W. (1998) Efficient activation of nucleoside phosphoramidites with 4,5-dicyanoimidazole during oligonucleotide synthesis.
Nucleic Acids Res. 26, 1046-1050.
Verhoeven, E.E., van Kesteren, M., Turner, J.J., van der Marel, G.A., van Boom, J.H., Moolenaar, G.F. and Goosen, N. (2002) The C-terminal region of Escherichia coli UvrC contributes to the flexibility of the UvrABC nucleotide excision repair system. Nucleic Acids Res., 30, 2492-2500.
Visse, R., de Ruijter, M., Moolenaar, G.F. and van de Putte, P. (1992) Analysis of UvrABC endonuclease reaction intermediates on cisplatin-damaged DNA using mobility shift gel electrophoresis. J. Biol. Chem., 267, 6736-6742.
Waters, T.R., Eryilmaz, J., Geddes, S. and Barrett, T.E. (2006) Damage detection by the UvrABC pathway:
crystal structure of UvrB bound to fluorescein-adducted DNA. FEBS Lett., 580, 6423-6427.
Wittschieben, B.Ø., Iwai, S. and Wood, R.D. (2005) DDB1-DDB2 (xeroderma pigmentosum group E) protein complex recognizes a cyclobutane pyrimidine dimer, mismatches, apurinic/apyrimidinic sites, and compound lesions in DNA. J. Biol. Chem., 280, 39982-39989.
