E. coli UvrA protein
Wagner, K.
Citation
Wagner, K. (2011, February 17). Nucleotide excision repair at the single-molecule level : analysis of the E. coli UvrA protein. Retrieved from https://hdl.handle.net/1887/16502
Version: Not Applicable (or Unknown)
License: Leiden University Non-exclusive license Downloaded from: https://hdl.handle.net/1887/16502
Note: To cite this publication please use the final published version (if applicable).
Role of the two ATPase domains of Escherichia coli UvrA in binding non- bulky DNA lesions and interaction with UvrB
Koen Wagner, Geri F. Moolenaar and Nora Goosen
Laboratory of Molecular Genetics, Leiden Institute of Chemistry, Leiden University
Einsteinweg 55, 2333 CC, Leiden, the Netherlands
DNA Repair, 9, 1176-1186 (2010)
ABSTRACT
The UvrA protein is the initial DNA damage-sensing protein in bacterial nucleotide excision repair and detects a wide variety of structurally unrelated lesions. After initial recognition of DNA damage, UvrA loads the UvrB protein onto the DNA. This protein then verifies the presence of a lesion, after which UvrA is released from the DNA. UvrA contains two ATPase domains, both belonging to the ABC ATPase superfamily. We have determined the activities of two mutants, in which a single domain was deactivated. Inactivation of either one ATPase domain in Escherichia coli UvrA results in a complete loss of ATPase activity, indicating that both domains function in a cooperative way. We could show that this ATPase activity is not required for the recognition of bulky lesions by UvrA, but it does promote the specific binding to the less distorting cyclo-butane pyrimidine dimer (CPD). The two ATPase mutants also show a difference in UvrB-loading, depending on the length of the DNA substrate. The ATPase domain I mutant was capable of loading UvrB on a lesion in a 50 bp fragment, but this loading was reduced on a longer substrate. For the ATPase domain II mutant the opposite was found: UvrB could not be loaded on a 50 bp substrate, but this loading was rescued when the length of the fragment was increased. This differential loading of UvrB by the two ATPase mutants could be related to different interactions between the UvrA and UvrB subunits.
Chapter
4
INTRODUCTION
UvrA is the initial DNA damage recognizing protein in the bacterial nucleotide excision repair (NER) pathway. In the current model for bacterial NER, the UvrA dimer and two UvrB molecules form the damage recognizing UvrA2B2-complex. In this complex, the UvrA subunits scan the DNA for potential lesions. Since the UvrA dimer preferentially binds not only to damaged DNA but also to single-stranded DNA [1,2] and the ends of linear DNA [3,4]it likely recognizes DNA sites that have a reduced base pairing.
Upon detection of such a site, UvrA will subsequently try to load UvrB to the lesion site.
After a successful loading, UvrA will dissociate from the DNA, leaving the UvrB-DNA pre- incision complex. This pre-incision complex is the binding target for UvrC, which subsequently incises the damage-containing DNA strand on both sides of the lesion (reviewed in [5 and 6]).
The initiation of DNA repair by UvrA and UvrB is a process that requires ATP hydrolysis in both UvrA and UvrB. The UvrB protein contains one ATPase site [7], whereas the UvrA monomer contains two ATPase domains [8,9], which both belong to the ABC (ATP Binding Cassette) type of ATPases. This type of ATPase is commonly found in the ABC-transporter protein family, a class of membrane transporter proteins, but is also present in the DNA repair proteins Rad50, MutS and UvrA [10]. The mechanism and function of the ABC ATPase have been subject of several studies, showing that ATP binding generates a large rearrangement in the structure of the ABC ATPase, through the interaction of its different sub-domains, from which the Walker A motif and the signature sequence are the most important. This structural rearrangement is translated into a change in the position of the functional domains in the protein, which for the ABC-transporter proteins results in the transport of their substrate across a membrane (reviewed in [11 and 12]).
The crystal structure of ADP-bound Bacillus stearothermophilus UvrA [13] shows the orientation of both ATPase domains in the UvrA protein (Figure 1). ATPase domain I consists of the N-terminal Walker A motif and the C-terminal signature sequence; ATPase domain II consists of the C-terminal Walker A motif and the N-terminal signature sequence.
Three additional domains, each coordinated by a zinc-binding module, are present in the UvrA structure: the UvrB-binding domain, the insertion domain (ID) and a zinc-finger motif [13,14].
The structure of UvrA also showed the presence of a conserved patch of positively charged amino acids, forming a DNA binding surface [13]. This DNA binding surface is proposed to coordinate the recognition of DNA damage, operating together with the insertion domain and the zinc-finger motif [13-16]. However, the interplay between these domains, as well as the role of ATP, in damage recognition remains largely unknown.
Mutants carrying a substitution of the conserved lysine in the Walker A motifs of the two Escherichia coli ATPase domains have been analyzed [8,9]. Both mutants were shown to be defective in repairing UV-induced lesions in vivo, but conflicting results as to which step of the repair reaction was affected by the mutations were reported. In the study of Thiagalingam and Grossman [9] the two mutants were reported to be deficient in damage recognition. Myles et al. [8] however, presented evidence that the two mutants were not affected in damage- specific binding but they were deficient in the loading of UvrB.
In this paper we have analyzed the same UvrA mutant proteins (UvrA K37A and UvrA K646A) using a combination of single-molecule and bulk techniques. We found that cooperativity exists between both ATPase domains, as inactivation of only one ATPase domain completely disabled the ATPase activity of UvrA. Furthermore we show that the ATPase of UvrA functions to enhance recognition of non-bulky lesions and to coordinate binding of UvrB.
Figure 1: Crystal structure of the ADP-bound dimer of B. stearothermophilus UvrA (PDB entry 2R6F) Monomer 1 is shown in gray; monomer 2 is shown in light blue.
ATPase domain I, containing residue K37 is shown in dark blue. ATPase domain II, containing residue K646 is shown in red. Bound ADP is shown as yellow spheres.
The three zinc-binding modules are colored black, bound Zn2+ is shown as a black sphere. The UvrB-binding domain coordinated by zinc-binding domain 1 (Zn1) is indicated in orange. The insertion domain coordinated by Zn2 is indicated in green. The zinc-finger motif coordinated by Zn3 is shown in violet.
This image was generated using PyMol, release 0.99 (DeLano Scientific).
MATERIALS AND METHODS
Plasmids
The pSC101-derived pNP120 plasmid, carrying the UvrA gene under control of its native promoter, has been described previously [17].
The mutations in the uvrA gene were introduced by PCR, by a similar procedure as described [17], using the pTTQ-A9 plasmid [18] as a template. After verification of the incorporation of the mutation into the uvrA gene, the mutants were inserted into pNP120.
UV-survival assay
The pNP120-derived plasmids expressing (mutated) UvrA were transformed into strain CS5865 (GM1 ∆uvrA, our lab). Transformants were grown to an OD600 of 0.3. This culture was diluted 10 times, after which 1 µl drops (containing around 30,000 bacteria) were deposited onto LB-agar plates, which were irradiated with the indicated doses of UV-light.
The plates were incubated overnight in the dark at 30 °C.
Proteins and chemicals
(Mutant) UvrA, UvrB and UvrC proteins were purified as described [19]. Ku70/80 was a gift form Roland Kanaar (Erasmus University, Rotterdam). ATP and ATPγS were purchased from Roche. DNA oligomers were purchased from EuroGentec.
DNA substrates
The 678 bp DNA fragments for AFM analysis, containing a Cholesterol, Menthol or CPD lesion in the center of the fragment, were prepared according to the method described in [4].
To prepare a radioactively labeled 678 bp substrate, the 50 nt oligomer containing the damage was radioactively labeled at the 5’ end prior to incorporation into the fragment. Incorporation of the Cholesterol and Menthol lesions at position 340, which is located in an EcoRV recognition sequence, was verified by restriction analysis. The presence of the CPD lesion at positions 335 and 336, which are not part of a recognition sequence, was verified by incision with UV damage endonuclease (UVDE) [20].
The 1020 bp DNA fragment for Ku70/80 binding was obtained by PCR on the URA-3 gene with forward primer U3 (GAAGGAAGAACGAAGGAAGGAGC) and reverse primer UH4 (TTTCCCGGGGGGCCCGGGTAATAACTGATATAATT).
The 50 bp DNA substrates, containing a lesion at position 27, were constructed and radioactively labeled at the 5’ end of the damaged strand as described [21]. Two 50 bp substrates with different sequences, as described in [21], were used. Substrate S1 had the following top-strand sequence (X indicates the position of the lesion):
GGGATTACTTACGGCCACATTACTACXGGAACTCAGAACGAGCTGACAGG
Substrate S2, with the lesion in same sequence context as in the 678 bp substrate, had this top- strand sequence:
ATTTGTTTACTAAAAACACATGTGGAXATCTTGACTGATTTTTCCATGGA.
The 96 bp Cholesterol substrate, containing the lesion at position 50 in the same sequence context as in S1, was constructed by directly hybridizing a 96 nt top-strand DNA oligomer with a 96 nt bottom-strand DNA oligomer and radioactively labeled as described [21].
ATPase assay
Samples were incubated for 20 minutes at 37 ºC in ATPase-endo buffer (50 mM Tris-HCl, pH 7.5, 15 mM MgCl2, 100 mM KCl, 0.1 mg/ml BSA, 10 % glycerol and 0.5 mM ATP plus 0.1 µCi γ32P-ATP (3000 Ci/mmol)). Reaction mixtures contained 30 nM UvrA. When indicated, supercoiled pUC18 DNA was added in a final concentration of 4 ng/µl. For the preparation of UV-damaged DNA, pUC18 was irradiated with 1000 J/m2 UV-light.
ATPase activity of UvrA was measured by counting the total amount of 32P-phosphate released from γ32P-ATP, as described [22].
Incision assays
Radioactively labeled DNA substrates were incubated for 30 minutes with 2.5 nM UvrA, 100 nM UvrB and 25 nM UvrC in 20 µl UV-endo buffer (50 mM Tris-HCl, pH 7.5, 10 mM MgCl2, 100 mM KCl, 0.1 mg/ml BSA and 1 mM ATP). When indicated non-irradiated pUC18 plasmid DNA was included in the mixture as competitor. After incision, the DNA was precipitated with ethanol and analyzed on 15 % denaturing polyacrylamide gels as described [21]. After incubation, samples containing the 678 bp substrate were digested with ApoI, shortening non-incised 678 nt DNA to 379 nt and 5’ incised DNA to 20 nt. The amounts of incised and unincised DNA were quantitated as described [21].
Gel retardation assay
Gel retardation assays were performed as described [19]. UvrA and, when indicated, UvrB were incubated in UV-endo buffer for 5 min at 37 °C, after which DNA was added to the samples. When indicated, non-irradiated pUC18 plasmid DNA was included in the mixture as competitor. After 10 min of incubation with DNA, samples were separated on a 3.5 % native polyacrylamide gel containing 1 mM ATP and 10 mM MgCl2.
Detection of UvrAB-complexes on a native gel
UvrA and UvrB (2.5 µM) were incubated in UV-endo buffer for 10 min at 37 °C with or without ATP as indicated. After incubation, samples were separated on a 3.5 % native poly- acrylamide gel containing 10 mM MgCl2.
The proteins were visualized by staining with Coommassie. For the Western blots 100 nM of UvrB was incubated with the indicated amounts of UvrA in the same buffer. Reaction mixtures were separated on a 3.5 % polyacrylamide native gel containing 10 mM MgCl2 and, when indicated, 1 mM ATP.
After blotting of the proteins on a nitrocellulose membrane, the membrane was incubated with rabbit UvrB antibody (described in [19]) and the secondary antibody goat anti-rabbit alkaline- phosphatase (GARAP) in blocking buffer (1 mg/ml BSA). The presence of UvrB-coupled GARAP was detected by staining with 0.4 mg/ml nitro blue tetrazolium (NBT) plus 0.19 mg/ml 5-bromo-4-chloro-3-indolyl phosphate (BCIP) in alkaline-phosphatase buffer (100 mM Tris-HCl, pH 9.5, 100 mM NaCl, 5 mM MgCl2).
Atomic Force Microscopy Imaging
For visualization of UvrA-DNA complexes, UvrA (20 nM) was incubated in 10 µl AFM-endo buffer (40 mM KCl, 10 mM MgCl2, 50 mM Tris-HCl, pH 7.5) with 50 ng of the damaged or undamaged 678 bp DNA for 10 minutes at 37 °C. When indicated, ATP or ATPγS was added in a concentration of 1 mM. Deposition of UvrA–DNA complexes was performed as described using a 10 mM MgCl2, 10 mM HEPES, pH 7.8 deposition buffer [4].
Ku70/80 heterodimer (50 nM) was incubated in 10 µl Ku-buffer (50 mM KCl, 10 mM MgCl2, 50 mM HEPES, pH 7.8) with 50 ng of 1020 bp DNA for 10 min. Simultaneous deposition of UvrA–DNA and Ku70/80–DNA complexes was performed as described [4].
To image UvrA monomers and dimers, the UvrA protein (20 nM) was incubated for 5 minutes at 37 °C in 20 µl AFM-UvrA buffer (100 mM KCl, 10 mM MgCl2, 50 mM Tris pH 7.5). When indicated, ATP was added in a concentration of 1 mM. Deposition was performed as described [4], by directly depositing 5 µl from the sample on freshly cleaved mica (Spruce Pine Mica Co.).
Imaging was performed with a Nanoscope III instrument (Digital Instruments), equipped with an E-scanner, using tapping mode in air. OMCL-AC240TS MicroCantilever tapping mode cantilevers (Olympus) with a spring constant of 2 N/m and a resonance frequency of 70 kHz were used for all imaging. All images of deposited protein-DNA complexes were collected at a scan rate of 2 Hz and a scan size of 2 x 2 µmrespectively.
Protein complex volumes were calculated by summing of the height at each pixel inside a circle around the mass center of a protein complex as described [4]. The percentage of site- specific binding and the binding specificity of UvrA for DNA ends and DNA damage was calculated as described [4,23].
RESULTS
The two ATPase domains operate in a cooperative fashion
To analyze the function of the two ATPase domains, we have constructed two mutants in which the central lysine of the Walker A motif in ATPase domain I (K37A) and II (K646A) of UvrA is replaced by alanine. To investigate the function of both ATPase domains on UV- survival in vivo, we have analyzed the ability of a ∆uvrA E. coli strain carrying a plasmid with the (mutated) UvrA gene under control of its native promoter, to survive exposure to UV- light. With Western blotting, we verified that each strain expressed a similar level of (mutant) UvrA (results not shown).
E. coli expressing wildtype UvrA can survive exposure of a dose of 50 J/m2 UV-light, whereas both ATPase mutants are extremely UV-sensitive, consistent to what was reported in literature [7,8]. UvrA K646A is almost completely UV-sensitive, tolerating no more than 2.5 J/m2 UV-light. UvrA K37A is slightly more UV-resistant than UvrA K646A, but does not tolerate exposure to more than 5 J/m2 UV-light (Figure 2). Apparently, the two ATPase domains of UvrA are very important for the repair of UV-damaged DNA. The low UV- tolerance of UvrA K37A indicates that this mutant still should be capable of performing some repair, although with a greatly reduced activity.
Figure 2: UV-survival of E. coli expressing wildtype and mutant UvrA proteins
E. coli strain CS5865 (GM1 ∆uvrA, our lab)
was transformed with a plasmid carrying the (mutant) UvrA gene under control of its native promoter. Transformants (about 30,000 cells) were spotted on LB-agar plates and exposed to UV-light. The dose of UV-light is indicated on top of each panel, in J/m2.
The purified wildtype UvrA protein shows ATPase activity both in the absence and in the presence of (damaged) DNA. In the presence of undamaged DNA however, wildtype UvrA shows roughly 50 % ATPase activity compared to when damaged DNA is present or in the absence of DNA (Table 1). Apparently, the ATPase activity of UvrA is inhibited when UvrA binds undamaged DNA, but increases upon binding a DNA lesion.
Notably, both ATPase mutants show a complete loss of ATPase activity either in the absence or the presence of (damaged) DNA (Table 1). This result demonstrates that the ATPase activity of UvrA requires the cooperative action of both ATPase domains.
Table 1: ATPase activity of UvrA and ATPase domain mutants
UvrA Effector
None DNA UV-DNA (1000 J/m2)
Wildtype 90.1 ± 4.3 45.3 ± 1.7 97.5 ± 5.5
K37A 0.5 ± 0.5 1.2 ± 1.4 0.8 ± 1.3
K646A 0.9 ± 0.9 0.7 ± 0.5 1.0 ± 1.0
(Mutant) UvrA protein was incubated in ATPase-endo buffer containing 0.5 mM ATP for 20 minutes at 37 ºC as described in Materials and Methods.
ATPase activity of UvrA was measured by counting the total amount of 32P-phosphate released from γ32P-ATP.
These values were corrected for the amount of 32P-phosphate released in the absence of protein.
The presented values show the average (± S.D.) ATP turnover rate of (mutant) UvrA in ATP·UvrA-1·min-1.
The two ATPase domains have different activities in the incision reaction
The activity of both ATPase mutants in the UvrABC repair reaction was further analyzed by performing incision assays. On a 50 bp substrate containing a Cholesterol lesion the incision level in the presence of UvrA K37A is the same as with wildtype UvrA (Figures 3A and B, lanes 1 and 2). On a 50 bp fragment containing a different type of lesion (Menthol) in a different sequence context, the amount of incision with K37A and wildtype UvrA is also similar (Figure 3C, lanes 1 and 2 and Figure 3D). Comparing the kinetics of the incision reaction did not reveal a significant difference between wildtype UvrA and K37A (Figures 4A and B). These results are unexpected as, based on its severely reduced UV-tolerance, a much lower incision activity could be expected for UvrA K37A.
In vivo however, DNA lesions will be obscured by undamaged genomic DNA, and this could affect repair activity. Therefore, we determined incision activity on a longer substrate which resembles more the in vivo situation: a 678 bp DNA fragment, containing a Menthol lesion in the same sequence context as in the 50 bp substrate (Figure 3C, lanes 4-6). On this longer substrate incision with K37A (~ 60 %) is significantly lower than with wildtype (about 90 %) (Figure 3D). This result suggests that the repair activity of K37A can be inhibited by the presence of undamaged DNA. Indeed, addition of undamaged competitor DNA to the 50 bp substrate has a stronger inhibiting effect on incision with K37A than with wildtype UvrA (Figures 3E and F).
Figure 3: UvrABC incision on DNA substrates with different lengths
(A) Incision and (B) quantitation of the incision on a 50 bp DNA fragment with a Cholesterol lesion in sequence S1. (C) Incision and (D) quantitation of the incision on a 50 bp (lanes 1-3) and 678 bp (lanes 4-6) DNA fragment with a Menthol lesion in sequence S2. (E) Incision and (F) quantitation of the incision on a 50 bp DNA fragment with a Menthol lesion in sequence S1, in the presence of indicated amounts of non-damaged competitor DNA.
500 ng DNA equals an 8000 times excess (in basepairs) of competitor DNA over the 50 bp DNA substrate. The DNA substrates (0.5 nM) were incubated with 2.5 nM UvrA, 100 nM UvrB and 25 nM UvrC. After incubation samples containing the 678 bp substrate were digested with restriction enzyme ApoI, shortening non-incised 678 bp DNA to 379 bp and 5’ incised DNA to 20 bp. Incision products are indicated with arrows. The presented values of the quantitations represent the average (± S.D.) of at least two independent experiments. Curve fits were made using OriginPro 8 software (OriginLab).
Surprisingly, with UvrA K646A the length of the DNA substrate has an opposite effect on the incision activity. With the 50 bp Cholesterol fragment, hardly any incision is observed (Figure 3A, lane 3). On the 50 bp Menthol fragment, incision with K646A is detectable (Figure 3C, lane 3), but much lower (~ 20 %) than with wildtype UvrA (~ 90 %) (Figure 3D).
When the same Menthol lesion is placed in the 678 bp fragment, however, K646A results in a remarkably high incision (~ 80 %), that is almost as high as with wildtype (~ 90 %) (Figure 3C, lanes 4 and 6 and Figure 3D). Also, the kinetics of the incision reaction on the 678 bp fragment show that on this substrate the incision rate of UvrA K646A is only slightly reduced compared to wildtype UvrA (Figures 4C and D). These results clearly show that K646A is perfectly capable of finding damage in a relatively long DNA fragment, indicating that this mutant is not disturbed in damage recognition. Apparently, it is affected in another step of the UvrABC reaction and this defect can be rescued by providing a longer substrate.
Figure 4: Kinetics of UvrABC incision
Incision on 50 bp (A) and 678 bp (C) DNA fragment containing a Menthol lesion (sequence S2) after different incubation times. The DNA substrates (0.5 nM) were incubated with 2.5 nM UvrA, 100 nM UvrB and 25 nM UvrC for the indicated amount of time. After incubation samples containing the 678 bp substrate were digested with restriction enzyme ApoI shortening non-incised 678 bp DNA to 379 bp and 5’ incised DNA to 20 bp.
Incision products are indicated with arrows. Quantitation of the incision activity on the 50 bp (B) and 678 bp (D) fragments is presented as the average (± S.D.) of at least two independent experiments.
Curve fits were made using OriginPro 8 software (OriginLab).
Both UvrA K37A and K646A are not disturbed in damage-specific binding
The reduced incision observed with K37A in the presence of undamaged DNA indicates that this mutant is disturbed in the damage recognition steps of the UvrABC reaction.
However, the high incision of K37A on the 50 bp Menthol fragment suggests that this mutant is not inhibited in binding damaged DNA per se. Indeed, UvrA K37A binds the 50 bp Menthol substrate with similar affinity as wildtype UvrA in a gel mobility shift assay (result not shown).
To test whether UvrA K37A has indeed difficulties in finding the damage on the 678 bp substrate, we analyzed binding of the mutant protein to this substrate using AFM (example images are shown in Figure 5). The volume of the UvrA K37A complexes on DNA was measured, showing that, like previously shown for UvrA wildtype [4], UvrA K37A only binds DNA as a dimer complex (data not shown).
Figure 5: AFM images of UvrA-DNA complexes
Fragments from representative AFM images of (mutant) UvrA-complexes on 678 bp undamaged DNA (A and B) and DNA with a Cholesterol lesion in the center of the substrate (C, D and E). In panel (E), a large amount of aggregated UvrA K646A is visible (example is indicated by a black arrow). Only UvrA K646A complexes with a volume similar to dimeric UvrA (example is indicated by a white arrow) were included in our analysis. In all images shown UvrA was incubated in the presence of ATP. The size of the image fragments is 1 × 1 µm.
Table 2: Site-specific binding on undamaged DNA
Cofactor % On End % On Non-specific Site
UvrA wildtype
None 56.2 ± 5.9 43.8 ± 5.9
ATP 50.5 ± 4.5 49.5 ± 4.5
ATPγS 54.2 ± 6.9 45.8 ± 6.9
UvrA K37A
None 50.8 ± 3.8 49.2 ± 3.8
ATP 56.2 ± 7.6 43.8 ± 7.6
(Mutant) UvrA was incubated with a 678 bp undamaged DNA fragment after which the complexes were visualized by AFM. The percentage of UvrA complexes bound at a DNA end was calculated as described [4].
The presented values represent the average (± S.D.) of at least two independent experiments. The values of specific binding of UvrA wild-type on undamaged DNA are taken from [4].
Table 3: Site-specific binding on DNA containing a Cholesterol lesion
Cofactor % on Damage % On End % On Non-specific Site
UvrA wildtype
None 96.4 ± 1.1 1.9 ± 0.5 1.7 ± 0.5
ATP 93.9 ± 2.4 2.5 ± 0.7 3.6 ± 1.1
ATPγS 40.0 ± 4.0 19.2 ± 3.6 40.8 ± 0.4
UvrA K37A
None 97.1 ± 4.3 0.5 ± 0.7 2.4 ± 1.5
ATP 95.0 ± 1.5 1.8 ± 1.1 3.2 ± 0.7
ATPγS 89.6 ± 3.6 5.2 ± 3.6 5.2 ± 0.2
UvrA K646A
ATP 87.7 7.7 4.6
(Mutant) UvrA was incubated with a 678 bp DNA fragment containing a Cholesterol lesion in the center and the complexes were visualized by AFM. The percentage of complexes bound at a specific site was calculated as described [4].The presented values represent the average (± S.D.) of at least two independent experiments.
On undamaged linear DNA, UvrA prefers to bind the ends of the substrate [4]. Both in the absence and the presence of ATP, UvrA K37A recognizes DNA ends with the same specificity as UvrA wildtype: for both proteins, approximately 50 % of the complexes are found at the DNA ends (Table 2). This is not unexpected, as it was previously shown [4] that the ability of UvrA to recognize a DNA end is not affected by binding or hydrolysis of ATP
On the damaged substrate, UvrA K37A also recognizes the Cholesterol lesion with a similar specificity as UvrA wildtype (Table 3). The specificity of UvrA K37A for the Cholesterol lesion is unaffected by ATP, as either in the absence of a cofactor or in the presence of ATP or the non-hydrolyzable ATPγS, similar percentages (about 95 %) of UvrA K37A-DNA complexes are bound to the lesion site (Table 3). In contrast, when wildtype UvrA is incubated with ATPγS, damage-specific binding is significantly reduced, since only 40 % of the complexes are at the lesion site (Table 3). The difference in damage specificity between K37A, in the presence of ATP (which cannot be hydrolyzed in the mutant) or ATPγS on one hand and wildtype UvrA containing the non-hydrolyzable ATP analog on the other hand suggests that the two ATPase domains of UvrA K37A are not fully occupied with cofactor. Likely, this mutant does not bind ATP (or ATPγS) in its mutated ATPase domain.
To summarize, the AFM results show that UvrA K37A is not affected in damage recognition on a 678 bp DNA fragment. The reduced incision on a similarly sized DNA substrate therefore must be due to a defect in a later step of the UvrABC reaction.
The incision assays already suggested that K646A is also not disturbed in finding a DNA lesion. In accordance with this, the mutant protein binds the 50 bp Menthol fragment with a similar affinity as wildtype UvrA in a gel mobility shift assay (not shown).
We have also tried to determine the damage specificity of UvrA K646A with AFM.
However, on our AFM images, only a very small amount of dimeric UvrA-DNA complexes could be detected with this mutant. Instead, we observed that the majority of this mutant protein has aggregated on the mica (Figure 5E). This aggregation of UvrA K646A is likely a problem specifically associated with the deposition of the protein on mica, as also in the absence of DNA this mutant aggregates on the mica surface (not shown).
From the small amount of UvrA K646A complexes that associated with the DNA as a dimer complex, it could be seen that this mutant is indeed not disturbed in damage recognition. On the undamaged DNA, 31 complexes are detected of which 16 are bound to the DNA ends. The percentage of UvrA K646A on a DNA end (52 %) is similar to that of UvrA wildtype (50 %). On the Cholesterol DNA, 130 dimeric K646A-DNA complexes could be analyzed. From these complexes, 114 (88 %) are bound at the lesion site. Due to the low amount of complexes that could be counted, it is unclear whether this result differs significantly from UvrA wildtype (94 %).
However, it does show that, in the presence of ATP, UvrA K646A has a higher damage specificity than UvrA wildtype in the presence of ATPγS (only 40 % on the lesion), indicating that, alike K37A, UvrA K646A is disturbed in binding ATP.
Both ATPase mutants are affected in binding UvrB
Since neither mutant was severely affected in damage discrimination, both mutants should be disturbed in another step of the UvrABC reaction, which would likely involve their interaction with UvrB. Therefore, we determined the ability of UvrA to bind UvrB in solution, using a native 3.5% polyacrylamide gel. On this gel UvrA and UvrB migrate at two distinct positions (Figure 6A) and these positions are independent on the absence (lanes 1 and 2) or presence (lanes 4 and 5) of ATP. Mixing equimolar amounts of UvrA and UvrB (2.5 µM) in the presence of ATP results in formation of a UvrAB-complex, which migrates at the same position as the UvrA protein (Figure 6A, lanes 4-6). The presence of UvrB in the higher migrating complex was confirmed using a Western blot with UvrB antibodies (Figure 6B, lane 2). Note that at the lower concentration of proteins (100 nM) that had to be used for the Western blot, complex formation between UvrA and UvrB still occurs. With the same concentrations of UvrA and UvrB in the absence of ATP, however, no UvrAB-complexes can be detected (Figure 6C), indicating that ATP promotes the interaction between both proteins.
At higher protein concentration (2.5 µM) some UvrAB-complexes are visible, but strikingly they migrate at a higher position in the gel (Figure 6A, lane 3). Apparently in the absence of ATP the UvrAB complex is not only less stable, but it also has a different conformation.
For both ATPase mutants the amount of UvrAB-complexes is reduced when ATP is present (Figure 6B, lanes 3 and 4). For K37A some UvrAB can be detected, but like the wildtype protein in the absence of cofactor this complex migrates at a higher position in the gel (lane 3). Given its low affinity for UvrB it is unlikely the higher migrating UvrAB- complex of K37A is caused by the association of an additional UvrB subunit. It rather reflects a different conformation of the UvrAB-complex. With K646A, no UvrAB-complexes can be detected at all (Figure 6B, lane 4).
Since the active form of UvrA is a dimer, a possible explanation for the reduced affinity of the ATPase mutants for UvrB might be that the dimeric form of these mutants in solution is less stable. We therefore tested the percentage of monomers and dimers formed by UvrA K37A using AFM. As shown before [4], dimer formation of the wildtype protein is enhanced by ATP, most likely because the cofactor promotes the formation of an ATP/ADP-mixed form (Table 4). The UvrA K37A mutant in the absence of ATP however, shows a much higher percentage of dimers than the wildtype protein (Table 4). The Walker A motif containing the K37A mutation is located in an alpha helix that contributes directly to the dimer interface [13].
Figure 6: Detection of UvrAB complexes on a native gel
(A) 2.5 µM of wt UvrA and UvrB were incubated in UV-endo buffer with or without 1 mM ATP as indicated.
The proteins were separated on a 3.5 % polyacrylamide native gel containing 10 mM MgCl2 after which they were visualized with Coomassie.
(B) 100 nM UvrB was incubated with 100 nM (mutant) UvrA in UV-endo buffer with 1 mM ATP. Reaction mixtures were separated on a 3.5 % polyacrylamide native gel containing 10 mM MgCl2 with 1 mM ATP. The gel was subsequently blotted on a nitrocellulose membrane and incubated with UvrB antibodies.
(C) 100 nM UvrB was incubated with the indicated amount of wildtype UvrA in UV-endo buffer without ATP.
Reaction mixtures were separated on a 3.5 % polyacrylamide native gel containing 10 mM MgCl2, after which the gel was blotted and incubated with UvrB antibodies.
Table 4. UvrA dimerization
Cofactor % Dimers
UvrA wildtype UvrA K37A
None 29 ± 3 88 ± 1
ATP 67 ± 2 63 ± 2
The (mutant) UvrA protein (20 nM) was pre-incubated with or without cofactor for 5 minutes at 37 ºC prior to deposition on mica for AFM visualization. The percentage of UvrA dimers was calculated as described [4]. The presented values represent the average (± S.D.) of at least two independent experiments.
Apparently, the mutation influences the conformation of this alpha helix such that the dimeric interface is stabilized, even in the absence of a cofactor. In the presence of ATP dimerization is reduced and similar to that of the wildtype protein with ATP, demonstrating that K37A does bind ATP and that this ATP causes a rearrangement of the dimeric interface of the mutant. Since this ATP is expected to be bound only in the non-mutated ATPase domain II, this means that also this second ATPase site (directly or indirectly) influences the dimer interface. Unfortunately, due to the aggregation of K646A we could not determine the efficiency of dimerization in solution of this mutant. We can only conclude that dimers of K646A can be formed since, as mentioned above, they were detected on the DNA.
Summarizing we show that, although the K37A mutation does affect the dimer interface, the reduced binding of UvrB cannot be ascribed to a quantitative difference in UvrA dimers.
Most likely the mutation causes a change in the orientation of the UvrB-binding domains, such that not only the contacts between UvrA and UvrB are reduced, but it also results in a different mobility of the UvrAB-complex in the gel.
The two ATPase domains of UvrA have a different contribution to loading UvrB to DNA damage
Next, we analyzed the loading of UvrB on a damaged site by the two ATPase mutants. On the 50 bp Menthol-DNA substrate UvrA K37A loads UvrB as efficient as UvrA wildtype (Figure 7A, lanes 1 and 2). This is not unexpected, since under the same conditions, K37A and wildtype also showed similar incision activities on this substrate (Figure 3E, lanes 1 and 5). However, when a lower amount of UvrB is used, UvrB-loading by K37A becomes less efficient than wildtype (Figure 7B), showing that the lower affinity of K37A for UvrB indeed affects its ability to load UvrB onto the DNA lesion.
Since on a longer substrate incision with UvrA K37A was reduced, we investigated whether also UvrB-loading by this mutant is decreased when the DNA fragment is longer. For that purpose we used a 96 bp fragment with a Cholesterol lesion.
Indeed on this substrate significantly less UvrB-DNA is formed with UvrA K37A than with wildtype UvrA (Figure 7C, lanes 2 and 3). Moreover, the gel also reveals the presence of K37A-DNA complexes that did not contain UvrB, whereas with wildtype UvrA, in the presence of UvrB only AB-DNA and B-DNA complexes are detected (Figure 7C, lanes 1-3).
This shows that the affinity of UvrA K37A for UvrB is also reduced in the presence of DNA.
(Note that on the 50 bp substrate there is no clear difference in the migration of the UvrA- DNA and UvrAB-DNA complexes. Therefore it could not be established whether also on this shorter substrate K37A forms UvrA-DNA complexes without UvrB.)
Since the incision assays with UvrA K37A showed an inhibition of the activity by additional undamaged DNA, we have determined the effect of undamaged competitor DNA on UvrB-loading by this mutant. The presence of competitor DNA indeed has a much stronger effect on UvrB loading with K37A than with wildtype (Figure 7D, lanes 1-8). This is likely due to the lower affinity of K37A for UvrB, since this will reduce the amount of available AB-complexes that can search for damaged DNA. Apparently, the amount of UvrAB-complexes is sufficient to load UvrB on the damaged site of a small 50 bp DNA substrate. However, when more undamaged DNA is present, either in the same substrate (in cis) or as competitor DNA (in trans), the amount of available UvrAB-complexes becomes limiting and this leads to a reduction in UvrB-loading.
In contrast with UvrA K37A, the UvrB-loading activity of UvrA K646A increases with the length of the DNA substrate, correlating with the incision activity of this mutant: on the 50 bp Menthol-DNA substrate, no UvrB-DNA complexes are visible in the presence of UvrA K646A (Figure 7A, lane 3), but on the longer 96 bp Cholesterol substrate a significant amount of UvrB-DNA complexes can be detected (Figure 7C, lane 4). Remarkably, although UvrA K646A has a greatly reduced affinity for UvrB in solution, only UvrAB- and UvrB-DNA are formed, but no UvrA-complexes without UvrB are detected on the DNA (Figure 7C, lane 4), indicating that, when bound to DNA, the affinity of K646A for UvrB is enhanced.
Figure 7: Loading of UvrB on damaged DNA
UvrB-loading on a 50 bp DNA fragment containing a Menthol lesion (sequence S1) (A) and on a 96 bp DNA fragment containing a Cholesterol lesion (sequence S2) (C).
UvrB-loading on the 50 bp fragment with Menthol was also determined in the presence of different amounts of UvrB (B) and in the presence of undamaged competitor DNA (D).
Unless indicated otherwise, 0.6 nM DNA was incubated with 1.25 nM UvrA and 100 nM UvrB in UV-endo buffer containing 1 mM ATP. In (C) 2 nM UvrA was used. In (D) 0.2 nM DNA was used and supercoiled pUC18 plasmid DNA was added as competitor DNA.
500 ng DNA equals an 8000 times excess (in basepairs) of competitor DNA over the 50 bp DNA substrate.
The more efficient UvrB-loading of K646A on a longer substrate could indicate that the activity of this mutant is stimulated by the presence of undamaged DNA. However, the addition of undamaged DNA in trans did not increase the amount of UvrB-complexes, but instead inhibited binding to the 50 bp substrate (Figure 7D, lanes 9-11). Apparently, the UvrB-loading activity of UvrA K646A is facilitated only when the additional DNA is flanking the lesion.
Recognition of CPD lesions requires ATP hydrolysis
The combined results of the in vitro analysis suggest that both ATPase mutants still have repair activities that are considerably higher than what would be expected from the UV- survival assays. One explanation could be that the activities of both mutants on UV-induced lesions differ from that on bulky lesions, such as the Menthol and Cholesterol used for the in vitro assays. To test this, we have analyzed the ability of UvrA to recognize a UV-induced DNA lesion, the cyclobutane-pyrimidine dimer (CPD), using AFM (example images are shown in Figure 8).
Wildtype UvrA recognizes the CPD lesion less efficiently than the Cholesterol lesion: in the presence of ATP, only 46 % of the UvrA-complexes are bound to the CPD site (Table 5), whereas, in the same conditions, 94 % of the UvrA-complexes bind the Cholesterol lesion (Table 3). From these percentages, the binding specificity of UvrA for both lesions was calculated (as described in [4 and 23]) finding that UvrA has a binding preference of approximately 15,000 for the Cholesterol lesion relative to undamaged DNA. This specificity of UvrA for Cholesterol is about 12 times higher than for the CPD lesion for which a specificity of 1250 was calculated.
In contrast to what is observed on the Cholesterol lesion, recognition of CPD-DNA by wildtype UvrA is stimulated by ATP: in the presence of ATP, 1.7-fold more UvrA-complexes are detected at the lesion site than in the absence of ATP (Table 5). This correlates to UvrA binding the CPD lesion with an approximately three times higher specificity in the presence of ATP (specificity 1250) than in the absence of ATP (specificity 450). Notably, a significant amount of UvrA-complexes are detected at the ends of the 678 bp CPD substrate (Figure 8 and Table 5). This demonstrates that, especially in the absence of ATP, there is only a slight preference for the CPD lesion over a DNA end (note that the DNA substrate contains one DNA lesion and two DNA ends). Recognition of the ends of the CPD substrate is not affected by binding or hydrolysis of ATP (Table 5), which is consistent with previously reported end-binding of UvrA on an undamaged substrate [4].
Figure 8: AFM images of UvrA-DNA complexes
Fragments from representative AFM images of UvrA wildtype (A) and UvrA K37A (B) on a 678 bp DNA fragment with a CPD lesion at the center of the substrate. UvrA was incubated in the presence of ATP. The size of the image fragments is 1 x 1 µm.
Table 5: Site-specific binding on DNA containing a CPD lesion
Cofactor % on Damage % On End % On Non-specific Site
UvrA wildtype
None 27.4 ± 2.3 31.3 ± 2.5 41.3 ± 0.3
ATP 46.3 ± 4.8 28.4 ± 0.6 25.3 ± 5.3
UvrA K37A
None 28.8 ± 2.4 36.2 ± 4.2 35.0 ± 4.8
ATP 26.6 ± 3.3 32.8 ± 6.5 40.6 ± 1.6
(Mutant) UvrA was incubated with a 678 bp DNA fragment containing a CPD lesion in the center and the complexes were visualized by AFM. The percentage of complexes bound at a specific site was calculated as described [4]. The presented values represent the average (± S.D.) of at least two independent experiments.
The importance of ATP for recognition of the CPD lesion also became clear when we tested UvrA K37A. Either in the absence or in the presence of ATP, this mutant has the same reduced specificity for CPD-DNA as UvrA wildtype in the absence of ATP (Table 5).
Unfortunately, due to the reduced specificity of UvrA for the CPD lesion and the aggregation of UvrA K646A on mica, we were not able to obtain a sufficient amount of complexes to determine the specificity of this mutant for CPD-DNA.
The increased specificity of UvrA for a CPD in the presence of ATP indicates that the ATPase activity of UvrA does contribute to the recognition of this type of non-bulky DNA lesion. Possibly the ATPase of UvrA is also involved in recognition of other types of UV- induced lesions, like the 6-4 photoproduct. This might explain why the ATPase mutants show a very low survival after UV irradiation in vivo, while still being able to incise long DNA fragments with a bulky lesion in vitro.
DISCUSSION
Our analysis of the two ATPase mutants suggests that both proteins are disturbed in binding ATP in their mutated ATPase domain, as both proteins have different activities compared to wildtype UvrA in the presence of ATPγS. In the presence of ATPγS both ATPase domains will be occupied by a non-hydrolyzable ATP, thus resembling the ATP- bound form of UvrA. In this form UvrA has reduced damage specificity [4] and a high affinity for UvrB [24]. These properties are not observed for the two ATPase mutants, not even in the presence of ATPγS.
The same mutation, replacement of lysine by alanine in the Walker A motif of the ATPase domain of MalK, an ABC-transporter protein with a single ABC ATPase domain, resulted in MalK not being able to bind ATP [25], indicating that this mutation indeed directly affects ATP binding. We have tried to determine the ATP binding of the wildtype and mutant UvrA proteins using the fluorescent ATP analog TNP-ATP (as described in [26 and 27]).
However, we have failed to detect significant amounts of TNP-ATP bound to either wildtype or mutated UvrA. Likely this is due to UvrA having a lower affinity for ATP than the generic ABC transporters OpuA and Pgp, which were shown to bind ATP with a Kd in the low micromolar range [27, 28].
In an earlier study ATP binding of UvrA wildtype and UvrA K37A was analyzed using equilibrium gel filtration [8]. Here it was shown that K37A is still capable of binding ATP, but approximately half as well as UvrA wildtype. This suggests that K37A only binds ATP in its non-mutated ATPase domain (i.e. ATPase domain II). ATP binding of UvrA K646A could not be analyzed in this study; therefore it is unclear whether this mutant can bind one ATP or none at all. It is however likely that like K37A this mutant binds ATP only in the non-mutated site (i.e.: ATPase domain I).
The apparent inability of both mutant proteins to hydrolyze ATP indicates that to perform its ATPase activity both ATPase domains of UvrA need to be occupied with ATP and that cooperativity exists between both ATPase sites. Cooperativity of the two ATPase sites of E.
coli UvrA has previously been reported by Myles et al. [8]. In contrast to our analysis however, in their study the authors detect residual ATPase activities in both the K37A and the K646A mutant. Also Thiagalingam and Grossman [9] reported ATPase activities of the same mutant proteins, albeit significantly lower than in the study of Myles et al. It is not clear what causes the difference between the ATPase activities of the mutants reported in literature and those measured in our study. The reaction conditions used in literature are very similar to ours using up to 650 µM or 500 µM ATP in both studies respectively.
We have assayed our mutants with ATP concentrations up to 1200 µM but still failed to detect any ATPase activity above background (results not shown). Fresh purifications of our mutant proteins did also not change their behaviors. The discrepancies might be explained by assuming that the mutants described in literature were contaminated with another ATPase.
The interdependent activity of the two ATPase domains suggests that UvrA utilizes a similar hydrolysis mechanism as Pgp, an ABC-transporter protein with two ABC ATPase domains. Like UvrA, deactivation of only one of the two ATPase domains in Pgp completely inhibited ATP hydrolysis [29]. In the ATPase mechanism of Pgp the two ATPase domains alternate in activity and only a single ATP molecule is hydrolyzed at a time [30,31]. Probably such a mechanism is also conserved in UvrA.
The ATPase sites play a role in the stabilization of the UvrA dimer. We have previously shown that ATP hydrolysis results in an ATP/ADP-mixed dimer form which is more stable than the dimer that does not contain a cofactor [4]. Here, we show that a mutation in ATPase domain I directly influences the dimer interface as without any cofactor the dimers of K37A are much more stable than those of the wildtype protein. Since Lys37 is located in an alpha helix that participates in the dimer interactions [13] it is very likely that the link between nucleotide binding state and dimer stability is mediated via this alpha helix. When ATP is present, the dimers of the wildtype protein and of K37A are equally stable, suggesting that either when UvrA contains a mixture of ATP and ADP (as in the wildtype) or when only domain II is occupied with ATP (as in K37A) a similar dimer interface is formed. This could indicate that in the ATP/ADP-mixed dimer of wildtype UvrA, ATPase domain II is occupied with ATP and ATPase domain I with ADP.
What is then the role of ATP binding and hydrolysis in the recognition of DNA lesions by UvrA? In a previous study, we have proposed that to effectively bind DNA damage the conformation of the UvrA dimer has to be such that both monomers can make contact with the DNA flanking the lesion [4]. The UvrA dimer has a conformation proficient for binding bulky lesions, such as Menthol or Cholesterol, either when both ATPase sites are empty or when they are occupied with a mixture of ATP and ADP [4]. The unaffected damage specificity of UvrA K37A and UvrA K646A in the presence of ATP shows that the binding of a single ATP to UvrA does not affect this proposed conformation of the DNA binding domains. However, when both ATPase sites are occupied with ATP, the conformation of the UvrA dimer is not proficient for binding DNA damage, as in the presence of ATPγS wildtype UvrA has a reduced specificity for DNA lesions [4].
Here, we show that recognition of the CPD lesion is enhanced in the presence of ATP.
One explanation for this could be that the conformation of the ATP/ADP-mixed form does differ from the ATP-free UvrA, or from UvrA in which a single ATPase domain is occupied with ATP. This different conformation would then only become apparent when UvrA binds to a non-bulky lesion. An alternative explanation for the enhanced recognition of CPD in the presence of ATP is however that the active ATP hydrolysis of UvrA facilitates binding to this lesion. We have shown that the ATPase activity of UvrA is enhanced in the presence of DNA containing (UV-induced) damage. Since apparently the ATPase is activated after recognition of the damage, it is most likely used for the subsequent step in the repair reaction, the loading of UvrB. We propose that the ATPase of UvrA serves to locally unwind the DNA helix in order to facilitate insertion of the β-hairpin of UvrB between the two DNA strands [32,33].
The local unwinding of DNA however, could in the case of a non-bulky lesion also stabilize the UvrA-DNA complex, as UvrA has a higher affinity for DNA structures that resemble single-stranded DNA [1,2,4]. On a bulky lesion, stabilization of the UvrA complex by ATP hydrolysis is not needed, since this type of DNA lesion which intercalates in the DNA already causes local unwinding by itself [34].
The difference in damage discrimination between a bulky lesion and a CPD might explain why in literature in one report it has been described that the ATPase mutants are proficient in damage discrimination [8], whereas another study showed them to be deficient in damage- specific binding [9]. Myles et al. [8] used DnaseI footprinting on DNA with a bulky psoralen monoadduct to show that the ATPase mutants are proficient in damage discrimination. This is consistent with the behavior of the same mutants on a Cholesterol or Menthol lesion described in this paper.
Thiagalingam and Grossman [9] on the other hand showed with filter-binding assays that the mutants could no longer discriminate between native DNA versus UV-irradiated DNA. This not only confirms the importance of the ATPase domains in specific binding of a CPD lesion as we show here, but in addition it indicates that recognition of other UV-induced lesions, like the 6-4 photoproduct, also requires active ATPase sites.
The ATPase domains of UvrA also play an important role in the interaction between UvrA and UvrB. It has previously been reported that the K37A and K646A mutations affect the delivery of UvrB to the damaged site [8]. Here we show that both ATPase mutants have a reduced affinity for UvrB but remarkably for UvrA K646A the ability to load UvrB onto a damaged site can be rescued by increasing the length of the DNA flanking the damage. It has been shown that upon binding of the UvrA2B2-complex to a lesion a region of about 72 bp 3’
to the lesion is wrapped around one of the two UvrB subunits [35].
On the 678 bp substrate there is enough flanking DNA present to accommodate a complete wrap around UvrB. On the 96 bp substrate the DNA can only partially be wrapped and on the 50 bp fragment wrapping will be even less or does not occur at all. The length- dependent UvrB-loading of UvrA K646A therefore suggests that DNA wrapping stabilizes the UvrAB-DNA complex of this mutant and allows the loading of UvrB to the DNA damage.
With UvrA K37A the UvrAB-DNA complex is not stabilized by the damage-flanking DNA.
Apparently the proposed DNA wrapping that stabilizes the UvrAB-complex of the K646A mutant does not occur with K37A. Possibly, the observed altered mobility of the UvrAB complex in the native gel reflects a different positioning of the UvrB subunit with respect to UvrA, or with respect to the second UvrB subunit, such that DNA wrapping is no longer possible.
The UvrB-binding domain of UvrA has been identified and it was shown that this domain makes an electrostatic interaction with domain 2 of UvrB [13,36,37]. In the crystal structure of UvrA the UvrB-binding domain occupies a solvent-exposed position, suggesting that this domain is easily accessible for UvrB [13]. However, in the crystal structure UvrA is in the ADP-bound form. Since in the absence of ATP UvrA has a much lower affinity for UvrB, it can be expected that in the ATP-free form the UvrB-binding domain of UvrA is less accessible. Our results indicate that, depending on whether ATPase domain I or II of UvrA is occupied with ATP, this domain is exposed differently, resulting in different conformations of the UvrAB-complex. These different conformations might reflect the changes in the interaction between UvrA and UvrB during the loading of UvrB at a damaged site and the subsequent release of UvrA.
ACKNOWLEDGEMENTS
This work was supported by the Netherlands Organization for Scientific Research (NWO) [grant number 700.52.706]. The authors would like to thank Federica Galli and Tjerk Oosterkamp for allowing our participation in the Bio-AFM lab at Leiden University.
REFERENCES
1. Van Houten, B., Gamper, H., Sancar, A. and Hearst, J.E. (1987) DNaseI footprint of ABC excinuclease. J.
Biol. Chem., 262, 13180-13187
2. Mazur, S. and Grossman, L. (1991) Dimerization of Escherichia coli UvrA and its binding to undamaged and ultraviolet damaged DNA. Biochemistry, 30, 4432-4443
3. Wang, H., Tessmer, I., Croteau, D.L,. Erie, D.A. and Van Houten, B. (2008) Functional characterization and atomic force microscopy of a DNA repair protein conjugated to a quantum dot. Nano Lett., 8, 1631- 1637
4. Wagner, K., Moolenaar, G., van Noort, J. and Goosen, N. (2009) Single-molecule analysis reveals two separate DNA binding domains in the Escherichia coli UvrA dimer. Nucleic Acids Res., 37, 1962-1972.
5. Van Houten, B., Croteau, D.L., DellaVecchia, M., Wang, H. and Kisker C. (2005) ‘Close fitting sleeves’:
DNA damage recognition by the UvrABC nuclease system. Mutat. Res., 577, 92-117.
6. Truglio, J.J., Croteau, D.L., Van Houten, B. and Kisker, C. (2006) Prokaryotic nucleotide excision repair:
the UvrABC system. Chem. Rev., 106, 233-252.
7. Seeley, T.M. and Grossman, L. (1989) Mutations in the Escherichia coli UvrB ATPase motif compromise excision repair capacity. Proc. Natl. Acad. Sci. U.S.A., 86, 6577-6581
8. Myles, G.M., Hearst, J.E. and Sancar, A. (1991) Site-specific mutagenesis of conserved residues within Walker A and B sequences of Escherichia coli UvrA protein. Biochemistry, 30, 3824-3834
9. Thiagalingam, S. and Grossman, L. (1991) Both ATPase sites of Escherichia coli UvrA have functional roles in nucleotide excision repair. J. Biol. Chem., 266, 11395-11401
10. Gorbalenya, A.E. and Koonin, E.V. (1990) Superfamily of UvrA-related NTP-binding proteins.
Implications for rational classification of recombination/repair systems. J. Mol. Biol., 213, 583-591
11. Moussatova, A., Kandt, C., O’Mara, M.L. and Tieleman, D.P. (2008) ATP-binding cassette transporters in Escherichia coli. Biochim. Biophyis. Acta, 1778, 1757-1771
12. Seeger, M.A. and van Veen, H.W. (2009) Molecular basis of multidrug transport by ABC transporters.
Biochim. Biophyis. Acta, 1794, 725-737
13. Pakotiprapha, D., Inuzuka, Y., Bowman, B.R., Moolenaar, G.F., Goosen, N., Jeruzalmi, D. and Verdine, G.L. (2008) Crystal structure of Bacillus stearothermophilus UvrA provides insight into ATP-modulated dimerization, UvrB interaction, and DNA binding. Mol. Cell., 29, 122-133
14. Timmins, J., Gordon, E., Caria, S., Leonard, G., Acajjaoui, S., Kuo, M.S., Monchois, V. and McSweeney, S. (2009) Structural and mutational analysis of Deinococcus radiodurans UvrA2 provide insight into DNA binding and damage recognition by UvrAs. Structure, 17, 547-558
15. Croteau, D.L., DellaVecchia, M.J., Wang, H., Bienstock, R.J., Mellon, M.A. and Van Houten, B. (2006) The C-terminal zinc-finger of UvrA does not bind DNA directly but regulates damage-specific DNA binding. J. Biol. Chem., 281, 26370-26381
16. Croteau, D.L., DellaVecchia, M.J., Perera, L. and Van Houten, B. (2008) Cooperative damage recognition by UvrA and UvrB: Identification of UvrA residues that mediate DNA binding. DNA Repair, 7, 392-404
17. Moolenaar, G.F., Moorman, C. and Goosen, N. (2000) Role of the Escherichia coli nucleotide excision repair proteins in DNA Replication, J. Bacteriol., 182, 5706-5714
18. Backendorf, C., Brandsma, J.A., Kartasova, T. and van de Putte, P. (1983) In vivo regulation of the UvrA gene: role of the “-10” and “-35” promoter regions. Nucleic Acids Res., 11, 5795-5810
19. 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
20. Paspaleva, K., Thomassen, E., Pannu, N.S., Iwai, S., Moolenaar, G.F., Goosen, N. and Abrahams, J.P.
(2007) Crystal structure of the DNA repair enzyme ultraviolet damage endonuclease. Structure, 15, 1316- 1324
21. Verhoeven, E.E.A., 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
22. Moolenaar, G.F., Visse, R., Ortiz-Buysse, M., Goosen N. and van de Putte, P. (1994) Helicase motifs V and VI of the Escherichia coli UvrB protein of the UvrABC endonuclease are essential for the formation of the preincision complex. J. Mol. Biol., 240, 294-307
23. Yang, Y., Sass, L.E., Du, C., Hsieh, P. and Erie, D.A. (2005) Determination of protein-DNA binding constants and specificities from statistical analyses of single molecules: MutS-DNA interactions. Nucleic Acids Res., 33, 4322-4334
24. 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
25. Panagiotidis, C.H., Reyes, M., Sievertsen, A., Boos, W. and Shuman, H.A. (1993) Characterization of the structural requirements for assembly an nucleotide binding of an ATP-binding cassette transporter. J.
Biol. Chem., 268, 23685-23696
26. Hiratsuka, T. (1976) Fluorescence properties of 2’ (or 3’)-O-(2,4,6-trinitrophenyl) adenosine 5’- triphosphate and its use in the study of binding to heavy meromyosin ATPase. Biochim. Biophys. Acta, 453, 293-297
27. Horn, C., Bremer, E. and Schmitt, L. (2003) Nucleotide dependent monomer/dimer equilibrium of OpuAA, the nucleotide-binding protein of the osmotically regulated ABC transportur OpuA from Bacillus subtilis. J. Mol. Biol., 334, 403-419
28. Kim, I.W., Peng, X.H., Sauna, Z.E., FitzGerald, P.C., Xia, D., Müller, M., Nandigama, K. and Ambudkar, S.V. (2006) The conserved tyrosine residues 401 and 1044 in ATP sites of human P-glycoprotein are critical for ATP binding and hydrolysis: evidence for a conserved subdomain, the A-loop in the ATP- binding cassette. Biochemistry, 45, 7605-7616
29. Urbatsch, I.L., Beaudet, L., Carrier, I. and Gros, P. (1998) Mutations in either nucleotide-binding site of P-glycoprotein (Mdr3) prevent vanadate trapping of nucleotide at both sites. Biochemistry, 37, 4592-4602
30. Urbatsch, I.L., Sankaran, B., Bhagat, S. and Senior, A.E. (1995) Both P-glycoprotein nucleotide-binding sites are catalytically active. J. Biol. Chem., 270, 26956-26961
31. Hrycyna, C.A., Ramachandra, M., Ambudkar, S.V., Ko, .H., Pedersen, P.L., Pastan, I. and Gottesman, M.
(1998) Mechanism of action of human P-glycoprotein ATPase activity. Photochemical cleavage during a catalytic transition state using orthovanadate reveals cross-talk between two ATPase sites. J. Biol. Chem., 273, 16631-16634
32. Zou, Y. and Van Houten, B. (1999) Strand opening by the UvrA2B complex allows dynamic recognition
33. Truglio, J.J., Karakas, E., Rhau, B., Wang, H., DellaVecchia, M.J., Van Houten, B. and Kisker C. (2006) Structural basis for DNA damage recognition and processing by UvrB. Nat. Struct. Mol. Biol., 13, 360- 364
34. Gómez-Pinto, I., Cubero, E., Kalko, S.G., Monaco, V., van der Marel, G., van Boom, J.H., Orozco, M.
and Gonzalez, C. (2004) Effect of bulky lesions on DNA. J. Biol. Chem., 279, 24552-24560
35. Verhoeven, E.E.A., Wyman, C., Moolenaar, G.F., Hoeijmakers, J.H.J. and Goosen, N. (2001) Architecture of nucleotide excision repair complexes: DNA is wrapped by UvrB before and after damage recognition. EMBO J., 20, 601-611
36. Truglio, J.J, Croteau, D.L., Skorvaga, M., DellaVecchia, M.J., Theis, K., Mandavili, B.S., Van Houten, B.
and Kisker, C. (2004) Interactions between UvrA and UvrB: the role of UvrB’s domain 2 in nucleotide excision repair. EMBO J., 23, 2498-2509
37. Pakotiprapha, D., Liu, Y., Verdine, G.L. and Jeruzalmi, D. (2009) A structural model for the damage sensing complex in bacterial nucleotide excision repair. J. Biol. Chem., 284, 12837-12844
