HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 281, Issue 22, 15227-15237, June 2, 2006

This Article Abstract Full Text (PDF) All Versions of this Article: 281/22/15227    most recent M601476200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wang, H. Articles by Van Houten, B. Search for Related Content PubMed PubMed Citation Articles by Wang, H. Articles by Van Houten, B. Social Bookmarking                      What's this?

UvrB Domain 4, an Autoinhibitory Gate for Regulation of DNA Binding and ATPase Activity^*

Hong Wang¹, Matthew J. DellaVecchia¹, Milan Skorvaga, Deborah L. Croteau, Dorothy A. Erie^¶, and Bennett Van Houten²

From the Laboratory of Molecular Genetics, NIEHS, National Institutes of Health, Department of Health and Human Services, Research Triangle Park, North Carolina 27709, the Department of Molecular Genetics, Cancer Research Institute, Slovak Academy of Sciences, 83391 Bratislava, Slovakia, and the ^¶Department of Chemistry, University of North Carolina, Chapel Hill, North Carolina 27599

Received for publication, February 15, 2006 , and in revised form, March 28, 2006.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
UvrB, a central DNA damage recognition protein in bacterial nucleotide excision repair, has weak affinity for DNA, and its ATPase activity is activated by UvrA and damaged DNA. Regulation of DNA binding and ATP hydrolysis by UvrB is poorly understood. Using atomic force microscopy and biochemical assays, we found that truncation of domain 4 of Bacillus caldotenax UvrB (UvrB4) leads to multiple changes in protein function. Protein dimerization decreases with an 8-fold increase of the equilibrium dissociation constant and an increase in DNA binding. Loss of domain 4 causes the DNA binding mode of UvrB to change from dimer to monomer, and affinity increases with the apparent dissociation constants on nondamaged and damaged single-stranded DNA decreasing 22- and 14-fold, respectively. ATPase activity by UvrB4 increases 14- and 9-fold with and without single-stranded DNA, respectively, and UvrB4 supports UvrA-independent damage-specific incision by Cho on a bubble DNA substrate. We propose that other than its previously discovered role in regulating protein-protein interactions, domain 4 is an autoinhibitory domain regulating the DNA binding and ATPase activities of UvrB.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Nucleotide excision repair (NER)³ is a DNA repair pathway conserved from bacteria to eukaryotes. DNA damage recognition and incision during NER in prokaryotes is a complex process involving UvrA, UvrB, and UvrC (1–3). UvrA and UvrB interact in solution, forming a UvrAB complex (4). It is believed that UvrA, as a dimer within the UvrAB complex, first recognizes helical distortions induced by DNA damage. Upon binding of UvrAB (either as UvrA₂B or UvrA₂B₂) to the site of DNA damage (4, 5), conformational changes in the UvrAB-DNA complex lead to transfer of DNA from UvrA to UvrB and dissociation of UvrA (6–10). After dissociation of UvrA from the protein-DNA complex, a very stable UvrB-DNA preincision complex is formed. UvrC recruitment by the UvrB-DNA precincision complex to the site of DNA damage leads to dual incisions on the damaged DNA strand (11–13).

UvrB plays a central role in bacterial NER. In the absence of UvrA, UvrB binds weakly to DNA and has no ATPase activity (2, 14). In the presence of DNA damage, through interaction with UvrA both DNA binding and ATPase activities of UvrB are activated. This allosteric control of UvrB activation is not well understood. UvrB is classified as a member of the helicase superfamily II (15). Crystal structures of the proteins from different organisms (16–18) and most recently the Y96A variant from Bacillus caldotenax (19) have shed light on structural components of UvrB. UvrB consists of five domains, namely 1a, 1b, 2, 3, and 4 (16). Even though UvrB contains all the elements of an intact helicase, including all residues implicated in coupling ATP hydrolysis to strand translocation, it is not a true helicase (2, 3, 20). Rather, the limited DNA unwinding activity of UvrAB complex is believed to be important in destabilizing duplex DNA and creating an "entry site" for UvrB to form a stable UvrB-DNA preincision complex (2, 20, 21). A unique structural element in UvrB is a flexible -hairpin that extrudes from domain 1a, near the wall of domain 1b. Based on superposition of the UvrB structure with known structures of UvrB-related helicase-DNA complexes, a padlock model for the UvrB-DNA preincision complex was proposed (16, 22). In this model, the -hairpin bifurcates the duplex DNA, and one of the single strands is locked between the -hairpin and domain 1b. This model was recently confirmed by a UvrB-DNA co-crystal structure (49). The C-terminal domain 4 of UvrB is not visible in any of the available full-length UvrB crystal structures. However, crystallography and NMR spectroscopy studies of the C-terminal fragment of E. coli UvrB have revealed that the last 46 amino acids of domain 4 adapt a coiled-coil structure and dimerize through specific hydrophobic and ionic interactions (23, 24). Previously, it has been demonstrated that domain 4 is involved in interacting with UvrC and promoting dimer formation of UvrB (5, 13, 25).

In this study, we provide quantitative measurement of the effect of domain 4 on UvrB self-association, and demonstrate that domain 4 functions as an autoinhibitory domain. Autoinhibitory domains are regions that negatively regulate the activity of a second, separable domain and can provide tight "on site" repression that restrains the targeted domain in a secure off state (26). Autoinhibition is a general method for achieving a high level of specificity and tight regulation in DNA binding site selection (26). Specifically, we show that truncation of domain 4 (residues Pro⁶¹² to Gly⁶⁵⁸) of UvrB leads to: 1) a change in UvrB DNA binding mode and increase in DNA binding affinity; 2) an increase in UvrB ATPase activity; and 3) UvrA-independent, damage-specific incision of a bubble substrate mediated by Cho (UvrC homolog). These data suggest that DNA repair enzymes may represent another important class of DNA-binding proteins that are regulated by autoinhibitory domains.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cloning, Expression, and Purification of B. caldotenax UvrA, UvrB, and Escherichia coli Cho Proteins—Genes encoding B. caldotenax UvrA, UvrB, and E. coli Cho were subcloned in the pTYB1 vector of the T7 IMPACT system (New England Biolabs). The domain 4 truncation mutation of B. caldotenax UvrB (UvrB4, sequence between Pro⁶¹² to Gly⁶⁵⁸ deleted) was constructed following the protocol of the QuikChange site-directed mutagenesis kit (Stratagene). The presence of the desired deletion and absence of additional mutations in the entire amplified uvrB gene were confirmed by DNA sequencing. Domain 2 deletion of B. caldotenax UvrB (UvrB2, sequence between Leu¹⁵⁷ and Pro²⁴⁵ deleted) was constructed as described previously (19). UvrA and UvrB proteins were expressed in Rosettagami(DE3)pLacI (Merck Biosciences) and BL21-CodonPlus®(DE3)-RIL (Stratagene) cells, respectively. Mutant UvrB proteins were expressed following the same procedures used for wild type (WT). All proteins were purified using the IMPACT^TM-CN system (New England Biolabs) as described previously (16). For purification of Cho, pooled material after elution from a chitin-binding column was purified further over a Superdex 200 HR 10/30 column (GE Healthcare) equilibrated with column buffer (20 mM Tris-HCl, pH 8.0, 500 mM NaCl, 10 mM MgCl₂). Proteins used in this study are more than 95% pure as judged by staining of SDS-PAGE protein gel with SimplyBlue^TM SafeStain (Invitrogen).

DNA Substrates—DNA substrates were synthesized by Sigma Genosys and 5' labeled using OptiKinase (U. S. Biochemical Corp.) and [-³²P]ATP (10 µCi/µl, GE Healthcare) or 3' labeled using terminal transferase (Roche Applied Science) and [-³²P]ATP (10 µCi/µl, GE Healthcare) according to the manufacturers' instructions. Reactions were terminated with 20 mM EDTA and heat denaturation (65 °C, 10 min). Free nucleotides were subsequently removed by gel filtration chromatography (Micro Biospin-6, Bio-Rad). Fully duplexed DNA (F50/NDB50) and bubble DNA (F50/B8) were constructed by annealing labeled F50 (4 pmol) with equal molar amounts of their respective "bottom" strands. Double-stranded character and homogeneity of the annealed DNA were examined on a 10% native polyacrylamide gel. The 5' end-labeled duplex DNA containing the UV-activated arylazido cross-linker moiety, XL2₂₅, was prepared as described previously (10). The DNA substrate (HS1F-M13mp19) used for the strand-destabilizing assays was labeled and constructed as described previously (27).

Atomic Force Microscopy—Proteins were diluted in buffer (50 mM Tris-HCl, pH 7.5, 100 mM KCl, 10 mM MgCl₂) and heated (65 °C, 10 min). Afterward, samples were deposited onto freshly cleaved ruby mica (Spruce Pine Mica Company) pre-warmed to 55 °C on a hot plate. Sample preparation and image collection using Nanoscope III (Veeco Instruments) operated in tapping mode in air were done as described previously (28). The atomic force microscopy (AFM)-derived volume for each protein was obtained as described previously (29). The standard equation employed for converting AFM volume (nm³) to molecular mass (M_r) is given by the following equation (30).

(Eq. 1)

Electrophoresis Mobility Shift Assay (EMSA)—Proteins and ssDNA substrates were preheated (65 °C, 10 min) prior to mixing. For single-stranded DNA (ssDNA), UvrB (WT or 4) was incubated with 1 nM DNA substrate (5'-³²P-labeled) in 20 µl of reaction buffer (50 mM Tris-HCl (pH 7.5), 50 mM KCl, 5 mM dithiothreitol) at 55 °C for 30 min. Half of the reaction was removed immediately after incubation and loaded onto an 8% native polyacrylamide gel (29:1). Gels were run (100 V, 4 h, 4 °C) in 1x TBE buffer (89 mM Tris, 89 mM boric acid, 2.5 mM EDTA). For bubble DNA, the procedures are the same except that the reactions were done in 1x ABC buffer (50 mM Tris-HCl, pH 7.5, 50 mM KCl, 10 mM MgCl₂, 1 mM ATP, and 5 mM dithiothreitol) containing 7 µg/ml bovine serum albumin. Reactions were loaded onto a 6 or 8% native polyacrylamide gel (29:1) containing 0.5x TBE, 10 mM MgCl₂, and 1 mM ATP. Gels were run (100 V, 2 h, 4 °C) in buffer containing 0.5x TBE, 10 mM MgCl₂, and 1 mM ATP. All gels were dried and exposed to a PhosphorImager screen (GE Healthcare) overnight. The B1- and B2-DNA complexes were resolved as distinct peaks and quantified using ImageQuant software (GE Healthcare). The concentration of protein dimer and monomer in each reaction was determined as described in Table 2. The apparent dissociation constants of B2-DNA for WT UvrB and B1-DNA complex for UvrB4 were determined by nonlinear regression analysis (Table 2) (31).

View this table: [in this window] [in a new window]   TABLE 2 Apparent dissociation constants of WT UvrB- and UvrB4-ssDNA complexes The concentration of WT UvrB dimer and UvrB4 monomer in each reaction was determined using the following equations: ; whereas [B1], [B2], [Ptot] are the monomer, dimer, total protein concentration, respectively; Kd is the equilibrium dissociation constant of protein dimer obtained from AFM images (Table I). Data from EMSA were analyzed as percentage of WT B2-DNA and 4 B1-DNA complexes versus calculated WT dimer and 4 monomer concentration, respectively, and were fitted by nonlinear regression analysis to the equation: ; whereas P is the protein concentration (calculated dimer for WT and monomer for 4); D is the total DNA concentration; Ka = 1/Kd(app); Kd(app) is the apparent dissociation constant (Schofield et al. (31)). The mean ± S.D was derived from three independent experiments.

ATP/GTP Hydrolysis Assay—Hydrolysis of ATP or GTP by UvrAB was monitored using a coupled enzyme assay system (32) as previously described with minor modifications (19). Briefly, ATP or GTP (Roche) was added to a final concentration of 1 mM in 100 µl containing: 50 mM Tris-HCl (pH 7.5), 55 mM KCl, 4 mM MgCl₂, 1 mM dithiothreitol, 20 units/ml L-lactic dehydrogenase (Sigma), 20 units/ml pyruvate kinase (Sigma), 2 mM phosphoenolpyruvate (Roche), 0.15 mM NADH (Roche), 50 nM UvrA, and 100 nM UvrB (WT or 4). Proteins were assayed without DNA (–DNA), or in the presence of 10 ng/µl UV-irradiated DNA (+UV DNA, prepared by exposure of 1 µg/µl pUC18 plasmid DNA to 200 J/m² for 1 min) or nondamaged 50-mer ssDNA (+ssDNA, complementary to NDT50). Rate of hydrolysis was calculated from the linear change in absorbance at 340 nm at 55 °C over 30 min, using a Beckman spectrophotometer. Data are reported as the mean rate (M/min) ± S.D. (n = 3).

Strand-destabilizing Assay—This assay was done as described previously and the incubations were done at 42 °C (27).

Incision Assays—The 3'-end labeled DNA (2 nM), F50/NDB50 or F50/B8, was incubated with 100 nM UvrB (WT or 4) with or without UvrA (20 nM), in 1x ABC buffer (18 µl), at 55 °C for 15 min. After reactions cooled to room temperature, E. coli Cho (2 µl, 0.5 µM) was added (37 °C, 30 min). Reactions were terminated with 200 mM EDTA (2 µl), heated to 85 °C (10 min), and cooled on ice. Incision products were resolved on an 18% denaturing polyacrylamide gel (375 V, 50 min, 1x TBE buffer). Gels were processed and developed. Incision products were quantified as described above.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
As part of our systematic analysis of the different structural domains of UvrB (10, 19, 27), we chose to further examine domain 4. We constructed a UvrB mutant gene product that has its last 47 amino acid residues (Pro⁶¹² to Gly⁶⁵⁸) deleted, UvrB4 (Fig. 1). The last 46 amino acids in E. coli UvrB have been shown to form a coiled-coil structure (23, 24). We examined the effects of domain 4 truncation on: 1) UvrB self-association; 2) UvrB single-stranded, bubble and duplex DNA binding; 3) UvrB ATPase activity; 4) strand-destabilizing activity of a UvrAB complex; and 5) ability of UvrB to support UvrA-independent incision mediated by Cho.

View larger version (13K): [in this window] [in a new window]   FIGURE 1. Sequence alignment of C-terminal E. coli and B. caldotenax UvrB. Proteolyzed E. coli UvrB from cell extracts yielded a 70-kDa fragment (UvrB*) (39) and mass spectrometric analysis suggested two possibilities for the UvrB* cleavage site(s), between Lys607 and Ala608 and between Lys609 and Gly610 (13). B. caldotenax UvrB4 has amino acids residues from Pro612 to Gly658 deleted.

To obtain the equilibrium dissociation constants of WT UvrB and UvrB4 dimers, we took the unique approach of utilizing the protein size information obtained from AFM images. Using topographical AFM images, AFM-derived volumes of proteins have been correlated to their molecular masses, permitting determination of oligomeric states and equilibrium dissociation constants of proteins (29, 30, 33). AFM offers the advantage of assaying proteins with low solubility or with tight binding constants. In addition, sample deposition for AFM imaging can be done over a wide range of temperatures, which is advantageous for studying thermophilic proteins.

Using AFM volume analysis, we found that B. caldotenax UvrA dimer has an equilibrium dissociation constant (K_d) of 3 nM (Table 1). These results are consistent with previously published data (K_d at 10 nM for E. coli UvrA) (34). Next, we analyzed WT UvrB. A representative field view AFM image of WT UvrB (20 nM) is shown in Fig. 2A. The inset is a surface plot of the boxed region (pink). The surface plot of WT UvrB clearly shows two distinct populations: proteins with a larger volume (Fig. 2A, green arrows) and those with a smaller volume (white arrows).

The histogram of AFM-derived volumes for UvrB4(20 nM) also displays two distinct peaks similar to WT UvrB, but located at 89 ± 21 and 175 ± 33 nm³ (Fig. 2C). These volumes correspond to protein molecular masses of 86 and 158 kDa, respectively, and are within error of the expected protein size for UvrB4 monomer (65 kDa) and dimer (130 kDa). Compared with the number of protein dimers observed for WT UvrB (217 of 390, 56%), the number of UvrB4 protein dimers observed is almost 50% less (126 of 390, 32%). The K_d for UvrB4 dimer is estimated to be 38 nM (Table 1), 8-fold higher than WT UvrB.

As a negative control, we imaged the domain 2 deletion mutant of UvrB, UvrB2. Domain 2 of UvrB is believed to interact with UvrA and is not known to play any role in UvrB self-association (19). The histogram of AFM-derived volumes of UvrB2 (20 nM) also displays two distinct peaks (located at 50 ± 20 and 120 ± 30 nm³, Fig. 2D), which correspond to protein molecular masses of 54 and 112 kDa, respectively, and are close to that expected for the UvrB2 monomer (60 kDa) and dimer (120 kDa). The K_d for UvrB2 dimer is estimated to be 4 nM (Table 1). Comparing the K_d values of WT, 2, and 4 UvrB dimers (Table 1) indicates that loss of C-terminal domain 4 significantly alters UvrB monomer-dimer equilibrium, whereas deletion of domain 2 does not have a detectable effect.

DNA Binding Affinity of WT UvrB and UvrB4 Measured by EMSA—Because UvrB has a higher affinity for ssDNA than for duplex DNA (14), we first examined ssDNA binding using EMSA to quantitatively compare the DNA binding affinity of WT UvrB and UvrB4. The nondamaged ssDNA substrate used for this assay was a 50-mer, NDT50 (Fig. 3A). The damaged ssDNA substrate, F50, has a fluorescein adduct attached to a centrally located thymine (Fig. 3A). Representative EMSA gels are shown in Fig. 4A. We observed that UvrB4 has greater affinity for ssDNA than WT UvrB. For example, at 1.5 µM WT UvrB, only 48% of F50 is bound (Fig. 4A, lane 3). Conversely, at the same protein concentration, close to 90% of F50 is bound by UvrB4 (Fig. 4A, lane 11).

Notably, two populations of UvrB-DNA complexes are observed via EMSA (faster and slower migrating complexes are labeled B1-DNA and B2-DNA, respectively, in Fig. 4A). These two bands are observed for both WT UvrB and UvrB4. The percentages of B1- and B2-DNA complexes versus total protein concentrations are plotted in Fig. 4A (bottom panel). For WT, at higher protein concentrations (>1 µM), the majority of the protein-DNA complexes are B2-DNA (Fig. 4A). For UvrB4, the majority of the complexes are B1-DNA. Given our observation of the effect of domain 4 on UvrB dimerization (Fig. 2), we interpret the B1- and B2-DNA complexes as protein monomerand dimer-DNA complexes, respectively. Because the percentages of WT B1-DNA or 4 B2-DNA are low (<25%), we assume WT UvrB binds DNA primarily as a dimer, whereas UvrB4 binds DNA primarily as a monomer, and calculate the apparent dissociation constants (K_d(app)) of each complex respectively. Furthermore, in our EMSA, the proteins were in large excess compared with DNA substrates, consequently DNA was not expected to significantly affect the monomer-dimer equilibrium of proteins. Based on this assumption, the concentration of WT dimer and 4 monomer present in each reaction was calculated using the equilibrium dissociation constants of protein dimers derived from AFM images (Table 1). To derive K_d(app), calculated protein dimer and monomer concentrations were plotted against the percentage of WT B2-DNA and 4 B1-DNA complexes, respectively (Fig. 4, B and C). Based on this analysis, WT UvrB dimer has a K_d(app) of 1300 nM on F50 and 2000 nM on NDT50 (Table 2), indicating that WT UvrB has a low specificity for damaged DNA. This is consistent with previous observations that UvrB has weak specificity for some DNA adducts on ssDNA (14, 35). UvrB4 monomer has a K_d(app) of 90 nM for F50 and NDT50 (Table 2). These binding constants are 14- and 22-fold tighter than those for WT dimer binding to damaged and nondamaged ssDNA, respectively. These data demonstrate that UvrB4asa monomer has increased affinity for damaged and nondamaged ssDNA than WT UvrB as a dimer. In contrast to the UvrB4 mutant, UvrB2 exhibited a lower affinity for ssDNA than WT UvrB (data not shown), indicating that the increased affinity for ssDNA is specifically a result of truncation of domain 4.

View larger version (51K): [in this window] [in a new window]   FIGURE 2. AFM image of UvrB and AFM volume analysis of WT UvrB, UvrB4, and UvrB2. A, a representative field view image of WT UvrB at 20 nM concentration. The actual image size is 1 x 1 µm. The color bar represents 0–2 nm height above the mica surface. The inset is a surface plot of the boxed region (pink). White and green arrows denote UvrB monomers and dimers, respectively. B–D, AFM volume analysis of WT UvrB, UvrB4, and UvrB2. The volume of each protein observed from the AFM image was obtained as described under "Materials and Methods." The total data points for each histogram are 390. The solid lines are the double Gaussian fits to the data. The R value for each curve fit is greater than 0.93. The predicted molecular mass of proteins from the AFM-derived volume was based on Equation 1 under "Materials and Methods."

View larger version (32K): [in this window] [in a new window]   FIGURE 3. DNA substrates used in this study. A, the DNA substrates used for EMSAs and incision assays. B, DNA substrate, XL225 used for photoaffinity cross-linking studies. XL in the gray oval denotes a photoreactive arylazido group covalently attached to the N4 position of cytosine (10). There is a DNA nick 3' to the cross-linker. C, DNA substrate used for strand-destabilizing assay.

View larger version (45K): [in this window] [in a new window]   FIGURE 4. DNA binding affinity of WT UvrB and UvrB4 for ssDNA. A, top panel, EMSAs comparing the binding of WT UvrB and UvrB4 to F50 ssDNA (1 nM). The reactions were done in buffer without ATP; bottom panel, quantification of WT and 4 B1- (black squares) and B2-DNA (white squares) complexes on F50 and NDT50 substrates. B and C, calculation of apparent dissociation constants of WT B2-DNA and 4 B1-DNA complexes. The derivation of protein dimer and monomer concentrations is described in Table 2. The data plotted were the results from three independent EMSA experiments. The lines in the plots are the curve fitting by using nonlinear regression analysis for WT B2-DNA and 4 B1-DNA complexes (31). For the curve fitting, R2 values are 0.96 and 0.87 for WT and 4 data, respectively. The sigmoidal shape of the 4 data is because of the loss of the DNA counts in the smear (not counted as protein-DNA complex) at lower protein concentrations.

The Effect of Nucleotide Cofactors on dsDNA Binding of WT UvrB and UvrB4—It is known that the affinity of WT UvrB for fully duplexed dsDNA is very low, therefore we used photoaffinity cross-linking to capture transient interactions between UvrB and dsDNA and study the effect of nucleotide cofactors, namely ATP, GTP, and AMPPNP, on the DNA binding affinity of WT UvrB and UvrB4. Our laboratory previously demonstrated that DNA containing arylazido nucleotide analogs can be used to capture UvrA- and UvrB-DNA complexes by UV-induced cross-linking (10). In this study, the DNA substrate used for cross-linking is XL2₂₅ (Fig. 3B). A photoreactive cytosine analog with a terminal arylazido group at the N4 position was incorporated into this substrate, and serves as both the site of DNA damage and the cross-linking reagent (10). In these experiments, UvrA (200 nM) was used as a positive control to monitor the efficiency of the cross-linking (Fig. 6, lane 1). Representative gels of the samples after cross-linking are shown in Fig. 6A, and the extent of cross-linking is summarized in Fig. 6B. Results from photoaffinity cross-linking show that in all cases (–NTP, +ATP, +GTP, or +AMPPNP), the amount of UvrB4 cross-linked to XL2₂₅ DNA is 2–3-fold higher than that of WT UvrB (Fig. 6, lanes 3 versus 2, 5 versus 4, 7 versus 6, and 9 versus 8).

View larger version (28K): [in this window] [in a new window]   FIGURE 5. DNA binding affinity of WT UvrB and UvrB4 for bubble DNA. A, electrophoresis mobility shift assay comparing the binding of WT UvrB and UvrB4 to F50/B8 (1 nM). The reactions were done in buffer containing 1 mM ATP. B, quantification of binding of WT- and 4-DNA complexes to F50/B8. The data plotted were the results from three independent EMSA experiments.

View larger version (51K): [in this window] [in a new window]   FIGURE 6. DNA binding of WT UvrB or UvrB4in the presence of different nucleotide cofactors. A, DNA substrate XL225 (2 nM) was incubated with the indicated protein for 30 min at 55 °C in the dark. Samples were exposed to 5 min of UV (365 nm) and processed as described under "Materials and Methods." B, percentage of DNA cross-linked to each protein (n = 3, mean ± S.D.): UvrA (hatched bar), WT UvrB (white bars), and UvrB4 (black bars).

ATPase Activities of WT UvrB and UvrB4—It has been shown previously that proteolyzed E. coli UvrB from cell extracts, UvrB* (Fig. 1), possesses a ssDNA-dependent ATPase activity (39). However, there is no direct comparison between the ATPase activity of WT UvrB and UvrB* in the absence of DNA. To address this question, we used an ATPase assay that is based on a coupled enzyme system to link the hydrolysis of ATP to the oxidation of NADH (32). We found that in the absence of DNA, truncation of domain 4 increases the basal level of ATPase activity of UvrB 9-fold (Fig. 7, column 4 versus 7, 2.3 x 10^–7 M/min to 2.1 x 10^–6 M/min). We observed that nondamaged ssDNA increased ATPase activity of UvrB4 9-fold compared with the basal level (Fig. 7, column 9 versus 7), and only increased ATPase activity of WT UvrB 5-fold (Fig. 7, column 6 versus 4), which is consistent with previous results (39). Overall, in the presence of 50-mer ssDNA, UvrB4 has 14-fold higher ATPase activity than WT UvrB (Fig. 7, column 9 versus 6). UV-irradiated plasmid DNA (in the absence of UvrA) had no significant effect on the ATPase activity of either WT UvrB or UvrB4 (Fig. 7, for WT, column 4 versus 5; for 4, column 7 versus 8), indicating that interaction with ssDNA is essential to activate the ATPase of UvrB.

View larger version (44K): [in this window] [in a new window]   FIGURE 7. The ATPase and GTPase activity of WT UvrB and UvrB4. Both WT UvrB and UvrB4 were assayed for their ATPase (A) and GTPase (B) activities individually and in the presence of UvrA at 55 °C. The DNA substrate in the reaction was also varied: –DNA (white bars), +UV irradiated DNA plasmid (hatched bars), and +nondamaged ssDNA (black bars). The columns labeled Expected and Observed are the values of the sum of the ATPase activities of individual proteins and the values measured when mixing two proteins together, respectively. n = 3, mean ± S.D.

The following question remains: is the stimulated ATPase activity of UvrAB and UvrAB4 complex in the presence of UV-irradiated plasmid due to further activation of the ATPase activity of UvrA or unmasking the ATPase activity of UvrB/UvrB4? To answer this question, a GTPase assay was performed to distinguish the activity contributed by each protein in the UvrAB complex. This experiment is based on the knowledge that UvrA can hydrolyze both ATP and GTP, but UvrB can only hydrolyze ATP (40). A summary of the GTPase assay results is shown in Fig. 7B. The presence of UV-irradiated plasmid or nondamaged ssDNA stimulated the GTPase activity of UvrA (Fig. 7, columns 23 and 24 versus 22). However, addition of either WT UvrB or UvrB4 does not stimulate the GTPase activity of UvrA further (Fig. 7, columns 32 versus 38, 33 versus 39, 35 versus 41, and 36 versus 42). These results support the notion that when UvrA and UvrB are combined together in the presence of damaged DNA, the stimulated ATPase activity above the sum of the individual protein activity is because of the DNA damage-dependent unlocking of the ATPase activity of UvrB by UvrA. Taken together, these data (Fig. 7, A and B) demonstrated that truncation of domain 4 increases: 1) the basal ATPase level; 2) ssDNA-dependent ATPase activity; and 3) UvrA- and DNA damage-dependent ATPase activity of UvrB.

Strand-destabilizing Activity of WT UvrAB and UvrAB4—The limited strand opening activity by the UvrAB complex is believed to be important in opening up the hydrophobic interior of DNA to allow appropriate positioning of UvrB for damage recognition (21, 41). This strand-destabilizing activity can be monitored by an assay that measures the release of radioactively labeled oligonucleotide (<27-mer) annealed to circular ssDNA (41, 42). It is worth noting that the UvrAB complex only destabilizes duplex DNA, and oligonucleotide is not released from the circular ssDNA until addition of a stop buffer containing SDS and EDTA. An early report stated that E. coli UvrB* was completely defective in the strand-destabilizing assay (43). However, more recently it was reported that UvrAB* complex could support incision mediated by Cho (44). Consequently, we decided to reevaluate whether or not the truncation of domain 4 impairs the strand-destabilizing activity. The DNA substrate used for this assay is shown in Fig. 3C. The results from this assay are shown in Fig. 8A with kinetics of the 26-mer release summarized in Fig. 8C. Contrary to previous results (43), we observed that UvrB4 supported destabilization of the fluorescein-containing 26-mer as efficiently as WT UvrB (t 16 min). In addition, the destabilization of small oligos annealed to circular ssDNA by UvrAB and UvrAB4 complexes is damage specific, because after a 60-min incubation under the same reaction conditions, less than 15% of a nondamaged 26-mer was released by either UvrAB or UvrAB4 complexes (data not shown). Furthermore, there was no detectable strand-destabilizing activity when only UvrB or UvrB4 was present (Fig. 8B).

View larger version (56K): [in this window] [in a new window]   FIGURE 8. Strand-destabilizing activities of WT UvrB and UvrB4. A, the kinetics of 26-mer release by UvrAB (lane 1-5) and UvrAB4 (lane 6-10) monitored by gel electrophoresis. B, monitoring the oligonucleotide release in the presence of only WT UvrB (lane 12) or UvrB4 (lane 13). C, quantitative comparison between the strand-destabilizing activities of WT UvrAB and UvrAB4. n = 3, mean ± S.D.

Initially, we tested the possibility that UvrB4 could support more efficient incision by using a 50-bp duplex substrate, F50/NDB50 (Fig. 3A). The results are shown in Fig. 9A. We observed two prominent Cho incision products (Fig. 9A, labeled P1 and P2). Previously, only one Cho incision product was observed on a 50-bp substrate containing a cholesterol lesion using E. coli UvrA and UvrB proteins (44). However, variation of the 3' incision site has been observed on substrates containing site-specifically modified N2-guanine BPDE adducts (45). Variation of the 3' incision site has been attributed to the stereochemistry and orientation of the BPDE adduct. Future studies are needed to address whether the presence of two 3' Cho incision sites is specific to B. caldotenax proteins or to the fluorescein containing substrate. For purposes of quantitative analysis, both bands (P1 and P2) were counted as incision products. In the presence of UvrA, WT UvrB and UvrB4 supported incision of 7 and 14% of the F50/NDB50 DNA substrate, respectively (Fig. 9, lanes 2 and 4). However, in the absence of UvrA, WT UvrB- and UvrB4-supported incisions were less than 1% of the total DNA substrate (Fig. 9, lanes 3 and 5). These data indicate that even though UvrB4 has higher affinity for duplex DNA, formation of the active form of UvrB4-DNA complex on fully duplexed DNA apparently requires UvrA to help facilitate unwinding DNA.

Next, we examined incision on a DNA substrate containing an 8-bp bubble, F50/B8. Under the same experimental conditions, in the presence of UvrA, both WT UvrB and UvrB4 supported more efficient incision of bubble substrate than of the fully duplexed DNA (Fig. 9, lane 7 versus 2 and lane 9 versus 4). In addition, UvrAB4 supported incision of F50/B8 with slightly higher efficiency than WT UvrAB (Fig. 9, lane 9 versus 7). A dramatic difference is seen on the bubble substrate in the absence of UvrA. In this case, whereas WT UvrB supported little incision of the bubble substrate (3%, Fig. 9, lane 8), UvrB4 supported incision of 41% of the bubble substrate (Fig. 9, lane 10). Similar results were observed with a 6-bp bubble (data not shown). It is worth noting that UvrB4 supported incision of the nondamaged bubble DNA substrate mediated by Cho was less than 1% (data not shown). These data imply that loss of domain 4 allows the UvrB4 mutant by itself to load properly onto DNA, and DNA damage is needed to trigger the conformational change in UvrB4 that can lead to incision. In summary, our results clearly show that: 1) the incision efficiency of Cho is higher on bubble substrate than on fully duplexed DNA (Fig. 9, lane 7 versus 2); 2) independent of UvrA, UvrB4 supports more efficient incision by Cho on bubble substrate (Fig. 9, lane 10 versus 8) than wild type protein.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The enhancement of an activity of a particular domain, such as DNA binding or ATPase, in the absence of some other region of the protein indicates autoinhibition (26). The discovery of an autoinhibitory domain is often through the characterization of the activity of fragments of a protein relative to the activity of the full-length protein. The enhancement of an activity of a truncated form of protein indicates existence of an autoinhibitory domain. Autoinhibition is a widespread regulatory strategy in regulation of protein function. In this study, using two different DNA binding assays, namely EMSAs and photoaffinity cross-linking, we made the surprising discovery that domain 4 of UvrB is an autoinhibitory domain (Fig. 10). Specifically, we report that UvrB4, as a monomer, has a 14- and 22-fold higher binding affinity for damaged and nondamaged ssDNA, respectively, than WT UvrB as a dimer (Fig. 4). In addition, our results demonstrated that this increased DNA binding is independent of nucleotide cofactors (Fig. 6). Domain 4 truncation also has a significant effect on ATPase activity of UvrB: it increases the basal ATPase activity 9-fold and ssDNA-dependent ATPase activity 14-fold (Fig. 7). Consistent with increased DNA binding affinity and ATPase activity, this mutant of UvrB supports more efficient UvrA-independent incision mediated by Cho (Fig. 9).

View larger version (50K): [in this window] [in a new window]   FIGURE 9. WT UvrB and UvrB4 supported incisions mediated by Cho. A, incisions of F50/NDB50 and F50/B8 substrates. B, graphic comparison of incision of F50/NDB50 and F50/B8 substrates mediated by Cho. n = 3, mean ± S.D.

View larger version (31K): [in this window] [in a new window]   FIGURE 10. Autoinhibition by domain 4 of UvrB. The red circle on the UvrB model represents the ATP binding site. The enhancement of DNA binding and ATPase activities in the absence of domain 4 indicates autoinhibition (top right). We propose that the autoinhibitory effect of domain 4 can be relieved by association with UvrA (bottom right), and this hypothesis is graphically demonstrated by showing domain 4 moved away from the rest of the protein. However, defining the mechanism of autoinhibition and understanding how autoinhibition is relieved by UvrA require further investigation.

We observed that after truncation of domain 4, the ATPase activity of UvrB was increased 9-fold in the absence of DNA. The ATP binding site of UvrB is located between domains 1a and 3. Based on the structural similarity between UvrB and helicases, it has been suggested that ATP hydrolysis in UvrB is coupled to the motion of domains 1a and 3 that close down on DNA. In helicase superfamily II, two arginines in motif VI are located near the scissile P–P bond during catalysis (16, 22). These residues are believed to participate in polarization of the scissile bond and stabilization of the developing negative charge at the -phosphate during ATP hydrolysis. In the crystal structure of B. caldotenax UvrB, the closest arginines, Arg⁵⁴³ and Arg⁵⁴⁰, are still too distant from the ATP molecule for direct interactions, thus explaining why the nucleotide cannot be hydrolyzed. It is possible that removal of domain 4 might change the conformation of the loop that connects domains 3 and 4 and the adenine binding pocket of UvrB, thus uncoupling ATPase activity and domain movement (22). In the context of intact UvrB, repositioning of domain 4 relative to other domains of UvrB through the interaction between UvrA and UvrB might alter the ATP binding pocket. Results based on using E. coli UvrB mutants that had a tryptophan reporter group at the ATP binding motif suggests that the accessibility of the ATP binding site in WT UvrB is comparable with UvrB* (46). These data also support a model of allosteric regulation of ATP hydrolysis by domain 4 and argue against physical masking of the ATP binding site by domain 4. These ideas are consistent with the recent finding by Barrett and co-workers (47) who have crystallized UvrB bound to oligo(dT)₃ in a ternary complex with ATP. In this structure domain 4 was removed by proteolysis during crystallization thus enabling ATP binding.

The common mechanisms for relieving autoinhibition include: displacement of the inhibitory domain by a second protein, proteolysis of the inhibitory domain, and post-translational modification or binding of small molecules that allosterically alter the inhibitory domain (26). If domain 4 is an autoinhibitory domain, it is reasonable to suggest that relieving the inhibition provided by domain 4 of UvrB involves the interaction between UvrB and UvrA. This suggestion is supported by our observation that removal of domain 4 enhanced the direct binding of UvrB to DNA (Figs. 4, 5, 6), but UvrA-dependent binding of WT UvrB and UvrB4 to duplex DNA containing damage are similar (data not shown). Previously, Hsu et al. (14), using maltose-binding protein fused with residues 547 to 673 from E. coli UvrB (containing the coiled-coiled domain), observed that E. coli UvrA directly interacts with this part of E. coli UvrB. Taking this observation together with our results, we propose that UvrA relieves the inhibitory effect of domain 4 through protein-protein interactions (Fig. 10).

In general, an autoinhibition domain presents a reversible barrier that prevents spurious activation of a biological pathway (26). In other words, an autoinhibition mechanism is a constitutive damper that serves to repress an activity, and only allows it to be executed in the presence of an activation signal that is sufficiently strong to overcome this barrier (26). In the case of the bacterial NER system, the presence of domain 4 as an autoinhibitory domain may be necessary for increasing the damage specificity of UvrB and for UvrB to be primed for recruiting UvrC only through interaction with UvrA. Unregulated direct binding of UvrB to DNA might provoke spurious incision reactions and could present a roadblock to various DNA transaction events, such as DNA replication, recombination, and other DNA repair pathways.

Another key issue in understanding the DNA damage recognition process during NER is the oligomeric state of UvrB. UvrB was proposed to function as a monomer from early gel filtration and sedimentation studies (4). However, evidence has emerged supporting the notion that UvrB can dimerize and it was suggested that dimerization might be essential for alternate scanning of both DNA strands (5, 25). Based on volume analysis of the proteins from AFM images, we estimated that the K_d for WT UvrB dimer is in the lower nanomolar range. Truncation of domain 4 of UvrB weakens the self-dimerization, increasing the K_d from 5 to 38 nM. We also discovered that domain 4 truncation leads to both increased DNA binding and ATPase activity. Furthermore, UvrB4 is more efficient than WT UvrB in supporting UvrA-independent incision of bubble DNA by Cho (Fig. 9). These data suggest that UvrB4 with its ability to dimerize significantly compromised is completely functional, and UvrB monomer could be the catalytically active form of UvrB. It is possible that after initial recognition of DNA damage by UvrA, verification of DNA damage by UvrB involves two steps. First, dimeric UvrB forms a weak complex with DNA. Second, to achieve tight DNA binding to form the preincision UvrB-DNA complex, one UvrB subunit would dissociate. A question that remains is: does UvrA regulate UvrB dimer formation? In an analogous system, it has been demonstrated that TFIIA promotes the dissociation of TATA-box-binding protein (TBP) dimers directly and accelerates the kinetics of DNA binding by TBP (48). Furthermore, domain 4 truncation seems to only destabilize the UvrB dimer, and not totally abolish dimerization (Fig. 2 and Table 1) (5, 25). Most likely, other residues in UvrB also participate in the dimerization and remain to be characterized. In summary, we propose that domain 4 of UvrB acts as an autoinhibitory gate for stepwise regulation of DNA binding, ATPase activity, and protein-protein interactions (Fig. 10).

	FOOTNOTES

 
^* This work was supported by the Intramural Research Program of the NIEHS, National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ Both authors contributed equally to this paper.

² To whom correspondence should be addressed: NIEHS, 111 Alexander Dr., P. O. Box 12233, Research Triangle Park, NC 27709. Tel.: 919-541-7752; Fax: 919-541-7593; E-mail: vanhout1{at}niehs.nih.gov.

³ The abbreviations used are: NER, nucleotide excision repair; WT, wild type; EMSA, electrophoresis mobility shift assay; BisTris, 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol; AMPPNP, 5'-adenylyl-,-imidodiphosphate; MOPS, 4-morpholinepropanesulfonic acid; ssDNA, single-stranded DNA; dsDNA, double-stranded DNA; AFM, atomic force microscopy.

	ACKNOWLEDGMENTS

 
We thank Drs. Leroy Worth, Jr. and William A. Beard for critical reading of this manuscript, Kolappan Subramaniapillai for technical support of purification of Cho, Paul Cacioppo for assistance with graphics, and Dr. Robert Petrovich and Lori Edwards of the NIEHS Protein Core Facility for assistance with production of the proteins.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Van Houten, B. (1990) Microbiol. Rev. 54, 18–51[Abstract/Free Full Text]
2. Goosen, N., and Moolenaar, G. F. (2001) Res. Microbiol. 152, 401–409[Medline] [Order article via Infotrieve]
3. Van Houten, B., Croteau, D. L., Della Vecchia, M. J., Wang, H., and Kisker, C. (2005) Mutat. Res. 577, 92–117[Medline] [Order article via Infotrieve]
4. Orren, D. K., and Sancar, A. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 5237–5241[Abstract/Free Full Text]
5. Verhoeven, E. E., Wyman, C., Moolenaar, G. F., and Goosen, N. (2002) EMBO J. 21, 4196–4205[CrossRef][Medline] [Order article via Infotrieve]
6. Kacinski, B. M., and Rupp, W. D. (1981) Nature 294, 480–481[CrossRef][Medline] [Order article via Infotrieve]
7. Bertrand-Burggraf, E., Selby, C. P., Hearst, J. E., and Sancar, A. (1991) J. Mol. Biol. 219, 27–36[CrossRef][Medline] [Order article via Infotrieve]
8. Shi, Q., Thresher, R., Sancar, A., and Griffith, J. (1992) J. Mol. Biol. 226, 425–432[CrossRef][Medline] [Order article via Infotrieve]
9. Verhoeven, E. E., Wyman, C., Moolenaar, G. F., Hoeijmakers, J. H., and Goosen, N. (2001) EMBO J. 20, 601–611[CrossRef][Medline] [Order article via Infotrieve]
10. Della Vecchia, M. J., Croteau, D. L., Skorvaga, M., Dezhurov, S. V., Lavrik, O. I., and Van Houten, B. (2004) J. Biol. Chem. 279, 45245–45256[Abstract/Free Full Text]
11. Sancar, A., and Rupp, W. D. (1983) Cell 33, 249–260[CrossRef][Medline] [Order article via Infotrieve]
12. Yeung, A. T., Mattes, W. B., Oh, E. Y., and Grossman, L. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 6157–6161[Abstract/Free Full Text]
13. Moolenaar, G. F., Franken, K. L., Dijkstra, D. M., Thomas-Oates, J. E., Visse, R., van de Putte, P., and Goosen, N. (1995) J. Biol. Chem. 270, 30508–30515[Abstract/Free Full Text]
14. Hsu, D. S., Kim, S. T., Sun, Q., and Sancar, A. (1995) J. Biol. Chem. 270, 8319–8327[Abstract/Free Full Text]
15. Gorbalenya, A. E., Koonin, E. V., Donchenko, A. P., and Blinov, V. M. (1989) Nucleic Acids Res. 17, 4713–4730[Abstract/Free Full Text]
16. Theis, K., Chen, P. J., Skorvaga, M., Van Houten, B., and Kisker, C. (1999) EMBO J. 18, 6899–6907[CrossRef][Medline] [Order article via Infotrieve]
17. Machius, M., Henry, L., Palnitkar, M., and Deisenhofer, J. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 11717–11722[Abstract/Free Full Text]
18. Nakagawa, N., Sugahara, M., Masui, R., Kato, R., Fukuyama, K., and Kuramitsu, S. (1999) J. Biochem. (Tokyo) 126, 986–990[Abstract/Free Full Text]
19. Truglio, J. J., Croteau, D. L., Skorvaga, M., Della Vecchia, M. J., Theis, K., Mandavilli, B. S., Van Houten, B., and Kisker, C. (2004) EMBO J. 23, 2498–2509[CrossRef][Medline] [Order article via Infotrieve]
20. Gordienko, I., and Rupp, W. D. (1997) EMBO J. 16, 880–888[CrossRef][Medline] [Order article via Infotrieve]
21. Zou, Y., and Van Houten, B. (1999) EMBO J. 18, 4889–4901[CrossRef][Medline] [Order article via Infotrieve]
22. Theis, K., Skorvaga, M., Machius, M., Nakagawa, N., Van Houten, B., and Kisker, C. (2000) Mutat. Res. 460, 277–300[Medline] [Order article via Infotrieve]
23. Sohi, M., Alexandrovich, A., Moolenaar, G., Visse, R., Goosen, N., Vernede, X., Fontecilla-Camps, J. C., Champness, J., and Sanderson, M. R. (2000) FEBS Lett. 465, 161–164[CrossRef][Medline] [Order article via Infotrieve]
24. Alexandrovich, A., Czisch, M., Frenkiel, T. A., Kelly, G. P., Goosen, N., Moolenaar, G. F., Chowdhry, B. Z., Sanderson, M. R., and Lane, A. N. (2001) J. Biomol. Struct. Dyn. 19, 219–236[Medline] [Order article via Infotrieve]
25. Hildebrand, E. L., and Grossman, L. (1999) J. Biol. Chem. 274, 27885–27890[Abstract/Free Full Text]
26. Pufall, M. A., and Graves, B. J. (2002) Annu. Rev. Cell Dev. Biol. 18, 421–462[CrossRef][Medline] [Order article via Infotrieve]
27. Skorvaga, M., Theis, K., Mandavilli, B. S., Kisker, C., and Van Houten, B. (2002) J. Biol. Chem. 277, 1553–1559[Abstract/Free Full Text]
28. Wang, H., Yang, Y., Schofield, M. J., Du, C., Fridman, Y., Lee, S. D., Larson, E. D., Drummond, J. T., Alani, E., Hsieh, P., and Erie, D. A. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 14822–14827[Abstract/Free Full Text]
29. Ratcliff, G. C., and Erie, D. A. (2001) J. Am. Chem. Soc. 123, 5632–5635[CrossRef][Medline] [Order article via Infotrieve]