Specific inhibition of the E.coli RecBCD enzyme by Chi sequences in si
http://www.100md.com
《核酸研究医学期刊》
Department of Chemistry and Biochemistry, University of Maryland, College Park, MD 20742, USA
* To whom correspondence should be addressed. Tel: +1 301 405 1821; Fax: +1 301 314 9121; Email: dj13@umail.umd.edu
ABSTRACT
RecBCD is an ATP-dependent helicase and exonuclease which generates 3' single-stranded DNA (ssDNA) ends used by RecA for homologous recombination. The exonuclease activity is altered when RecBCD encounters a Chi sequence (5'-GCTGGTGG-3') in double-stranded DNA (ds DNA), an event critical to the generation of the 3'-ssDNA. This study tests the effect of ssDNA oligonucleotides having a Chi sequence (Chi+) or a single base change that abolishes the Chi sequence (Chio), on the enzymatic activities of RecBCD. Our results show that a 14 and a 20mer with Chi+ in the center of the molecule inhibit the exonuclease and helicase activities of RecBCD to a greater extent than the corresponding Chio oligonucleotides. Oligonucleotides with the Chi sequence at one end, or the Chi sequence alone in an 8mer, failed to show Chi-specific inhibition of RecBCD. Thus, Chi recognition requires that Chi be flanked by DNA at either end. Further experiments indicated that the oligonucleotides inhibit RecBCD from binding to its dsDNA substrate. These results suggest that a specific site for Chi recognition exists on RecBCD, which binds Chi with greater affinity than a non-Chi sequence and is probably adjacent to non-specific DNA binding sites.
INTRODUCTION
The RecBCD enzyme of Escherichia coli is a multifunctional enzyme involved in the repair of double-strand DNA (dsDNA) breaks by RecA-dependent homologous recombination . RecBCD enzyme generates a 3'-ended stretch of single-stranded DNA (ssDNA) which can be used as a substrate by the RecA and single-stranded DNA binding (SSB) proteins to carry out homologous recombination (2). The RecBCD holoenzyme is a DNA-dependent ATPase (3), ATP-dependent helicase (4–6), an exonuclease (3' to 5' and 5' to 3') on single- and double-stranded DNA (3,7) and a ssDNA endonuclease (3). Of the three subunits of the enzyme, RecB is a DNA helicase, ATPase and nuclease (8–11), RecD is an ATPase and helicase (12–14) and RecC is speculated to play some role in Chi sequence recognition .
Critical to RecBCD-mediated generation of 3'-ended ssDNA is the encounter with and recognition of a specific octameric regulatory sequence called the Chi sequence (2,16,17). This sequence is recognized by RecBCD only when the enzyme encounters it from its 3'-end in dsDNA (19). The enzyme cleaves the Chi-containing strand 4 to 6 nt to the 3' side of Chi and continues to unwind beyond Chi, generating a 3'-terminated single strand (19,20). RecA protein is loaded specifically onto this 3'-terminated strand to initiate homologous recombination (21–23). Chi sites are thus hot-spots for recombination, stimulating recombination specifically to the 5' side of the sequence (17). Single base changes within the Chi sequence reduce or abolish its effectiveness both in vitro and in vivo (18,24,25).
The biochemical nature of the RecBCD–Chi interaction is not well understood. The effects of Chi on the enzyme found in previous in vitro work are varied and depend on the reaction conditions, particularly the ATP and magnesium ion concentrations. RecBCD has greater exonuclease activity on the 3'-ended strand, as compared to the 5'-ended strand, of a dsDNA substrate (21,26). However, once it encounters Chi, the translocating enzyme pauses (27), and subsequent cutting on the 3'-strand is attenuated leading to production of a Chi-specific DNA fragment, the length of which is equal to the length of the DNA from Chi to the 5' end of the strand (20,26). Also observed under some conditions is enhanced cleavage of the 5'-ended strand (containing the complement of the Chi sequence) after Chi (7,28,29). The enhanced cutting of this strand by the 5' to 3' exonuclease activity leads to the production of a second, upstream Chi-specific fragment (7,29). Based on these and other observations, two models were proposed to explain the effect of Chi on RecBCD nuclease activity. In one (30), there is a conformational change in the enzyme after it encounters and binds to Chi, which re-positions the nuclease active site in the RecB subunit to the opposite strand of the dsDNA substrate. A second model (31) suggests that the 3' Chi-ended strand is bound in a Chi-specific site and is thereby excluded from the nuclease active site in that process. The 5' strand loops around and can be bound to the nuclease active site of RecB after the Chi encounter, where the DNA strand can be cut by the enzyme. Both models suggest that the loss of exonuclease activity on the Chi-containing strand is because of a change in positioning of the nuclease active site relative to the substrate.
At the other end of the spectrum, there is evidence suggesting that there can be general inactivation of the enzyme and complete disassembly of the subunits of RecBCD upon or after the Chi encounter. These experiments, done at relatively low magnesium ion concentrations, show that the encounter with Chi in one dsDNA substrate prevents nicking at a second Chi in the same substrate (32) and inactivates RecBCD as a helicase and nuclease towards a second DNA substrate (32–34). Further, analysis by glycerol gradient sedimentation indicated that the three subunits dissociate after the enzyme encounters Chi in dsDNA under these conditions (34) and hence there is overall inactivation of the enzyme.
Most previous studies done on the effects of Chi on RecBCD involved the use of double-stranded DNA substrates containing Chi. However, a Chi sequence can be recognized as a single strand in a double-stranded substrate (35). Hence, in this study we have tested the effects of Chi in single-stranded oligonucleotides on the biochemical activities of RecBCD. We tested for specific recognition of Chi by looking for Chi-dependent inhibition of the exonuclease and helicase activities, and of dsDNA substrate binding. We also tested for up-regulation of the 5' to 3' exonuclease activity on dsDNA. For each Chi oligonucleotide tested, a corresponding control oligonucleotide with a single nucleotide change from the Chi sequence (5'-GCAGGTGG-3', called Chio) was also tested to establish specificity for the Chi sequence. The Chio sequence is not recognized in dsDNA in vitro, and does not function as Chi in vivo (24,25).
Our results show inhibition of the dsDNA exonuclease and DNA unwinding activities by the Chi sequence, but only in those oligonucleotides where the Chi sequence is flanked by DNA at both ends. There was no evidence that Chi in an oligonucleotide could prevent Chi recognition in dsDNA, nor was enhanced cleavage of the strand containing the complement of Chi observed. Instead, the results are consistent with the Chi-containing oligonucleotides binding to the enzyme and preventing binding of other DNA substrates. These Chi-containing oligonucleotides may be useful for further studies to gain insight into the physical effects of Chi on the enzyme and to examine where Chi is bound by RecBCD.
MATERIALS AND METHODS
Materials
Synthetic oligodeoxyribonucleotides were purchased from Invitrogen Corp. Some batches of the oligonucleotides were desalted by the manufacturer and were used without further purification. Others were purified by high performance liquid chromatography (HPLC) by the manufacturer, and were purified further using a QIAquick Nucleotide Removal Kit (Qiagen Corp.). Essentially identical results were obtained with the separate batches of oligonucleotides. Oligonucleotide stocks were prepared in deionized distilled water and their concentrations were determined from absorbance measurements at 260 nm. Absorption coefficients were calculated by the nearest-neighbor method using the Bench Mate program (http://www.roche-applied-science.com/benchmate) and parameters given in (36). Tritiated E.coli chromosomal DNA (4040 c.p.m./nmol nucleotide residues) was purified as described (37). Plasmid pTZ18R DNA (2860 bp; 122 300 c.p.m./nmol nucleotide) was tritiated in vivo as described (38) and was purified using the QIAfilter Plasmid Maxi Kit (QIAgen Corp.) following the manufacturer's instructions. pBR322+F plasmid DNA (18) was a gift from Dr Gerald Smith, Fred Hutchinson Cancer Research Center, and was isolated by the QIAfilter Plasmid Maxi Kit.
RecBCD enzyme was purified as described (11,37).
Methods
Double-stranded DNA exonuclease assay
Exonuclease reaction mixtures contained 50 mM Tris–HCl, pH 8.5, 10 mM MgCl2, 1 mM ATP, 0.67 mM DTT and 0.2 mg/ml BSA, with varied concentrations of oligonucleotides. The DNA substrate was either pTZ18R DNA (linearized by cleavage with SmaI, 5 μM nucleotide residues) or E.coli chromosomal DNA (40 μM nucleotide). The reactions were started by adding RecBCD and were incubated at 37°C. At pre-determined time intervals, 20 μl aliquots were removed, quenched in 100 μl ice-cold 10% trichloroacetic acid and 5 μl calf thymus DNA (0.5 mg/ml), and the concentration of acid-soluble nuclease reaction products was determined as described (37).
Pre-incubation of RecBCD with oligonucleotides
Some experiments were carried out by first incubating RecBCD with the oligonucleotide to be tested as an inhibitor, and then adding that to an exonuclease reaction mixture that contained the E.coli DNA substrate. The basic pre-incubation conditions were those used in (34) and contained 20 mM MOPS–KOH, pH 7.0, 3 mM magnesium acetate, 5 mM ATP, 20 mM DTT, 0.01 mg/ml BSA, 50 nM RecBCD, and varying concentrations of oligonucleotide. An aliquot of the pre-incubation mixture (typically 0.8 μl) was added to a dsDNA exonuclease reaction mixture (85 μl total volume) containing the reaction components given above and acidsoluble products were measured as above. A number of variations were made, including different incubation times, temperatures, and ATP and Mg2+ concentrations (see Results).
Reactions with Chi-containing double-stranded DNA substrate
pBR322+F plasmid DNA was linearized by treatment with HindIII restriction endonuclease, the 5' phosphate was removed with shrimp alkaline phosphatase (U. S. Biochemicals Corp.), and the DNA was purified using the QIAprep Spin Miniprep Kit (QIAgen Corp.). The linearized plasmid was then 5'-32P-end-labeled using polynucleotide kinase (New England Biolabs) and ATP (Amersham Biosciences, 3000 Ci/mmol) under conditions given in (39). The reaction products were repurified using the same kit and the amount of labeled DNA obtained was estimated by comparison of the band intensity to a DNA standard (High DNA Mass Ladder, Invitrogen Corp.) on a 1% agarose gel stained with ethidium bromide.
RecBCD reaction mixtures contained 25 mM Tris–acetate buffer, pH 7.5, 1 mM magnesium acetate, 1 mM ATP, 1 mM DTT, with 1.15 nM pBR322+F DNA molecules, 0.31 nM RecBCD enzyme, and varying concentration of the oligonucleotides as indicated in the figure legends. The reactions were started by adding the enzyme, incubated at 37°C, and 10 μl aliquots were quenched at predetermined times in 2.5 μl of quench solution (30% glycerol, 90 mM EDTA, 0.2% bromophenol blue and 1.3% SDS). The samples were run on a 1% agarose gel at 4.2 V/cm for 2 h or at 1.0 V/cm overnight, in 1 x TBE (40 mM Tris–borate, 1 mM EDTA, pH 8.0). The gel was dried and the radioactivity was detected using a Phosphorimager (Molecular Dynamics).
A variation of the above assay was carried out to test if the oligonucleotides can inhibit RecBCD that was pre-bound to the dsDNA substrate. RecBCD was mixed with the DNA and other reaction components except for ATP to allow the enzyme to bind the dsDNA. After a 2 min incubation at 37°C, a zero time-point aliquot was removed and quenched as above, and ATP alone or a mixture of ATP and the oligonucleotide was added to start the reaction. The final reaction mixtures contained 3 nM RecBCD, 2.3 nM DNA molecules, and other reactants as above. Samples were removed, quenched and analyzed on agarose gels as above.
Helicase assay
The helicase substrate was made by annealing two complementary DNA oligonucleotides (32 nt each): oligo1 and oligo2 . Oligo2 was 5' end-labeled using ATP and polynucleotide kinase as described in (39). The labeled oligo2 was purified (QIAgen Nucleotide Removal Kit), and analyzed by thin layer chromatography on polyethyleneimine–cellulose plates (J. T. Baker) in 1 M potassium phosphate buffer (pH 3.0) to ensure separation of residual ATP from the labeled oligonucleotide. The concentration of the purified labeled oligonucleotide was estimated from the amount of radioactivity (determined by scintillation counting) of samples taken before and after purification.
The double-stranded helicase substrate was prepared by mixing oligo2 (5 nM) with 95 nM of unlabeled oligo2 and 102 nM of its unlabeled complementary strand (oligo1) in 0.5 M NaCl, 0.2 M Tris–acetate, pH 7.5, and 10 mM MgCl2. The mixture was heated to 95–100°C for 2 min, cooled gradually to 28°C in the same water bath, placed on ice for 10–15 min, and stored at 4°C or used in the assay below.
The standard reaction was carried out in 25 mM Tris–acetate buffer, pH 7.5, 1 mM magnesium acetate, 1 mM ATP, 1 mM DTT, 0.31 nM RecBCD enzyme, varied concentrations of oligonucleotide, and 3 nM dsDNA substrate. The control reaction (no oligonucleotide) had 0.1 nM RecBCD. The reactions were incubated at 37°C and 5 μl aliquots were added to 2 μl of a quench solution containing 10% glycerol, 25 mM EDTA, 0.6% SDS, 0.03% bromophenol blue and a 5-fold excess of unlabeled oligo2. The quenched samples were run on 15% polyacrylamide gels in 1 x TBE. The gels were pre-run for 30 min at 12 V/cm, the samples were loaded and the gels were run at 12 V/cm for 2 h. The gels were dried and analyzed using a Phosphorimager.
Quantitation for reactions with Chi-containing double-stranded DNA and helicase assays
The amount of dsDNA unwound and of Chi fragment produced in these assays was quantitated using the ImageQuant software (Molecular Dynamics) on the Phosphorimager. Objects were drawn to encircle the bands to be quantitated and the volume of the band was integrated using the software. The background was determined by drawing an object on a part of the gel that did not contain sample and its integrated volume was subtracted from the values obtained for each band. The percentage values were then calculated relative to the volume of the dsDNA band in the zero time-point lane.
RESULTS
We designed the series of single-stranded oligonucleotides shown in Table 1 to test for Chi-specific effects on RecBCD enzyme. These consist of an 8-mer which is just the Chi sequence itself (8-merChi+), as well as sets of 14 and 20mers that are Chi with additional flanking DNA sequence. The larger oligonucleotides have Chi either in the center (14-merChi+, 20-merChi+), or at one or the other end (e.g. 3'14-merChi+, 5'14-merChi+, etc.). These Chi-ended oligonucleotides were tested because RecBCD normally sees Chi as it approaches Chi from one end (the 3' end) in one strand of a double-stranded DNA molecule. For each Chi+ oligonucleotide, we also tested a corresponding Chio control molecule with a single base change that inactivates the Chi-sequence (25).
Table 1. Oligonucleotides used as inhibitors
Inhibition of RecBCD exonuclease activity on dsDNA by Chi+ oligonucleotides
We first did experiments with the 8mers, 14-merChi+/o and 3'20-merChi+/o (see Table 1) to find the concentrations required to see inhibition of RecBCD exonuclease activity on a linear double-stranded DNA substrate (Figure 1). The longer oligomers are somewhat more potent inhibitors than the shorter oligomers (i.e. greater inhibition by the 20 and 14mers than the 8mers, at a given concentration). There was little difference in the degree of inhibition between Chio and Chi+ in the 8mers and 3'20mers, but the 14-merChi+ inhibited somewhat more than the 14-merChio, at a given concentration of each oligomer.
Figure 1. Concentration dependence of inhibition of exonuclease activity by oligodeoxyribonucleotides. Reaction mixtures contained 50 mM Tris–HCl, pH 8.5, 10 mM MgCl2, 40 μM ATP, 40 μM E. coli chromosomal DNA substrate (4040 c.p.m./nmol), and 0.31 nM RecBCD. The mixtures were incubated at 37°C and aliquots were quenched at the indicated times in ice-cold 10% trichloroacetic acid. The concentration of acid-soluble nucleotides was determined as described in Materials and Methods. (A) 8-merChio (left panel) or 8-merChi+ (right panel) were present at 0 μM (circles), 5 μM (triangles), 30 μM (squares) or 50 μM (diamonds). (B) 14-merChio (left panel) or 14-merChi+ (right panel) were present at 0 μM (circles), 1.5 μM (triangles), 5 μM (squares) or 10 μM (diamonds). (C) 3'20-merChio (left panel) or 3'20-merChi+ (right panel) were present at 0 μM (circles), 0.5 μM (triangles), 5 μM (squares) or 10 μM (diamonds).
These oligonucleotide concentrations were then used in the experiments in Figures 2–4, and in the assays of other RecBCD enzyme activities shown in subsequent figures. The effect of the 14mer oligonucleotide series on RecBCD-catalyzed nuclease reactions on linear double-stranded DNA is shown in Figure 2. The 14mer with Chi in the center and flanked by 3 nt of additional DNA on both ends (14-merChi+) was a more potent inhibitor of the RecBCD reaction than was an identical oligonucleotide except having the single base change to make the Chio sequence (Figure 2A). Both oligonucleotides (Chi+ and Chio) are inhibitors at the concentrations tested, but the reaction rate with the 14-merChi+ was about 2.2-fold less than that when the 14-merChio was present. We also tested a 14mer that is a scrambled version of the 14-merChi+, as an inhibitor in these experiments. This scrambled 14mer (at 15 μM) inhibited to a similar degree as the same concentration of 14-merChio, and it inhibited less than the 14-merChi+ (data not shown). The results with the Chi-ended 14mers are shown in Figure 2B and C. These molecules are also inhibitors of the RecBCD-catalyzed reactions, but there is little or no difference in the efficacy of the Chi+ and Chio oligonucleotides, when the Chi-sequence is situated entirely at one or the other end (3' or 5') of the DNA molecule. The results in Figure 2 are consistent with there being Chi-specific inhibition of RecBCD by the 14mers, but only when the Chi sequence is flanked by additional DNA at both ends.
Figure 2. Effect of 14mers on the exonuclease activity of RecBCD. Reactions containing 50 mM Tris–HCl, pH 8.5, 10 mM MgCl2, 1 mM ATP, 0.67 mM DTT, 0.2 mg/ml BSA, linear pTZ18R DNA (5 μM nucleotides) and oligonucleotides were started by adding 0.03 nM RecBCD. The concentration of acid-soluble nucleotides was determined as described in Materials and Methods, and values were normalized against the averaged 6 min time point for the corresponding control reaction. Reactions were done in triplicate and the error bars show the standard deviation of the averaged individual determinations. (A) Reaction mixtures contained 15 μM 14-merChi+ (open triangles), 15 μM 14-merChio (black squares) or no oligonucleotide (black circles). (B) Reaction mixtures contained 15 μM 5'14-merChi+ (open triangles), 15 μM 5'14-merChio (black squares) or no oligonucleotide (black circles). (C) Reaction mixtures contained 15 μM 3'14-merChi+ (open triangles), 15 μM 3'14-merChio (black squares), or no oligonucleotide (black circles).
Figure 3. Effect of 20mers on the exonuclease activity of RecBCD. Reactions were as in Figure 2 except that the DNA substrate was E.coli DNA (40 μM nt) and they were started by adding 0.3 nM RecBCD. Reactions were done in triplicate and normalized as in Figure 2. The error bars show the standard deviation of the averaged individual determinations. (A) Reaction mixtures contained 2 μM 20-merChi+ (open triangles), 2 μM 20-merChio (black squares), or no oligonucleotide (black circles). (B) Reaction mixtures contained 2 μM 5'20-merChi+ (open triangles), 2 μM 5'20-merChio (black squares), or no oligonucleotide (black circles). (C) Reaction mixtures contained 2 μM 3'20-merChi+ (open triangles), 2 μM 3'20-merChio (black squares) or no oligonucleotide (black circles).
Figure 4. Effect of 8mers on the exonuclease activity of RecBCD. Reactions were as in Figure 2 with linear pTZ18R DNA (5 μM nt), 0.03 nM RecBCD, and 50 μM 8-merChi+ (open triangles) 50 μM 8-merChio (black squares) or no oligonucleotide (black circles). Reactions were done in triplicate and normalized as in Figure 2. Error bars are the standard deviations.
A similar pattern is observed with Chi sequences in the 20mers. Again, the Chi+ 20mer is a more potent inhibitor that the Chio 20mer, when Chi is in the center of the molecule (20-merChi+ versus 20-merChio; Figure 3A). As with the 14mers, there is little if any difference in the degree of inhibition when the Chi+ and Chio sequences are at either the 5' end (Figure 3B) or the 3' end (Figure 3C) of the molecule. A second 20-merChio with a different mutant Chi sequence was also tested in this assay. It inhibited similar to the 20-merChio shown in Figure 3A, and was less potent than the 20-merChi+ (data not shown). This second mutant Chi sequence was also shown to be non-functional as Chi in previous work (24,25).
We also tested the 8-merChi+, which is simply the Chi sequence with no additional flanking DNA, and 8-merChio, in experiments such as that shown in Figure 4. As with the larger oligonucleotides with Chi at one end (Figures 2B and C and 3B and C), there is no difference in the inhibition by Chi+ versus Chio in the 8mers (Figure 4).
Exonuclease assays were also done using RecBCD enzyme that was first pre-incubated with the oligonucleotides and then diluted into the exonuclease reaction mixtures (see Materials and Methods). Previous experiments have shown that RecBCD that is allowed to react first with a Chi-containing dsDNA molecule under some conditions, loses activity towards a second dsDNA substrate (32–34). We decided to test whether such a ‘trans’ effect of Chi could be detected using the single-stranded Chi-containing oligonucleotides. The enzyme and oligonucleotides were pre-incubated at various concentrations, temperatures (0, 23 or 37°C), Mg2+, ATP, and oligonucleotide concentrations, and for various times (5 min to overnight) and then diluted into the nuclease reaction mixture. The oligonucleotide concentrations in most pre-incubations were 0.5–2.5 μM, with 50 nM RecBCD. We expect substantial binding of the oligonucleotides to the enzyme under these conditions, given previously measured dissociation constants for single-stranded DNA molecules of similar size (40). (The oligonucleotide concentration was much lower in the exonuclease reaction mixture itself than in the pre-incubation, due to the nearly 100-fold dilution.) In no case was there a significant difference in the amount of nuclease activity obtained with enzyme incubated with a Chi+ versus a Chio oligonucleotide (data not shown; the 14-merChi+/o, the 8-merChi+/o and 3'20-merChi+/o were tested). Chi-specific inhibition was seen only in reactions using RecBCD that was incubated with the 14-merChi+ (20 μM) and then diluted 2-fold into the nuclease reaction mixture. Here the degree of inhibition was the same as that when 10 μM 14-merChi+ was added directly to the nuclease reaction mixture (data not shown). Thus we do not observe Chi-dependent inactivation of RecBCD as has been observed when RecBCD encounters Chi in dsDNA under some reaction conditions (32–34).
Reactions with Chi-containing double-stranded DNA
The oligonucleotides were next tested for their ability to inhibit the production of a Chi-specific fragment that arises from attenuation of nuclease activity on the Chi containing (3'-ended) strand of dsDNA. The substrate used in this reaction, pBR322F+, is a 4.3 kb plasmid with a Chi site 1463 bp from one end after cleavage with HindIII . Recognition of this site by RecBCD, followed by attenuation of its nuclease activity at that point, leads to the generation of a 1463 nt 5'-32P-labeled fragment of ssDNA, the downstream Chi fragment (Figure 5A). Activation of opposite strand cleavage at the Chi sequence would lead to production of a 3000 nt upstream 5'-32P-labeled Chi fragment (7,28). (The upstream Chi fragment and uncleaved full-length ssDNA are indistinct at the ATP and Mg2+ concentrations used for our experiments and because of the absence of SSB protein, which stabilizes the single-stranded products . Experiments were carried out to test whether a Chi-containing oligonucleotide could affect Chi recognition on a competing dsDNA substrate (as measured by the amount of downstream Chi-fragment obtained). The reaction could be altered in the presence of a Chi+ oligonucleotide in one of two ways: qualitatively (e.g. different relative amounts of the downstream and upstream Chi fragments) or quantitatively (less overall reaction of the dsDNA).
Figure 5. Effect of Chi-centered 14mers on the reaction with linear, Chi-containing dsDNA. (A) HindIII-linearized pBR322F+ plasmid substrate (4.3 kb) used in the reaction. Cutting 4–6 nt on the 3' side of the Chi site in the top strand (by a RecBCD enzyme approaching from the right end as drawn) results in a 1.46 kb 5'-32P-labeled downstream Chi fragment; cutting of the lower strand after Chi results in a 3 kb 5'-32P-labeled upstream Chi fragment. (B) Reactions contained 25 mM Tris–acetate, pH 7.5, 1 mM magnesium acetate, 1 mM ATP, 1 mM DTT, 1.15 nM (molecules) of pBR322F+ plasmid substrate and 10 μM Chi-centered 14-merChi+ or 14-merChio. The control reaction contained no oligonucleotide. Reactions were started by adding 0.3 nM RecBCD and incubated at 37°C. Aliquots were quenched at the indicated times, run on a 1% agarose gel and the gel scanned using a Phosphorimager. (C) The percentage dsDNA remaining at each time point was quantitated as described in Materials and Methods. 14-merChi+ (open triangles), 14-merChio (black squares), control reaction (black circles). Reactions containing oligonucleotide were done in duplicate and the error bars show the range of the individual determinations. (Some error bars are obscured by the plot symbols.) (D) The amount of Chi fragment produced with time (as a % of the dsDNA at time zero) was calculated as described in Materials and Methods. 14-merChi+ (open triangles), 14-merChio (black squares), control reaction (black circles). Error bars show the range of the individual determinations.
The results show a decrease in the extent of DNA unwinding (helicase activity) and of production of the downstream Chi fragment. The results for the 14-merChi+/o are shown in Figure 5B. Figure 5C shows a plot of the percentage of the dsDNA remaining with time during the reaction. While essentially all of the DNA is unwound and degraded within 2 min in the control reaction, and 70% has disappeared in 3 min in the presence of the 14-merChio, 30% of the dsDNA has reacted in 3 min in the 14-merChi+ reaction. The total amount of downstream Chi fragment obtained is also less when the 14-merChi+ is present (Figure 5D) presumably because of the inhibition of the unwinding reaction shown in Figure 5C. Thus, there is a quantitative difference in the degree of inhibition by the Chi+ versus Chio oligonucleotides.
The results with the 20-merChi+/o are shown in Figure 6. Again, the 20-merChi+ inhibits the helicase activity much better than the 20-merChio. After 3 min 15% of the dsDNA substrate is unwound in the presence of the 20-merChi+ whereas 70% is unwound in the presence of the 20-merChio (Figure 6B). The 20-merChi+ also inhibited production of the downstream Chi band product more than did the 20-merChio (Figure 6C).
Figure 6. Effect of Chi centered 20mers on the reaction with linear, Chi-containing dsDNA. (A) Reaction conditions were as described for Figure 5, with 2 μM 20-merChi+, 20-merChio or no oligonucleotide (control reaction). (B) The percentage dsDNA remaining versus time with the 20-merChi+ (open triangles), 20-merChio (black squares) and control reaction (black circles). Reactions containing oligonucleotide were done in duplicate and the error bars show the range of the individual determinations. (C) Amount of Chi fragment was calculated as in Figure 5D. 20-merChi+ (open triangles), 20-merChio (black squares), control reaction (black circles). Error bars show the range of the individual determinations.
Results with the 8mer are shown in Figure 7. There is little difference between the reactions containing the 8-merChi+ and 8-merChio oligonucleotide, again confirming that the Chi site does not seem to be specifically recognized or distinguished from the Chio site in the 8mer constructs. There is also little difference in the inhibition by the oligomers (14 or 20mers) having Chi+ or Chio at either the 3' or 5' end, in this assay (data not shown).
Figure 7. Effect of 8mers on the reaction with linear, Chi-containing dsDNA. (A) Reaction conditions were as described for Figure 5 with 25 μM 8-merChi+, 8-merChio, or no oligonucleotide. (B) Percentage dsDNA remaining versus time with the 8-merChi+ (open triangles), 8-merChio (black squares) and control reaction (black circles). Reactions containing oligonucleotide were done in duplicate and the error bars show the range of the individual determinations. (C) Amount of Chi fragment was calculated as described in Figure 5D. 8-merChi+ (open triangles), 8-merChio (black squares), control reaction (black circles). Error bars show the range of the individual determinations.
Effect of oligonucleotides on RecBCD helicase activity
Since the above set of assays revealed inhibition of unwinding by the oligonucleotides with Chi+ in the center (14 and 20mers), we decided to test these constructs for inhibition of DNA unwinding directly using a small (32 bp) dsDNA substrate. Results for the 14mers are as shown in Figure 8. While 75% of the substrate is unwound in 3 min in the presence of the 14-merChio oligonucleotide, only 30% is unwound in the reaction containing the 14-merChi+ (Figure 8B).
Figure 8. Effect of Chi-centered 14mers on the helicase activity of RecBCD. (A) Reactions contained 25 mM Tris–acetate, pH 7.5, 1 mM magnesium acetate, 1 mM ATP, 1 mM DTT, 3 nM 5'-32P-labeled double-stranded 32 bp substrate (see Materials and Methods) and 15 μM Chi-centered 14-merChi+ or 14-merChio. Reactions were started by adding 0.3 nM RecBCD and were incubated at 37°C. Aliquots were quenched at the indicated times and run on a 15% polyacrylamide gel. The control reaction contained 0.1 nM RecBCD and no oligonucleotide. (B) The amount of double-strand band in each lane was quantitated as described in Materials and Methods and plotted as a function of time. 14-merChi+ (open triangles), 14-merChio (black squares), no oligonucleotide (black circles). Reactions containing oligonucleotide were done in duplicate and the error bars show the range of the individual determinations.
Effect of oligonucleotides on pre-formed RecBCD-dsDNA complexes
The observed inhibition of the exonuclease and helicase activities might arise if oligonucleotides interfere with binding of RecBCD to the dsDNA substrate. In order to test for this effect, the Chi-cleavage assay was repeated with an initial pre-incubation of the enzyme and the 4.3 kb plasmid substrate for 2 min at 37°C. This pre-incubation would allow the enzyme to bind the duplex ends first without interference by the competing oligonucleotides (41). The reactions were started by adding either ATP alone (to the control reaction) or an ATP–oligonucleotide mixture and samples were removed within the first 30 s. A higher enzyme:substrate ratio was used for this set of reactions than for those shown above, to ensure that most of the duplex ends have an enzyme molecule bound to them. The concentrations used and other details are as described in the Materials and Methods section and in the figure legend.
The 20mers inhibit very strongly when RecBCD is added to reaction mixtures containing both the DNA substrate and the oligonucleotide inhibitor without pre-incubation (Figure 9A). All of the linear substrate is consumed within 20 s in the control reaction, while very little substrate has been consumed in the first 20 s in those reactions containing the oligonucleotide inhibitors. In contrast, the enzyme that is allowed to bind first to the dsDNA substrate (Figure 9B) is impervious to inhibition by the oligonucleotides, and most of the substrate is consumed by the first time point whether or not an oligonucleotide is present in the mixture. This shows that allowing the enzyme to bind the dsDNA substrate first ensures that oligonucleotides, even those with the Chi sequence, are unable to affect the enzyme once it initiates its helicase/nuclease activities on the dsDNA. The same result was obtained with the 8mers (data not shown).
Figure 9. Effect of pre-incubation of dsDNA substrate with RecBCD. (A) Reactions done without pre-incubation were started by adding RecBCD enzyme to mixtures that contained all other reaction components. The final reaction mixture contained HindIII-linearized pBR322F+ (2.3 nM molecules), 3 nM RecBCD, 2 μM 20-merChi+ or 20-merChio, and other components as in Figure 5. The control reaction contained no oligonucleotide. Aliquots were quenched at the indicated times and run on a 1% agarose gel. (B) For pre-incubated reactions, HindIII-linearized pBR322F+ (2.3 nM molecules) was incubated for 2 min at 37°C with 3 nM RecBCD and all other reaction components except ATP and the oligonucleotides. The reaction was started by adding a mixture of ATP and 20-merChi+ or 20-merChio (final concentration 2 μM).
DISCUSSION
The results in this paper show that all of the oligonucleotides tested act as inhibitors of RecBCD. The overall inhibition is length dependent, the longer the oligonucleotide the more effective the inhibition. More significantly, the Chi+ oligonucleotides inhibit the activities of RecBCD much more than do the Chio oligonucleotides, and this distinction is only seen in constructs where the Chi+ or Chio site is flanked by sequence at either end. This suggests that recognition of the Chi sequence may be a location-specific effect and flanking sequences may play an important role in this interaction. The oligonucleotides all prevent RecBCD binding to the dsDNA substrate, but they apparently do not cause any qualitative or physical changes in the enzyme as has been observed previously with Chi-containing dsDNA (32–34).
These results suggest that a specific site exists on RecBCD that recognizes Chi and binds to it with greater affinity than to a non-Chi sequence. The greater inhibition seen due to the Chi+ sequence only in the Chi-centered constructs suggests RecBCD needs a DNA ‘handle’ region to be able to recognize Chi. The flanking DNA in the 14-mers and 20-mers would serve as this handle enabling Chi recognition. We envision an elongated DNA binding site, a part of which binds non-specifically to the DNA handle. This binding site merely serves as an anchor to enable better contact between RecBCD and the oligonucleotide but it does not recognize Chi (Figure 10A). A site adjacent to the non-specific site may be specific for Chi recognition and binding. The DNA flank in the oligonucleotide would then be important as it would ensure that the Chi sequence itself does not become the handle and go unrecognized. This may explain the lack of Chi recognition in the Chi-ended 14 and 20mers, and also the 8mer. The Chi site in all these constructs becomes part of the handle sequence itself and hence goes unrecognized (Figure 10B).
Figure 10. Model for location-specific recognition of the Chi+ sequence. (A) Interaction of RecBCD with the Chi sequence in Chi-centered oligonucleotide. Proposed ‘handle’ region on the ends of the oligonucleotide is contacted by non-specific DNA binding site in the N-terminal domain of RecB (+++). This site grips the handle, but cannot recognize the Chi sequence. The Chi site is recognized by a specific recognition site adjacent to the non-specific site. (B) Interaction of RecBCD with Chi sequence in Chi-ended oligonucleotide and the 8mer. The Chi sequence at the 3' end now becomes the handle itself and hence goes unrecognized. The same is true of the Chi+ in the 8mer.
The fact that the oligonucleotides (whether Chi+ or Chio) do not inhibit the enzyme which has been allowed to bind to the dsDNA substrate first (Figure 9) shows that their binding site(s) become inaccessible to the oligonucleotide when the enzyme has bound to the dsDNA. Thus our results do not provide evidence for a Chi-specific site that can be filled by a Chi-containing oligonucleotide once the enzyme starts to act on dsDNA. Instead, the non-specific site that binds the ‘handle’ sequence may be part of or adjacent to a non-specific DNA binding site that is also involved in the helicase activity. This would allow the enzyme to scan the DNA for the Chi site as the DNA is being unwound. The requirement for a DNA flank also suggests that if RecBCD were to start unwinding a DNA having a Chi site at the very end, it would be likely to skip the site and continue DNA degradation until it encounters the next Chi.
It is still not known which subunit(s) contains the Chi-specific binding site. Mutations have been isolated in both the recB and recC genes that affect the response to Chi in vivo (42,43). The recC mutant enzyme responds to a variant of the Chi sequence as might be expected if the mutation has changed a Chi-specific binding site (15,44), suggesting that RecC has the Chi-specific binding site. However, in a photocrosslinking experiment probing RecBCD bound to the end of linear dsDNA, RecC was crosslinked to the 5'-ended strand, while RecB was found to bind the 3'-ended strand of dsDNA (45). Since Chi must be in the 3'-ended strand for recognition by the translocating RecBCD enzyme, this would suggest that RecB might have the Chi-specific site. RecB must contain non-specific DNA binding site(s), since the protein is a DNA-dependent ATPase, helicase and endonuclease (8,11). The isolated RecC protein does not bind DNA (46), but RecC stimulates the ATP hydrolysis and helicase activities of RecB (40,47,48). A simple model is that the non-specific DNA binding sites shown in Figure 10 reside in RecB. A DNA binding region in RecC that is specific for Chi could be at an interface with RecB, and contribute greater affinity for Chi than for non-specific DNA to the RecBCD complex. In preliminary photocrosslinking experiments using RecBCD and the 14-merChi+, we find that the RecC protein appears to be the main subunit crosslinked (A. Kulkarni and D. A. Julin, unpublished results). Efforts are underway to identify more precisely the specific amino acid residues involved in the crosslinking.
ACKNOWLEDGEMENTS
This work was supported by grant # GM39777 from the NIH.
REFERENCES
Kuzminov,A. ( (1999) ) Recombinational repair of DNA damage in Escherichia coli and bacteriophage lambda. Microbiol. Mol. Biol. Rev., , 63, , 751–813.
Kowalczykowski,S.C. ( (2000) ) Initiation of genetic recombination and recombination-dependent replication. Trends Biochem. Sci., , 25, , 156–165.
Goldmark,P.J. and Linn,S. ( (1972) ) Purification and properties of the recBC DNase of Escherichia coli K-12. J. Biol. Chem., , 247, , 1849–1860.
MacKay,V. and Linn,S. ( (1974) ) The mechanism of degradation of duplex deoxyribonucleic acid by the recBC enzyme of Escherichia coli K-12. J. Biol. Chem., , 249, , 4286–4294.
Taylor,A. and Smith,G.R. ( (1980) ) Unwinding and rewinding of DNA by the RecBC enzyme. Cell, , 22, , 447–457.
Roman,L.J. and Kowalczykowski,S.C. ( (1989) ) Characterization of the helicase activity of the Escherichia coli RecBCD enzyme using a novel helicase assay. Biochemistry, , 28, , 2863–2873.
Anderson,D.G. and Kowalczykowski,S.C. ( (1997) ) The recombination hot spot is a regulatory element that switches the polarity of DNA degradation by the RecBCD enzyme. Genes Dev., , 11, , 571–581.
Boehmer,P.E. and Emmerson,P.T. ( (1992) ) The RecB subunit of the Escherichia coli RecBCD enzyme couples ATP hydrolysis to DNA unwinding. J. Biol. Chem., , 267, , 4981–4987.
Hsieh,S. and Julin,D.A. ( (1992) ) Alteration by site-directed mutagenesis of the conserved lysine residue in the consensus ATP-binding sequence of the RecB protein of Escherichia coli. Nucleic Acids Res., , 20, , 5647–5653.
Yu,M., Souaya,J. and Julin,D.A. ( (1998) ) The 30-kDa C-terminal domain of the RecB protein is critical for the nuclease activity, but not the helicase activity, of the RecBCD enzyme from Escherichia coli. Proc. Natl Acad. Sci. USA, , 95, , 981–986.
Yu,M., Souaya,J. and Julin,D.A. ( (1998) ) Identification of the nuclease active site in the multifunctional RecBCD enzyme by creation of a chimeric enzyme. J. Mol. Biol., , 283, , 797–808.
Chen,H.-W., Ruan,B., Yu,M., Wang,J. and Julin,D.A. ( (1997) ) The RecD subunit of the RecBCD enzyme from Escherichia coli is a single-stranded DNA-dependent ATPase. J. Biol. Chem., , 272, , 10072–10079.
Dillingham,M.S., Spies,M. and Kowalczykowski,S.C. ( (2003) ) RecBCD enzyme is a bipolar DNA helicase. Nature, , 423, , 893–897.
Taylor,A.F. and Smith,G.R. ( (2003) ) RecBCD enzyme is a DNA helicase with fast and slow motors of opposite polarity. Nature, , 423, , 889–893.
Handa,N., Ohashi,S., Kusano,K. and Kobayashi,I. ( (1997) ) *, a -related 11-mer sequence partially active in an E. coli recC* strain. Genes Cells, , 2, , 525–536.
Smith,G.R. and Stahl,F.W. ( (1985) ) Homologous Recombination Promoted by Chi Sites and RecBC Enzyme of Escherichia coli. BioEssays, , 2, , 244–249.
Smith,G.R., Amundsen,S.K., Dabert,P. and Taylor,A.F. ( (1995) ) The initiation and control of homologous recombination in Escherichia coli. Phil. Trans. R. Soc. Lond. B., , 347, , 13–20.
Smith,G.R., Kunes,S.M., Schultz,D.W., Taylor,A. and Triman,K.L. ( (1981) ) Structure of chi hotspots of generalized recombination. Cell, , 24, , 429–436.
Taylor,A.F., Schultz,D.W., Ponticelli,A.S. and Smith,G.R. ( (1985) ) RecBC enzyme nicking at Chi sites during DNA unwinding: location and orientation-dependence of the cutting. Cell, , 41, , 153–163.
Ponticelli,A.S., Schultz,D.W., Taylor,A.F. and Smith,G.R. ( (1985) ) Chi-dependent DNA strand cleavage by RecBC enzyme. Cell, , 41, , 145–151.
Dixon,D.A. and Kowalczykowski,S.C. ( (1991) ) Homologous pairing in vitro stimulated by the recombination hotspot, Chi. Cell, , 66, , 361–371.
Dixon,D.A. and Kowalczykowski,S.C. ( (1995) ) Role of the Escherichia coli recombination hotspot, , in RecABCD-dependent homologous pairing. J. Biol. Chem., , 270, , 16360–16370.
Anderson,D.G. and Kowalczykowski,S.C. ( (1997) ) The translocating RecBCD enzyme stimulates recombination by directing RecA protein onto ssDNA in a Chi-dependent manner. Cell, , 90, , 77–86.
Cheng,K.C. and Smith,G.R. ( (1984) ) Recombinational hotspot activity of Chi-like sequences. J. Mol. Biol., , 180, , 371–377.
Cheng,K.C. and Smith,G.R. ( (1987) ) Cutting of chi-like sequences by the RecBCD enzyme of Escherichia coli. J. Mol. Biol., , 194, , 747–750.
Dixon,D.A. and Kowalczykowski,S.C. ( (1993) ) The recombination hotspot is a regulatory sequence that acts by attenuating the nuclease activity of the E. coli RecBCD enzyme. Cell, , 73, , 87–96.
Spies,M., Bianco,P.R., Dillingham,M.S., Handa,N., Baskin,R.J. and Kowalczykowski,S.C. ( (2003) ) A molecular throttle: the recombination hotspot chi controls DNA translocation by the RecBCD helicase. Cell, , 114, , 647–654.
Taylor,A.F. and Smith,G.R. ( (1995) ) Strand specificity of nicking of DNA at Chi sites by RecBCD enzyme. J. Biol. Chem., , 270, , 24459–24467.
Anderson,D.G., Churchill,J.J. and Kowalczykowski,S.C. ( (1997) ) Chi-activated RecBCD enzyme possesses 5'-3' nucleolytic activity, but RecBC enzyme does not: evidence suggesting that the alteration induced by Chi is not simply ejection of the RecD subunit. Genes Cells, , 2, , 117–128.
Anderson,D.G. and Kowalczykowski,S.C. ( (1998) ) SSB protein controls RecBCD enzyme nuclease activity during unwinding: a new role for looped intermediates. J. Mol. Biol., , 282, , 275–285.
Wang,J., Chen,R. and Julin,D.A. ( (2000) ) A single nuclease active site of the Escherichia coli RecBCD enzyme catalyzes single-stranded DNA degradation in both directions. J. Biol. Chem., , 275, , 507–513.
Taylor,A.F. and Smith,G.R. ( (1992) ) RecBCD enzyme is altered upon cutting DNA at a Chi recombination hotspot. Proc. Natl Acad. Sci. USA, , 89, , 5226–5230.
Dixon,D.A., Churchill,J.J. and Kowalczykowski,S.C. ( (1994) ) Reversible inactivation of the Escherichia coli RecBCD enzyme by the recombination hotspot in vitro: evidence for functional inactivation or loss of the RecD subunit. Proc. Natl Acad. Sci. USA, , 91, , 2980–2984.
Taylor,A.F. and Smith,G.R. ( (1999) ) Regulation of homologous recombination: Chi inactivates RecBCD enzyme by disassembly of the three subunits. Genes Dev., , 13, , 890–900.
Bianco,P.R. and Kowalczykowski,S.C. ( (1997) ) The recombination hotspot Chi is recognized by the translocating RecBCD enzyme as the single strand of DNA containing the sequence 5'-GCTGGTGG-3'. Proc. Natl Acad. Sci. USA, , 94, , 6706–6711.
Borer,P.N. ( (1975) ) In Fasman,G. D. (ed.), Handbook of Biochemistry and Molecular Biology, 3rd edn. CRC Press, Cleveland, OH, Vol. 1, p. 589.
Korangy,F. and Julin,D.A. ( (1992) ) Alteration by site-directed mutagenesis of the conserved lysine residue in the ATP-binding consensus sequence of the RecD subunit of the Escherichia coli RecBCD enzyme. J. Biol. Chem., , 267, , 1727–1732.
Korangy,F. and Julin,D.A. ( (1992) ) Enzymatic effects of a lysine-to-glutamine mutation in the ATP-binding consensus sequence in the RecD subunit of the RecBCD enzyme from Escherichia coli. J. Biol. Chem., , 267, , 1733–1740.
Sambrook,J., Fritsch,E.F. and Maniatis,T. ( (1989) ) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Chamberlin,M. and Julin,D.A. ( (1996) ) Interactions of the RecBCD enzyme from Escherichia coli and its subunits with DNA, elucidated from the kinetics of ATP and DNA hydrolysis with oligothymidine substrates. Biochemistry, , 35, , 15949–15961.
Korangy,F. and Julin,D.A. ( (1992) ) A Mutation in the Consensus ATP-Binding Sequence of the RecD Subunit Reduces the Processivity of the RecBCD Enzyme from Escherichia coli. J. Biol. Chem., , 267, , 3088–3095.
Amundsen,S.K., Neiman,A.M., Thibodeaux,S.M. and Smith,G.R. ( (1990) ) Genetic dissection of the biochemical activities of RecBCD enzyme. Genetics, , 126, , 25–40.
Schultz,D.W., Taylor,A.F. and Smith,G.R. ( (1983) ) Escherichia coli RecBC pseudorevertants lacking chi recombinational hotspot activity. J. Bacteriol., , 155, , 664–680.
Arnold,D.A., Handa,N., Kobayashi,I. and Kowalczykowski,S.C. ( (2000) ) A novel, 11 nucleotide variant of chi, chi*: one of a class of sequences defining the Escherichia coli recombination hotspot chi. J. Mol. Biol., , 300, , 469–479.
Ganesan,S. and Smith,G.R. ( (1993) ) Strand-specific binding to duplex DNA ends by the subunits of the Escherichia coli RecBCD enzyme. J. Mol. Biol., , 229, , 67–78.
Hickson,I.D., Atkinson,K.E., Hutton,L., Tomkinson,A.E. and Emmerson,P.T. ( (1984) ) Molecular amplification and purification of the E. coli recC gene product. Nucleic Acids Res., , 12, , 3807–3819.
Korangy,F. and Julin,D.A. ( (1993) ) Kinetics and processivity of ATP hydrolysis and DNA unwinding by the RecBC enzyme from Escherichia coli. Biochemistry, , 32, , 4873–4880.
Phillips,R.J., Hickleton,D.C., Boehmer,P.E. and Emmerson,P.T. ( (1997) ) The RecB protein of Escherichia coli translocates along single-stranded DNA in the 3' to 5' direction: a proposed ratchet mechanism. Mol. Gen. Genet., , 254, , 319–329.(Avanti Kulkarni and Douglas A. Julin*)
* To whom correspondence should be addressed. Tel: +1 301 405 1821; Fax: +1 301 314 9121; Email: dj13@umail.umd.edu
ABSTRACT
RecBCD is an ATP-dependent helicase and exonuclease which generates 3' single-stranded DNA (ssDNA) ends used by RecA for homologous recombination. The exonuclease activity is altered when RecBCD encounters a Chi sequence (5'-GCTGGTGG-3') in double-stranded DNA (ds DNA), an event critical to the generation of the 3'-ssDNA. This study tests the effect of ssDNA oligonucleotides having a Chi sequence (Chi+) or a single base change that abolishes the Chi sequence (Chio), on the enzymatic activities of RecBCD. Our results show that a 14 and a 20mer with Chi+ in the center of the molecule inhibit the exonuclease and helicase activities of RecBCD to a greater extent than the corresponding Chio oligonucleotides. Oligonucleotides with the Chi sequence at one end, or the Chi sequence alone in an 8mer, failed to show Chi-specific inhibition of RecBCD. Thus, Chi recognition requires that Chi be flanked by DNA at either end. Further experiments indicated that the oligonucleotides inhibit RecBCD from binding to its dsDNA substrate. These results suggest that a specific site for Chi recognition exists on RecBCD, which binds Chi with greater affinity than a non-Chi sequence and is probably adjacent to non-specific DNA binding sites.
INTRODUCTION
The RecBCD enzyme of Escherichia coli is a multifunctional enzyme involved in the repair of double-strand DNA (dsDNA) breaks by RecA-dependent homologous recombination . RecBCD enzyme generates a 3'-ended stretch of single-stranded DNA (ssDNA) which can be used as a substrate by the RecA and single-stranded DNA binding (SSB) proteins to carry out homologous recombination (2). The RecBCD holoenzyme is a DNA-dependent ATPase (3), ATP-dependent helicase (4–6), an exonuclease (3' to 5' and 5' to 3') on single- and double-stranded DNA (3,7) and a ssDNA endonuclease (3). Of the three subunits of the enzyme, RecB is a DNA helicase, ATPase and nuclease (8–11), RecD is an ATPase and helicase (12–14) and RecC is speculated to play some role in Chi sequence recognition .
Critical to RecBCD-mediated generation of 3'-ended ssDNA is the encounter with and recognition of a specific octameric regulatory sequence called the Chi sequence (2,16,17). This sequence is recognized by RecBCD only when the enzyme encounters it from its 3'-end in dsDNA (19). The enzyme cleaves the Chi-containing strand 4 to 6 nt to the 3' side of Chi and continues to unwind beyond Chi, generating a 3'-terminated single strand (19,20). RecA protein is loaded specifically onto this 3'-terminated strand to initiate homologous recombination (21–23). Chi sites are thus hot-spots for recombination, stimulating recombination specifically to the 5' side of the sequence (17). Single base changes within the Chi sequence reduce or abolish its effectiveness both in vitro and in vivo (18,24,25).
The biochemical nature of the RecBCD–Chi interaction is not well understood. The effects of Chi on the enzyme found in previous in vitro work are varied and depend on the reaction conditions, particularly the ATP and magnesium ion concentrations. RecBCD has greater exonuclease activity on the 3'-ended strand, as compared to the 5'-ended strand, of a dsDNA substrate (21,26). However, once it encounters Chi, the translocating enzyme pauses (27), and subsequent cutting on the 3'-strand is attenuated leading to production of a Chi-specific DNA fragment, the length of which is equal to the length of the DNA from Chi to the 5' end of the strand (20,26). Also observed under some conditions is enhanced cleavage of the 5'-ended strand (containing the complement of the Chi sequence) after Chi (7,28,29). The enhanced cutting of this strand by the 5' to 3' exonuclease activity leads to the production of a second, upstream Chi-specific fragment (7,29). Based on these and other observations, two models were proposed to explain the effect of Chi on RecBCD nuclease activity. In one (30), there is a conformational change in the enzyme after it encounters and binds to Chi, which re-positions the nuclease active site in the RecB subunit to the opposite strand of the dsDNA substrate. A second model (31) suggests that the 3' Chi-ended strand is bound in a Chi-specific site and is thereby excluded from the nuclease active site in that process. The 5' strand loops around and can be bound to the nuclease active site of RecB after the Chi encounter, where the DNA strand can be cut by the enzyme. Both models suggest that the loss of exonuclease activity on the Chi-containing strand is because of a change in positioning of the nuclease active site relative to the substrate.
At the other end of the spectrum, there is evidence suggesting that there can be general inactivation of the enzyme and complete disassembly of the subunits of RecBCD upon or after the Chi encounter. These experiments, done at relatively low magnesium ion concentrations, show that the encounter with Chi in one dsDNA substrate prevents nicking at a second Chi in the same substrate (32) and inactivates RecBCD as a helicase and nuclease towards a second DNA substrate (32–34). Further, analysis by glycerol gradient sedimentation indicated that the three subunits dissociate after the enzyme encounters Chi in dsDNA under these conditions (34) and hence there is overall inactivation of the enzyme.
Most previous studies done on the effects of Chi on RecBCD involved the use of double-stranded DNA substrates containing Chi. However, a Chi sequence can be recognized as a single strand in a double-stranded substrate (35). Hence, in this study we have tested the effects of Chi in single-stranded oligonucleotides on the biochemical activities of RecBCD. We tested for specific recognition of Chi by looking for Chi-dependent inhibition of the exonuclease and helicase activities, and of dsDNA substrate binding. We also tested for up-regulation of the 5' to 3' exonuclease activity on dsDNA. For each Chi oligonucleotide tested, a corresponding control oligonucleotide with a single nucleotide change from the Chi sequence (5'-GCAGGTGG-3', called Chio) was also tested to establish specificity for the Chi sequence. The Chio sequence is not recognized in dsDNA in vitro, and does not function as Chi in vivo (24,25).
Our results show inhibition of the dsDNA exonuclease and DNA unwinding activities by the Chi sequence, but only in those oligonucleotides where the Chi sequence is flanked by DNA at both ends. There was no evidence that Chi in an oligonucleotide could prevent Chi recognition in dsDNA, nor was enhanced cleavage of the strand containing the complement of Chi observed. Instead, the results are consistent with the Chi-containing oligonucleotides binding to the enzyme and preventing binding of other DNA substrates. These Chi-containing oligonucleotides may be useful for further studies to gain insight into the physical effects of Chi on the enzyme and to examine where Chi is bound by RecBCD.
MATERIALS AND METHODS
Materials
Synthetic oligodeoxyribonucleotides were purchased from Invitrogen Corp. Some batches of the oligonucleotides were desalted by the manufacturer and were used without further purification. Others were purified by high performance liquid chromatography (HPLC) by the manufacturer, and were purified further using a QIAquick Nucleotide Removal Kit (Qiagen Corp.). Essentially identical results were obtained with the separate batches of oligonucleotides. Oligonucleotide stocks were prepared in deionized distilled water and their concentrations were determined from absorbance measurements at 260 nm. Absorption coefficients were calculated by the nearest-neighbor method using the Bench Mate program (http://www.roche-applied-science.com/benchmate) and parameters given in (36). Tritiated E.coli chromosomal DNA (4040 c.p.m./nmol nucleotide residues) was purified as described (37). Plasmid pTZ18R DNA (2860 bp; 122 300 c.p.m./nmol nucleotide) was tritiated in vivo as described (38) and was purified using the QIAfilter Plasmid Maxi Kit (QIAgen Corp.) following the manufacturer's instructions. pBR322+F plasmid DNA (18) was a gift from Dr Gerald Smith, Fred Hutchinson Cancer Research Center, and was isolated by the QIAfilter Plasmid Maxi Kit.
RecBCD enzyme was purified as described (11,37).
Methods
Double-stranded DNA exonuclease assay
Exonuclease reaction mixtures contained 50 mM Tris–HCl, pH 8.5, 10 mM MgCl2, 1 mM ATP, 0.67 mM DTT and 0.2 mg/ml BSA, with varied concentrations of oligonucleotides. The DNA substrate was either pTZ18R DNA (linearized by cleavage with SmaI, 5 μM nucleotide residues) or E.coli chromosomal DNA (40 μM nucleotide). The reactions were started by adding RecBCD and were incubated at 37°C. At pre-determined time intervals, 20 μl aliquots were removed, quenched in 100 μl ice-cold 10% trichloroacetic acid and 5 μl calf thymus DNA (0.5 mg/ml), and the concentration of acid-soluble nuclease reaction products was determined as described (37).
Pre-incubation of RecBCD with oligonucleotides
Some experiments were carried out by first incubating RecBCD with the oligonucleotide to be tested as an inhibitor, and then adding that to an exonuclease reaction mixture that contained the E.coli DNA substrate. The basic pre-incubation conditions were those used in (34) and contained 20 mM MOPS–KOH, pH 7.0, 3 mM magnesium acetate, 5 mM ATP, 20 mM DTT, 0.01 mg/ml BSA, 50 nM RecBCD, and varying concentrations of oligonucleotide. An aliquot of the pre-incubation mixture (typically 0.8 μl) was added to a dsDNA exonuclease reaction mixture (85 μl total volume) containing the reaction components given above and acidsoluble products were measured as above. A number of variations were made, including different incubation times, temperatures, and ATP and Mg2+ concentrations (see Results).
Reactions with Chi-containing double-stranded DNA substrate
pBR322+F plasmid DNA was linearized by treatment with HindIII restriction endonuclease, the 5' phosphate was removed with shrimp alkaline phosphatase (U. S. Biochemicals Corp.), and the DNA was purified using the QIAprep Spin Miniprep Kit (QIAgen Corp.). The linearized plasmid was then 5'-32P-end-labeled using polynucleotide kinase (New England Biolabs) and ATP (Amersham Biosciences, 3000 Ci/mmol) under conditions given in (39). The reaction products were repurified using the same kit and the amount of labeled DNA obtained was estimated by comparison of the band intensity to a DNA standard (High DNA Mass Ladder, Invitrogen Corp.) on a 1% agarose gel stained with ethidium bromide.
RecBCD reaction mixtures contained 25 mM Tris–acetate buffer, pH 7.5, 1 mM magnesium acetate, 1 mM ATP, 1 mM DTT, with 1.15 nM pBR322+F DNA molecules, 0.31 nM RecBCD enzyme, and varying concentration of the oligonucleotides as indicated in the figure legends. The reactions were started by adding the enzyme, incubated at 37°C, and 10 μl aliquots were quenched at predetermined times in 2.5 μl of quench solution (30% glycerol, 90 mM EDTA, 0.2% bromophenol blue and 1.3% SDS). The samples were run on a 1% agarose gel at 4.2 V/cm for 2 h or at 1.0 V/cm overnight, in 1 x TBE (40 mM Tris–borate, 1 mM EDTA, pH 8.0). The gel was dried and the radioactivity was detected using a Phosphorimager (Molecular Dynamics).
A variation of the above assay was carried out to test if the oligonucleotides can inhibit RecBCD that was pre-bound to the dsDNA substrate. RecBCD was mixed with the DNA and other reaction components except for ATP to allow the enzyme to bind the dsDNA. After a 2 min incubation at 37°C, a zero time-point aliquot was removed and quenched as above, and ATP alone or a mixture of ATP and the oligonucleotide was added to start the reaction. The final reaction mixtures contained 3 nM RecBCD, 2.3 nM DNA molecules, and other reactants as above. Samples were removed, quenched and analyzed on agarose gels as above.
Helicase assay
The helicase substrate was made by annealing two complementary DNA oligonucleotides (32 nt each): oligo1 and oligo2 . Oligo2 was 5' end-labeled using ATP and polynucleotide kinase as described in (39). The labeled oligo2 was purified (QIAgen Nucleotide Removal Kit), and analyzed by thin layer chromatography on polyethyleneimine–cellulose plates (J. T. Baker) in 1 M potassium phosphate buffer (pH 3.0) to ensure separation of residual ATP from the labeled oligonucleotide. The concentration of the purified labeled oligonucleotide was estimated from the amount of radioactivity (determined by scintillation counting) of samples taken before and after purification.
The double-stranded helicase substrate was prepared by mixing oligo2 (5 nM) with 95 nM of unlabeled oligo2 and 102 nM of its unlabeled complementary strand (oligo1) in 0.5 M NaCl, 0.2 M Tris–acetate, pH 7.5, and 10 mM MgCl2. The mixture was heated to 95–100°C for 2 min, cooled gradually to 28°C in the same water bath, placed on ice for 10–15 min, and stored at 4°C or used in the assay below.
The standard reaction was carried out in 25 mM Tris–acetate buffer, pH 7.5, 1 mM magnesium acetate, 1 mM ATP, 1 mM DTT, 0.31 nM RecBCD enzyme, varied concentrations of oligonucleotide, and 3 nM dsDNA substrate. The control reaction (no oligonucleotide) had 0.1 nM RecBCD. The reactions were incubated at 37°C and 5 μl aliquots were added to 2 μl of a quench solution containing 10% glycerol, 25 mM EDTA, 0.6% SDS, 0.03% bromophenol blue and a 5-fold excess of unlabeled oligo2. The quenched samples were run on 15% polyacrylamide gels in 1 x TBE. The gels were pre-run for 30 min at 12 V/cm, the samples were loaded and the gels were run at 12 V/cm for 2 h. The gels were dried and analyzed using a Phosphorimager.
Quantitation for reactions with Chi-containing double-stranded DNA and helicase assays
The amount of dsDNA unwound and of Chi fragment produced in these assays was quantitated using the ImageQuant software (Molecular Dynamics) on the Phosphorimager. Objects were drawn to encircle the bands to be quantitated and the volume of the band was integrated using the software. The background was determined by drawing an object on a part of the gel that did not contain sample and its integrated volume was subtracted from the values obtained for each band. The percentage values were then calculated relative to the volume of the dsDNA band in the zero time-point lane.
RESULTS
We designed the series of single-stranded oligonucleotides shown in Table 1 to test for Chi-specific effects on RecBCD enzyme. These consist of an 8-mer which is just the Chi sequence itself (8-merChi+), as well as sets of 14 and 20mers that are Chi with additional flanking DNA sequence. The larger oligonucleotides have Chi either in the center (14-merChi+, 20-merChi+), or at one or the other end (e.g. 3'14-merChi+, 5'14-merChi+, etc.). These Chi-ended oligonucleotides were tested because RecBCD normally sees Chi as it approaches Chi from one end (the 3' end) in one strand of a double-stranded DNA molecule. For each Chi+ oligonucleotide, we also tested a corresponding Chio control molecule with a single base change that inactivates the Chi-sequence (25).
Table 1. Oligonucleotides used as inhibitors
Inhibition of RecBCD exonuclease activity on dsDNA by Chi+ oligonucleotides
We first did experiments with the 8mers, 14-merChi+/o and 3'20-merChi+/o (see Table 1) to find the concentrations required to see inhibition of RecBCD exonuclease activity on a linear double-stranded DNA substrate (Figure 1). The longer oligomers are somewhat more potent inhibitors than the shorter oligomers (i.e. greater inhibition by the 20 and 14mers than the 8mers, at a given concentration). There was little difference in the degree of inhibition between Chio and Chi+ in the 8mers and 3'20mers, but the 14-merChi+ inhibited somewhat more than the 14-merChio, at a given concentration of each oligomer.
Figure 1. Concentration dependence of inhibition of exonuclease activity by oligodeoxyribonucleotides. Reaction mixtures contained 50 mM Tris–HCl, pH 8.5, 10 mM MgCl2, 40 μM ATP, 40 μM E. coli chromosomal DNA substrate (4040 c.p.m./nmol), and 0.31 nM RecBCD. The mixtures were incubated at 37°C and aliquots were quenched at the indicated times in ice-cold 10% trichloroacetic acid. The concentration of acid-soluble nucleotides was determined as described in Materials and Methods. (A) 8-merChio (left panel) or 8-merChi+ (right panel) were present at 0 μM (circles), 5 μM (triangles), 30 μM (squares) or 50 μM (diamonds). (B) 14-merChio (left panel) or 14-merChi+ (right panel) were present at 0 μM (circles), 1.5 μM (triangles), 5 μM (squares) or 10 μM (diamonds). (C) 3'20-merChio (left panel) or 3'20-merChi+ (right panel) were present at 0 μM (circles), 0.5 μM (triangles), 5 μM (squares) or 10 μM (diamonds).
These oligonucleotide concentrations were then used in the experiments in Figures 2–4, and in the assays of other RecBCD enzyme activities shown in subsequent figures. The effect of the 14mer oligonucleotide series on RecBCD-catalyzed nuclease reactions on linear double-stranded DNA is shown in Figure 2. The 14mer with Chi in the center and flanked by 3 nt of additional DNA on both ends (14-merChi+) was a more potent inhibitor of the RecBCD reaction than was an identical oligonucleotide except having the single base change to make the Chio sequence (Figure 2A). Both oligonucleotides (Chi+ and Chio) are inhibitors at the concentrations tested, but the reaction rate with the 14-merChi+ was about 2.2-fold less than that when the 14-merChio was present. We also tested a 14mer that is a scrambled version of the 14-merChi+, as an inhibitor in these experiments. This scrambled 14mer (at 15 μM) inhibited to a similar degree as the same concentration of 14-merChio, and it inhibited less than the 14-merChi+ (data not shown). The results with the Chi-ended 14mers are shown in Figure 2B and C. These molecules are also inhibitors of the RecBCD-catalyzed reactions, but there is little or no difference in the efficacy of the Chi+ and Chio oligonucleotides, when the Chi-sequence is situated entirely at one or the other end (3' or 5') of the DNA molecule. The results in Figure 2 are consistent with there being Chi-specific inhibition of RecBCD by the 14mers, but only when the Chi sequence is flanked by additional DNA at both ends.
Figure 2. Effect of 14mers on the exonuclease activity of RecBCD. Reactions containing 50 mM Tris–HCl, pH 8.5, 10 mM MgCl2, 1 mM ATP, 0.67 mM DTT, 0.2 mg/ml BSA, linear pTZ18R DNA (5 μM nucleotides) and oligonucleotides were started by adding 0.03 nM RecBCD. The concentration of acid-soluble nucleotides was determined as described in Materials and Methods, and values were normalized against the averaged 6 min time point for the corresponding control reaction. Reactions were done in triplicate and the error bars show the standard deviation of the averaged individual determinations. (A) Reaction mixtures contained 15 μM 14-merChi+ (open triangles), 15 μM 14-merChio (black squares) or no oligonucleotide (black circles). (B) Reaction mixtures contained 15 μM 5'14-merChi+ (open triangles), 15 μM 5'14-merChio (black squares) or no oligonucleotide (black circles). (C) Reaction mixtures contained 15 μM 3'14-merChi+ (open triangles), 15 μM 3'14-merChio (black squares), or no oligonucleotide (black circles).
Figure 3. Effect of 20mers on the exonuclease activity of RecBCD. Reactions were as in Figure 2 except that the DNA substrate was E.coli DNA (40 μM nt) and they were started by adding 0.3 nM RecBCD. Reactions were done in triplicate and normalized as in Figure 2. The error bars show the standard deviation of the averaged individual determinations. (A) Reaction mixtures contained 2 μM 20-merChi+ (open triangles), 2 μM 20-merChio (black squares), or no oligonucleotide (black circles). (B) Reaction mixtures contained 2 μM 5'20-merChi+ (open triangles), 2 μM 5'20-merChio (black squares), or no oligonucleotide (black circles). (C) Reaction mixtures contained 2 μM 3'20-merChi+ (open triangles), 2 μM 3'20-merChio (black squares) or no oligonucleotide (black circles).
Figure 4. Effect of 8mers on the exonuclease activity of RecBCD. Reactions were as in Figure 2 with linear pTZ18R DNA (5 μM nt), 0.03 nM RecBCD, and 50 μM 8-merChi+ (open triangles) 50 μM 8-merChio (black squares) or no oligonucleotide (black circles). Reactions were done in triplicate and normalized as in Figure 2. Error bars are the standard deviations.
A similar pattern is observed with Chi sequences in the 20mers. Again, the Chi+ 20mer is a more potent inhibitor that the Chio 20mer, when Chi is in the center of the molecule (20-merChi+ versus 20-merChio; Figure 3A). As with the 14mers, there is little if any difference in the degree of inhibition when the Chi+ and Chio sequences are at either the 5' end (Figure 3B) or the 3' end (Figure 3C) of the molecule. A second 20-merChio with a different mutant Chi sequence was also tested in this assay. It inhibited similar to the 20-merChio shown in Figure 3A, and was less potent than the 20-merChi+ (data not shown). This second mutant Chi sequence was also shown to be non-functional as Chi in previous work (24,25).
We also tested the 8-merChi+, which is simply the Chi sequence with no additional flanking DNA, and 8-merChio, in experiments such as that shown in Figure 4. As with the larger oligonucleotides with Chi at one end (Figures 2B and C and 3B and C), there is no difference in the inhibition by Chi+ versus Chio in the 8mers (Figure 4).
Exonuclease assays were also done using RecBCD enzyme that was first pre-incubated with the oligonucleotides and then diluted into the exonuclease reaction mixtures (see Materials and Methods). Previous experiments have shown that RecBCD that is allowed to react first with a Chi-containing dsDNA molecule under some conditions, loses activity towards a second dsDNA substrate (32–34). We decided to test whether such a ‘trans’ effect of Chi could be detected using the single-stranded Chi-containing oligonucleotides. The enzyme and oligonucleotides were pre-incubated at various concentrations, temperatures (0, 23 or 37°C), Mg2+, ATP, and oligonucleotide concentrations, and for various times (5 min to overnight) and then diluted into the nuclease reaction mixture. The oligonucleotide concentrations in most pre-incubations were 0.5–2.5 μM, with 50 nM RecBCD. We expect substantial binding of the oligonucleotides to the enzyme under these conditions, given previously measured dissociation constants for single-stranded DNA molecules of similar size (40). (The oligonucleotide concentration was much lower in the exonuclease reaction mixture itself than in the pre-incubation, due to the nearly 100-fold dilution.) In no case was there a significant difference in the amount of nuclease activity obtained with enzyme incubated with a Chi+ versus a Chio oligonucleotide (data not shown; the 14-merChi+/o, the 8-merChi+/o and 3'20-merChi+/o were tested). Chi-specific inhibition was seen only in reactions using RecBCD that was incubated with the 14-merChi+ (20 μM) and then diluted 2-fold into the nuclease reaction mixture. Here the degree of inhibition was the same as that when 10 μM 14-merChi+ was added directly to the nuclease reaction mixture (data not shown). Thus we do not observe Chi-dependent inactivation of RecBCD as has been observed when RecBCD encounters Chi in dsDNA under some reaction conditions (32–34).
Reactions with Chi-containing double-stranded DNA
The oligonucleotides were next tested for their ability to inhibit the production of a Chi-specific fragment that arises from attenuation of nuclease activity on the Chi containing (3'-ended) strand of dsDNA. The substrate used in this reaction, pBR322F+, is a 4.3 kb plasmid with a Chi site 1463 bp from one end after cleavage with HindIII . Recognition of this site by RecBCD, followed by attenuation of its nuclease activity at that point, leads to the generation of a 1463 nt 5'-32P-labeled fragment of ssDNA, the downstream Chi fragment (Figure 5A). Activation of opposite strand cleavage at the Chi sequence would lead to production of a 3000 nt upstream 5'-32P-labeled Chi fragment (7,28). (The upstream Chi fragment and uncleaved full-length ssDNA are indistinct at the ATP and Mg2+ concentrations used for our experiments and because of the absence of SSB protein, which stabilizes the single-stranded products . Experiments were carried out to test whether a Chi-containing oligonucleotide could affect Chi recognition on a competing dsDNA substrate (as measured by the amount of downstream Chi-fragment obtained). The reaction could be altered in the presence of a Chi+ oligonucleotide in one of two ways: qualitatively (e.g. different relative amounts of the downstream and upstream Chi fragments) or quantitatively (less overall reaction of the dsDNA).
Figure 5. Effect of Chi-centered 14mers on the reaction with linear, Chi-containing dsDNA. (A) HindIII-linearized pBR322F+ plasmid substrate (4.3 kb) used in the reaction. Cutting 4–6 nt on the 3' side of the Chi site in the top strand (by a RecBCD enzyme approaching from the right end as drawn) results in a 1.46 kb 5'-32P-labeled downstream Chi fragment; cutting of the lower strand after Chi results in a 3 kb 5'-32P-labeled upstream Chi fragment. (B) Reactions contained 25 mM Tris–acetate, pH 7.5, 1 mM magnesium acetate, 1 mM ATP, 1 mM DTT, 1.15 nM (molecules) of pBR322F+ plasmid substrate and 10 μM Chi-centered 14-merChi+ or 14-merChio. The control reaction contained no oligonucleotide. Reactions were started by adding 0.3 nM RecBCD and incubated at 37°C. Aliquots were quenched at the indicated times, run on a 1% agarose gel and the gel scanned using a Phosphorimager. (C) The percentage dsDNA remaining at each time point was quantitated as described in Materials and Methods. 14-merChi+ (open triangles), 14-merChio (black squares), control reaction (black circles). Reactions containing oligonucleotide were done in duplicate and the error bars show the range of the individual determinations. (Some error bars are obscured by the plot symbols.) (D) The amount of Chi fragment produced with time (as a % of the dsDNA at time zero) was calculated as described in Materials and Methods. 14-merChi+ (open triangles), 14-merChio (black squares), control reaction (black circles). Error bars show the range of the individual determinations.
The results show a decrease in the extent of DNA unwinding (helicase activity) and of production of the downstream Chi fragment. The results for the 14-merChi+/o are shown in Figure 5B. Figure 5C shows a plot of the percentage of the dsDNA remaining with time during the reaction. While essentially all of the DNA is unwound and degraded within 2 min in the control reaction, and 70% has disappeared in 3 min in the presence of the 14-merChio, 30% of the dsDNA has reacted in 3 min in the 14-merChi+ reaction. The total amount of downstream Chi fragment obtained is also less when the 14-merChi+ is present (Figure 5D) presumably because of the inhibition of the unwinding reaction shown in Figure 5C. Thus, there is a quantitative difference in the degree of inhibition by the Chi+ versus Chio oligonucleotides.
The results with the 20-merChi+/o are shown in Figure 6. Again, the 20-merChi+ inhibits the helicase activity much better than the 20-merChio. After 3 min 15% of the dsDNA substrate is unwound in the presence of the 20-merChi+ whereas 70% is unwound in the presence of the 20-merChio (Figure 6B). The 20-merChi+ also inhibited production of the downstream Chi band product more than did the 20-merChio (Figure 6C).
Figure 6. Effect of Chi centered 20mers on the reaction with linear, Chi-containing dsDNA. (A) Reaction conditions were as described for Figure 5, with 2 μM 20-merChi+, 20-merChio or no oligonucleotide (control reaction). (B) The percentage dsDNA remaining versus time with the 20-merChi+ (open triangles), 20-merChio (black squares) and control reaction (black circles). Reactions containing oligonucleotide were done in duplicate and the error bars show the range of the individual determinations. (C) Amount of Chi fragment was calculated as in Figure 5D. 20-merChi+ (open triangles), 20-merChio (black squares), control reaction (black circles). Error bars show the range of the individual determinations.
Results with the 8mer are shown in Figure 7. There is little difference between the reactions containing the 8-merChi+ and 8-merChio oligonucleotide, again confirming that the Chi site does not seem to be specifically recognized or distinguished from the Chio site in the 8mer constructs. There is also little difference in the inhibition by the oligomers (14 or 20mers) having Chi+ or Chio at either the 3' or 5' end, in this assay (data not shown).
Figure 7. Effect of 8mers on the reaction with linear, Chi-containing dsDNA. (A) Reaction conditions were as described for Figure 5 with 25 μM 8-merChi+, 8-merChio, or no oligonucleotide. (B) Percentage dsDNA remaining versus time with the 8-merChi+ (open triangles), 8-merChio (black squares) and control reaction (black circles). Reactions containing oligonucleotide were done in duplicate and the error bars show the range of the individual determinations. (C) Amount of Chi fragment was calculated as described in Figure 5D. 8-merChi+ (open triangles), 8-merChio (black squares), control reaction (black circles). Error bars show the range of the individual determinations.
Effect of oligonucleotides on RecBCD helicase activity
Since the above set of assays revealed inhibition of unwinding by the oligonucleotides with Chi+ in the center (14 and 20mers), we decided to test these constructs for inhibition of DNA unwinding directly using a small (32 bp) dsDNA substrate. Results for the 14mers are as shown in Figure 8. While 75% of the substrate is unwound in 3 min in the presence of the 14-merChio oligonucleotide, only 30% is unwound in the reaction containing the 14-merChi+ (Figure 8B).
Figure 8. Effect of Chi-centered 14mers on the helicase activity of RecBCD. (A) Reactions contained 25 mM Tris–acetate, pH 7.5, 1 mM magnesium acetate, 1 mM ATP, 1 mM DTT, 3 nM 5'-32P-labeled double-stranded 32 bp substrate (see Materials and Methods) and 15 μM Chi-centered 14-merChi+ or 14-merChio. Reactions were started by adding 0.3 nM RecBCD and were incubated at 37°C. Aliquots were quenched at the indicated times and run on a 15% polyacrylamide gel. The control reaction contained 0.1 nM RecBCD and no oligonucleotide. (B) The amount of double-strand band in each lane was quantitated as described in Materials and Methods and plotted as a function of time. 14-merChi+ (open triangles), 14-merChio (black squares), no oligonucleotide (black circles). Reactions containing oligonucleotide were done in duplicate and the error bars show the range of the individual determinations.
Effect of oligonucleotides on pre-formed RecBCD-dsDNA complexes
The observed inhibition of the exonuclease and helicase activities might arise if oligonucleotides interfere with binding of RecBCD to the dsDNA substrate. In order to test for this effect, the Chi-cleavage assay was repeated with an initial pre-incubation of the enzyme and the 4.3 kb plasmid substrate for 2 min at 37°C. This pre-incubation would allow the enzyme to bind the duplex ends first without interference by the competing oligonucleotides (41). The reactions were started by adding either ATP alone (to the control reaction) or an ATP–oligonucleotide mixture and samples were removed within the first 30 s. A higher enzyme:substrate ratio was used for this set of reactions than for those shown above, to ensure that most of the duplex ends have an enzyme molecule bound to them. The concentrations used and other details are as described in the Materials and Methods section and in the figure legend.
The 20mers inhibit very strongly when RecBCD is added to reaction mixtures containing both the DNA substrate and the oligonucleotide inhibitor without pre-incubation (Figure 9A). All of the linear substrate is consumed within 20 s in the control reaction, while very little substrate has been consumed in the first 20 s in those reactions containing the oligonucleotide inhibitors. In contrast, the enzyme that is allowed to bind first to the dsDNA substrate (Figure 9B) is impervious to inhibition by the oligonucleotides, and most of the substrate is consumed by the first time point whether or not an oligonucleotide is present in the mixture. This shows that allowing the enzyme to bind the dsDNA substrate first ensures that oligonucleotides, even those with the Chi sequence, are unable to affect the enzyme once it initiates its helicase/nuclease activities on the dsDNA. The same result was obtained with the 8mers (data not shown).
Figure 9. Effect of pre-incubation of dsDNA substrate with RecBCD. (A) Reactions done without pre-incubation were started by adding RecBCD enzyme to mixtures that contained all other reaction components. The final reaction mixture contained HindIII-linearized pBR322F+ (2.3 nM molecules), 3 nM RecBCD, 2 μM 20-merChi+ or 20-merChio, and other components as in Figure 5. The control reaction contained no oligonucleotide. Aliquots were quenched at the indicated times and run on a 1% agarose gel. (B) For pre-incubated reactions, HindIII-linearized pBR322F+ (2.3 nM molecules) was incubated for 2 min at 37°C with 3 nM RecBCD and all other reaction components except ATP and the oligonucleotides. The reaction was started by adding a mixture of ATP and 20-merChi+ or 20-merChio (final concentration 2 μM).
DISCUSSION
The results in this paper show that all of the oligonucleotides tested act as inhibitors of RecBCD. The overall inhibition is length dependent, the longer the oligonucleotide the more effective the inhibition. More significantly, the Chi+ oligonucleotides inhibit the activities of RecBCD much more than do the Chio oligonucleotides, and this distinction is only seen in constructs where the Chi+ or Chio site is flanked by sequence at either end. This suggests that recognition of the Chi sequence may be a location-specific effect and flanking sequences may play an important role in this interaction. The oligonucleotides all prevent RecBCD binding to the dsDNA substrate, but they apparently do not cause any qualitative or physical changes in the enzyme as has been observed previously with Chi-containing dsDNA (32–34).
These results suggest that a specific site exists on RecBCD that recognizes Chi and binds to it with greater affinity than to a non-Chi sequence. The greater inhibition seen due to the Chi+ sequence only in the Chi-centered constructs suggests RecBCD needs a DNA ‘handle’ region to be able to recognize Chi. The flanking DNA in the 14-mers and 20-mers would serve as this handle enabling Chi recognition. We envision an elongated DNA binding site, a part of which binds non-specifically to the DNA handle. This binding site merely serves as an anchor to enable better contact between RecBCD and the oligonucleotide but it does not recognize Chi (Figure 10A). A site adjacent to the non-specific site may be specific for Chi recognition and binding. The DNA flank in the oligonucleotide would then be important as it would ensure that the Chi sequence itself does not become the handle and go unrecognized. This may explain the lack of Chi recognition in the Chi-ended 14 and 20mers, and also the 8mer. The Chi site in all these constructs becomes part of the handle sequence itself and hence goes unrecognized (Figure 10B).
Figure 10. Model for location-specific recognition of the Chi+ sequence. (A) Interaction of RecBCD with the Chi sequence in Chi-centered oligonucleotide. Proposed ‘handle’ region on the ends of the oligonucleotide is contacted by non-specific DNA binding site in the N-terminal domain of RecB (+++). This site grips the handle, but cannot recognize the Chi sequence. The Chi site is recognized by a specific recognition site adjacent to the non-specific site. (B) Interaction of RecBCD with Chi sequence in Chi-ended oligonucleotide and the 8mer. The Chi sequence at the 3' end now becomes the handle itself and hence goes unrecognized. The same is true of the Chi+ in the 8mer.
The fact that the oligonucleotides (whether Chi+ or Chio) do not inhibit the enzyme which has been allowed to bind to the dsDNA substrate first (Figure 9) shows that their binding site(s) become inaccessible to the oligonucleotide when the enzyme has bound to the dsDNA. Thus our results do not provide evidence for a Chi-specific site that can be filled by a Chi-containing oligonucleotide once the enzyme starts to act on dsDNA. Instead, the non-specific site that binds the ‘handle’ sequence may be part of or adjacent to a non-specific DNA binding site that is also involved in the helicase activity. This would allow the enzyme to scan the DNA for the Chi site as the DNA is being unwound. The requirement for a DNA flank also suggests that if RecBCD were to start unwinding a DNA having a Chi site at the very end, it would be likely to skip the site and continue DNA degradation until it encounters the next Chi.
It is still not known which subunit(s) contains the Chi-specific binding site. Mutations have been isolated in both the recB and recC genes that affect the response to Chi in vivo (42,43). The recC mutant enzyme responds to a variant of the Chi sequence as might be expected if the mutation has changed a Chi-specific binding site (15,44), suggesting that RecC has the Chi-specific binding site. However, in a photocrosslinking experiment probing RecBCD bound to the end of linear dsDNA, RecC was crosslinked to the 5'-ended strand, while RecB was found to bind the 3'-ended strand of dsDNA (45). Since Chi must be in the 3'-ended strand for recognition by the translocating RecBCD enzyme, this would suggest that RecB might have the Chi-specific site. RecB must contain non-specific DNA binding site(s), since the protein is a DNA-dependent ATPase, helicase and endonuclease (8,11). The isolated RecC protein does not bind DNA (46), but RecC stimulates the ATP hydrolysis and helicase activities of RecB (40,47,48). A simple model is that the non-specific DNA binding sites shown in Figure 10 reside in RecB. A DNA binding region in RecC that is specific for Chi could be at an interface with RecB, and contribute greater affinity for Chi than for non-specific DNA to the RecBCD complex. In preliminary photocrosslinking experiments using RecBCD and the 14-merChi+, we find that the RecC protein appears to be the main subunit crosslinked (A. Kulkarni and D. A. Julin, unpublished results). Efforts are underway to identify more precisely the specific amino acid residues involved in the crosslinking.
ACKNOWLEDGEMENTS
This work was supported by grant # GM39777 from the NIH.
REFERENCES
Kuzminov,A. ( (1999) ) Recombinational repair of DNA damage in Escherichia coli and bacteriophage lambda. Microbiol. Mol. Biol. Rev., , 63, , 751–813.
Kowalczykowski,S.C. ( (2000) ) Initiation of genetic recombination and recombination-dependent replication. Trends Biochem. Sci., , 25, , 156–165.
Goldmark,P.J. and Linn,S. ( (1972) ) Purification and properties of the recBC DNase of Escherichia coli K-12. J. Biol. Chem., , 247, , 1849–1860.
MacKay,V. and Linn,S. ( (1974) ) The mechanism of degradation of duplex deoxyribonucleic acid by the recBC enzyme of Escherichia coli K-12. J. Biol. Chem., , 249, , 4286–4294.
Taylor,A. and Smith,G.R. ( (1980) ) Unwinding and rewinding of DNA by the RecBC enzyme. Cell, , 22, , 447–457.
Roman,L.J. and Kowalczykowski,S.C. ( (1989) ) Characterization of the helicase activity of the Escherichia coli RecBCD enzyme using a novel helicase assay. Biochemistry, , 28, , 2863–2873.
Anderson,D.G. and Kowalczykowski,S.C. ( (1997) ) The recombination hot spot is a regulatory element that switches the polarity of DNA degradation by the RecBCD enzyme. Genes Dev., , 11, , 571–581.
Boehmer,P.E. and Emmerson,P.T. ( (1992) ) The RecB subunit of the Escherichia coli RecBCD enzyme couples ATP hydrolysis to DNA unwinding. J. Biol. Chem., , 267, , 4981–4987.
Hsieh,S. and Julin,D.A. ( (1992) ) Alteration by site-directed mutagenesis of the conserved lysine residue in the consensus ATP-binding sequence of the RecB protein of Escherichia coli. Nucleic Acids Res., , 20, , 5647–5653.
Yu,M., Souaya,J. and Julin,D.A. ( (1998) ) The 30-kDa C-terminal domain of the RecB protein is critical for the nuclease activity, but not the helicase activity, of the RecBCD enzyme from Escherichia coli. Proc. Natl Acad. Sci. USA, , 95, , 981–986.
Yu,M., Souaya,J. and Julin,D.A. ( (1998) ) Identification of the nuclease active site in the multifunctional RecBCD enzyme by creation of a chimeric enzyme. J. Mol. Biol., , 283, , 797–808.
Chen,H.-W., Ruan,B., Yu,M., Wang,J. and Julin,D.A. ( (1997) ) The RecD subunit of the RecBCD enzyme from Escherichia coli is a single-stranded DNA-dependent ATPase. J. Biol. Chem., , 272, , 10072–10079.
Dillingham,M.S., Spies,M. and Kowalczykowski,S.C. ( (2003) ) RecBCD enzyme is a bipolar DNA helicase. Nature, , 423, , 893–897.
Taylor,A.F. and Smith,G.R. ( (2003) ) RecBCD enzyme is a DNA helicase with fast and slow motors of opposite polarity. Nature, , 423, , 889–893.
Handa,N., Ohashi,S., Kusano,K. and Kobayashi,I. ( (1997) ) *, a -related 11-mer sequence partially active in an E. coli recC* strain. Genes Cells, , 2, , 525–536.
Smith,G.R. and Stahl,F.W. ( (1985) ) Homologous Recombination Promoted by Chi Sites and RecBC Enzyme of Escherichia coli. BioEssays, , 2, , 244–249.
Smith,G.R., Amundsen,S.K., Dabert,P. and Taylor,A.F. ( (1995) ) The initiation and control of homologous recombination in Escherichia coli. Phil. Trans. R. Soc. Lond. B., , 347, , 13–20.
Smith,G.R., Kunes,S.M., Schultz,D.W., Taylor,A. and Triman,K.L. ( (1981) ) Structure of chi hotspots of generalized recombination. Cell, , 24, , 429–436.
Taylor,A.F., Schultz,D.W., Ponticelli,A.S. and Smith,G.R. ( (1985) ) RecBC enzyme nicking at Chi sites during DNA unwinding: location and orientation-dependence of the cutting. Cell, , 41, , 153–163.
Ponticelli,A.S., Schultz,D.W., Taylor,A.F. and Smith,G.R. ( (1985) ) Chi-dependent DNA strand cleavage by RecBC enzyme. Cell, , 41, , 145–151.
Dixon,D.A. and Kowalczykowski,S.C. ( (1991) ) Homologous pairing in vitro stimulated by the recombination hotspot, Chi. Cell, , 66, , 361–371.
Dixon,D.A. and Kowalczykowski,S.C. ( (1995) ) Role of the Escherichia coli recombination hotspot, , in RecABCD-dependent homologous pairing. J. Biol. Chem., , 270, , 16360–16370.
Anderson,D.G. and Kowalczykowski,S.C. ( (1997) ) The translocating RecBCD enzyme stimulates recombination by directing RecA protein onto ssDNA in a Chi-dependent manner. Cell, , 90, , 77–86.
Cheng,K.C. and Smith,G.R. ( (1984) ) Recombinational hotspot activity of Chi-like sequences. J. Mol. Biol., , 180, , 371–377.
Cheng,K.C. and Smith,G.R. ( (1987) ) Cutting of chi-like sequences by the RecBCD enzyme of Escherichia coli. J. Mol. Biol., , 194, , 747–750.
Dixon,D.A. and Kowalczykowski,S.C. ( (1993) ) The recombination hotspot is a regulatory sequence that acts by attenuating the nuclease activity of the E. coli RecBCD enzyme. Cell, , 73, , 87–96.
Spies,M., Bianco,P.R., Dillingham,M.S., Handa,N., Baskin,R.J. and Kowalczykowski,S.C. ( (2003) ) A molecular throttle: the recombination hotspot chi controls DNA translocation by the RecBCD helicase. Cell, , 114, , 647–654.
Taylor,A.F. and Smith,G.R. ( (1995) ) Strand specificity of nicking of DNA at Chi sites by RecBCD enzyme. J. Biol. Chem., , 270, , 24459–24467.
Anderson,D.G., Churchill,J.J. and Kowalczykowski,S.C. ( (1997) ) Chi-activated RecBCD enzyme possesses 5'-3' nucleolytic activity, but RecBC enzyme does not: evidence suggesting that the alteration induced by Chi is not simply ejection of the RecD subunit. Genes Cells, , 2, , 117–128.
Anderson,D.G. and Kowalczykowski,S.C. ( (1998) ) SSB protein controls RecBCD enzyme nuclease activity during unwinding: a new role for looped intermediates. J. Mol. Biol., , 282, , 275–285.
Wang,J., Chen,R. and Julin,D.A. ( (2000) ) A single nuclease active site of the Escherichia coli RecBCD enzyme catalyzes single-stranded DNA degradation in both directions. J. Biol. Chem., , 275, , 507–513.
Taylor,A.F. and Smith,G.R. ( (1992) ) RecBCD enzyme is altered upon cutting DNA at a Chi recombination hotspot. Proc. Natl Acad. Sci. USA, , 89, , 5226–5230.
Dixon,D.A., Churchill,J.J. and Kowalczykowski,S.C. ( (1994) ) Reversible inactivation of the Escherichia coli RecBCD enzyme by the recombination hotspot in vitro: evidence for functional inactivation or loss of the RecD subunit. Proc. Natl Acad. Sci. USA, , 91, , 2980–2984.
Taylor,A.F. and Smith,G.R. ( (1999) ) Regulation of homologous recombination: Chi inactivates RecBCD enzyme by disassembly of the three subunits. Genes Dev., , 13, , 890–900.
Bianco,P.R. and Kowalczykowski,S.C. ( (1997) ) The recombination hotspot Chi is recognized by the translocating RecBCD enzyme as the single strand of DNA containing the sequence 5'-GCTGGTGG-3'. Proc. Natl Acad. Sci. USA, , 94, , 6706–6711.
Borer,P.N. ( (1975) ) In Fasman,G. D. (ed.), Handbook of Biochemistry and Molecular Biology, 3rd edn. CRC Press, Cleveland, OH, Vol. 1, p. 589.
Korangy,F. and Julin,D.A. ( (1992) ) Alteration by site-directed mutagenesis of the conserved lysine residue in the ATP-binding consensus sequence of the RecD subunit of the Escherichia coli RecBCD enzyme. J. Biol. Chem., , 267, , 1727–1732.
Korangy,F. and Julin,D.A. ( (1992) ) Enzymatic effects of a lysine-to-glutamine mutation in the ATP-binding consensus sequence in the RecD subunit of the RecBCD enzyme from Escherichia coli. J. Biol. Chem., , 267, , 1733–1740.
Sambrook,J., Fritsch,E.F. and Maniatis,T. ( (1989) ) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Chamberlin,M. and Julin,D.A. ( (1996) ) Interactions of the RecBCD enzyme from Escherichia coli and its subunits with DNA, elucidated from the kinetics of ATP and DNA hydrolysis with oligothymidine substrates. Biochemistry, , 35, , 15949–15961.
Korangy,F. and Julin,D.A. ( (1992) ) A Mutation in the Consensus ATP-Binding Sequence of the RecD Subunit Reduces the Processivity of the RecBCD Enzyme from Escherichia coli. J. Biol. Chem., , 267, , 3088–3095.
Amundsen,S.K., Neiman,A.M., Thibodeaux,S.M. and Smith,G.R. ( (1990) ) Genetic dissection of the biochemical activities of RecBCD enzyme. Genetics, , 126, , 25–40.
Schultz,D.W., Taylor,A.F. and Smith,G.R. ( (1983) ) Escherichia coli RecBC pseudorevertants lacking chi recombinational hotspot activity. J. Bacteriol., , 155, , 664–680.
Arnold,D.A., Handa,N., Kobayashi,I. and Kowalczykowski,S.C. ( (2000) ) A novel, 11 nucleotide variant of chi, chi*: one of a class of sequences defining the Escherichia coli recombination hotspot chi. J. Mol. Biol., , 300, , 469–479.
Ganesan,S. and Smith,G.R. ( (1993) ) Strand-specific binding to duplex DNA ends by the subunits of the Escherichia coli RecBCD enzyme. J. Mol. Biol., , 229, , 67–78.
Hickson,I.D., Atkinson,K.E., Hutton,L., Tomkinson,A.E. and Emmerson,P.T. ( (1984) ) Molecular amplification and purification of the E. coli recC gene product. Nucleic Acids Res., , 12, , 3807–3819.
Korangy,F. and Julin,D.A. ( (1993) ) Kinetics and processivity of ATP hydrolysis and DNA unwinding by the RecBC enzyme from Escherichia coli. Biochemistry, , 32, , 4873–4880.
Phillips,R.J., Hickleton,D.C., Boehmer,P.E. and Emmerson,P.T. ( (1997) ) The RecB protein of Escherichia coli translocates along single-stranded DNA in the 3' to 5' direction: a proposed ratchet mechanism. Mol. Gen. Genet., , 254, , 319–329.(Avanti Kulkarni and Douglas A. Julin*)