Pre-steady-state DNA unwinding by bacteriophage T4 Dda helicase reveals a monomeric molecular motor

Abstract

Helicases are molecular motor enzymes that unwind and translocate nucleic acids. One of the central questions regarding helicase activity is whether the process of coupling ATP hydrolysis to DNA unwinding requires an oligomeric form of the enzyme. We have applied a pre-steady-state kinetics approach to address this question with the bacteriophage T4 Dda helicase. If a helicase can function as a monomer, then the burst amplitude in the pre-steady state might be similar to the concentration of enzyme, whereas if the helicase required oligomerization, then the amplitude would be significantly less than the enzyme concentration. DNA unwinding of an oligonucleotide substrate was conducted by using a Kintek rapid quench-flow instrument. The substrate consisted of 12 bp adjacent to 12 nucleotides of single-stranded DNA. Dda (4 nM) was incubated with substrate (16 nM) in buffer, and the unwinding reaction was initiated by the addition of ATP (5 mM) and Mg²⁺ (10 mM). The reaction was stopped by the addition of 400 mM EDTA. Product formation exhibited biphasic kinetics, and the data were fit to the equation for a single exponential followed by a steady state. The amplitude of the first phase was 3.5 ± 0.2 nM, consistent with a monomeric helicase. The burst amplitude of product formation was measured over a range of enzyme and substrate concentrations and remained consistent with a functional monomer. Thus, Dda can rapidly unwind oligonucleotide substrates as a monomer, indicating that the functional molecular motor component of a helicase can reside within a single polypeptide.

Molecular motors are enzymes that transduce chemical energy to mechanical energy. A type of molecular motor that is required for almost all aspects of nucleic acid metabolism is helicase. Helicases are ubiquitous enzymes that unwind or transport double-stranded nucleic acids during replication, recombination, transcription, and DNA repair. The mechanism of this class of enzymes has received a great deal of attention during the past decade (1–4). Helicases are classified in several superfamilies (SFs) based on sequence homology. Because multiple DNA-binding sites are required for processive DNA unwinding, oligomerization can provide a means to provide these sites. Some helicases such as those in the DNA-B-like SF clearly function as hexamers (4). Hexameric helicases can be highly processive, because the DNA passes through a central channel created by the doughnut shape of the enzyme. The fact that many helicases function as hexamers as well as consideration of the necessity for multiple DNA-binding sites has led to the suggestion that oligomerization may be necessary for helicase function. Direct, functional evidence has not been presented for a monomeric helicase that can transduce the chemical energy available from hydrolyzing ATP to mechanical energy for DNA unwinding.

Attempts to address this question have relied on biophysical and biochemical techniques such as sedimentation equilibrium, size-exclusion chromatography, and chemical crosslinking. Negative results from such approaches have been interpreted typically in terms of a monomeric helicase (5–7). However, these approaches may fail to provide evidence for oligomerization, because protein–protein interactions that are required for optimal activity may be transient in nature (8). Others have used functional approaches in which the activity of a helicase is measured in the presence of an inactive, mutant form of the enzyme. Formation of heterooligomers between the wild-type helicase and mutated helicase can lead to reduction in activity, thus providing evidence for oligomerization (9). Although this approach may detect transient interactions due to the sensitivity of the kinetic measurements, the assumption that a mutated helicase will interact with the wild-type enzyme may not always hold. Point mutations may lead to alterations in the folding pattern of a mutant form of an enzyme, which could reduce or eliminate protein–protein interactions. Thus, weak protein–protein interactions may actually occur despite negative results from the array of experiments that are designed to investigate such interactions.

Positive results for a monomeric form of several helicases have been provided by x-ray crystallographic structures (10–12). The crystal structures of PcrA, NS3, and RecG helicases reveal monomers; however, crystal structures may not reveal transient interactions that are required for activity. Therefore, the crystal structure may not represent the optimally active form of the enzyme.

Thus, results from several laboratories using different approaches have led to different conclusions regarding the need, or lack thereof, for oligomerization. The debate has centered on the largest class of helicases, SF1. One enzyme from this family that has been studied intensively is Rep from Escherichia coli. Results from chemical crosslinking and gel-permeation chromatography led to the conclusion that DNA binding by Rep induces dimerization (13). Subsequent studies of Rep binding to single-stranded DNA (ssDNA) and double-stranded DNA as well as the effects of nucleotide cofactors on DNA binding led to the formulation of an “active rolling” mechanism for dimeric Rep helicase (reviewed in ref. 1). A different SF1 helicase, PcrA from Bacillus stearothermophilus, has been studied by x-ray crystallography and biochemical methods (reviewed in ref. 3). The crystal structure of PcrA revealed only a monomer even when bound to ssDNA. Additionally, chemical crosslinking and gel-permeation chromatography failed to provide evidence for dimerization even in the presence of ssDNA. Based on these studies, an “inchworm” mechanism was proposed for PcrA. Because PcrA shares 41% sequence identity with Rep, similar mechanisms for these two enzymes might be expected. However, it is possible that these enzymes exhibit some significant differences in their DNA-unwinding activity, which may account for the different results. A similar debate has surrounded NS3, an SF2 helicase from the hepatitis C virus. The crystal structure reveals a monomeric enzyme even when bound to ssDNA (11), whereas biochemical evidence suggests that an oligomeric form of the enzyme is required for optimal DNA-unwinding activity (9).

Pre-steady-state kinetic analysis has proven to be an extremely powerful tool for studying enzymes that act on DNA such as polymerases (14). This method measures the first cycle of product formation in the presence of excess substrate. The quantity of product formed during the first cycle reflects the quantity of substrate bound by the enzyme that satisfies the kinetic requirements for product formation. This quantity is referred to as the “burst amplitude” of the enzymatic reaction, which may indicate the quantity of active enzyme that is present in the reaction mixture. If a helicase can function as a monomer and is sufficiently processive, then the burst amplitude of product observed during DNA unwinding should be similar to the concentration of enzyme. If a helicase requires oligomerization or is highly nonprocessive, then the burst amplitude will be significantly less than the enzyme concentration.

Recently, pre-steady-state unwinding experiments were performed with Rep helicase under a variety of conditions (8). The results are most consistent with a requirement of multiple Rep molecules for optimal unwinding of an 18-bp duplex rather than a monomer. Similar experiments have yet to be performed for PcrA. Thus, definitive data supporting the function of a monomeric helicase in DNA unwinding is lacking. Here, we report pre-steady-state DNA-unwinding experiments with an SF1 helicase, termed Dda, from bacteriophage T4. The results provide clear evidence that a helicase from SF1 can function as a monomer.

Materials and Methods

Reagents.

Hepes, glycerol, EDTA, NaCl, isopropyl β-D-thiogalactoside, dextrose, and KOAc were from Fisher. T4 polynucleotide kinase was purchased from New England Biolabs. [γ-³²P]ATP was purchased from New England Nuclear. All oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA) and purified by PAGE (6) followed by HPLC analysis. No UV-absorbing species were present other than the DNA. The absorbance of the DNA was measured at 260 nm under conditions in which the DNA is denatured (0.2 M KOH), and the concentration was determined by calculating the extinction coefficient of the DNA (15). Dda was expressed and purified as described (6). The protein concentration was determined by UV absorbance in 6 M guanidine by using a calculated extinction coefficient (16) and by quantitative amino acid analysis (Vanderbilt University Protein Chemistry Facility, Nashville, TN). Each method resulted in similar values for the protein concentration, and the higher value, which was determined by UV absorbance, was used for all experiments reported here.

DNA Unwinding.

DNA-unwinding assays were performed by using a Kintek rapid chemical quench-flow instrument (Kintek, Austin, TX) maintained at 25°C (14). All concentrations listed are after mixing unless noted otherwise. Dda (at the concentrations indicated in the figure legends) was incubated with the indicated concentration of ³²P-labeled DNA substrate in reaction buffer (25 mM Hepes, pH 7.5/10 mM KOAc/0.1 mM EDTA/2 mM β-mercaptoethanol/0.1 mg/ml BSA). The reaction was initiated by the addition of ATP (5 mM) and Mg(OAc)₂ (10 mM). For some experiments, 5 μM poly(dT) was included with the ATP as a protein trap. The reaction was quenched at various times with 400 mM EDTA (concentration before mixing). The sample was either collected in a tube that contained an annealing trap or the annealing trap was added with the ATP as indicated in the figure legends. Samples were analyzed by 20% native PAGE, visualized with a Molecular Dynamics PhosphorImager, and quantitated with IMAGEQUANT software (Molecular Dynamics).

Results and Discussion

Single-Turnover Kinetics.

The Dda helicase from bacteriophage T4 has been suggested to play a role in early events in T4 replication and in recombination (17, 18). Dda is classified in helicase SF1 (19) and readily unwinds oligonucleotide substrates in a 5′-to-3′ direction (20–22). A substrate (24:12-mer) was designed for DNA-unwinding experiments that contained 12 nucleotides of ssDNA adjacent to 12 bp (Table 1). Similar to many helicases, Dda requires an ssDNA overhang adjacent to the duplex to initiate unwinding. The overhang is believed to be the strand on which the enzyme translocates (21, 22). Dda-catalyzed unwinding of the 24:12-mer was measured under single-turnover conditions with excess enzyme to determine the rate of unwinding and to measure the quantity of substrate that could be unwound in a single binding event. To ensure single-turnover conditions, Dda was incubated with 24:12-mer, and the reaction was initiated by the addition of ATP and Mg²⁺, along with poly(dT) (5 μM) as depicted in Fig. 1A. The poly(dT) was included in the reaction to serve as a protein trap such that any Dda that dissociated from the substrate would bind to the poly(dT). Reaction mixtures were analyzed by native PAGE to separate ssDNA products from duplex reactants (Fig. 1B). The quantity of ssDNA formed over time was determined by PhosphorImage analysis (23) and is plotted in Fig. 1C. The amplitude of product was ≥90% of the total substrate at each concentration of Dda, indicating that the DNA substrate was bound productively. The rate of unwinding was 31.7 ± 7.7 s⁻¹ at 100 nM Dda and 43.7 ± 9.5 s⁻¹ at 300 nM Dda. Thus, Dda unwound the 24:12-mer substrate very rapidly and efficiently under single-turnover conditions in the presence of excess enzyme.

Table 1.

DNA-unwinding substrates

DNA substrate	Sequence
24:12-mer	*5′-TTTTTTTTTTTTCGCTGATGTCGC-3′
	3′-GCGACTACAGCG-5′
28:16-mer	*5′-TTTTTTTTTTTTCGCTGATGTCGCCTGG-3′
	3′-GCGACTACAGCGGACC-5′

Fig 1.

DNA unwinding under single-turnover conditions. (A) Dda was incubated with ³²P-labeled DNA substrate (24:12-mer, 2 nM) in reaction buffer. ATP (5 mM) and Mg(OAc)₂ (10 mM) were added to initiate the unwinding reaction. poly(dT) (5 μM) was added at the same time as the ATP to serve as a protein trap (see Results and Discussion). At varying times, a quench solution (400 mM EDTA) was added to stop the reaction. The sample was collected in a Microcentrifuge tube that contained a 12-mer oligonucleotide that served as the annealing trap (see Results and Discussion). (B) Product (ssDNA) was separated from substrate (double-stranded DNA) by 20% native PAGE. (C) Product formation plotted vs. time under single-cycle conditions. The lines are the best fit of the data to a single exponential (Eq. 1) by using the program KALEIDAGRAPH.

Three experiments were performed in triplicate at each concentration of Dda providing an average and standard deviation for the amplitude (A) and the rate constant (k₁). In the presence of 100 nM Dda (•) the amplitude was 1.80 ± 0.08 nM and the rate was 31.7 ± 7.7 s⁻¹. In the presence of 300 nM Dda (□) the amplitude was 1.82 ± 0.09 nM and the rate was 43.7 ± 9.5 s⁻¹.

Pre-Steady-State Kinetics.

To observe a burst of product formation, the concentration of substrate must exceed that of the enzyme and must exceed the K_D such that the majority of the enzyme will be bound to the DNA before initiation of the reaction. These conditions present special problems for helicase assays, because the products of the reaction (ssDNA) can spontaneously hybridize to reform the substrate (23, 24). An oligonucleotide that is complementary to one of the ssDNA products can be included in the reaction mixture to serve as an annealing trap. However, the annealing trap can bind to the helicase; therefore, its presence in the reaction mixture might alter the kinetics for unwinding (24). Low DNA concentrations can be used to reduce the rate of hybridization, thereby precluding the need for an annealing trap in the reaction mixture (23).

To establish appropriate conditions for pre-steady-state unwinding reactions, a fluorescein-labeled 12-mer oligonucleotide was used for fluorescence polarization binding assays as described (6). Polarization was measured by using a Beacon (Panvera, Madison, WI) polarization system. Titration of 0.5 nM oligonucleotide with increasing concentrations of Dda in helicase unwinding buffer resulted in a K_D of 0.43 ± 0.19 nM (data not shown). The low K_D allowed for a relatively low concentration of 24:12-mer (16 nM) for initial experiments, although the concentration was kept well above the K_D for DNA binding. Hybridization of reaction products occurs slowly at this concentration relative to the unwinding reaction, allowing the annealing trap to be introduced immediately after the unwinding reaction was quenched by placing the trap in the receiving vial as shown in Fig. 1A.

Dda (4 nM) was incubated with the 24:12-mer (16 nM) in reaction buffer, and the unwinding reaction was initiated after rapid mixing with ATP and Mg²⁺. At varying times, the reaction was stopped by the addition of quencher (400 mM EDTA). The product was separated from substrate by gel electrophoresis (Fig. 2A), and the quantity of ssDNA was determined. Under these conditions, the progress curve for unwinding exhibited biphasic kinetics (Fig. 2B). Fitting the data to a single exponential followed by a steady-state rate (Eq. 2) resulted in a pseudo first-order rate constant of 21.3 ± 3.4 s⁻¹ for the first phase of the reaction, which was similar to the observed rate constant obtained under single-cycle conditions in Fig. 1. This strongly suggests that the first phase of the reaction under pre-steady-state conditions represents the first cycle for unwinding of productively bound substrate. Significantly, the amplitude of the first phase was 3.5 ± 0.2 nM, which is very similar to the concentration of the enzyme (4 nM) in the reaction mixture. The simplest explanation for this result is that Dda can function as a monomeric helicase.

Fig 2.

DNA unwinding under pre-steady-state conditions. (A) PhosphorImage of products after separation of ssDNA from double-stranded DNA on a 20% native polyacrylamide gel. Unwinding of the 24:12-mer substrate (16 nM) by Dda (4 nM) was initiated after the addition of ATP (5 mM) and Mg(OAc)₂ (10 mM) followed by the addition of quench solution (400 mM EDTA) after varying times. (B) The ssDNA formed under pre-steady-state conditions was plotted vs. time. Data were fit to an equation describing a single exponential followed by a steady-state rate (Eq. 2). The amplitude and rate for the burst phase of the reaction were 3.5 ± 0.2 nM and 21.3 s⁻¹, respectively.

DNA Unwinding at Varying DNA and Helicase Concentrations.

To examine further the biphasic kinetics for unwinding, it was necessary to extend the conditions in which the pre-steady-state unwinding could be observed. For example, oligomerization of Dda might occur at much higher protein concentrations than in Fig. 2, which would require higher substrate concentrations. Annealing of ssDNA products would be expected to occur much faster at higher substrate concentrations; therefore, conditions were sought that would allow increasing protein and substrate concentrations to be tested. To determine whether the first phase of unwinding was sensitive to the presence of excess ssDNA, excess poly(dT) (5 μM) was included in the reaction. Dda (4 nM) was incubated with 24:12-mer (16 nM), and unwinding was initiated with ATP, Mg²⁺, and poly(dT). As shown in Fig. 3A, the rate constant and amplitude of the burst phase were very similar to the rate and amplitude from Fig. 2; however, the steady-state phase of the reaction was essentially eliminated. This indicates that high concentrations of ssDNA did not effect the first cycle of unwinding but did prevent Dda from undergoing multiple rounds of unwinding after completion of the first cycle. An additional test of the reaction conditions was performed in which the annealing trap was included in the reaction mixture, along with ATP and Mg²⁺, rather than being introduced after the reaction was quenched. Dda (4 nM) was incubated with 24:12-mer (16 nM) in reaction buffer, and the unwinding reaction was initiated after mixing with ATP, Mg²⁺, and the annealing trap. The rate constant and amplitude of the first phase of the reaction were very similar to the reaction in which the annealing trap was added after the reaction was quenched (Fig. 3B). Thus, the presence of the annealing trap in the reaction mixture does not influence the first phase of the reaction.

Fig 3.

DNA unwinding under pre-steady-state conditions in the presence of a protein trap or annealing trap. (A) The time course for product formation under the same conditions as described for Fig. 2 except that a protein trap [poly(dT), 5 μM] was added along with the ATP and Mg(OAc)₂ at the start of the reaction. The data were fit to Eq. 2. The average amplitude and rate of the burst phase from three experiments were 3.6 ± 0.1 nM and 30.5 ± 5.5 s⁻¹, respectively. (B) The progress curve for product formation under the same conditions as described for Fig. 2 except that the annealing trap was placed with the ATP and Mg(OAc)₂ rather than in the receiving vial. The data were fit to Eq. 2. The average amplitude and rate of the burst phase from three experiments were 3.8 ± 0.2 nM and 25.0 ± 3.3 s⁻¹, respectively.

To determine whether oligomerization of Dda occurs at higher protein concentrations, a series of experiments was performed under pre-steady-state conditions at increasing protein concentration. The burst amplitude of product formation remained >75% of the total protein over a range of 4–100 nM Dda when the substrate was kept in 4-fold excess of the enzyme (Fig. 4A). Importantly, the rate of the first phase is similar over the entire range of enzyme concentrations tested (Fig. 4B). The rates vary over a 2-fold range, which is common for pre-steady-state DNA-unwinding assays (8); however, no trend in the variation is observed. This indicates that the enzyme activity does not increase at higher concentrations, reducing the possibility that transient protein–protein interactions are required for unwinding to occur.

Fig 4.

DNA unwinding under conditions of varying concentrations of Dda and DNA (24:12-mer). DNA concentration was in 4-fold excess of Dda. For example, at 100 nM Dda, the 24:12-mer concentration was 400 nM. The burst amplitude and burst rate were determined as described for Fig. 2 except the annealing trap was placed with the ATP and Mg(OAc)₂. (A) The amplitude of the burst phase divided by the concentration of Dda is plotted as a function of the concentration of Dda. (B) The rate of the burst phase is plotted as a function of Dda concentration. The error bars in each plot represent the standard deviation for three measurements.

If Dda oligomerization requires binding to DNA, as reported for Rep helicase (8, 25), then the burst amplitude or rate should change as the ratio of protein to DNA is altered (8). Unwinding reactions were performed at relatively high substrate concentration (200 nM) to greatly exceed the measured K_D. When the concentration of Dda was varied from 20 to 150 nM, the burst amplitude was >75% of the enzyme concentration in each experiment, indicating that oligomerization was not required for rapid unwinding to occur (Fig. 5; Table 2). The rate constant for the burst phase was similar at each protein concentration, further suggesting that oligomerization is not required for Dda-catalyzed DNA unwinding (Table 2).

Fig 5.

DNA unwinding with varying concentrations of Dda in the presence of 200 nM DNA substrate (24:12-mer). Progress curves are plotted at 20 (○), 40 (⋄), 50 (•), 75 (▪), 100 (▵), and 150 nM (▴) Dda. The lines represent the fit of the data to a single exponential followed by a steady-state rate (Eq. 2) with the program KALEIDAGRAPH. The burst amplitude and burst rate for each reaction are listed in Table 2.

Table 2.

DNA unwinding with varying concentrations of Dda at a fixed concentration of DNA

Reaction conditions		Amplitude, nM	Burst rate, s−1	Amplitude/ [Dda]
Dda, nM	DNA, nM
20	200	17.2  ± 2.5	29.9  ± 13.3	0.86
40	200	38.9  ± 2.7	25.7  ± 5.0	0.97
50	200	41.2  ± 1.4	23.4  ± 2.3	0.82
75	200	58.3  ± 2.7	28.6  ± 3.9	0.78
100	200	85.4  ± 4.3	22.5  ± 3.1	0.85
150	200	115.3  ± 5.7	25.1  ± 3.6	0.77

The amplitudes, rates, and standard errors of the burst phase were determined from the fit of the data in Fig. 5.

^*The amplitude of the burst phase divided by the concentration of Dda.

Multiple Steps Are Required for Unwinding of Longer Substrates.

Helicases are considered molecular motors, in which the enzyme translocates while unwinding the DNA through multiple cycles or “steps.” Processive unwinding of oligonucleotide substrates by helicases may be indicated by a lag phase when multiple steps are necessary for complete melting of the duplex (26). No clearly observable lag phase is apparent with the 24:12-mer substrate used in these studies. Twelve base pairs may be too short to discern multiple steps of unwinding for this helicase; therefore, results shown thus far may be the result of a single step of helicase activity.

To determine whether monomeric Dda can undergo multiple steps of unwinding during the burst phase, a longer substrate was designed containing 16 bp adjacent to the 12-nt single-stranded overhang (Table 1). Single-turnover conditions were used in the presence of excess enzyme to determine the quantity of product formed in a single binding event. A distinct lag in the burst phase is apparent with the longer substrate (Fig. 6A). The lag is consistent with a multistep sequential mechanism as has been described for UvrD helicase (26, 27). Two steps did not provide a good fit (not shown), and thus an equation describing a sequential three-step mechanism was used to fit the data (Eq. 3; ref. 27). Scheme depicts association of helicase with DNA followed by a three-step mechanism, where k_u is the rate constant for unwinding, and k_d is the rate constant for dissociation of the helicase from the DNA. Dissociation can occur at any step in the unwinding cycle such that the amplitude of product formation may be reduced under single-turnover conditions. Indeed, the amplitude for product formation was 1.22 nM ± 0.07 nM or ≈60% of the total DNA. Presumably, Dda dissociates from the DNA during unwinding, and the protein trap [poly(dT)] prevents reassociation of the helicase with the substrate.

Fig 6.

DNA unwinding by Dda of a 16-bp substrate. (A) A single-turnover unwinding experiment was performed as described for Fig. 1 by using the 28:16-mer as substrate (see Table 1 for sequence). Product formation exhibited a lag in the burst phase. The data were fit to a three-step sequential mechanism (Eq. 3), where A is the amplitude of product formation and k_obs is equal to (k_u + k_d) from Scheme (26). Results from three experiments were averaged to provide an amplitude of 1.22 ± 0.07 nM and a k_obs of 57.1 ± 2.2 s⁻¹.

(B) Unwinding of the 28:16-mer substrate was measured under pre-steady-state conditions. Dda (4 nM) was incubated with 28:16-mer (16 nM), and unwinding was initiated after the addition of ATP (5 mM) and Mg(OAc)₂ (10 mM). The multiphase time course was fit to Eq. 4, which describes a three-step sequential mechanism followed by a steady-state rate (solid line).

Results from four experiments were averaged to provide an amplitude (A) of 2.40 ± 0.3 nM and a rate (k_obs) of 73.7 ± 12.1 s⁻¹.

Scheme 1.

These results indicate that under pre-steady-state conditions, only ≈60% of the initially bound helicase would be expected to form product during the first unwinding cycle. Unwinding under pre-steady-state conditions was conducted as described for the 12-bp DNA substrate. Dda (4 nM) was incubated with the 28:16-mer (16 nM) followed by rapid mixing with ATP, Mg²⁺, and annealing trap to initiate the reaction. As shown in Fig. 6B, a lag in the burst phase could be detected, which indicated that multiple steps of enzymatic activity were required for complete unwinding. Fitting the data to an equation for a three-step sequential mechanism followed by a steady-state rate (Fig. 6B; Eq. 4) yielded a burst amplitude of 2.4 ± 0.3 nM. This amplitude is ≈60% of the total Dda, consistent with a monomeric enzyme even on a substrate that is long enough to require multiple steps for complete unwinding. The lower amplitude observed with the 28:16-mer compared with the 24:12-mer likely results from dissociation of Dda from the substrate before complete unwinding due to the relatively low processivity of the enzyme (24).

Conclusions

Many helicases clearly function as hexamers, but controversy exists as to whether any helicase can function as a monomer. Here we report a direct functional assay that strongly indicates that a monomeric helicase is able to rapidly unwind oligonucleotide substrates. The significance of this observation is that the catalytic cycle of ATP hydrolysis-coupled DNA unwinding is performed by a single polypeptide, consistent with an inchworm mechanism, possibly as proposed for PcrA (10). Dda need not oligomerize to function. It is important to note that the lack of a requirement for oligomerization does not preclude the possibility that multiple monomers might align along the nucleic acid and function in a cooperative manner. Dda is capable of displacing streptavidin from the 3′ end of biotin-labeled oligonucleotides in a reaction that is enhanced when the oligonucleotide length is increased (22), suggesting that multiple Dda monomers may assist one another in the displacement reaction. Also, the in vitro results here do not preclude the possibility that multiple enzyme molecules might be required for some functions in vivo such as removing proteins in the path of a replication complex.

Abbreviations

• SF, superfamily

• ssDNA, single-stranded DNA