Escherichia coli Helicase I1 (uruD Gene Product) Translocates Unidirectionally in a 3’ to 5’ Direction*

Escherichia coli helicase 11, product of the uvrD gene, is a single-stranded DNA-dependent nucleoside 5’-triphosphatase with helicase activity. As a DNA-dependent ATPase, helicase I1 translocates proces- sively along single-stranded DNA (S. W. Matson, un-published results). The direction of translocation has been determined using a helicase assay that directly measures the ability of helicase I1 to catalyze the dis- placement of a labeled DNA fragment from one end of a single-stranded linear DNA molecule. The translo- cation of helicase I1 along single-stranded DNA is unidirectional and in the 3’ to 5’ direction with respect to the DNA strand on which the enzyme is bound. A kinetic analysis of the displacement of a labeled DNA fragment annealed to a linear single-stranded DNA molecule is also consistent with unidirectional trans- location in the 3’ to 5’ direction. These results are contrary to results previously obtained using an indirect helicase assay (Kuhn, B., Abdel-Monem, M., Krell, H., and Hoffmann-Berling, H. (1979) J. Biol. Chem. 254,11343-11350). The unwinding of duplex DNA to yield single-stranded for use thought to be catalyzed by has same apparent value for same sensitivity to ATP analog inhibitors of the ATPase activity as pre- described addition, of helicase I1 and purified rep protein does not migrate to the same position compared on polyacrylamide gel run in the presence of sodium dodecyl sulfate. these criteria, this preparation of helicase I1 contains the same polypeptide as that used Kuhn

The unwinding of duplex DNA to yield single-stranded template DNA for use by DNA polymerase is thought to be catalyzed by a class of enzymes called the helicases (1,2). Helicases have been isolated from bacteriophage-infected cells, bacterial cells (for review see Ref. l), and yeast (3). Helicases from bacteriophage T4 and phage T7 have been shown to have roles in DNA replication both as helicases and as part of the primase complex (4)(5)(6). The specific role(s) of the helicases isolated from bacterial cells and yeast is not yet known.
The methods used to determine the direction of translocation of the helicases have varied widely. The unidirectional 5' to 3' translocation of the phage T7 gene 4 protein was inferred from an analysis of the frequency of utilization of priming sites by gene 4 protein primase activity (9). Studies of the gene 4 protein helicase activity support this direction of translocation (13). The phage T4 gene 41 protein was shown to move in a 5' to 3' direction by measuring the displacement of short, labeled DNA fragments from a linear partial duplex DNA molecule (6). Kuhn et al. (10,11) suggested that E. coli helicases I and I1 translocate in a 5' to 3' direction. The DNA substrate used in these studies was a linear, duplex DNA molecule that had been partially digested with exonuclease I11 or X exonuclease. Exonuclease I11 catalyzes the hydrolysis of one strand of a duplex moving in a 3' to 5' direction. The X exonuclease catalyzes a similar reaction but moves in the opposite direction. Since only the DNA digested with exonuclease I11 served as a substrate in the helicase reaction, it was inferred that the helicase bound the single-stranded region of DNA created by nuclease digestion and translocated in the 5' to 3' direction into duplex DNA. The E. coli helicase I11 enzyme was shown to move in a 5' to 3' direction by determining which of two labeled DNA fragments was displaced by the enzyme from a linear DNA molecule (12). Displacement of a labeled DNA fragment, in this case, was assayed by the release of DNA susceptible to digestion by S1 nuclease. The rep protein of E. coli was shown to move in a 3' to 5' direction by the same method (8).
Since the phage helicases, T7 gene 4 protein and T4 gene 41 protein, also participate in the synthesis of RNA primers on the lagging strand side of the replication fork, it is necessary that these enzymes translocate in a 5' to 3' direction (4)(5)(6)9). Similarly, rep protein has a role in the replication of bacteriophage 4x174 that suggests a 3' to 5' direction of translocation (14). Roles for the other helicases in E. coli are currently unknown, and therefore no obligatory direction of translocation is suggested by function for these enzymes. In this communication, a direct helicase assay has been used to determine the direction of translocation for helicase 11. This enzyme moves unidirectionally in a 3' to 5' direction with respect to the strand of DNA on which it is bound. This direction of translocation is opposite to that reported by Kuhn et al. (10,11) and is consistent with a role for helicase I1 in excision repair (15,16).

Materinls
DNA and Nucleotides-Bacteriophage M13mp7 replicative form I DNA and single-stranded DNA were prepared as described (17). All unlabeled nucleotides were from P-L Biochemicals. [a3'P]dCTP and [y3*P]ATP were from ICN Radiochemicals.
Enzymes--Restriction endonucleases, DNA polymerase I (large fragment), and bacteriophage T4 polynucleotide kinase were purchased from New England Biolabs; the reaction conditions used were those suggested by the supplier. E. coli helicase I1 was purified from E. coli harboring a multicopy plasmid carrying the structural gene for helicase I1 as described (18) with the following modifications. After harvesting, the cells (80 g) were resuspended in 50 mM Tris-HC1 (pH 7.5), 10% sucrose, 10 mM EDTA and frozen in liquid nitrogen. The cells were lysed by the addition of lysozyme to 250 pg/ml and NaCl to 0.1 M followed by incubation at 0 "C for 45 min. The cell suspension was heated in a 37 "C water bath until the temperature reached 20 "C and then transferred to an ice bath until the temperature reached 10 "C. Lysed cells were centrifuged at 21,000 rpm for 60 min and the supernatant recovered as fraction I. Fraction 11, 111, and IV were prepared as described (18). Fraction IV was applied to a singlestranded DNA cellulose column equilibrated with 20 mM Tris-HC1 (pH 7.5), 10% glycerol, 0.1 mM EDTA, and 5 mM 2-mercaptoethanol (buffer A) containing 100 mM NaCI. Helicase I1 was eluted with buffer A containing 1 M NaCI. The enzyme was greater than 95% pure as judged by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate and has a relative molecular mass of 76,000 g/mol. This is exactly the same size as helicase I1 purified by . Moreover, this preparation of helicase I1 has essentially the same apparent K, value for ATP and the same sensitivity to ATP analog inhibitors of the ATPase activity as previously described (19). In addition, this preparation of helicase I1 and purified rep protein does not migrate to the same position when directly compared on a polyacrylamide gel run in the presence of sodium dodecyl sulfate. By these criteria, this preparation of helicase I1 contains the same polypeptide as that used by Kuhn et al. (10,ll). Bacteriophage gene 41 protein and gene 61 protein were the generous gift of Dr. Nancy Nossal and Dr. Deborah Hinton (National Institutes of Health) (20,21).

Methods
Helicase Substrate Preparation-The DNA substrates used in helicase assays consist of the complementary strand of a radioactively labeled M13mp7 replicative form I H o e 1 1 1 restriction fragment annealed to M13mp7 single-stranded DNA to form a partial duplex. The substrates were constructed by incubating the appropriate restriction fragment (approximately 100 ng) with M13mp7 singlestranded DNA (2 pg) in 40 mM Tris-HCI (pH 7.5), 10 mM MgCl', 1 mM dithiothreitol, 50 mM NaCl at 95 "C for 5 min, followed by 65 "C for 20 min, and then 23 "C for 20 min. The resulting partial duplex was labeled at the 3'-OH of the restriction fragment in a reaction containing 5 units of DNA polymerase I (large fragment) and [a3'P] dCTP in the above buffer. Incubation was for 20 min at 23 "C followed by the addition of 50 p~ unlabeled dCTP and incubation at 23 "C for an additional 20 min. After phenol extraction, the reaction mixture was filtered through an agarose A-5m column (1.5 ml) equilibrated with 10 mM Tris-HCI (pH 7.5), 1 mM EDTA, and 100 mM NaCI. The void volume containing the labeled DNA was collected and used directly as helicase substrate. It should be noted that this substrate may be contaminated with single-stranded DNA containing no annealed restriction fragment.
The DNA substrate used to determine the direction of translocation of helicase I1 was constructed as above except that after annealing the complementary strand of the 341-base pair (bp') H o e 1 1 1 restriction fragment with circular single-stranded M13mp7 DNA, the partial duplex was cleaved with ClaI restriction endonuclease. This restriction endonuclease cleaves once within the duplex region of the DNA produced in the annealing reaction to generate DNA fragments 200 and 141 bases in length. All available 3"OH termini then were labeled using DNA polymerase I (large fragment), dGTP, and [a"P]dCTP as above. This procedure will label both DNA fragments produced by cleavage with ClaI and the single-stranded linear M13mp7 DNA molecule.
Helicase Assay-The helicase assay measures the displacement of a labeled DNA fragment from a partial duplex DNA molecule catalyzed by helicase 11. The reaction mixture (20 pl) contained 40 mM Tris-HCI (pH 7.5), 4 mM MgCl', 1 mM dithiothreitol, 50 pg/ml bovine serum albumin, approximately 2 p~ DNA substrate, 1.8 mM ATP, 20-40 m M NaCI, and helicase 11. Incubations were at 37 "C for 10-30 min. The reaction was terminated by the addition of 10 pl of 50 mM EDTA, 40% glycerol, 0.6% sodium dodecyl sulfate, 0.1% bromphenol blue, 0.1% xylene cyanol, and the products were separated on an 8% nondenaturing polyacrylamide gel. Electrophoresis was at 5-15 V/cm for 14 h. Polyacrylamide gels were analyzed by film autoradiography ' The abbreviation used is: bp, base pairs. Other Methods-Nondenaturing polyacrylamide gel electrophoresis was carried out as described (22). DNA concentrations were determined by directly measuring the absorbance at 260 nm and are expressed as nucleotide equivalents.

RESULTS
E. coli helicase I1 is a single-stranded DNA-dependent ATPase with helicase activity that translocates processively along single-stranded DNA (10,11,19).' A processive mechanism for translocation suggests that translocation is unidirectional. Previous studies (10, 11) have suggested that helicase I1 translocates unidirectionally along single-stranded DNA i n a 5' to 3' direction with respect to the strand on which it is bound. The question of the direction of translocation has been addressed in this report using a more direct helicase assay which measures the displacement of a labeled DNA fragment from a partial duplex DNA molecule.
The Helicase Activity of Helicase 11-The helicase reaction catalyzed by helicase I1 has been demonstrated using an in uitro assay that measures the displacement of a labeled DNA fragment from a single-stranded circular DNA molecule to which the labeled fragment has been annealed. Three different partial duplex DNA molecules were tested as helicase substrates (Fig. 1). In Fig. 1, lanes 1-5, the DNA substrate used in the helicase reaction consists of a 71-nucleotide long, 3'-end-labeled DNA fragment annealed to single-stranded M13mp7 DNA (71-bp partial duplex). Helicase I1 displaces greater than 80% of t h e labeled fragment at the highest concentration of enzyme used (Fig. 1, lane 5 ) . Increasing the ' S. W. Matson  time of incubation or the helicase I1 concentration does not increase the amount of the 71-nucleotide fragment displaced. In Fig. 1, lanes 6-10, the length of duplex DNA has been increased by annealing a 119-nucleotide long, 3'-end-labeled DNA fragment to M13mp7 DNA (119-bp partial duplex). Helicase I1 is capable of displacing this DNA fragment from the single-stranded circle, although the extent of displacement is less than that observed with the 71-bp partial duplex. Approximately 70% of the fragment was displaced at the highest concentration of helicase I1 used (Fig. 1, lane IO). Lanes 11-15 in Fig. 1 contain a partial duplex DNA substrate with a duplex region that is 343 nucleotides in length (343-bp partial duplex). Helicase I1 also catalyzes the displacement of this DNA fragment. Again, the extent of the helicase reaction is less than that observed with either the 71-bp partial duplex or the 119-bp partial duplex. At the highest concentration of helicase I1 used, approximately 60% of this DNA fragment was displaced from the single-stranded circular DNA molecule (Fig. 1, lane 15). All of the displacement reactions are dependent on the presence of ATP in the reaction mixture (data not shown). It should be noted that although a smaller fraction of labeled fragment is displaced by helicase I1 as the length of the duplex region increases, the total amount of duplex DNA that is unwound by the enzyme remains essentially constant irrespective of the substrate used a t low enzyme concentrations. 3 Helicase 11 Translocates in a 3' to 5' Direction-To directly determine the direction of translocation of helicase 11, the strategy depicted in Fig. 2   the 341-bp partial duplex4 DNA molecule. After annealing the DNA fragment to the circular DNA molecule, the duplex region was cleaved with ClaI. Cleavage with ClaI generates three 3"OH groups that can be labeled using DNA polymerase I (large fragment) and the appropriate [a"'P]dNTPs, one on the 200 nucleotide fragment, one on the 141 nucleotide fragment, and one on the linear M13mp7 DNA molecule. After labeling all available 3'-OH groups, the DNA substrate shown in Fig. 2 was used in helicase reactions catalyzed by helicase I1 (Fig. 3). Helicase I1 catalyzes the displacement of the 143nucleotide long fragment from this linear DNA substrate but fails to catalyze the displacement of the 202-nucleotide long fragment (Fig. 3, lanes 2 and 3). When ATP is omitted from the reaction mixture, there is no displacement of the fragment (Fig. 3, lane 4), confirming the ATP dependence of the reaction. As a control, the 343-bp partial duplex circular DNA substrate was used as a helicase substrate under the same conditions (Fig. 3, lane 7). As expected, helicase I1 catalyzes the displacement of the 343-nucleotide long fragment from the circular DNA molecule. These data suggest that helicase I1 translocates unidirectionally, but in the 3' to 5' direction.
Results obtained previously by others (10, 11) have suggested that helicase I1 translocates in a 5' to 3' direction, opposite to the direction of translocation reported here. Translocation in the 5' to 3' direction is predicted to cause the displacement of the 3'-end of a DNA fragment first. As a control for the experiments presented above, this prediction was tested using the linear helicase substrates shown in Fig.   4A. The construction of these DNA substrates takes advantage of restriction sites located within the duplex DNA of the 343-bp partial duplex DNA substrate. There is a ClaI site located 202 nucleotides from the 3'-end of the annealed ["PI DNA fragment and a RsaI site located 163 nucleotides from the 3'-end of the annealed ["PIDNA fragment (Fig. 4A). The linear helicase substrates shown were constructed by incubating 343-bp partial duplex DNA (after labeling the 3'-end) with the appropriate restriction endonuclease. The resulting DNA substrate is a linear molecule with a labeled DNA fragment of discreet length annealed to one end. A second, unlabeled DNA fragment is annealed at the opposite end of the linear M13mp7 DNA molecule.
An enzyme that translocates unidirectionally in the 5' to 3' direction should displace the labeled DNA fragment from ' The HaeIII fragment isolated after digestion of M13rnp7 replicative form I DNA is 341 base pairs in length. After the 3'-end labeling reaction in the presence of [a3*P]dCTP, the fragment is 343 base pairs in length. the linear substrates shown in Fig. 4A. Each linear DNA substrate was incubated with helicase I1 under conditions that are optimal for the helicase reaction. Displacement of the labeled DNA fragment from the linear substrate catalyzed by helicase I1 was not observed for either of the DNA substrates tested (Fig. 4B, lanes 6 and lo), although the full-length 343nucleotide long fragment was displaced from the singlestranded circular substrate under these conditions (Fig. 4B,  lane 2). As a positive control for this experiment, the bacteriophage T 4 gene 41 protein-gene 61 protein complex was used in a helicase reaction (Fig. 4B, lanes 3, 7, and 11). The gene 41 protein-gene 61 protein complex is a helicase/primase that has been shown to move in a 5' to 3' direction along single-stranded DNA (6).

E. coli Helicase 11 Translocates in a 3' to 5' Direction
As expected, the 3'-end-labeled DNA fragments were displaced from the linear substrates and the 343-nucleotide DNA fragment was displaced from the single-stranded circular substrate (Fig. 4B, lanes 3, 7, and 11). This confirms the direction of translocation of the gene 41 protein-gene 61 protein complex and further suggests that helicase I1 does not move in the 5' to 3' direction but in the 3' to 5' direction. 5. Helicase I1 displaces the 5'-end of an annealed DNA fragment first. The DNA substrates used in helicase reactions were constructed as follows. The 341-bp H a d 1 restriction fragment was 5'-end-labeled using bacteriophage T4 polynucleotide kinase and [y3*P]ATP. The complementary strand was annealed onto M13mp7 phage DNA and the resulting 341-bp partial duplex purified as described under "Experimental Procedures." This DNA substrate is slightly contaminated with the strand of the 341-bp restriction fragment which has the same sense as the phage DNA. This is reflected by the presence of a labeled band migrating at the position expected for a 341-nucleotide long DNA fragment. The 341-bp circular partial duplex DNA substrate was converted into a linear molecule containing duplex DNA at both ends by cleavage with either BgnI or ClaI restriction endonuclease. The resulting linear helicase substrate contains duplex DNA at both ends of the linear molecule; the DNA fragment annealed at the 5'-end of the phage DNA is labeled, and the DNA fragment annealed at the opposite end is not labeled. In lanes 1 3 ,  Reciprocal experiments were performed in which a 341-bp partial duplex DNA substrate was labeled at the 5'-end using bacteriophage T 4 polynucleotide kinase prior to incubation with the appropriate restriction endonuclease (Fig. 5). The linear DNA substrates obtained after restriction digestion contain a single, labeled DNA fragment of discreet length but, in this case, the labeled DNA fragment is on the 5'-end of the linear DNA molecule (refer to Fig. 4A). As suggested by the results presented above, the 5'-end-labeled fragment was displaced by helicase I1 using this DNA substrate (Fig. 5,  lanes 2 and 5 ) . Taken together, these results strongly suggest that helicase I1 translocates in the 3' to 5' direction along single-stranded DNA.

FIG.
In addition to the experiments presented above, a kinetic analysis of the displacement of a DNA fragment from a linear molecule has been carried out using the DNA substrates shown in Fig. 6. In each case, a 3'-end-labeled circular DNA substrate has been converted into a linear DNA substrate by cleaving within the single-stranded region of the circular DNA molecule using EcoR1.S One DNA substrate has a singlestranded 3'-tail that is 817 nucleotides long and a duplex region of 71 bp, and the other DNA substrate has a singlestranded 3'-tail that is approximately 7000 nucleotides long and a duplex region of 119 bp. Each DNA substrate was incubated with helicase 11, and the amount of fragment displaced from the linear DNA substrate has been compared with the amount of fragment displaced from a circular subs Single-stranded circular M13mp7 DNA can be cut with EcoRI to yield a single-stranded linear molecule. The 71-bp partial duplex DNA substrate and the 119-bp partial duplex DNA substrates were constructed and labeled as described under "Experimental Procedures." Linear DNA substrates were obtained by incubating the circular DNA substrate with the restriction endonuclease EcoRI which can cleave at the hairpin present in M13mp7 phage DNA. The resulting 3' singlestranded regions are 817 nucleotides and about 7000 nucleotides in length. Asterisk denotes position of radioactive label.

TABLE I
A kinetic analysis of the direction of translocation of helicase ZZ Reaction mixtures were as described under "Experimental Procedures" using the indicated amounts of E. coli helicase 11. Incubations were for 20 min at 37 "C. The amount of labeled fragment displaced was determined by slicing the polyacrylamide gel into 1-cm sections and counting in a liquid scintillation counter. strate with the same length of duplex DNA (Table I).
If helicase I1 translocates in a 3' to 5' direction, the linear DNA substrate with a short single-stranded 3'-tail is expected to be a relatively poor helicase substrate compared to a circular DNA molecule. This is expected because the long 5'tail will bind enzyme molecules more frequently than the short 3'-tail, since it is %fold longer, allowing 3' to 5' translocation but no displacement reaction. Only when the enzyme binds the relatively short 3'-tail will it encounter the labeled DNA fragment after 3' to 5' translocation and cause a displacement reaction. This is in contrast to the situation encountered on a circular DNA molecule where binding at any site and unidirectional translocation will ultimately bring the enzyme to duplex DNA. Conversely, the linear DNA substrate with the long single-stranded 3'-tail should allow binding with subsequent 3' to 5' translocation resulting in a displacement reaction almost as frequently as circular DNA substrate.
The results presented in Table I confirm these predictions. Comparison of the fraction of labeled DNA fragment displaced from the linear DNA substrate with a short 3'-tail and the corresponding circular substrate shows that a displacement reaction occurs on the linear molecule less frequently than on the circular molecule. However, when the labeled DNA fragment is moved to the opposite end of the linear molecule, creating a long single-stranded 3'-tail, a displacement reaction is catalyzed by helicase I1 more frequently on the linear substrate than observed using the corresponding circular DNA substrate. These results are consistent with unidirectional translocation in the 3' to 5' direction. The displacement of more labeled DNA fragment from the linear substrate with a long single-stranded 3'-tail as compared to the circular DNA substrate reflects a processive mechanism of translocation along single-stranded DNA (see "Discussion").

DISCUSSION
Many enzymes that interact with DNA do so with a specific polarity. This is true for the DNA polymerases and the RNA polymerases that synthesize nucleotide chains in the 5' to 3' direction, for a number of exonucleases which degrade DNA in a specific direction, and for many other enzymes (23). The helicases, which catalyze the unwinding of duplex DNA, also exhibit a directionality in their interaction with DNA. Each individual helicase translocates in either the 5' to 3' direction or the 3' to 5' direction with respect to the DNA strand on which it is bound. In this context, the E. coli rep protein unwinds DNA in the 3' to 5' direction (8) and E. coli helicases I and I11 unwind DNA in the 5' to 3' direction (10-12).
In this paper, it has been demonstrated that helicase I1 translocates along single-stranded DNA in the 3' to 5' direction. This direction of translocation is opposite to that reported previously (10, 11). The conclusions reported here are based on direct observation of the end products of a helicase I1 reaction using polyacrylamide gels to resolve reaction products. In the experimental protocol used (Fig. 2), helicase I1 is provided with a DNA substrate on which it should be able to translocate in either the 5' to 3' direction or the 3' to 5' direction to reach duplex DNA. The helicase reaction is followed by observing the labeled DNA fragments that are displaced from the linear partial duplex DNA substrate by helicase 11. The results suggest that helicase I1 migrates along single-stranded DNA in a 3' to 5' direction and displaces the DNA fragment on the 5'-end of the linear substrate molecule while failing to displace the DNA fragment on the 3'-end of the linear substrate molecule. This interpretation is based on two facts. First, helicase I1 translocates along single-stranded DNA processively' and therefore in a single direction. Second, helicase I1 requires single-stranded DNA for binding (10); this condition is not met at either end of the linear DNA molecule shown in Fig. 2. Therefore, helicase I1 will bind to the interior, single-stranded region and translocate in one direction, ultimately displacing a labeled DNA fragment of specific length. The results shown in Fig. 3 clearly demonstrate translocation in the 3' to 5' direction.
In addition to the direct analysis of a helicase reaction, a kinetic analysis of the displacement of DNA fragments from linear DNA molecules was consistent with 3' to 5' translocation. In one case, the linear partial duplex substrate had a DNA fragment annealed at a position 817 nucleotides from the 3'-end of the linear DNA molecule. Compared with the length of single-stranded DNA on the other side of the duplex region (approximately 6300 nucleotides), this target for initial binding of helicase I1 is small. Therefore, only a small percentage of enzyme molecules will bind and initiate 3' to 5' translocation into a duplex region of DNA. The remainder will bind and translocate to the end of the DNA molecule without encountering the duplex region of the substrate. The results obtained in comparing displacement of the fragment from the linear substrate uersus the circular substrate were DNA pol I dNTPs FIG. 7. Model for helicase I1 action during uvrABC nuclease-mediated excision repair. Subsequent to incision on both sides of the pydrimidine dimer catalyzed by the uvrABC enzyme complex, helicase I1 and DNA polymerase I may interact to catalyze strand resynthesis and displacement of the DNA fragment containing the pyrimidine dimer. In this model, helicase I1 translocates along the template strand in a 3' to 5' direction. consistent with this idea. Even at high helicase I1 concentrations, less labeled DNA fragment was displaced from the linear molecule with a short 3'-tail than from the circular molecule. When the DNA fragment is annealed on the 5'-end of the linear DNA molecule, the target for initial binding on the 3' side of the duplex region is large. Consequently, most of the enzyme molecules will bind in this region of singlestranded DNA and, if translocation is in the 3' to 5' direction, translocate toward duplex DNA. On this linear DNA substrate, helicase I1 catalyzed the displacement of nearly 40% more labeled DNA fragment than when a circular DNA substrate was used. Two conclusions can be drawn from this result. First, the initial binding rate must be nearly as large on this linear substrate molecule as it is on the circular molecule. Second, the unidirectional translocation is processive. A processive mechanism for translocation suggests that helicase I1 will remain bound on the circular DNA substrate after displacing the labeled DNA fragment and continue unidirectional translocation. On the linear DNA substrate, an end will necessarily be reached and the enzyme will dissociate and bind a new substrate molecule. Since the new substrate molecule will be a partial duplex, another displacement event will be catalyzed. The result will be a greater displacement of labeled DNA fragments from the linear substrate than from the circular substrate during the initial stages of the reaction. This is consistent with results we have obtained in the analysis of the helicase I1 ATPase reaction.* Although the kinetic results only suggest that helicase I1 may translocate in the 3' to 5' direction, coupled with the direct analysis of translocation direction, it is clear that helicase I1 translocates in a 3' to 5' direction along the DNA strand on which it is bound.
The results presented here indicate a direction of translocation that is opposite to that reported by others (10, 11). Previous conclusions relied on an indirect assay of the helicase reaction catalyzed by helicase 11. In those experiments, a linear duplex DNA molecule was incubated with either exonuclease 111 or X exonuclease prior to incubation with helicase 11. This provided a single-stranded DNA binding site for helicase I1 with a 5' polarity or a 3' polarity. The DNA substrates were subsequently incubated with helicase 11, and the amount of single-stranded DNA formed was determined after digestion with S1 nuclease. The results suggested that a suitable substrate for helicase I1 was generated when the DNA substrate was partially degraded with exonuclease I11 but not with X exonuclease. Since exonuclease I11 digests DNA in a 3' to 5' direction, the single-stranded DNA tail remaining on the duplex would have the correct polarity for an enzyme that translocates in the 5' to 3' direction. It was inferred from these experiments that helicase I1 translocated in the 5' to 3' direction. This conclusion was apparently confirmed by experiments utilizing substrates similar to the ones used in this report. However, the DNA fragments displaced by the helicase were analyzed on sucrose gradients rather than polyacrylamide gels. Subsequent hybridization analysis confirmed the identity of the fragment displaced, which was consistent with 5' to 3' translocation. Presently, I have no explanation for the discrepancy in results. The experiments reported here are more direct and of higher resolution than those reported previously.
The direction of translocation of several prokaryotic helicases is suggested by their function in the cell (4-6, 9, 14). Translocation in the 3' to 5' direction is consistent with a role for helicase I1 in excision repair mediated by the uvrABC incision nuclease (15, 16), although 5' to 3' translocation is also possible. A model for helicase I1 interaction with DNA in such a repair scheme is shown in Fig. 7. In this model, helicase I1 could interact with DNA polymerase I and/or the uvrABC enzyme complex and move along the template strand in the 3' to 5' direction. Such an interaction with DNA polymerase I has been suggested (15). Translocation of helicase I1 through the duplex could facilitate displacement of the fragment that has been incised by the uvrABC nuclease allowing that enzyme to turn over and providing DNA template for DNA polymerase I. In this model, translocation in a 3' to 5' direction is consistent with an interaction with DNA polymerase I and it maintains helicase I1 on the template DNA rather than bound to the fragment being displaced. Many aspects of this model remain to be elucidated and clarified. For instance, is there an interaction between uvrABC and helicase II? How does helicase I1 bind the nicked DNA-protein complex and how does it dissociate to leave a nick for ligase to seal? These questions await further experimental testing.