Purification and Characterization of a DNA Helicase from Saccharomyces cerevisiae"

A novel DNA helicase, scHelI, has been purified from whole cell extracts of Saccharomyces cerevisiae using biochemical assays to monitor the fractionation. The enzyme unwinds partial duplex DNA substrates, as long as 343 base pairs in length, in a reaction that is dependent on either ATP or dATP hydrolysis. scHelI also catalyzes a single-stranded DNA-dependent ATP hydrolysis reaction; the apparent K,,, for ATP is 325 PM. The unwinding reaction on circular partial duplex substrates is biphasic, with a fast component occurring within 5 min of the initiation of the reaction and a slow component continuing to 60 min. This is in contrast to the ATP hydrolysis reaction, which exhibits linear kinetics for 60 min. The direction of the unwinding reaction is 5' to 3' with respect to the strand of DNA on which the enzyme is bound. The unwinding reaction is strongly stimulated by the addition of Escherichia coli single-stranded DNA-binding protein when long partial duplex substrates are used. The enzymatic activity of scHelI copurifies with a polypeptide of 135 kDa as determined by polyacryl- amide gel electrophoresis in the presence of sodium dodecyl sulfate. The polypeptide sediments as a mon- omer in a glycerol gradient in the presence of 0.2 M NaCl.

and three DNA helicases from the yeast Saccharomyces cereuisiue; Rad3 protein (4,5), Pifl protein (6,7), and Srs2 protein (8,9). The three yeast helicases are encoded by the cellular genome, although Pifl is thought to function in the mitochondria (7). In each case, the gene was identified first by mutation (4,6,8) and subsequently, using biochemical assays, the gene product was shown to catalyze a helicase reaction (5,7,9). In addition to the helicases described above, several other eukaryotic DNA helicases have been identified by biochemical assay. A listing of these helicases has been presented elsewhere (10) and includes enzymes from calf thymus (11-14) and cultured human cells (15)(16)(17). None of these proteins have been characterized genetically. Even more numerous are the putative helicases that have been identified by molecular genetic techniques. In each case, sequence homology suggests that the gene may encode a helicase, but gene products have not yet been shown to have helicase activity by biochemical assay (18)(19)(20)(21). It is apparent that both genetic and biochemical information are required to identify a precise role for these proteins in nucleic acid metabolism.
The yeast S. cerevisiae has been well characterized genetically, is highly amenable to molecular genetic analysis, and can be used for the biochemical purification of enzymes. For this reason, we have chosen yeast as a eukaryotic model system in which to identify and characterize individual DNA helicases both biochemically and genetically. This communication reports the purification and initial biochemical characterization of a new DNA helicase from yeast.
DNA and Nucleotides-Bacteriophage M13 mp7 replicative form I (RF I) and ssDNA' were prepared as previously described (22).
Enzymes-Restriction endonucleases and DNA polymerase I (large fragment) were from New England Biolabs or U. S. Biochemical. The reaction conditions used were those suggested by the supplier. Proteinase K was from Boehringer Mannheim. E. coli helicase I1 was The abbreviations used are: ssDNA, single-stranded DNA; dsDNA, double-stranded DNA; scHelI, Saccharomyces cereuisiae helicase I; SSB, Escherichia coli single-stranded DNA binding protein; bp, base pair(s); ATP+, adenosine 5'-O-(thiotriphosphate); HPLC, high pressure liquid chromatography. 21783 purified as previously described (24). E. coli helicase I, purified as previously described (25), was the generous gift of B. Morton (this lab). E. coli SSB was purified from an overproducing strain of E. coli. Both the E. coli strain and the purification procedure were kindly provided by Dr. R. McMacken (Johns Hopkins University, Baltimore).
Other Materials-Glass beads (0.5 mm), heparin-agarose, and ATP-agarose, linked through ribose hydroxyls, were from Sigma. Double-strand DNA-cellulose (12.6 mg of DNA/g of cellulose) was from U. S. Biochemical. Poly(ethy1enimine) (Polymin-P) was from Aldrich and phosphocellulose (P-cell) was from Whatman. Anion exchange high pressure liquid chromatography (HPLC) was performed on a Rainin system using a HydroDore Ax column (1.0 X 10.0 cm) from Rainin.
Methods DNA Helicase Substrates-The helicase substrates used in this study were either circular or linear partial duplex DNA molecules as previously described (24,26). Circular partial duplex substrates were constructed by annealing the complementary strand of a specific M13 mp7 RF1 Hoe111 restriction fragment to M13 mp7 ssDNA and labeling the 3' terminus using [a-32P]dCTP and DNA polymerase I (large fragment) as previously described (26). The linear partial duplex substrate used to determine the direction of unwinding was constructed as previously described (24).
Helicase Activity Assay-The helicase activity assay measures the displacement of a [32P]DNA fragment from a partial duplex DNA substrate (26). The reaction mixture (20 pl) contained 40 mM Tris-HCl (pH 7.5), 4 mM MgC12, 1 mM dithiothreitol, 50 pg/ml bovine serum albumin, 2 mM ATP, and DNA substrate at a concentration carried out at 30 "C and stopped by the addition of EDTA and loading of approximately 2 p M (nucleotide equivalents). Helicase assays were dyes, and the displaced DNA fragment was resolved from the substrate DNA on a nondenaturing polyacrylamide gel as described (26). The unwound product was visualized by autoradiography and quantitated by slicing it from the gel and counting in a liquid scintillation counter.
ATPase Activity Assay-ATPase activity was measured as previously described (27). Reaction mixtures were identical with those used for the helicase activity assays except that [3H]ATP was used at a concentration of 0.54 mM in assays to monitor purification and at 2.1 mM during subsequent characterization. ATPase activity was measured at 30 'C in the presence of M13 mp7 ssDNA added to a final concentration of 31 p M (nucleotide equivalents), unless otherwise specified.
Purification of a DNA Helicase from Yeast-All of the following purification steps were performed at 4 "C. Approximately 1.8 kg of yeast cells (wet weight) were suspended 1:l (v/v) in lysis buffer with added proteinase inhibitors. Cells were lysed using glass beads in a BioSpec bead beater, 6 X 30-s cycles with 2-min cooling periods between disruption cycles. The crude lysate was clarified by centrifugation at 30,000 X g for 90 min. The amount of protein recovered in the supernatant (fraction I) was equivalent to 3.4% of the wet weight of cells lysed.
Nucleic acids and associated proteins were precipitated by the dropwise addition of a 10% solution of Polymin-P (pH 8.0, in water) to a final concentration of 0.2%. Stimng was continued for 15 min, and the precipitate was collected by centrifugation at 30,000 X g for 20 min. The pellet was suspended using a Dounce homogenizer in buffer A with 1 M KC1 and proteinase inhibitors. The volume of the suspension buffer was adjusted to 1 m1/2 g of lysed cells. The resus-pended mixture was clarified by centrifugation for 20 min at 30,000 X g (fraction 11).
Fraction I1 was dialyzed against buffer A until the conductivity was equivalent to that of buffer A. Approximately 30% of the protein precipitated during dialysis and was removed by centrifugation. The supernatant was divided into four aliquots, and each was loaded onto a 160-ml phosphocellulose column (5 X 8 cm) equilibrated with buffer A. Each column was washed with 2 column volumes of buffer A + 0.2 M KC1 and eluted with a linear 10-column volume gradient from 0.2 to 0.8 M KC1 in buffer A. Fractions were assayed for ATPase activity in the presence of ssDNA, and fractions containing a peak of ATPase activity eluting between 0.55 and 0.65 M KC1 were pooled (fraction 111).
Fraction 111 was dialyzed against buffer B until the conductivity was equivalent to that of buffer B and loaded onto a 20-ml dsDNAcellulose column (2.5 X 5 cm) equilibrated with buffer B. The column was eluted with a linear 15-column volume gradient from 0.1 to 1.0 M NaCl in buffer B. Two peaks of ATPase activity were resolved; one eluted at 0.32-0.42 M NaCl, and the other eluted at 0.47-0.60 M NaCl. Fractions containing the second peak of activity were pooled and dialyzed against buffer C (fraction IV). Fraction IV was loaded onto a 4-ml heparin-agarose column (1.0 X 4.5 cm) equilibrated with buffer C. The column was washed with 5 volumes of buffer C containing 0.2 M NaCl and developed with a linear 15-column volume gradient from 0.2 to 0.7 M NaCl in buffer C. Fractions containing peak ATPase activity, eluting between 0.47 and 0.57 M NaCl, were pooled (fraction V) and dialyzed against buffer D until the conductivity was equivalent to that of buffer D. After dialysis fraction V was clarified by centrifugation at 30,000 X g for 20 min, and the supernatant was loaded onto an anion exchange HPLC column (1.0 X 10.0 cm) and eluted with a 60-ml linear gradient from 0.075 to 0.5 M NaCl in buffer D. ATPase activity eluted in a broad peak from 0.1 to 0.25 M NaCl, and fractions in this range were pooled (fraction VI). After dialysis against buffer E, fraction VI was loaded onto a 0.6-ml ATP-agarose column (0.8 X 1.2 cm) that had been equilibrated in buffer E. The column was washed with 5 volumes of buffer E containing 0.2 M NaCl and eluted with a 15-column volume gradient from 0.2 to 0.5 M NaCl in buffer E. Peak ATPase activity eluted at 0.27-0.35 M NaC1, and fractions containing the activity were pooled (fraction VII) and dialyzed against buffer C containing 10% glycerol and 0.2 M NaCl. Fraction VI1 was loaded onto a 0.5-ml heparin-agarose column (0.8 x 1.0 cm) equilibrated in buffer C containing 0.2 M NaCl and 10% glycerol. The column was washed with 5 volumes of buffer C containing 0.4 M NaCl and 10% glycerol and developed using a 15-column volume linear gradient from 0.4 to 0.7 M NaCl in buffer C containing 10% glycerol. Fractions eluting at 0.52-0.60 M NaCl contained the peak of both ATPase and helicase activity and were pooled as fraction VIII. Fraction VI11 was divided into 0.2-ml aliquots, which were layered onto continuous glycerol gradients (5 ml, 15-35% glycerol in buffer C with 0.2 M NaC1). Centrifugation was at 55,000 rpm, 4 "C for 25 h in a Beckman SWTi55 rotor. Each centrifugation run also contained parallel gradients with sedimentation markers catalase (11.2 s), aldolase (7.4 s), and bovine serum albumin (4.3 s). The ssDNA-dependent ATPase and helicase activities cosedimented during glycerol gradient ultracentrifugation. Fractions containing the ATP hydrolysis-dependent helicase were pooled and supplemented to 50% glycerol in buffer c containing 0.2 M NaC1. These were stored at -20 'C and used in subsequent studies. The enzyme was stable for at least 6 months.
Other Methods-DNA concentrations were determined by measuring the absorbance at 260 nm and are expressed as nucleotide based on the known concentration of DNA in the labeling reaction equivalents. The concentration of helicase substrate was estimated and assuming a 75% recovery of the DNA after removal of unincorporated nucleotide by gel filtration (26). Protein concentrations were determined by the method of Bradford (28), with bovine serum albumin as the standard. Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate was by the method of Laemmli (29).

RESULTS
Purification of DNA Helicase I (scHelI) from Yeast-The purification of S. cereuisiae DNA helicase I (scHel1) is summarized in Table I. Whole cell lysates were initially fractionated using polymin-P (0.2% final concentration) to precipitate nucleic acids and associated proteins. It has been reported that Rep protein from E. coli sediments with the nucleic acid pellet in polymin-P precipitations (30). For this reason, the polymin-P pellet from a yeast lysate was extracted with high salt (1 M KCl) and assayed for DNA-dependent ATPase activity. There was significant activity in the high salt eluate, and this activity was monitored in subsequent purification steps.
The supernatant derived from high salt extraction of the polymin-P precipitate was dialyzed and chromatographed on phosphocellulose as described under "Experimental Procedures." This step resolved three peaks of ATPase activity (Fig. 1). Further fractionation of the second and third peaks of DNA-dependent ATPase activity revealed three distinct DNA helicases (data not shown). The second peak contains scHelIII, and the third peak has a mixture of two DNA helicases, scHelII and scHelI. In subsequent purification steps, both ATPase and helicase activity were monitored. However, contaminating nuclease activity prevented quantitation of helicase activity until fraction VIII, the second heparin-agarose pool.
The DNA-dependent ATPase activity eluting from the phosphocellulose column at approximately 0.6 M KC1 was further fractionated on dsDNA-cellulose, which resolved two peaks of DNA-dependent ATPase activity (data not shown). The first activity peak, which eluted at 0.4 M NaCl, had DNA helicase activity but was heavily contaminated with nuclease activity at this stage of the purification. This activity has been further purified and characterized as scHel11.' The ssDNA-dependent ATPase eluting at 0.47-0.60 M NaCl had DNA helicase activity and appeared as a single broad peak. Further fractionation on heparin-agarose, an anion exchange matrix, and ATP-agarose resolved the DNA helicase activity from most of the contaminating nuclease activity and increased the ATPase specific activity 3-fold (Table I). After fractionation on a second heparin-agarose column, aliquots from each fraction were resolved on a polyacrylamide gel run in the presence of sodium dodecyl sulfate. Both helicase and ssDNA-dependent ATPase activity cochromatographed with a polypeptide of approximately 135 kDa, but high levels of a low molecular weight polypeptide were present in fractions containing peak enzymatic activity (data not shown). Preparative glycerol gradient ultracentrifugation resolved the 135-' D. W. Bean and S. W. Matson, manuscript in preparation. kDa polypeptide from the low molecular weight polypeptide (Fig. 2) and from several minor contaminants between 60 and 100 kDa. Both helicase and ATPase activity cosedimented with the 135-kDa polypeptide, which was the predominant polypeptide present in peak activity fractions (Fig. 2, lower panel, fractions 9-11). Nuclease activity, measured as degradation of the helicase substrate, was not detected in the gradient. The helicase/ATPase activity sedimented between the 7.4 s and the 4.3 s markers consistent with a monomeric protein with a relative molecular mass of approximately 135,000. We conclude that scHelI is a monomeric protein with an M, = 135,000.
Reaction Requirements-Purified scHelI unwound the double-stranded region of a 71-bp partial duplex molecule in a reaction that required ATP or dATP and MgC12 or MnC12 (Table 11). None of the remaining NTPs (dNTPs) could effectively substitute for ATP when present at a concentration of 1.0 mM. In the absence of MgC12, or in the presence of 10 mM EDTA, helicase activity was below detectable levels. DNA helicase activity was dependent on MgClz with an optimum concentration between 1.0 and 2.0 mM (Fig. 3B) in the presence of 2.0 mM ATP. Titrations using MnC12, ZnC12, or CaC12, substituted for MgC12, indicated that MnClz could substitute for MgClz but was less effective (data not shown). ZnC12 and CaClz could not substitute for MgClz in the helicase reaction. The poorly hydrolyzed ATP analog, ATPyS, could not substitute for ATP in the unwinding reaction, suggesting a requirement for concomitant NTP hydrolysis. ATPyS has also been shown to be a competitive inhibitor of the unwinding reaction (data not shown). NaCl concentrations greater than 40 mM inhibited the unwinding reaction (Fig. 3C), and, at 200 mM NaCl, the unwinding reaction was diminished by more than 95% (Fig. 3C and Table 11).
The pH optimum for the unwinding reaction ranged broadly, from 6.5 to 8.0 (Fig. 3A). Unwinding dramatically decreased below pH 6.5 and was barely detectable at pH 5.5. The activity above pH 8.0 was only slightly diminished. Three different buffers were used to span the pH range, and no significant buffer effects were measured at the points of overlap (pH 6.5 and 7.5).
The Unwinding Reaction Catalyzed by scHelI Is Dependent on Protein Concentration-scHelI catalyzed the displacement of DNA fragments of various lengths that had been annealed to M13 mp7 ssDNA (Figs. 4 and 6). The unwinding reaction was proportional to protein concentration on substrates with either 71-or 216-bp partial duplex regions (Fig. 4). As the length of the duplex region increased, the fraction of the substrate unwound by scHelI decreased. Unwinding of a 343bp partial duplex substrate was barely detectable (see Fig. 6).   The Unwinding Reaction Catalyzed by scHetI Is Time-dependent-The kinetics of the unwinding reaction catalyzed by scHelI exhibited both a fast and a slow component (Fig.  6). During the first 5 min, the rate of unwinding was rapid with both the 71-bp substrate and the 343-bp partial duplex substrate plus E. coli SSB. Unwinding of the 343-bp substrate, in the presence of SSB, continued at a linear, but much slower, rate out to 60 min. Further unwinding of the 71-bp partial duplex substrate could not be detected after 20 min.

E. coli SSB
Without SSB, the unwinding of the 343-bp partial duplex substrate was barely detectable, even after a 60-min incubation. The kinetics of unwinding were the same whether the reactions were initiated by the addition of enzyme or by the addition of ATP (data not shown).
ScHelI unwinds longer duplex regions less effectively than short duplex regions (Figs. 4 and 6), which has also been reported for a number of prokaryotic helicases (26,32). However, the addition of SSB increased the fraction of the 343bp partial duplex substrate unwound to that observed using the 71-bp partial duplex substrate. The addition of SSB served to stimulate the unwinding of a 71-bp partial duplex substrate only modestly (data not shown).
The ATPase Activity of scHelI Is ssDNA-dependent-As expected, scHelI also catalyzed the hydrolysis of ATP in a reaction that was ssDNA-dependent (Fig. 7). In the absence of ssDNA, ATP hydrolysis was barely detectable. Both M13 mp7 ssDNA and linear poly(dT) (approximately 1100 nucleotides in length) served as effectors in the ATP hydrolysis reaction. Interestingly, the rate of ATP hydrolysis in the presence of poly(dT) was nearly twice that measured in the presence of M13 mp7 ssDNA. Supercoiled dsDNA, linear dsDNA, and poly(A) supported a low level of ATP hydrolysis. The titration curves presented in Fig. 7 indicate a Kerf value for these three nucleic acids at least 20-fold higher than that of ssDNA, where Kerf is defined as the nucleic acid concentration required to achieve ATP hydrolysis at one-half the maximum velocity, at saturating concentrations of ATP. Poly(C) and tRNA (data not shown) failed to support the ATPase reaction. In reactions containing 2 PM M13 mp7 ssDNA, the addition of 13 nM E. coli SSB (tetramer), sufficient to achieve about 35% saturation of the ssDNA, had little or no effect on the ATPase activity of scHelI. The addition of 130 nM SSB reduced ATPase activity by 30% (data not shown). This concentration of SSB is more than three times that required to fully coat the ssDNA. The apparent K , for rATP was determined to be 325 PM under standard reaction conditions (data not shown).
[ in the presence of ssM13 mp7, displayed linear kinetics throughout a 60-min incubation (Fig. 8). This is in contrast to the biphasic reaction kinetics observed when the DNA unwinding reaction was monitored. scHelI Unwinds Duplex DNA in a 5' to 3' Direction-To determine the polarity of the unwinding reaction catalyzed by scHelI, the DNA substrate shown in Fig. 9A was constructed. This molecule contains a 143-bp duplex region at the 5'-end, a 202-bp region at the 3'-end, and a long internal region of ssDNA to which the helicase may initially bind. An enzyme that unwinds duplex DNA with a 5' to 3' polarity with respect to the ssDNA on which it initially binds is expected to displace the 202-nucleotide DNA fragment; an enzyme with the opposite polarity will unwind the 143-nucleotide DNA fragment.
scHelI, in the presence of E. coli SSB (130 nM, tetramer), catalyzed the unwinding of the 202-nucleotide DNA fragment with no detectable unwinding of the 143-nucleotide DNA fragment (Fig. 9B, lane 4 ) indicating a 5' to 3' polarity for the unwinding reaction. ScHelI without added SSB was not capable of efficiently displacing the partial duplex direction substrate (Fig. 9B, lane 5 ) . By increasing the amount of scHelI 10-fold, unwinding of the 202-nucleotide fragment could be detected without SSB (data not shown). Thus, the addition of SSB did not alter the polarity of unwinding but served to stimulate the unwinding reaction. Control experiments were performed using E. coli DNA helicase I1 (Fig. 9B, lune 3 ) , a 3' to 5' helicase; E. coli DNA helicase I (Fig. 9B, Zane Z), a 5' to 3' helicase; and SSB in the absence of a helicase, which was incapable of displacing either DNA fragment (Fig. 9B,  lane 1 ). We conclude that scHelI catalyzes a 5' to 3' unwinding reaction.

DISCUSSION
We have used a direct, in vitro helicase assay to identify, purify, and characterize a previously undescribed ATP-dependent DNA helicase activity from whole cell lysates of S. cerevkiae. The purified enzyme consists of a single polypeptide with a molecular mass of 135 kDa as determined by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. The sedimentation velocity of the enzyme in a continuous glycerol gradient suggests the protein exists as a monomer in solution. This enzyme has been designated scHelI. scHelI unwinds duplex DNA in a reaction that depends on the addition of both MgC12 and ATP (or dATP). The poorly hydrolyzed ATP analog, ATPrS, could not substitute for ATP in the unwinding reaction, indicating a requirement for concomitant ATP hydrolysis as has been observed for helicase enzymes isolated from other sources (1). Unwinding is in the 5' to 3' direction with respect to the strand of DNA on which the enzyme initially binds.
Kinetic studies indicate the unwinding reaction catalyzed by scHelI consists of a rapid phase lasting approximately 5 min, during which most of the unwinding occurs, followed by a slow phase in which duplex DNA is steadily displaced for at least 60 min. The fraction of substrate molecules unwound during the rapid phase is proportional to the ratio of substrate to enzyme. When the concentration of substrate is decreased by one-half, the fraction of unwound molecules is nearly doubled. This is in contrast to the kinetics of ATP hydrolysis, which are linear for at least 60 min over a range of DNA effector concentrations, ATP concentrations, and enzyme concentrations. These observations suggest a rapid associa- tion of the enzyme with the partial duplex substrate followed by unwinding of the duplex region and a continued hydrolysis of ATP, while the enzyme remains associated with the ssDNA circle. Dissociation and association with other partial duplex substrate molecules is apparently slow, resulting in the slow phase of unwinding kinetics. Unwinding of other substrates, such as linear partial duplex DNA, fully duplex linear DNA, and DNARNA hybrid substrates, must be tested to further characterize the unwinding reaction. The purification of scHelI was monitored using a partial duplex substrate with 71 bp of duplex DNA. When the length of the duplex region on the partial duplex substrate was increased, the fraction of the substrate unwound by scHelI decreased. In fact, unwinding could barely be detected when the duplex region was 343 bp in length (Fig. 6). However, the unwinding reaction measured using longer partial duplex sub- strate. In addition, only slight stimulation of unwinding was achieved by incubation of the 71-bp partial duplex substrate with SSB. The stimulatory effects of SSB are likely to have an impact on a step or steps in the unwinding reaction that become more important as the length of the duplex region is increased. Since SSB is not able to catalyze the unwinding of duplex DNA, it seems likely that SSB functions by coating the ssDNA present in the reaction to enhance unwinding. We presume that SSB is bound to both the ssDNA along which the enzyme translocates to move to the duplex region and the ssDNA fragment produced during the course of the unwinding reaction. Interestingly, the addition of SSB has little effect A. on the ATP hydrolysis reaction catalyzed by scHelI. This is in contrast to what has been observed for the majority of prokaryotic helicases that have been described (1). The stimulatory effect of SSB on the unwinding reaction catalyzed by scHelI is similar to that previously reported for a minority of the prokaryotic enzymes. For example, the unwinding reaction catalyzed by DnaB protein from E. coli is stimulated 6fold by the addition of SSB (33), while PriA protein (34) shows nearly absolute dependence on SSB for unwinding of long duplex regions. Among helicases identified from eukaryotes, several of the calf thymus helicases are stimulated by E. coli SSB or the eukaryotic SSB analogs (11, 12), while a DNA helicase isolated from a human cell line is dependent on eukaryotic SSB for unwinding activity (17).
Five DNA helicases, isolated from yeast, have been described biochemically as follows: Rad3 protein (5), ATPase I11 (35), Pifl protein (7), Srs2 protein (9), and an RF-Cassociated helicase (36). scHelI appears to be distinct from each of these previously described helicases on the basis of biochemical and/or physical criteria. The molecular masses of Rad3 (89 kDa), Pifl (97 kDa) and Srs2 (134 kDa) have been deduced from DNA sequence data and confirmed for the purified proteins. ATPase I11 (63 kDa) and RF-C-associated helicase (60 kDa) have been characterized on polyacrylamide gels run in the presence of sodium dodecyl sulfate. Only the SRS2 gene product is large enough to be scHelI; however, biochemical criteria distinguish the two enzymes. The Srs2 protein has been shown to display a 3' to 5' polarity of unwinding (9) using a partial duplex substrate similar to the one used in this study (Fig. 9). ScHelI, in contrast, exhibits a 5' to 3' polarity in the unwinding reaction.
A number of genes in S. cerevkiae have been shown to share sequences that are found in DNA helicases. Several of these genes, including FUN30 (18), STHl (19), and RAD5 (20) are of approximately the correct size to encode scHelI. Of these, only rad5 mutants have thus far been shown to display phenotypes consistent with a helicase function. The RAD5 gene has been placed in the RAD6 epistasis group of DNA repair proteins, and it encodes a polypeptide of 134 kDa (20). Thus, Rad5 protein and scHelI could be equivalent, but, since the gene product of RAD5 has not been characterized biochemically and the gene encoding scHelI has not yet been cloned, we can only speculate on this relationship. We conclude, on the basis of both polypeptide size and biochemical properties, that scHelI is a previously undescribed helicase. Any speculation as to the role played by this protein in nucleic acid metabolism must await cloning of the gene encoding this protein and a description of the phenotype of scHelI mutants. Thommes .