Cloning and simplified purification of Escherichia coli DNA gyrase A and B proteins.

We have transferred the Escherichia coli gyrA and gyrB genes onto plasmids that allow the overproduction of the DNA gyrase A and B proteins and have designed relatively simple purification procedures for both proteins. The pure proteins are obtained in good yield; from 2 liters of culture (12 g of cells), one can recover 25 mg of GyrA or 3 mg of GyrB protein.

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact, added separately to a reaction mixture which contains all the other components. In a variant assay procedure, which gives a severalfold higher specific activity for GyrB (see "Results" and Table I), the Gyrh and GyrB proteins are preincubated together at high concentration (0.05-0.25 mg/ml of each protein) in TGED-2 buffer containing 50% (w/v) glycerol and 70 mM KC1 for 30 min at 20 "C. The enzyme is then diluted as usual and the assay is started promptly.
Cloning of gyrA and gyrB Genes onto Plasmids The plasmid vector for cloning gyrA was pKC16 ( l l ) , composed essentially of the large EcoRI-BamHI fragment of pBR322 and the N-cI-0-P region of phage X. A BamHI fragment (approximately 11 kilobases) containing the gyrA gene was isolated from AdnalA (3) and was inserted at the BarnHI site of pKC16. Transformants in E. coli strain RW1053 recA Aka1 attA bio) were screened for overproduction of GyrA protein. Strain N4186 = RW1053 (pMK9O) was used for the enzyme purification described below.
In cloning gyrB, we took account of the fact the the genome region near gyrB contains genes (e.g. dmA and dnaN) whose presence on high-copy-number plasmids could lead to poor growth of the host cells. (In the present case, this precaution may have been unnecessary; see below.) We therefore constructed a composite plasmid vector that could be maintained at either high-or low-copy number by replication from either the pBR322 origin or from the F factor origin. In a polA host strain, the pBR322 origin is inactive (12) and the plasmid copy number drops to that of the F factor (about three copies/cell). To construct this vector, pMM121, we joined the large EcoRI fragment ofpKC16 to the EcoRI-5 fragment of FA(0-15) (13); the small BamHI fragment of EcoRI-5 was deleted in the process. Due to the loss of the small EcoRI fragment of pKC16 that contained phage X replication functions, pMM121 cannot replicate from the A origin. Plasmid pMM121 was linearized by partial digestion with BarnHI nuclease. It was then ligated to a BamHI fragment (approximately 10 kilobases) which had been isolated from Atna 552 (14). Ampicillin-resistant transformants were selected first in E. coli N1069 polA and screened for the proper orientation of the inserted fragment relative to the PL promoter of the vector. The DNA of one such transformant plasmid, pMM115, was then used to transform strain RW1053 (pol'). Transformants in the polA and pol+ strains grew at similar rates; the plasmid was therefore maintained in RW1053.
To reduce the size of the cloned segment, a Hind111 fragment containing the gyrB, recF, dnaN, dmA, and AN genes was trimmed with nuclease Ba131 and then cut with BarnHI and inserted at the BamHI site of pKC16. After transformation of RW1053 and selection for ampicillin resistance, the resulting clones were screened for the orientation of the inserted fragment and for high production of GyrB protein. One such transformant, MK47, clearly produced more GyrB than RW1053 (pMM115) and was therefore used as a source of GyrB in the purification procedure described below. The plasmid contained in this strain is designated pMK47.
Growth and Lysk of Cells Cells were grown in a 100-or 300-liter fermentor in a medium containing, per liter, 10 g of Bacto-tryptone (Difco), 5 g of yeast extract (Difco), 10 g of NaCI, 5 g of glucose, 5 ml of 1 M potassium phosphate (pH 7.01, and 1 mg of biotin. The last three ingredients were added separately after autoclaving. The culture was aerated at 32 "C until A690 reached 0.6. The temperature was raised to 42 "C for 15 min and then lowered to 37 "C; aeration was continued throughout.

9199
The pH of the culture was continuously adjusted to 7.0 by adding 5 M KOH. At A5* = 4.0, the culture was chilled and the cells were collected by centrifugation (about 600 g of packed cells from a 100liter culture).
The cells were resuspended with 0.05 M Tris-HC1 (pH 7.5), 10% sucrose (66 m1/100 g of cells) a t low speed in a Waring blender, frozen in 40-ml aliquots in liquid nitrogen, and stored at -70 "C.
Cell lysates were prepared by the action of lysozyme and Brij-58, as described (15). All the purification steps were carried out at 0-4 "C. All centrifugations were at 10,000 X g for 15-20 min.
Purification of DNA Gyrase A Protein Polymin P Fractionation-23 ml of a crude extract of strain N4186 (fraction 1; protein concentration of 21.4 mg/ml) were thawed, diluted to a final protein concentration of 8 mg/ml with 0.05 M Tris-HC1 (pH 7.5),10% sucrose, and brought to 0.2 M NaCl with 4 M NaCl. To the resulting 62 ml of diluted extract, 4.3 ml(O.07 volume) of 5% Polymin P were added over 10 min with stirring, and stirring was continued for 15 min. After centrifugation, the bottles were drained and the pellets were resuspended in 25 ml of 0.45 M NaCI, 0.05 M Tris-HC1 (pH 7.5), 1 mM Na3EDTA, 1 mM dithiothreitol with the use of a stirring rod. After further stirring for 15 min, the suspension was centrifuged, the supernatant was discarded, and the pellets were resuspended in 25 ml of 1 M NaCI, 0.05 M Tris-HC1 (pH 7.5), 1 mM Na3EDTA, 1 mM dithiothreitol. After stirring for 15 min, the suspension was again centrifuged and the supernatant solution was collected. Solid ammonium sulfate (0.31 g/g of supernatant) was added, and the suspension was stirred for 15 min. The precipitate was collected by centrifugation and resuspended in 4 ml of 0.1 M KCl, 0.05 M Tris-HC1 (pH 7.5), 1 mM Na3EDTA, 1 mM dithiothreitol (fraction 2; 6.0 ml, 90 mg of protein).
DEAE-Sepharose Chromatography-Fraction 2 was dialyzed for 4 h against 2 liters of TGED buffer, centrifuged to remove a precipitate that formed during dialysis, and then diluted with 2 volumes of TGED buffer. The sample was loaded onto a DEAE-Sepharose column (bed volume of 8 ml) previously equilibrated with TGED buffer containing 0.025 M NaCl. The column was washed with 40 ml of the same solution, and the protein was eluted with a 240-ml linear gradient of 0.025-0.5 M NaCl in TGED buffer. Active fractions were eluted around 0.2 M NaCl in a peak that contained the bulk of the protein applied to the column (fraction 3; 22.0 ml, 62.4 mg of protein).
Valine-Sepharose Chromatography-A 1.7-ml portion of fraction 3 was diluted in small aliquots with 3 volumes of 2 M potassium phosphate (pH 7.5), 1 mM Na,EDTA, 5 mM dithiothreitol and loaded onto a column of valine-Sepharose (bed volume of 1 ml) previously washed with TGED buffer and then equilibrated with 1.5 M potassium phosphate (pH 7.5) in TGED buffer. The column was washed with 8 ml of this solution, and the protein was eluted with a 30-ml linear gradient of 1.5-0.0 M potassium phosphate (pH 7.5) in TGED buffer. GyrA activity was eluted around 0.85 M potassium phosphate. Active fractions were frozen in liquid nitrogen and stored at -70 "C (fraction 4; 14.5 ml, 3.0 mg of protein). Fractions 1-4 were all stable in storage a t -70 "C for at least 1 year.
Purification of DNA Gyrase B Protein Streptomycin-Ammonium Sulfate Fractionation-A crude extract of strain MK47 (33.5 ml and 39 mg/ml of protein) was diluted to a protein concentration of 20 mg/ml with 0.05 M Tris-HC1 (pH 7.51, 1 mM NasEDTA, 2 mM dithiothreitol (final volume of 65 ml). A 20% (w/v) solution of streptomycin sulfate was added dropwise, with stirring, to a final concentration of 4%. After stirring for an additional 15 min, the mixture was centrifuged and the supernatant solution was retained. Solid ammonium sulfate (0.31 g/g of supernatant) was added with stirring. After 15 min of stirring, the precipitate was collected by centrifugation and redissolved in 9 ml of TGED buffer (fraction 2; 15 ml, 670 mg of protein).
h against 2 liters of TGED-3 buffer and diluted with 6 volumes of the Heparin-Agarose Chromatography-Fraction 2 was dialyzed for 3.5 same buffer to reduce the conductivity to 1.5 times that of TGED-3 buffer. The sample was loaded onto a 30-ml column of heparinagarose (0.55 mg of heparin/ml) equilibrated with the same buffer. After the column was washed with 3 volumes of TGED-3 buffer, GyrB activity was eluted with 0.3 M KC1 in the same buffer (fraction 3; 20.8 ml and 110 mg of protein). DEAE-Sepharose Chromatography-An 11-ml column of DEAE-Sepharose was pretreated with 20 ml of bovine serum albumin (5 mg/ ml in TGED-2 buffer containing 0.025 M NaCI), washed with several column volumes of 0.3 M NaCl in TGED-2, and then equilibrated with 0.025 M NaCl in TGED-2. (This pretreatment improved the recovery of GyrB considerably.) Fraction 3 was dialyzed for 2.5 h against TGED-2 buffer, diluted with 1 volume of the same buffer, and loaded onto the column. The column was developed with a 400ml linear gradient of 0.025-0.3 M NaCl in TGED-2 buffer. Fractions with GyrB activity were eluted around 0.11 M NaCl (fraction 4; 32 ml, 21.2 mg of protein).
Leucine-Agarose Chromatography-A 2-ml column of leucine-agarose was washed with TGED buffer and equilibrated with 1.5 M potassium phosphate (pH 7.5) in TGED. A 3.75-m1 portion of fraction 5 was diluted in small aliquots with 3 volumes of 2 M potassium phosphate (pH 7.5), 1 mM Na3EDTA, 5 mM dithiothreitol and loaded onto the column. The column was washed with 8 ml of the equilibration buffer, and the protein was eluted with a 60-ml linear gradient of 1.5-0.5 M potassium phosphate (pH 7.5) in TGED buffer. GyrB activity was eluted around 1.15 M potassium phosphate (fraction 6; 15 ml, 1.6 mg of protein).
Fractions 1-6 were all stable at -70 "C for at least 1 year. Table I summarizes the purification procedures. The two activities are purified with an overall recovery of about 50% for GyrA and 40% for GyrB. Both protein preparations are nearly homogeneous when displayed in sodium dodecyl sulfate-polyacrylamide electrophoresis (Fig. 1). Very minor contaminants (<2% total) that could be seen in the original gel are not visible on the photographic print.

RESULTS AND DISCUSSION
The strains that carry gyrA or gyrB plasmids, combined with the purification method described above, are a useful source of both DNA gyrase proteins. From 12 g of packed cells (the yield of about 2 liters of culture), one can recover more than 25 mg of purified GyrA protein or 3 mg of GyrB protein. The higher yield of GyrA protein reflects its higher 6. Leucine-agarosea 65.1 7.0 7 X lo5 5 X lo6 *These steps were run using an aliquot of the preceding fraction, as described under "Materials and Methods." The numbers in this table have been adjusted accordingly.
The values in parentheses refer to the modified assay in which the GyrA and GyrB proteins are preincubated together at high concentration; the specific activity of each protein is determined with an excess (in units) of the other (see "Materials and Methods").  Table I).
content in the crude extract, where at least 10% of the total protein is seen to be GyrA (Fig. 1). Even the lower content of GyrB reflects a t least a 20-fold overproduction by MK47 compared to a normal E. coli strain.
Neither protein preparation had any detectable DNA supercoiling activity in the absence of the complementing protein (assayed with 100 units), indicating the absence of crosscontamination of the A and B subunits. Both protein preparations assayed singly were also free of contaminating topoisomerase and endonuclease activity under the gyrase assay conditions (none detected with 100 units of GyrA or GyrB protein). The ATPase activity of the gyrase complex was inhibited more than 95% by novobiocin in the absence or presence of DNA, indicating that there is little contamination by nonspecific ATPases.
For some uses, the penultimate fractions (fraction 3 of GyrA and fraction 5 of GyrB protein) are pure enough. They are substantially free of contaminants visible in gel electrophoresis, and their specific activities in supercoiling are nearly the same as those of the final preparations. However, these earlier fractions do contain low levels of contaminating enzymatic activities (for example, novobiocin-insensitive ATPase) which are removed at the final step.
A peculiar feature of this gyrase preparation and of that resulting from at least one previous purification of the separate A and B subunits (8) is that the specific activity of GyrB in the supercoiling reaction is consistently lower than that of GyrA. If, as is believed, the two proteins work in DNA supercoiling only as an equimolar complex (5,6, 16) and both are fully active, one would expect to find that they have equal molar activities. Indeed this is the result that is found when ATP hydrolysis or DNA relaxation catalyzed by our preparations of the subunits is assayed.' Because ATPase assays are relatively insensitive, and because DNA gyrase has much less DNA relaxing activity than supercoiling activity (15), the protein concentration in these reactions is about 100-fold higher than in the supercoiling assay. However, in the normal supercoiling assay, where both proteins are added to DNA at very low concentration, the specific activity of GyrB assayed in the presence of excess GyrA is a t least 10-fold lower than that of GyrA assayed with excess GyrB (Table I), and even with a very large molar excess of GyrA (>lOO-fold) the GyrBspecific activity does not increase further.' One might therefore conclude either that the GyrB preparation is largely inactive in supercoiling, although competent in ATP hydrolysis and DNA relaxation, or that the extra factor of dilution in the supercoiling assay is responsible. We believe the latter to be the case, because when the two proteins are preincubated together a t high concentration in the variant assay procedure described under "Materials and Methods," the specific activity of GyrB increases notably, as is shown by the values in parentheses in Table I. The increase, which has ranged from 4 to 6-fold, is optimal if the subunits are preincubated at concentrations above 0.1 mg/ml. The increase of activity is not changed by the addition of magnesium ion, ATP, DNA, or any combination of them to the preincubation. When GyrA is in molar excess, the increase of the specific activity of GyrB is independent of molar ratio; when GyrB is in molar excess, there is no increase of GyrA activity.
When assayed in this way, our preparation of GyrB has a specific activity 2.5-fold less than GyrA. As a practical measure, the preincubation makes it possible to use the two subunits at comparable molar concentrations without a great sacrifice in activity. The specific activities of the subunits measured in this way are at least as high as those previously reported for GyrA (8), GyrB (7, 8), or the purified complex (5) when adjusted for different definitions of units. The mechanism of the activation is still obscure. It is reversed after dilution of the enzyme into the assay medium if the tubes are kept a t 0 "C for 30 min before being transferred to 25 "C to start the supercoiling reaction, a result which again implicates the state of aggregation of the enzyme. Further studies on the process are being carried out.