(Received for publication, November 8, 1996, and in revised form, March 9, 1997)
From the Laboratory of Molecular Biology, NIDDK, National Institutes of Health, Bethesda, Maryland 20892-0540
Reverse gyrases are ATP-dependent
type I 5-topoisomerases that positively supercoil DNA. Reverse gyrase
from Methanopyrus kandleri is unique as the first
heterodimeric type I 5
-topoisomerase described, consisting of a
138-kDa subunit involved in the hydrolysis of ATP (RgyB) and a 43-kDa
subunit that forms the covalent complex with DNA during the
topoisomerase reaction (RgyA). Here we report the reconstitution of
active reverse gyrase from the two recombinant proteins overexpressed
in Escherichia coli. Both proteins have been purified by
column chromatography to >90% homogeneity. RgyB has a
DNA-dependent ATPase activity at high temperature
(80 °C) and is independent of the presence of RgyA. RgyA alone has
no detectable activity. The addition of RgyA to RgyB reconstitutes positive supercoiling activity, but the RgyB and RgyA subunits form a
stable heterodimer only after being heated together. This is the first
case in which it has been possible to reconstitute an active
heterodimeric enzyme of a hyperthermophilic prokaryote from recombinant
proteins.
In the 1980s organisms were discovered that thrive at temperatures >100 °C. These hyperthermophiles have provided a host of enzymes for scientific study and potential industrial application (1). However, many archaea fail to grow in the laboratory (2), and the majority that can be grown do so inefficiently (3). To fully realize the potential benefits of these enzymes and to make them easily accessible for study in the laboratory, it is advantageous to use recombinant DNA technology to express hyperthermophilic proteins in heterologous hosts and then recover active enzymes (1, 4). Although many active thermostable enzymes have been recovered after expression in Escherichia coli, until now it had not been possible to reconstitute an active heteromultimeric hyperthermophilic enzyme from recombinant proteins (1).
Reverse DNA gyrases, a novel class of type I 5-topoisomerases,
were discovered in hyperthermophiles (5-11). These enzymes, which
positively supercoil DNA in the presence of ATP or dATP, have been
found in all major branches of hyperthermophilic archaea as well as in
hyperthermophilic eubacteria (12, 13). Reverse gyrase is thus
considered a molecular marker for hyperthermophilic prokaryotes (14).
Although the physiological role of the activity remains obscure, it has
been postulated that positive supercoiling helps to stabilize the DNA
duplex against denaturation at the extreme temperatures where these
organisms grow. In the eubacteria and the crenarchaeota branch of
the archaea, reverse gyrases isolated to date are large monomeric
proteins (12, 13). Sequence analysis of reverse gyrase from
Sulfolobus acidocaldarius revealed that the type I
5
-topoisomerase domain was fused to a domain reminiscent of a nucleic
acid helicase (10). Helicases are nucleic acid-dependent ATPases as is reverse gyrase (15). However, it is important to note
that this domain of reverse gyrase may not function as a helicase
during the enzymatic reaction and the conserved amino acid motifs in
this enzyme may serve another function.
Recently, we described a reverse gyrase from Methanopyrus
kandleri, the first type I 5-topoisomerase isolated from the
euryarchaeota branch of the archaea (16-18). Unlike any other type I
topoisomerase, this enzyme is a heterodimer with subunits of 43 kDa
(RgyA) and 138 kDa (RgyB). It was demonstrated that RgyA forms the
covalent link with DNA during the strand cleavage-religation reactions, whereas RgyB is involved in the hydrolysis of ATP (17). We expected the
two protomers of M. kandleri reverse gyrase to consist of the conserved domains found in S. acidocaldarius reverse
gyrase but as separate proteins. However, the cloning and sequence
analysis of the genes encoding the two subunits of M. kandleri reverse gyrase revealed the unexpected result that the
universally conserved type I 5
-topoisomerase domain was not continuous
but instead was shared between the two subunits.
To study the activities of the enzyme and its subunits in greater detail, it is useful to generate larger amounts of protein than can conveniently be made in M. kandleri. Here, we report growth conditions for the recovery of soluble recombinant RgyA and RgyB from E. coli and the reconstitution of active, stable reverse gyrase from the purified proteins.
Materials
Restriction enzymes, T4 DNA ligase, Vent DNA polymerase, calf
alkaline phosphatase, and deoxyribonucleotides were purchased from New
England Biolabs Inc. and used as described by the manufacturer. pET-3
and pET-16b protein expression vectors were from Novagen. Polyethyleneimine (Sigma) was used as described (19).
Isopropyl-1-thio--D-galactopyranoside was from Gold
Biotechnology Inc. HiTrap chelating and heparin columns and a Sephadex
S200HR 10/30 sizing column were from Pharmacia Biotech Inc. High
resolution hydroxylapatite (HTH)1 was from
Calbiochem. Phosphocellulose P-11 was from Whatman. Both L-broth and
SuperbrothMDRV were purchased from Advanced Biotechnologies Inc.
Methods
Column Chromatography, Protein Gel Electrophoresis, and Determination of Protein PurityColumn chromatography was performed using a Pharmacia fast protein liquid chromatography apparatus. All chromatography steps were performed at 4 °C except for the elution of the HTH column of RgyA, which was done at room temperature. Analytical gel filtration was performed using a Pharmacia Smart system with a Superdex 200 3.2/30 column that was standardized using the Pharmacia high and low molecular weight gel filtration kits. For electrophoretic analysis, protein samples were diluted as indicated in the figure legends, mixed 1:1 with Tris/glycine SDS sample buffer (Novex), and separated on 4-24% gradient SDS-PAGE gels (Novex). Proteins were visualized by staining with Coomassie Brilliant Blue (0.25% in 45% methanol, 10% glacial acetic acid) followed by destaining with 10% glacial acetic acid. Protein purity was determined by scanning gels of protein samples with a Microtek Scanmaker II and creating a computer file with Adobe Photoshop and then analyzing band intensity with the NIH Image computer program.
Assay of ATPase ActivityRgyB was first preheated in its
storage buffer for 5 min at 80 °C. Then 0.1 µg of the protein was
combined with 0.1 µg of supercoiled pBR322 in a 20-µl volume of
reaction buffer (50 mM Tris, pH 8, 200 mM NaCl,
10 mM MgCl2, 1 mM DTT, 10% w/v
glycerol) containing 50 µM [-32P]ATP
(600 mCi/mmol, DuPont NEN). Samples were incubated at the temperatures
and times indicated in the figure legends, and the reactions were
terminated by rapid cooling and dilution into 100 mM EDTA,
pH 8. A 1-µl aliquot of each sample was spotted on a 20 × 20-cm
polyethyleneimine cellulose plastic plate (Alltech Associates Inc.),
dried, and developed by ascending chromatography for 2 h with 0.75 M KH2PO4. Radioactivity was
quantitated with a PhosphorImager (Molecular Dynamics, Inc.).
To reconstitute positive supercoiling activity, we mixed amounts of RgyA and RgyB as indicated in their storage buffer (50 mM Tris, pH 8, 1 M NaCl, 1 mM EDTA, 10% glycerol) and heated for 5 min at 80 °C.
For tests of positive supercoiling activity, 1 µl of enzyme (0.1 µg of either the A subunit, the B subunit, or the reconstituted reverse gyrase) was mixed with 0.1 µg of pBR322 DNA (either relaxed or negatively supercoiled, as specified) in 20 µl of reaction buffer (50 mM Tris, pH 8, 200 mM NaCl, 5% glycerol, 10 mM MgCl2) and incubated at 80 °C for 15 min. The reactions were terminated by rapidly cooling to 0 °C and adding 0.2 volume of 5% SDS, 50 mM EDTA, 50% glycerol, 0.5% bromphenol blue. The topoisomerization products were analyzed by electrophoresis in a 1.2% agarose gel in the presence or absence of chloroquine at 3 V/cm for 12 h.
Construction of Expression VectorsRgyA and RgyB were
produced separately in E. coli using the pET
expression system of Studier et al. (20). The
rgyB coding sequence was amplified by polymerase chain
reaction from a recombinant phage carrying the entire gene and
cloned into the NdeI-BamHI site of pET-16b,
placing a 22-amino acid leader containing a poly(His) sequence onto the
amino terminus of the protein. rgyA was amplified by
polymerase chain reaction from a recombinant
phage that carried the
full-length gene. The first codon of the open reading frame was changed
from TTG to ATG, and a second alternative translation start site in a
different reading frame adjacent to the first codon was removed by
making a silent mutation in a Asn codon (changing AAT to AAC).
rgyA was cloned into the NdeI-BamHI
site of pET3.
Four liters of Superbroth were
inoculated with BL21::DE3 pLysE carrying the RgyA or RgyB
expression plasmid. Cells were grown at 37 °C to an
A600 of 0.8 with aeration, then
isopropyl-1-thio-
-D-galactopyranoside (final
concentration, 400 µM) was added, and the culture
immediately shifted to 20 °C for protein expression. After 8 h
the cells were harvested, yielding approximately 16 g of cells
that were then resuspended in 40 ml of freezing buffer (50 mM Tris, pH 8, 10% sucrose, 2 mM EDTA, 100 mM NaCl), frozen in liquid nitrogen, and stored at
70 °C. The cell paste was thawed on ice, and an equal volume of
freezing buffer containing lysozyme (0.2 µg/ml) was added. After
stirring for 30 min at 4 °C, an equal volume of lysis buffer was
added (50 mM Tris, pH 8, 2 mM DTT, 5 mM EDTA, 200 mM NaCl, 0.4% Brij 58) and
stirred for 30 min followed by shearing of the cell lysate with a
rotary homogenizer (Janke and Kunkel, Ultra-Turrax T25) to reduce
viscosity. Cellular debris was removed by centrifugation at 32,000 × g for 30 min at 4 °C, after which a 0.07 volume of 5%
polyethyleneimine chloride, pH 8, was slowly added to the cleared
lysate and then stirred for 30 min. The resulting aggregate was
collected by centrifugation and washed with 50 ml of buffer A (50 mM Tris, 2 mM EDTA, 2 mM DTT)
containing 300 mM NaCl to remove loosely bound protein.
Proteins were removed from the polyethyleneimine aggregate with 50 ml
of buffer A containing 1 M NaCl and precipitated by the
addition of solid ammonium sulfate to a concentration of 50% (w/v). At
this point the purifications of RgyA and RgyB diverged.
The RgyA ammonium sulfate pellet was
resuspended in 25 ml of buffer A (with 1 M NaCl) and
dialyzed overnight against 2 liters of Buffer A (with 300 mM NaCl) containing 10% glycerol. Under these conditions,
RgyA preferentially precipitates out of solution. This precipitate was
collected and then redissolved in 15 ml of buffer B (50 mM
Tris, pH 8, 1 M NaCl, 2 mM DTT, 10% glycerol) plus 10 mM potassium phosphate and 0.1 mM EDTA.
The sample was pumped through a 6-ml phosphocellulose column directly
onto a 2-ml HTH column that had been pre-equilibrated with buffer B
with 10 mM potassium phosphate. The phosphocellulose column
was removed, and RgyA was eluted from the HTH column with 20 column
volumes of a linear 0.01-0.5 M potassium phosphate
gradient, pH 7.4, in buffer B. Under these conditions, RgyA eluted as
two separate peaks centered on fractions 7 and 13 (Fig.
1). The protein in both peaks behaves identically in the
assays shown below. The second peak contains more pure RgyA protein
than peak 1. Fractions corresponding to peak 2 were pooled, dialyzed
against 2 liters of storage buffer (buffer A plus 1 M NaCl,
10% glycerol), and used in subsequent experiments. The pooled
fractions of peak 2 yielded 2.7 mg of protein that was >90% pure
(Table I).
|
The ammonium sulfate-precipitated RgyB
was resuspended in 20 ml of buffer C (50 mM Tris, pH 8, 1 M NaCl, 2 mM -mercaptoethanol) containing 10 mM imidazole and 0.1 mM EDTA, pumped through a
5-ml phosphocellulose column, and loaded directly onto a 3-ml
Ni2+-Sepharose column. The phosphocellulose column was
removed, and the Ni2+ column was washed with 75 ml of
buffer C containing 0.05 M imidazole followed by a gradient
from 0.05 to 2 M imidazole in buffer C (30 ml total).
Fractions containing RgyB were pooled and dialyzed overnight against 2 liters of buffer D (50 mM Tris, pH 8, 100 mM
NaCl, 0.1 mM EDTA, and 2 mM DTT). Some
precipitate was observed that was removed by centrifugation. The sample
was loaded onto a 5-ml HiTrap heparin column at 0.1 ml/min and washed
with 25 ml of buffer D, and the protein was eluted with a linear
gradient of 0.1-1 M NaCl in buffer D (50 ml total, Fig.
2). Fractions in which the major protein observed was
full-length RgyB were pooled, adjusted to 10 mM potassium
phosphate, and loaded directly onto a 2-ml HTH column, and the sample
was washed with 10 ml of Buffer E (50 mM Tris, pH 8, 200 mM NaCl, 10% glycerol) plus 10 mM potassium phosphate. The protein was eluted with a linear gradient (20 ml total)
from 10 to 600 mM potassium phosphate, pH 7.4, in buffer E. Fractions in which the major protein component was full-length RgyB
were pooled. The sample was split into three separate aliquots, and
each aliquot was separated on a Superdex 200HR 10/30 sizing column
pre-equilibrated with buffer A with 1 M NaCl and 10% w/v glycerol. Samples of full-length RgyB were pooled, yielding 0.61 mg of
protein (Fig. 2, Table I).
RgyA protein was insoluble when overexpressed in E. coli at 37 °C, but solubility was improved by expressing the protein at 20 °C. After induction of the cells and lysis as described under "Experimental Procedures," approximately 50% of the overexpressed protein remained in the soluble fraction of the cleared lysate. The same proportion of RgyA stayed in the soluble fraction even when the salt concentration of the lysis buffer was lowered to 100 mM NaCl. The production of soluble RgyA may rely on the cold shock response of E. coli. Alternatively, translation at a lower temperature may proceed slowly enough to allow regions of secondary structure sufficient time to fold. Improved solubility of other recombinant proteins from hyperthermophiles by growth at lower temperatures has been observed (1). The solubility of recombinant RgyB was also improved by expressing the protein at 20 °C (data not shown).
RgyB undergoes considerable proteolytic cleavage and premature translation termination during expression in E. coli, creating a very heterogeneous sample (Fig. 2A, lane 2). However, by being selective in which fractions were pooled during purification, it was possible to isolate a sample that was almost exclusively full-length RgyB.
Using a single-column purification protocol for RgyA and a four-column purification protocol for RgyB, each subunit was purified to >90% homogeneity as determined by scanning analysis of stained SDS-PAGE gels of each sample.
Analysis of RgyAAlthough RgyA eluted as two separate peaks from an HTH column (Fig. 1), these two fractions had biochemical properties that were indistinguishable by several tests. In a supercoiling assay of 1 µl of each column fraction of RgyA from the HTH column together with 0. 25 µg of purified RgyB revealed two peaks of enzymatic activity corresponding with the two peaks of RgyA protein that eluted from the column. RgyA samples from peak 1 and peak 2 both elute in the excluded volume of an S200 sizing column, indicating a tendency to form high molecular mass aggregates (see below). Finally, protein from either peak of RgyA can form a stable heterodimer with RgyB at high temperature (see below). We speculate that these two peaks may represent two different folding intermediates of RgyA that have different affinity for the HTH packing material.
In the early steps of the purification procedure, RgyA had a strong tendency to precipitate out of solution at lower salt concentrations. We used this property to our advantage in the purification by collecting the precipitate that formed during dialysis from 1 M NaCl to 300 mM NaCl (Fig. 1). This precipitate readily dissolved in buffers containing 1 M NaCl and did not behave like denatured protein. After passage of the sample through a phosphocellulose column, RgyA remained in solution on dialysis into buffers containing as little as 100 mM NaCl. We believe this precipitation of early fractions is due to residual polyethyleneimine in the sample that is later removed by passage through the phosphocellulose column (19).
RgyB Is a DNA-dependent ATPaseBecause all
previously isolated reverse gyrases were single-chain enzymes (14, 15),
it had not been possible to test whether the DNA-dependent
ATPase activity was associated with the helicase-like domain. In the
reverse gyrase from M. kandleri, the conserved amino acid
motifs shared with other DNA-dependent ATPases are in RgyB
(18). Recombinant RgyB displayed a DNA-dependent ATPase activity but only at elevated temperatures (Fig. 3).
Efficient hydrolysis of ATP was not observed until 60 °C, with a
large increase in activity from 60 to 80 °C (Fig.
4B). Preheating RgyB with DNA at 80 °C
before incubation with ATP at lower temperatures did not enhance ATPase
activity (data not shown). RgyA by itself did not hydrolyze ATP, even
in the presence of DNA (data not shown).
Reconstitution of Reverse Gyrase Activity
Positive supercoiling activity was reconstituted from the recombinant RgyA and RgyB protomers by mixing equal volumes of the subunits in storage buffer (at concentrations of 0.47 mg/ml for RgyA and 0.14 mg/ml for RgyB), heating for 5 min at 80 °C, and assaying as described (Fig. 4, lanes 12 and 24). Analysis of the reaction products on agarose gels in the presence of chloroquine demonstrated that either relaxed or negatively supercoiled DNA became positively supercoiled (Fig. 4, lower panel, lanes 12 and 24). In the absence of ATP, but with Mg2+ present, RgyA/B was able to bind and nick the substrate DNA but was not able to complete the ligation reaction (Fig. 4, lower panel (Relaxed Substrate), compare lanes 22-24). In the absence of Mg2+, RgyA/B nicking of the supercoiled substrate was greatly reduced.
Recombinant RgyA and RgyB Form a Heterodimer after HeatingWhen purified from M. kandleri, the reverse
gyrase heterodimer was very stable, with the two subunits copurifying
through four separate chromatographic steps (17). Denaturing SDS-PAGE was the only effective method found to separate the two subunits. To
determine if a stable heterodimer could be reconstituted from the
recombinant subunits, a mixture of RgyA and RgyB was dialyzed against
storage buffer at 4 °C overnight, in parallel with separate samples
of RgyA and RgyB. The samples were split, and half of each sample was
heated to 80 °C for 5 min. Samples were then analyzed using gel
filtration chromatography at room temperature. Both heated and unheated
samples of RgyB eluted at a position consistent with a molecular mass
of 150 kDa, indicating that this subunit is monomeric at a
concentration of 5 µg/ml (Fig. 5). In contrast, unheated RgyA eluted in the excluded volume of the S200 column (Fig. 5;
see also Fig. 6). After heating to 80 °C,
approximately 20% of the material eluted at a position consistent with
a monomer (40 kDa). The elution profile of the unheated mixed RgyA/B
sample resembled a superposition of the unheated RgyA and RgyB
profiles. In contrast, the elution profile of the heated RgyA/B sample
showed that the peak corresponding to the RgyA aggregate decreased in amount, and the peak corresponding to RgyB shifted from 150 to 180 kDa
(Fig. 5). SDS-PAGE analysis of fractions demonstrated that both RgyA
and RgyB were present in the 180-kDa peak in approximately a 1:1 ratio,
with very little RgyA in the excluded volume (Fig. 6). Enzyme isolated
from this peak had a specific activity of 2.1 × 105
units/mg, which is very similar to the activity of the complete enzyme
prepared from M. kandleri.
Prolonged incubation of a mixture of RgyA and RgyB at low temperature
did not yield a stable heterodimer (Fig. 6) nor did heating the
subunits separately at 80 °C for 5 min followed by cooling and
mixing at room temperature (see Fig. 7). However, heating (80 °C for 5 min) the two subunits together led to the formation of a stable heterodimer. The formation of an active heterodimer upon coincubation of the two subunits at high temperature may be due to the release of RgyA from a misfolded state that resulted
from its growth in E. coli. Other proteins from extreme thermophiles that are multimers when expressed in E. coli
have also required heating to become active. Work by DiRuggiero and Robb (21) demonstrated that approximately 50% of the hexameric enzyme
glutamate dehydrogenase from the hyperthermophile Pyrococcus furiosus is in the form of inactive monomers when overexpressed and purified from E. coli. Heating the inactive monomers
converts these to the active multimeric complex.
To our knowledge, reverse gyrase from M. kandleri is the first heterodimeric enzyme from a hyperthermophile to be reconstituted in vitro (1). Now that an active reverse gyrase can be reconstituted from a recombinant source, both deletion analysis and site-directed mutagenesis can be used to investigate the reaction mechanism of this unusual enzyme.
We thank members of our laboratory for helpful and insightful discussions during the course of this work. We also thank Alison B. Hickman and Joanne Hesse for reading and helpful comments on the manuscript.