|
|
|
|||||||
|
|||||||||
(Received for publication, November 2,
1994; and in revised form, December 16, 1994) From the
Escherichia coli DNA polymerase III holoenzyme forms a
stable initiation complex with RNA-primed template in the presence of
ATP. To determine the linear arrangement of the holoenzyme subunits
along the primer-template duplex region, we cross-linked holoenzyme to
a series of photo-reactive primers. Site-specific photo-cross-linking
revealed that the Analyses of the interaction of photo-activatible
primer-templates with the preinitiation complex proteins (
The multisubunit enzyme DNA polymerase III holoenzyme
(holoenzyme) ( The effects of ATP, SSB, and salt upon DNA polymerase III are
influenced by the presence or absence of the auxiliary subunits. A
distinction between holoenzyme and its simpler subassembly forms is the
ability of holoenzyme to form a stable initiation complex with a primed
template at the expense of ATP hydrolysis. The auxiliary subunits act
to lock the core polymerase onto a primer-template in an ATP (or
dATP)-dependent reaction creating a complex capable of replicating a
natural chromosome in a rapid and highly processive manner (Wickner and
Kornberg, 1973; Wickner, 1976; Johanson and McHenry, 1982; Burgers and
Kornberg, 1982). In the absence of the auxiliary subunits, ATP is
ineffective in inducing the formation of stable complexes (Johanson and
McHenry, 1982) dramatically decreasing processivity. Both the
polymerase and 3` In vitro, holoenzyme can be loaded onto the primer-template
in two distinct stages. In the first stage (preinitiation complex
formation), the The availability of special
photo-activatible probes strategically positioned within a
primer-template provides a means to determine protein-DNA contact
points in a site-specific manner. One such system designed by Gibson
and Benkovic(1987) has been used to identify the molecular interactions
within the DNA polymerase I-primer-template complex and between T4 DNA
polymerase holoenzyme and its primer-template (Catalano et al., 1990; Capson et al., 1991) The T4 replicative
complex, also a multienzyme system, contains the polymerase gene 43 and
its associated ancillary subunits, genes 44/62 and 45. Capson and
colleagues(1991) determined that within the T4 initiation complex, gene
43 contacted position -4 upstream of the 3`-primer terminus
followed by gene 62 at -9 and gene 45 at
-14 In this study we have used the same
nitrophenyl azide probes developed by Benkovic to identify E. coli DNA Pol III holoenzyme contacts with primer-template. Using a
series of seven photo-activatible primers annealed to a derivative of
M13mp19, we found that within the initiation complex,
Figure 1:
Chemical steps in the preparation of
phenylazide photo-cross-linking primers. A, oligonucleotide
with trifluoroacetate-protected propylamine deoxyuridine at position
-2. B, oligonucleotide after removal of the
trifluoroacetate-protecting group. C, primer after
derivatization with N-hydroxysuccimidyl-5-azido-nitrobenzoate. D, primer after 3` extension with
[
Figure 2:
Nucleotide sequence and position of the
photo-reactive probe for each of the seven derivatized primers. Seven
different primers containing uniquely positioned photo-reactive aryl
azides were annealed to a derivative of M13 and labeled at the primer
3`-terminus with [
In the presence of ATP, the multi-subunit E. coli DNA Pol III holoenzyme ( Our previous work from DNase I footprint
studies indicated that holoenzyme protects To map the linear arrangement of the holoenzyme subunits
along the DNA helix, we photo-irradiated holoenzyme-primer-template
complexes bearing a photo-reactive group within the primer strand.
Seven different derivatized primers were separately annealed to
M13mp19, and the 3` terminus of each primer was radiolabeled. The
photo-reactive probe position for each of the seven primers was 3, 9,
13, 18, 22, 27, or 46 nucleotides upstream from the primer 3` terminus (Fig. 2). By changing the position of the photo-reactive probe
relative to the 3` terminus of the primer, photolysis of holoenzyme
primer-template complexes permitted the identification of different
subunit-DNA contacts along the primer-template in a linear fashion. Photo-products were resolved on a 5-20% SDS-polyacrylamide gel
with no stacking gel. Each of the gels showed two radioactive bands
corresponding to primer-template that did not enter the gel (top of gel) or to cross-linked SSB (bottom of gel) (Fig. 3). Holoenzyme subunits that were cross-linked to the
primer were determined by the gel mobility of the cross-linked species
relative to free protein. Cross-linked photo-products exhibited gel
mobilities slightly slower than their non-cross-linked counterparts due
to the molecular weight of the covalently attached primer. In addition
to non-cross-linked molecular weight markers, the individually
cross-linked proteins
Figure 3:
Subunit-DNA adducts formed after
cross-linking E. coli DNA polymerase III holoenzyme to six
different photo-reactive primer-templates. Reaction conditions were as
described under ``Materials and Methods.'' Initiation
complexes were formed using 3 units of holoenzyme/fmol of
primer-template and 1.0 mM ATP prior to photolysis. A control
reaction for each probe position was performed by photo-irradiating
each primer-template in the absence of holoenzyme. The position of the
photo-reactive group relative to the 3` terminus of the primer is
indicated above each pair of lanes.
To monitor the
3`
We then cross-linked holoenzyme to the
remaining derivatized primers (Fig. 3). Primer-template with the
aryl azide positioned at -9 resulted in one highly radioactive
band that corresponded to cross-linked SSB photo-cross-linked to all of the
annealed primers used in these experiments. All complexes were isolated
by gel filtration to eliminate free primers, but we cannot absolutely
rule out the possibility that trace quantities dissociated after
isolation. Presumably, SSB can interact with the duplex region, albeit
weakly. Previously, we observed that SSB did not protect annealed
primers from DNase I digestion. When interpreting these results it
should be kept in mind that one cannot compare the cross-linking
intensity for two separate proteins to estimate their binding strength.
Cross-linking efficiency varies markedly between proteins and is
determined by the proximity of reactive amino acids to the
photoreactive group. Capson and co-workers (et al., 1991) made
similar observations with the T4 gene 32 protein cross-linking to
annealed photoreactive primers. Several subassembly forms of DNA
polymerase III are known to interact with primer-template in an
ATP-independent mode. However, maximum synthetic and 3`
Photo-cross-linking the components of the preinitiation
complex to primer-template in the presence of both ATP and SSB resulted
in the covalent attachment of the
Figure 4:
Subunit-DNA adducts formed after
photo-cross-linking components of the preinitiation complex (
Photo-cross-linking of the
Surprisingly
Figure 5:
Photo-products generated using individual
components of DNA polymerase III holoenzyme initiation complexes.
Reaction conditions were as described under ``Materials and
Methods.'' The primer was radiolabeled at the 5` terminus. The
photo-reactive probe was located 2 nucleotides upstream of the 3`
terminus. Enzyme-primer-template complexes were formed using 3 units of
holoenzyme, 10 units of
Figure 6:
Subunit-DNA adducts occur with
primer-template and not with primer alone. Reaction conditions were as
described under ``Materials and Methods.'' The primer was
radiolabeled at the 5` terminus. The position of the photo-reactive
probe was 9 nucleotides upstream of the 3` terminus.
Enzyme-primer-template complexes were formed using 3 units of
holoenzyme, 10 units of
Figure 7:
Excess unlabeled primer-template
competitively inhibits subunit-DNA adducts with radioactive
primer-template. Reaction conditions were as described under
``Materials and Methods.'' Lane 1, excess unlabeled
primer-template (P:T) (100 nM), was incubated with radioactive
photo-reactive primer-template (4.0 nM) 50 mM HEPES,
pH 7.5, 500 µM ATP, 10 mM magnesium acetate,
0.01% (v/v) Nonidet P-40, 80 µg/ml bovine serum albumin, and 2.0
µg SSB/nmol nucleotide (25 µl) for 5 min at 30 °C prior to
the addition of subsaturating levels of holoenzyme (0.5 unit/fmol
primer-template), incubated for 1 min at 30 °C, and
photo-irradiated. Lanes 2 and 3, excess unlabeled
single-stranded circular template strands (500 nM) or
unannealed primer (500 nM) were added to labeled
photo-reactive primer-template (4.0 nM). After 5 min at 30
°C, holoenzyme (0.5 unit/fmol primer-template) was added to the
reaction, incubated for 1 min at 30 °C and photo-irradiated. Lane 4, positive control where no competitor was added before
holoenzyme-primer-template complexes were formed. Lane 5,
negative control where no competitor holoenzyme were
added.
The purpose of this study was to map the linear arrangement
of the holoenzyme subunits along the duplex region of the
primer-template at the initiation site. To accomplish this goal, we
used photo-cross-linking to site-specifically identify holoenzyme
subunit contacts relative to a DNA primer. These studies revealed that
the
Figure 8:
The
linear alignment of E. coli DNA polymerase III holoenzyme
subunits relative to the duplex region of the primer-template at the
initiation site. Schematic representation of the holoenzyme subunit-DNA
contacts defined by site-specific photo-cross-linking. The
We previously reported the use of
fluorescence energy transfer to map the position of the Interestingly, the T4 DNA polymerase
genes 43, 44/62, and 45, which are analogous to the E. coli DNA polymerase III DNase I footprint analysis and
fluorescence energy transfer studies also suggested that subunit
rearrangements occur during formation of holoenzyme-primer-template
complexes (Griep and McHenry, 1992; Reems and McHenry, 1994). We,
therefore, photo-cross-linked partially reconstituted complexes in
order to isolate and analyze intermediate subunit-DNA interactions.
Photo-cross-linking preinitiation complexes to primer-template
indicated that Our map of the linear
alignment of the holoenzyme subunits along the DNA helix provides
structural information central to defining the functional activities of
the polymerase subunits within the initiation complex. Our data
indicating that
Volume 270,
Number 10,
Issue of March 10, 1995 pp. 5606-5613
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
, 
, and
Directly Contact the Primer-Template (*)
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
, 
, and 
subunits formed ATP-dependent
contacts with the primer-template. The polymerase catalytic
subunit covalently attached to nucleotide positions -3, -9,
and -13 upstream of the primer terminus, with the most efficient
adduct formation occurring at position -9. The subunit
contacted the primer at positions -13, -18, and -22,
with the strongest -primer interactions occurring at position
-18. The subunit predominated in cross-linking at position
-22. Thus, within the initiation complex, contacts roughly
the first 13 nucleotides upstream of the 3`-primer terminus followed by
at -18 and at -22, and the subunit
remains a part of the initiation complex, bridging the and
subunits.-complex
(-
-`-
-
) and subunit) revealed the
subunit within the preinitiation complex covalently attached to
primer at position -3. However, addition of core DNA polymerase
III to preinitiation complex, fully reconstituting holoenzyme resulted
in replacement of by at the primer terminus. These data
indicate that assembly of holoenzyme onto a primer-template can occur
in distinct stages and results in a structural rearrangement during
initiation complex formation.
)is the major replicative enzyme responsible
for the synthesis of the Escherichia coli chromosome. This
enzyme is comprised of at least 10 different subunits (, 
,
, 
, , , `, , , and 
)
(McHenry and Kornberg, 1977; McHenry and Crow, 1979; McHenry, 1982,
1988; McHenry et al., 1986; Maki and Kornberg, 1988a;
O'Donnell and Studwell, 1990). Due to the tendency of holoenzyme
to dissociate during the purification process, three functionally
distinct subassembly forms, Pol III
, Pol III`, and core Pol
III, have been isolated. The tightly associated , , and
subunits comprise the simplest holoenzyme form, core DNA Pol
III. The and subunits possess the catalytic sites for the
polymerization and 3`
5` exonuclease activities, respectively,
while the function of the 
subunit remains unclear (Slater et
al., 1994). The association of the auxiliary subunits
with core Pol III defines the subassemblies Pol III` (core Pol III plus
the 
subunit) and Pol III* (holoenzyme minus the subunit). 5` exonuclease activities are inhibited by high
salt concentrations and SSB (Fay et al., 1981, 1982; Mok and
Marians, 1987; Griep and McHenry, 1989; Reems et al., 1991) in
the absence of auxiliary subunits. Thus, the auxiliary subunits appear
to reverse these inhibitory effects in the simpler polymerase forms. -complex (, , `, , and
subunits) directs the subunit to the initiation site in an
ATP-dependent reaction (Wickner, 1976; O'Donnell, 1987). In the
second stage, an initiation complex is formed by the addition of core
DNA polymerase III to the preinitiation complex creating a complex
capable of rapid and highly processive DNA replication
(O'Donnell, 1987).-20., , and
subunits covalently attach to the primer-template in an
ATP-dependent reaction and that a change in subunit-DNA contacts occurs
during the transition from a preinitiation to an initiation complex.
Proteins and Enzymes
E. coli DNA Pol
III holoenzyme (
)(700,000 units/mg) was prepared by the
method of Oberfelder and McHenry(1987), DNA Pol III` (
)(400,000 units/mg) by the method of McHenry(1982), and
core DNA Pol III (850,000 units/mg) by the method of
McHenry and Crow(1979). The -complex (18.7 
10
units/mg) was provided by Dr. A. J. Hughes, Jr. of this
laboratory and contains the , , `, , and
subunits. subunit (10.0 10 units/mg) was
isolated by standard procedure (Johanson and McHenry, 1980). These
proteins are also available from Enzyco, Inc., Denver, CO. SSB (0.7
mg/ml) was isolated from an E. coli overproducer, RLM727
(constructed by and gift of Roger McMacken) and purified by a
modification (Griep and McHenry, 1989) of the method described by Meyer et al. (1980). Sequenase version 2.0 T7 DNA polymerase was
purchased from United States Biochemical Corp. (Cleveland, OH).Nucleotides and Nucleic Acids
Unlabeled ATP was
obtained from Pharmacia and [-
P]ddATP
(>5000 Ci/mmol) was purchased from Amersham Corp. M13mp19
single-stranded DNA template was isolated by the procedure of Johanson
and McHenry(1984) and provided by Mary Ann Franden of this laboratory.Reagent Preparation
Dichloromethane (Aldrich) was
dried as described by Perrin and Armarego(1988), refluxed over
CaH (Fisher) for 2 h, and the distillate was collected at
35 °C. 4,4`-Dimethoxytrityl chloride was dried overnight in an
abderhalden drying apparatus containing PO
(Aldrich).Synthesis of 5-(3-Aminopropyl)-2`-deoxyuridine
phosphoramidite
Freshly distilled N-propargyltrifluoroacetamide, prepared from propargylamine
(Aldrich) and trifluoroacetic anydride (Aldrich), was coupled in good
yield with 5-iodo-2`-deoxyuridine (Fluka) according to the method of
Robins et al.(1990). The resulting crude product was purified
by flash chromatography on silica gel utilizing a step gradient (1
column volume each of 75/25, 50/50, and 75/25 methylene chloride/ethyl
acetate). The structure of the product was confirmed by proton NMR (200
MHz). The propynyl intermediate (3.6 g) was dissolved in 60 ml of
absolute ethanol containing 20 ml of ethyl acetate. Following the
addition of 300 mg of hydrogenation catalyst (5% Pd on C), the reaction
was placed on a Parr hydrogenation apparatus and was complete in 2 h.
The catalyst and solvent were removed, and the product was crystallized
from absolute ethanol (2.6 g, 86%). Proton NMR revealed that the
reduction had proceeded as desired. The 5` OH of the reduced nucleoside
was dimethoxytritylated by treatment with demethoxytrityl chloride
(Aldrich) as outlined by Jones(1984) and purified by silica gel
chromatography (2.5 18 cm; 0-10% methanol in methylene
chloride). The 3`-phosphoramidite derivative was prepared from the 5`
blocked compound according to the methods described (Barone et
al., 1984; Sinha, 1984; Atkinson and Smith, 1984). The resulting
3`-phosphoramidite was precipitated upon the dropwise addition of an
ethyl acetate solution of the product to a precooled (-80 °C)
solution of reagent grade hexane. The dried, precipitated
3`-phosphoramidite compound was sufficiently pure (thin layer
chromatography) for direct incorporation into oligonucleotides.Synthesis and Purification of the
Oligonucleotide
Two 24-mers, four 30-mers, and one 50-mer
(complementary to M13mp19 at nucleotides 991-1015, 991-1021, and
991-1041, respectively) (Yanisch-Perron et al., 1985) were
synthesized on a BioSearch 8600 DNA synthesizer. A cytidine-controlled
pore glass support (1 µmol) (MilliGen/Biosearch) was used in order
to position a ribonucleotide at the 3` terminus of each synthesized
deoxynucleotide primer (Fig. 1A). The terminal
5`-dimethoxytrityl protecting group was left on the oligomers to
facilitate purification by reversed phase HPLC. The
trifluoroacetate-protected
5-(3-trifluoroacetamide-1-propyl)-2`-deoxyuridine was incorporated into
each of the seven oligomers at a unique site. Six of the oligomers had
a modified residue positioned within a 30-nucleotide region to
accommodate the occupancy site of holoenzyme as defined by DNase I
footprint analysis (Reems and McHenry, 1994) at positions 2, 8, 12, 17,
21, and 26 upstream from the 3`-primer terminus; the seventh oligomer
had the modified residue positioned 45 nucleotides upstream of the
3`-primer terminus (Fig. 2). Cleavage from the controlled pore
glass support and removal of all protecting groups except the
5`-dimethoxytrityl group were carried out as described (Hagerman,
1985). The trityl-protected oligonucleotides were purified by HPLC on a
Waters µBondapak C
column (3.9 300 mm) using a
gradient of acetonitrile in 90 mM triethylammonium acetate, pH
7.0 (flow rate, 0.8 ml/min; Program: 0-4 min, 0% acetonitrile
wash; 4-54 min, concave gradient to 50% acetonitrile; 54-59
min, concave gradient to 100% acetonitrile; 59-68 min wash).
Continuous monitoring at 260 nm (Waters 994 photodiode array detector)
indicated that the trityl-protected oligonucleotides eluted at 58 min
for the 24-mers, 46 min for the 30-mers, and 51 min for the 50-mer. The
tritylated oligomers were deprotected with 6% acetic acid, 10 min at
room temperature, and purified by HPLC using the same conditions as for
the tritylated oligomers (Fig. 1B). Retention time
after detritylation for each of the oligomers was 47 min. Stock primer
concentrations were determined using extinction coefficients derived by
the summation of the individual base extinction coefficients (Borer,
1975). Average yield from a 1-µmol oligonucleotide synthesis for
purified detritylated oligomers was between 5-10%. The oligomers
were evaporated to dryness in a Speed-Vac (Savant Instruments).
P]ddATP, placing the photo-reactive probe at
position -3.
P]ddATP using
deoxynucleotidyl transferase as described under ``Materials and
Methods.'' Each of the seven derivatized primer-templates
possesses a 3`-terminal dideoxyadenosine, a 3`-penultimate
ribonucleotide (cytidine), and a single photo-reactive probe. The final
probe position after annealing the derivatized primer to M13 and
radiolabeling the 3` terminus is indicated to the left of each primer.
The photo-reactive probe is positioned either 3, 9, 13, 18, 22, 27, or
46 (-3, -9, -13,
-18, -22, -27, and
-46, respectively) nucleotides upstream from the primer
3` terminus. U* = propylamine-deoxyuridine derivatized
with phenylnitroazide.
Derivatization of Primers with
N-Hydroxy-succinimidyl-5-azido-2- nitrobenzoate
The detritylated
oligonucleotides were derivatized with the cross-linking reagent N-hydroxy-succinimidyl-5-azido-2-nitrobenzoate (Pierce) as
described by Gibson and Benkovic(1987) with the following modification.
The dried detritylated oligomers were dissolved in 100 mM sodium carbonate, pH 9.5, prior to treatment with the
cross-linking reagent. Derivatized oligomers were purified on the
Waters C18 reverse-phase column using the same conditions and gradient
as described above and eluted at 47 min. Peak fractions containing
photo-reactive oligomers exhibited UV spectra (Waters 994 photodiode
array detector) maximum absorption at 258 nm and a shoulder peak at 320
nm, characteristic of the ring system containing the azido group.Exonuclease Assay
Two 24-mers, the first with the
photo-reactive aryl azide positioned at -2 and the second without
the photo-reactive group, were radiolabeled at the 5` terminus using T4
polynucleotide kinase and [-P]ATP according
to described procedures (Maniatis et al., 1982). A premix
containing 50 mM HEPES, pH 7.5, 100 mM potassium
glutamate, 10 mM dithiothreitol, 500 µM ATP, 10
mM magnesium acetate, 2.0 nM primed template, and 2.0
µg SSB/nmol nucleotide was incubated for at least 2 min at 30
°C. Three aliquots (12.5 µl) were removed prior to the addition
of 3-4 units of holoenzyme/fmol of primer-template. After
holoenzyme addition, 12.5-µl aliquots were withdrawn at various
time points and quenched with a final concentration of 140 mM EDTA. Quenched reaction mixtures were transferred onto Whatman
DE-81 filters and batch-washed (three times, 5 min each in 0.3 M ammonium formate, pH 7.8, 10 mM sodium pyrophosphate;
once in deionized water; and once in 95% ethanol). Filters were dried,
and the amount of radiolabel remaining was measured by scintillation
counting. Infinity points were determined by adding 3-4 units of
holoenzyme/fmol primer-template and allowing the reaction to proceed
for 10 half-lives. An additional amount of holoenzyme (3-4
units/fmol primer-template) was added, and the reaction was allowed to
proceed for another 10 half-lives. Three aliquots (12.5 µl) were
removed and the radiolabel quantitated as described above. Data were
analyzed as described (Griep et al., 1990).Elongation Assay
The elongation of a
primer-template by holoenzyme was determined by using a modification of
previously described procedures (Johanson and McHenry, 1984; Griep and
McHenry, 1989). Briefly, the final reaction mixture contained either
2.0 nM primer-template in 50 mM HEPES, pH 7.5, 500
µM ATP, 10 mM magnesium acetate, 0.01% (v/v)
Nonidet P-40, 80 µg/ml bovine serum albumin, or 1.0 nM primer-template, 2.0 µg SSB/nmol nucleotide, and 500
µM each of dATP, dCTP, dGTP, and dTTP. Each reaction
mixture was equilibrated to 30 °C for at least 2 min prior to the
addition of 3-4 units of holoenzyme/fmol primer-template. After a
5-min incubation, the reaction was quenched with 90% formamide and
loaded onto an 8% polyacrylamide, 8 M urea sequencing gel.Exonuclease-resistant Photoreactive
Primer-Templates
The presence of a terminal dideoxyadenosine
coupled to a penultimate ribonucleotide reduces the 3`5`
exonuclease activity of holoenzyme nearly 1000-fold (Griep et
al., 1990). To generate radioactive exonuclease-resistant primers,
each of the photo-reactive oligonucleotides containing a ribonucleotide
positioned at the 3` terminus was extended with
[-P]ddATP after hybridizing each oligomer
to M13mp19. This was accomplished by mixing oligomer (800 nM)
with M13mp19 (400 nM) in 50 mM HEPES (pH 7.5), 200
mM NaCl at 75 °C for 5 min followed by slow cooling (1.5
°C/min) to room temperature. Primer-template (200 nM) was
incubated with [-P]ddATP (1.0
µM), 3 units of Sequenase/pmol of primer-template in 10
mM magnesium acetate, 50 mM HEPES (pH 7.5) for 30 min
at 37 °C. Enzyme was thermally inactivated (10 min, 65 °C). The
labeled primer-template was separated from free primer and
unincorporated nucleotide on a Bio Gel A-5 m column (1.0 3.0
cm) (Bio-Rad) equilibrated in 50 mM HEPES (pH 7.5), 100 mM potassium glutamate, 1 mM EDTA. Primer-template
concentrations were determined spectrophotometrically at 260 nm and
were presented in terms of primer 3`-hydroxyl termini. Seven
oligonucleotides containing aryl azide-substituted U at position 3, 9,
13, 18, 22, 27, or 46 upstream of the 3`-ddA terminus were prepared (Fig. 2).Site-specific Photo-cross-linking of Enzyme to
Primer-Template
Saturating levels of enzyme (see figure legends)
were added to 50 mM HEPES (pH 7.5), 500 µM ATP,
10 mM magnesium acetate, 0.01% (v/v) Nonidet P-40, 80
µg/ml bovine serum albumin, 4.0 nM photo-reactive
primer-template, and 2.0 µg SSB/nmol nucleotide (25 µl).
Enzyme
DNA complexes were incubated at 30 °C for 1 min prior
to photo-irradiation. Photolysis was accomplished using an SLM-Aminco
48000 spectrofluorometer. The samples were photolyzed at a wavelength
of 315 nm (450 W xenon arc lamp, 16 nm bandpass) at a distance of 145
cm for 10 min (15 milliwatts/cm). An equal volume of sample
buffer (50% glycerol, 10 mM dithiothreitol, 0.2% bromphenol
blue, 0.1 M Tris, pH 6.8, 1.0% SDS) was added to each reaction
prior to loading the samples onto a 5-20% SDS-polyacrylamide gel.
After electrophoresis, gels were silver-stained, transferred to Whatman
3MM paper, and dried. To visualize the resolved cross-linked products,
gels were exposed to Kodak film (SB or X-OMAT AR) at room temperature
or exposed to a PhosphoroImager screen (Molecular Dynamics).
Cross-linked subunits were identified by their molecular weights and
gel mobilities relative to a holoenzyme standard. Molecular weights and R
values were calculated for each radioactive
photo-product and for each subunit present in the holoenzyme standard.Site-specific Photo-cross-linking of Enzyme to Unannealed
Primer
A 24-mer with the reporter group positioned 9 nucleotides
upstream of the 3`-primer terminus was 5`-phosphorylated using T4
polynucleotide kinase and [-P]ATP according
to a procedure by Maniatis et al.(1982). Radiolabeled
oligonucleotide was gel-filtered and quantitated as described above.
Saturating levels of enzyme (see figure legends) were incubated with
unannealed primer (4.0 nM) in 50 mM HEPES (pH 7.5),
500 µM ATP, 10 mM magnesium acetate, 0.01% (v/v)
Nonidet P-40, 80 µg/ml bovine serum albumin, and 2.0 µg
SSB/nmol nucleotide (25 µl). After 1 min at 30 °C, samples were
photo-irradiated, and the cross-linked products were identified as
described for enzyme-primer-template complexes.
, , , , , ,
`, , , and 
subunits) forms a highly stable
initiation complex with a primed template (Wickner and Kornberg, 1973;
Wickner, 1976; Burgers and Kornberg, 1982a, 1982b; Johanson and
McHenry, 1980, 1982). To identify the subunit-DNA interactions that
occur within the initiation complex, we mapped the linear arrangement
of the holoenzyme subunits along the DNA helix using site-specific
photo-cross-linking.
30 nucleotides of
primer (Reems and McHenry, 1994). Based on the footprint data, we
synthesized seven different primers with a single photo-reactive aryl
azide group uniquely placed within the holoenzyme binding region (Fig. 2), adapting the synthetic methods originally developed by
Gibson and Benkovic(1987) to synthesize primers containing uniquely
positioned photo-affinity labels. This method involved the introduction
of a trifluoroacetate-protected propylamine to deoxyuridine, synthesis
of protected phosphoramidite, and insertion of the
trifluoroacetate-protected propylamine deoxyuridine phosphoramidite
into an oligonucleotide chain. Standard deblocking procedures removed
the trifluoroacetate protecting group from the propylamine which was
then derivatized with the photo-reactive agent N-hydroxy-succinimidyl-5-azido-nitrobenzoate in preparation
for the photo-cross-linking studies (Fig. 1). Each of the
primers was complementary to single-stranded circular bacteriophage
M13mp19. and SSB were used as markers.
Primer-Templates Containing a Photo-reactive Aryl Azide Group
Are Effective Substrates for Holoenzyme
To determine whether
primers containing a photo-reactive aryl azide were suitable substrates
for holoenzyme, we monitored enzyme function using derivatized
primer-templates. A derivatized primer bearing a photo-reactive group
at position -2 and a radioactive label at the 5` terminus was
annealed to M13mp19 and incubated with holoenzyme in the presence of
all four dNTPs. Primer extension was monitored using a 6% sequencing
gel. In the presence of all four dNTPs, holoenzyme elongated both
derivatized and non-derivatized primers to full-length product with no
detectable intermediate fragments (data not shown).5` exonuclease activity of holoenzyme, we compared the
enzyme's excisional activities using primers with or without a
photo-reactive group. When holoenzyme was incubated with
primer-template in the absence of all four dNTPs, the enzyme fully
degraded derivatized primers in 90 s and non-derivatized primers in 45
s. Even though there was a 2-fold reduction in exonuclease activity
using derivatized primers (data not shown), the primer was fully
degraded within the time frame of our photo-cross-linking experiments.
Thus, the aryl azide did not significantly interfere with
holoenzyme's synthetic or proofreading activities. However, to
overcome the technical difficulties for our experiments presented by
holoenzyme's ability to rapidly degrade DNA primers, we made two
alterations in the primer to reduce its susceptibility to the
enzyme's 3`5` exonuclease activity. First, we rendered the
primer exonuclease-resistant by positioning a penultimate
ribonucleotide and a terminal dideoxyribonucleotide residue at the 3`
terminus. These modifications reduced the exonuclease activity to
stabilize holoenzyme-DNA primer-template interactions nearly 1000-fold
(Griep et al., 1990). Second, we radiolabeled the 3` rather
than the 5` terminus of the primer, ensuring that the 3`5`
exonuclease activity would cleave the radiolabel in the form of a
mononucleotide, thus eliminating ambiguity in cross-link assignments
since only photo-products that occurred prior to excision would be
detectable. These alterations made it possible for us to detect
photo-reactive products due only to the formation of static
holoenzyme-primer-template complexes, which is necessary in determining
the linear alignment of holoenzyme subunits relative to the
primer-template.E. coli DNA Polymerase III Holoenzyme Subunits,
Photolysis of holoenzyme incubated with
derivatized primer-template containing an aryl azide at the -3
position resulted in a radioactive band that migrated to the position
expected for cross-linked , , and Directly Contact the
Primer-Template (Fig. 3) and required the
presence of ATP. (
). Photo-irradiation of
derivatized primer-template, aryl azide positioned at -13,
resulted in faint photo-products that corresponded to the gel positions
expected for cross-linked and . Holoenzyme-primer-template
complexes photo-irradiated with the aryl azide positioned at -18
resulted in only one highly radioactive photo-product whose gel
migration corresponded to cross-linked . At position -22,
the photo-products migrated to positions expected for cross-linked
and , and at position -27, there were no apparent
photo-products. Cross-linking at position -46 was also not
detected (data not shown).5`
exonuclease activities for holoenzyme are achieved only when ATP is
present. Analysis of ATP requirements to confirm that the cross-linked
products were due to holoenzyme and not subassembly forms revealed that
cross-linking of the , , and subunits to primer
required ATP (data not shown).The
To identify subunit contacts made with the primer
during preinitiation complex formation, we photo-cross-linked the
components of the preinitiation complex (, (or `), and Subunits of the
Preinitiation Complex Directly Cross-link to Primer in an ATP-dependent
Reaction plus -complex
(, , `, , and )) to a derivatized
primer-template. Since exonuclease activity is associated with the
subunit which is not part of the preinitiation complex, primers
with a radiolabel at the 5` terminus instead of the 3` terminus were
used. The aryl azide was positioned 2 nucleotides upstream of the 3`
terminus., , and (or `)
subunits (Fig. 4A, lane 1; in the system used,
the and ` subunits cannot be distinguished). By eliminating
ATP from the reaction, both the and subunit interactions
with the primer-template were precluded (Fig. 4A,
compare lanes 1 and 5), whereas (or `)
primer-template interactions remained essentially unchanged (Fig. 4A, compare lanes 1 and 5). The
radioactive band, designated , between the and
subunits was apparent even in the absence of any enzyme (Fig. 4A, lanes 4 and 8), indicating
that this photo-product was not due to cross-linked holoenzyme
subunits.
plus
-complex subunits) to primer-template in the presence or absence
of SSB. Reaction conditions were described under ``Materials and
Methods.'' The primer was radiolabeled at the 5` terminus. The
photo-reactive probe was positioned 2 nucleotides upstream of the 3`
terminus. Enzyme-primer-template complexes were formed with 10 units of
-complex, and/or 70 units of subunit/fmol of primer-template
in the presence or absence of 500 µM ATP. A,
photo-products obtained after irradiating enzyme-primer-template in the
presence of 2.0 µg SSB/nmol of nucleotide. B,
photo-products obtained after irradiating enzyme-primer-template in the
absence of SSB.
-complex in the absence of
the subunit still resulted in and (or `)
interactions with the primer. In the presence of ATP, within the
-complex cross-linked to the primer and did not require the
addition of the subunit (Fig. 4A, lane
2). In the absence of ATP, interactions with the primer were
almost eliminated (Fig. 4A, lane 6).
Photo-cross-linking of the subunit without the -complex did
not result in -DNA adduct formation (Fig. 4A, lane 3). Further, no apparent -DNA adducts were
obtained even when using excess and a primer containing an aryl
azide positioned at -22 (data not shown). Thus,
interactions with primer required both -complex and ATP (Fig. 4A, compare lane 1 to lanes 4 and 7), results which are consistent with the proposal
that the -complex transfers the subunit to primer-template
in an ATP-dependent reaction (Wickner 1976; O'Donnell, 1987).SSB Interferes with
SSB has several known functions during DNA replication,
including its ability to enhance polymerase processivity and to promote
polymerase binding to primer-template (for review, see Meyer and Laine,
1990). To investigate the role of SSB in the binding of holoenzyme to
primer-template, we cross-linked preinitiation complexes to
primer-template in the presence or absence of SSB (Fig. 4).
Photo-cross-linking components of the preinitiation complex to
primer-template in the presence of ATP without SSB resulted in the
covalent attachment of the (or `) ATP-independent
Binding, , and subunits (Fig. 4B, lane 9). However,
cross-linking was more prevalent than in the absence of SSB,
while formation of and -adducts to the primer was equivalent
in the presence of SSB. By eliminating ATP from the reaction, both
and interactions with the primer-template were precluded (Fig. 4B, lane 13). (or
`)-primer interactions were dramatically different in the absence
or presence of SSB. In the absence of SSB, the amount of (or
`) that cross-linked to the primer was increased 5-10-fold
(compare Fig. 4, A to B). Further, (or
`) interactions with the primer-template were essentially the same
with or without ATP (Fig. 4B, compare lanes 9 and 10 to lanes 13 and 14). These
results indicate that SSB competes with the (or `) subunit
for binding to the primer-template at a position 2 nucleotides upstream
of the 3` terminus of the primer. Further, (or `)
interactions with the primer-template occur in an ATP-independent
fashion.Holoenzyme Undergoes a Conformational Change during
Initiation Complex Formation
To identify subunit-DNA contacts
that occur during the formation of holoenzyme-primer-template
complexes, we photo-cross-linked partially reconstituted complexes and
analyzed intermediate subunit-DNA interactions. Photo-irradiation of
primer-template incubated with either core Pol III or Pol III` resulted
in 10-fold less -DNA adducts than with
holoenzyme-primer-template complexes (Fig. 5, compare lanes
2 and 3 to lane 7). Within the preinitiation
complex, the , , and subunits cross-linked to the
primer 2 nucleotides upstream from the 3` terminus (Fig. 5, lane 4). However, with the addition of core Pol III or Pol
III` to the preinitiation complex, the , , and (or
`) subunits were replaced by the subunit (Fig. 5, lanes 5 and 6). These results suggest that a
conformational change occurs upon the binding of either core Pol III or
Pol III` to preinitiation complexes.
-complex, 70 units of , 3 units of
core Pol III, and/or 3 units of Pol III`/fmol of primer-template in the
presence of 500 µM ATP. These additions resulted in
enzyme/primer molar ratios of 6:1, 18:1, and 20:1 for holoenzyme, core
pol III and pol III`, respectively. and complex, when
included, were present at enzyme/primer ratios of 18:1 and 86:1,
respectively. Enzyme additions are indicated above each
lane.
Holoenzyme Subunits
To ensure that
cross-linked photo-products directly reflect holoenzyme interactions
with primer-template and not with unannealed primer, we examined
holoenzyme's ability to interact with unannealed primer using two
approaches. The first approach involved incubating holoenzyme or
, , and Cross-link to
Primer-Template and Not to Unannealed Primer-complexes with derivatized unannealed primers and then
photo-irradiating the enzyme-primer complexes; no apparent
cross-linking of the subunit occurred with unannealed primer when
using holoenzyme (Fig. 6, compare lanes 3-6),
and no apparent cross-linking of the , , (or `) or
subunits occurred to unannealed primer when using the -complex (Fig. 6, compare lane 2 to 5). The second
approach was a competition assay in which a nonspecific competitor, i.e. unannealed primer-template or M13mp19 alone, or a
specific competitor, i.e. non-derivatized primer annealed to
M13, was added before holoenzyme to the primer-template. A 10-fold
molar excess of specific competitor successfully competed
cross-linking to a primer with the photo-reactive group at position
-3 (Fig. 7, lane 1), whereas a 100-fold molar
excess of nonspecific competitor failed to compete cross-linking
to the primer (Fig. 7, lanes 2 and 3). Thus,
the subunit-DNA photo-products were detected only when enzyme
interacted with primer annealed to a template strand.
-complex, and 70 units of
subunit/fmol of primer-template in the presence of 500 µM ATP. Photo-products obtained after irradiating enzyme that had
been incubated with either primer alone or
primer-template.
subunit cross-linked to positions -3, -9, and
-13, with the most efficient cross-linking occurring at position
-9. The subunit cross-linked to positions -13,
-18, and -22, with the strongest -primer interactions
occurring at position -18. The subunit was the predominant
protein cross-linked at position -22, while at positions
-27 and -45, essentially no holoenzyme contacts were noted.
Together these results indicate that within the initiation complex,
contacts roughly the first 13 nucleotides upstream of the
3`-primer terminus, followed by at -18 and at
-22 (Fig. 8).
subunit, possessing the polymerase catalytic site, covalently attaches
to positions -3, -9, and -13, with the most efficient
cross-linking occurring at position -9. The subunit
cross-links to positions -13, -18, and -22, with the
strongest -primer interactions occurring at position -18.
The subunit is the predominant subunit covalently attached to the
primer at position -22. Thus, within the initiation complex,
contacts roughly the first 13 nucleotides upstream of the
3`-primer terminus followed by at -18 and at
-22. remains part of the initiation complex connecting the
and subunits.
subunit
within the initiation complex (Griep and McHenry, 1992). In that study,
a donor-acceptor distance of 65 Å was determined between a donor
fluorophore located at the -3 position of the primer and an
acceptor fluorophore positioned within the subunit at
Cys
, a distance which positions the subunit
approximately 19 nucleotides from the donor fluorophore or 22
nucleotides upstream of the 3`-primer terminus. This distance is
consistent with the present results which show that covalently
attaches to the primer 22 nucleotides upstream of the 3`-primer
terminus. DNase I footprint analysis (Reems and McHenry, 1994) of
holoenzyme binding to the primer-template showed that holoenzyme
contacts 30 nucleotides of primer-template; photo-cross-linking
results indicate that holoenzyme contacts at least 22 nucleotides of
primer. Since DNase I cuts occur at 4-nucleotide intervals and
would be expected to be sterically hindered from cutting immediately
adjacent to a protein on DNA, it is reasonable to find a slightly
larger region defined by DNase I footprint analysis than the site-size
defined by photo-cross-linking., -complex, and subunits,
respectively, map to similar positions. Gene 43, which is responsible
for nucleotide incorporation in the T4 DNA polymerase replication
complex, contacts position -4 upstream of the 3`-primer terminus.
Gene 62, which is part of the ATP accessory complex, contacts the
primer at position -9, and gene 45, the T4 processivity factor,
contacts the primer at positions -14 and -20 (Capson et
al., 1991). The similar structural position and order of the T4
replication machine components is consistent with their functional
equivalences, predicting that this order might be conserved among all
replication machines. To extend this prediction to eukaryotic
replication enzymes, one might expect polymerase to interact with
the primer terminus, a component of 5-protein RFC to interact at an
adjacent position, and the PCNA sliding clamp to reside at the most
distal position in eukaryotic replication complexes. Of course, extra
interactions may be present to accommodate other interactions including
the polymerase-primase complex., , and subunits contact the primer at
position -2. However, with the addition of core Pol III or Pol
III` to the preinitiation complex, the subunit replaces ,
, and subunits at position -2. These results provide
direct evidence that the binding of core polymerase to preinitiation
complexes induces a subunit rearrangement. contacts at least 13 nucleotides of the primer
upstream of the 3`-primer terminus positions this subunit, which
possesses the catalytic site for the polymerization reaction, at the
primer-template junction poised to incorporate dNTPs at the 3`hydroxyl
terminus of the primer. At positions -18 and -22,
respectively, and contact the primer. The subunit,
which confers processivity to the core polymerase (Fay et al., 1981) and functions as a ``sliding clamp'' (Stukenberg et al., 1991), is situated upstream of the and
subunits, placing the processivity factor approximately two helical
turns upstream from the 3`-primer-terminus assuming a -form DNA
conformation. The -complex recognizes the primer-template and
couples ATP hydrolysis to clamp at the initiation site (Onrust et al., 1991). The positioning of between and
suggests a role for in locking and linking the processivity
and elongation factors. Several laboratories have reported that the
complex functions to load onto the
primer-template and then dissociates. Our results suggest a
participatory role for the complex in elongation.
), , , ,
, , `, , , and 
subunits); SSB, E.
coli single-stranded DNA-binding protein; Pol III*, E. coli DNA polymerase III* (holoenzyme minus the subunit); Pol
III`, E. coli DNA polymerase III` (, , , and
subunits); core Pol III, E. coli DNA polymerase III
core (, , and 
subunits); -complex, , ,
`, , and subunits; preinitiation complex, plus
-complex; HPLC, high performance liquid chromatography.))). This band was highly variable in most of
the experiments reported in this paper and was further confused by
appearance even when holoenzyme was deleted (see lane
22(-), Fig. 3) The extra band could be due to DNA-DNA
cross-links or a related artifact. Thus, we cannot make any definitive
statement about the appearance of .
We thank Mark Seville, John Hughes, and Doug Dellinger
for helpful discussions. We are especially grateful to Steve Benkovic
and Katherine Gibson for their advice and for supplying derivatized
primers for our preliminary studies.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  N. Yao, F. P. Leu, J. Anjelkovic, J. Turner, and M. O'Donnell DNA Structure Requirements for the Escherichia coli gamma Complex Clamp Loader and DNA Polymerase III Holoenzyme J. Biol. Chem., April 6, 2000; 275(15): 11440 - 11450. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Ju. Mozzherin, C.-K. Tan, K. M. Downey, and P. A. Fisher Architecture of the Active DNA Polymerase delta {middle dot}Proliferating Cell Nuclear Antigen{middle dot}Template-Primer Complex J. Biol. Chem., July 9, 1999; 274(28): 19862 - 19867. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Hingorani and M. O'Donnell ATP Binding to the Escherichia coli Clamp Loader Powers Opening of the Ring-shaped Clamp of DNA Polymerase III Holoenzyme J. Biol. Chem., September 18, 1998; 273(38): 24550 - 24563. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Bambara, R. S. Murante, and L. A. Henricksen Enzymes and Reactions at the Eukaryotic DNA Replication Fork J. Biol. Chem., February 21, 1997; 272(8): 4647 - 4650. [Full Text] [PDF]
|
|
|
|
|
|
D. R. Kim and C. S. McHenry Biotin Tagging Deletion Analysis of Domain Limits Involved in Protein-Macromolecular Interactions. MAPPING THE tau BINDING DOMAIN OF THE DNA POLYMERASE III alpha SUBUNIT J. Biol. Chem., August 23, 1996; 271(34): 20690 - 20698. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Kim and C. S. McHenry Identification of the beta -binding Domain of the alpha Subunit of Escherichia coli Polymerase III Holoenzyme J. Biol. Chem., August 23, 1996; 271(34): 20699 - 20704. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. W. Olson, H. G. Dallmann, and C. S. McHenry DnaX Complex of Escherichia coli DNA Polymerase III Holoenzyme J. Biol. Chem., December 8, 1995; 270(49): 29570 - 29577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-S. Song, P. T. Pham, M. Olson, J. R. Carter, M. A. Franden, R. M. Schaaper, and C. S. McHenry The delta and delta ' Subunits of the DNA Polymerase III Holoenzyme Are Essential for Initiation Complex Formation and Processive Elongation J. Biol. Chem., September 7, 2001; 276(37): 35165 - 35175. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |