(Received for publication, October 18, 1996, and in revised form, June 25, 1997)
From the § Laboratory of Molecular Genetics, NICHD,
National Institutes of Health, Bethesda, Maryland 20892, the
To understand how ribonucleases H recognize
RNA-DNA hybrid substrates, we analyzed kinetic parameters of binding of
Escherichia coli RNase HI to RNA-DNA hybrids ranging in
length from 18 to 36 base pairs (bp) using surface plasmon resonance
(BIAcoreTM). The kon and
koff values for the binding of the enzyme to
the 36-bp substrate were 1.5 × 106
M Escherichia coli ribonuclease HI (RNase HI) degrades
only the RNA strand of an RNA-DNA hybrid (1) and is composed of a single polypeptide chain of 155 amino acid residues (2). It requires
divalent cations such as Mg2+ or Mn2+ for
activity (1). The involvement of the amino acid residues Asp10, Glu48, Asp70 (3),
His124 (4), and Asp134 (5) in the catalytic
function was established by site-directed mutagenesis for the catalytic
function of the enzyme. Two alternative mechanisms have been proposed:
one is a two-metal ion mechanism (6) and the other is a
carboxyl-hydroxyl relay mechanism (4, 7-9).
X-ray crystallographic analyses of E. coli RNase HI (6, 10,
11) and the RNase H domain of
HIV-11 reverse transcriptase
(12) showed that these RNases H have a similar structural topology,
with the exception of the presence of a handle (6) or basic protrusion
region (11) in E. coli RNase HI. The importance of this
region for substrate binding has been demonstrated in several studies.
The RNase HI domain isolated from HIV-1 reverse transcriptase is
enzymatically inactive (13, 14), whereas that from murine leukemia
virus reverse transcriptase, which has a part of this region, is active
(15, 16). Site-directed mutagenesis indicated that the positively charged residues in this region are important for substrate binding (17). Incorporation of the basic protrusion of E. coli RNase HI at the equivalent position of the RNase H domain of HIV-1 reverse transcriptase resulted in the production of the active HIV-1 RNase H
domain (18, 19). In addition, Cys13, Asn16,
Asn44, Asn45 (7), and Thr43 (17)
have been shown to be important for substrate binding in E. coli RNase HI. Thus, it seems likely that all amino acid residues
that are involved in catalytic function and substrate binding have been
identified.
However, it is not fully understood how RNase H binds to its substrate.
A kinetic study using synthetic nucleosides with modifications of their
2 Recently, kinetic analyses using RNA-DNA hybrids, under conditions in
which the hybrid was cleaved at a unique site (22), suggest that DNA
residues complementary to the RNA residues located six or seven
residues upstream of the cleavage site interact with the basic
protrusion region of the enzyme. Such an interaction seems to require a
conformational change in the enzyme or substrate, or in both.
Determination of kinetic parameters and stoichiometry of RNase HI
molecules bound to substrates of various lengths would provide
additional information about the binding of enzyme to substrate.
Magnesium ions may also modulate protein-nucleic acid interactions and
participate in catalysis. For example, the binding of
Saccharomyces cerevisiae RNase HI to double-stranded RNA is influenced by Mg2+ concentration, with tight binding being
detected at low concentrations and little binding in the presence of 5 mM Mg2+ (23), and the DNA binding specificity
of EcoRV is increased in the presence of Mg2+
(24). Therefore, it is of great interest to investigate the influence
of magnesium ions on binding of RNase HI to RNA-DNA substrates.
In this study, we have analyzed the interaction between E. coli RNase HI and RNA-DNA hybrids using the BIAcoreTM
system, an instrument based on surface plasmon resonance
technology. This technology is useful for obtaining kinetic data on the
interaction between enzyme and substrate, particularly when the enzyme
is inactive. This system enabled us to distinguish between inability to
bind the substrate and inability to cleave. Furthermore, utilization of
a 2 The BIAcoreTM instrument was
manufactured by Pharmacia Biosensor AB (Uppsala, Sweden). Sensor chips
CM5, Tween P20, and the amine coupling kit containing
N-hydroxysuccinimide,
N-ethyl-N The plasmid pJAL600 (26), used for
overproduction of E. coli RNase HI, was constructed
previously. Competent cells of E. coli HB101
(F The RNA
oligonucleotides biotinylated on the 5
E. coli RNase HI wild-type protein (28) and the
mutant protein D134A, in which Asp134 is replaced by Ala
(5), were prepared as described. The enzyme with an N-terminal
6×His-Tag (6×His-RNase HI) was obtained as described (16, 23).
A group of plasmids for overproduction of the RNase HI variants with
randomized sequences, in which Ile82 and Val101
are connected by four random amino acid residues resulting in deletion
of the basic protrusion, was constructed as follows. Because the
plasmid pJAL600 has a unique BamHI site in the sequence encompassing amino acid residues 81-83, and a unique SalI
site about 50-base pairs downstream of the termination codon of the rnhA gene, the rnhA gene in plasmid pJAL600 was
amplified by polymerase chain reaction using BamHI
site-containing 5 RNase H activity was determined either
in the presence of 10 mM MgCl2 or 1 mM MnCl2 by measuring the radioactivity of the acid-soluble digestion product of the substrate, 3H-labeled
M13 DNA-RNA hybrid, as described previously (16), or by degradation of
32P-labeled poly(rA)·poly(dT) as described (30).
Protein concentrations were
determined from the UV absorption, assuming that all mutant proteins,
except for The CD spectra were measured on
a J-600 spectropolarimeter (Japan Spectroscopic Co., Ltd.) at 25 °C
in 20 mM sodium acetate buffer, pH 5.5, containing 0.1 M NaCl, as described previously (33).
RNA-DNA hybrids were prepared so that the RNA strand was
biotinylated at the 5 All the proteins and
poly(rA)· poly(dT) were dissolved in TBS buffer or TBS buffer
containing 10 mM MgCl2 instead of EDTA (TBSM
buffer). Samples were injected at 25 °C at a flow rate of 5 µl/min
onto the sensor chip surface on which the RNA-DNA hybrid had been
immobilized, or onto a control surface on which streptavidin had been
immobilized and blocked with biotin. Binding surfaces were regenerated
by washing with 2 M NaCl.
Immobilization was carried out as described (25). The CM5
sensor chip was modified with NTA. Ni2+ was chelated by
injecting NiCl2 onto the NTA-modified surface, which was
followed by injection of 6×His-RNase HI to immobilize the protein by
chelation, giving a change in RU value of 800. The 36-bp RNA-DNA hybrid
was then injected onto the surface. Binding surfaces were regenerated
by washing with 0.5 M EDTA.
Sensorgrams for
the interaction of RNase HI and RNA-DNA hybrids were analyzed with
BIAlogue kinetics evaluation software, as described in the standard
model (34), after subtraction of the values for the interactions
between the samples and streptavidin-coated control surface. The
association rate constants (kon) were calculated from the association phases of the sensorgrams at various
concentrations of the analyte, using Equation 1.
Department of Material and Life Sciences,
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
1 s
1 and 3.2 × 10
2 s
1, respectively. Similar values were
obtained with the shorter substrates. Using uncleavable
2
-O-methylated RNA-DNA substrates, values for
kon and koff were
2.1 × 105 M
1
s
1 and 1.3 × 10
1 s
1 in
the absence of Mg2+ that were further reduced in the
presence of Mg2+ to 7.4 × 103
M
1 s
1 and 2.6 × 10
2 s
1. Kinetic parameters similar to the
wild-type enzyme were obtained using an active-site mutant enzyme,
Asp134 replaced by Ala, whereas a greatly reduced on-rate
was observed for another inactive mutant enzyme, in which the basic
protrusion is eliminated, thereby distinguishing between poor catalysis
and inability to bind to the substrate. Stoichiometric analyses of RNase HI binding to substrates of 18, 24, 30, and 36 bp are consistent with previous reports suggesting that RNase HI binds to 9-10 bp of
RNA-DNA hybrid.
-hydroxyl groups revealed the importance of the 2
-hydroxyl group of
the nucleoside on the 3
-side of the cleaved phosphodiester and that of
the second nucleoside 5
to the cleaved phosphodiester for hydrogen
bonding (20). Models for the binding of the enzyme to an RNA-DNA hybrid
have been proposed based upon computer docking of the structure of
E. coli RNase HI (free from its substrate) with an RNA-DNA
hybrid whose structure was assumed to be an A form (6, 11), was found
by NMR to be an A form (7), or that was neither A nor B (21). In these
models, the RNA strand upstream of the cleavage site interacts with the
enzyme. None of these models assumes that either the enzyme or the
substrate alters its conformation upon binding.
-O-methylated substrate permitted analysis of the effect of the Mg2+ ions on substrate binding.
Instrumentation and Reagents
-(3-dimethylaminopropyl)carbodiimide, and ethanolamine hydrochloride were from Pharmacia Biosensor AB. Nickel
nitrilotriacetic acid (NTA ligand) was a kind gift from S. Khilko (25).
Poly(rA)·poly(dT) was made by annealing poly(rA) with poly(dT)
obtained from Pharmacia. Annealing was performed by mixing 100 µg of
poly(rA) and poly(dT) in HBS buffer (10 mM Hepes-NaOH, 150 mM NaCl, 1 mM EDTA, 0.005% Tween P20, pH 7.0), boiling for 2 min, and cooling slowly to room temperature.
hsdS20(rB
,
mB
) recA13 ara-13 proA2
lacY1 galK2 rpsL20(Smr)
xyl-5 mtl-1 supE44
)
were from Takara Shuzo Co., Ltd. Cells were grown in Luria Bertani medium (27) containing 100 mg/liter ampicillin.
-amino group (Fig.
1) were obtained from Integrated
DNA Technologies and Oligos, Etc. DNA oligonucleotides (Fig. 1) were
obtained from Integrated DNA Technologies. All concentrations are
expressed as moles of molecules and not bp. RNA and DNA
oligonucleotides (1 µM) were annealed in HBS buffer
by boiling for 2 min and allowed to cool to room temperature to form
RNA-DNA hybrids.
Fig. 1.
The sequence of RNA-DNA hybrids. RNA-DNA
hybrids used for the BIAcoreTM analysis are listed. The 5-ends of the
RNA strands are biotinylated.
[View Larger Version of this Image (17K GIF file)]
-mutagenic primer, 5
-CCGGATCCNSNNSNNSNNSGTCGATCTCTGGCAACGTCTTGATGC-3
and SalI site-containing 3
-primer
5
-GGGTCGACCAATTCGCAGGCGGTTGG-3
. In these sequences, N represents A, G, C, or T and S represents G or C, and restriction sites are underlined. Polymerase chain reaction was performed in 25 cycles with a Perkin-Elmer DNA Thermal Cycler (model PJ2000) using a
Gene Amp kit (Takara Shuzo Co., Ltd.) according to the procedure
recommended by the supplier. These oligodeoxyribonucleotides were
synthesized by Sawady Technology Co., Ltd. After digestion of the
polymerase chain reaction fragment with BamHI and
SalI, the resultant 280-base pair fragment and large
BamHI-SalI fragment of pJAL600 were ligated to
construct pJAL600
BP. The mutant proteins were overproduced in
E. coli HB101 harboring plasmid pJAL600 derivatives as
described previously (26). The cellular production levels of these
proteins were estimated by subjecting whole cell lysates to
SDS-polyacrylamide gel electrophoresis on a 15% polyacrylamide gel
(29), followed by staining with Coomassie Brilliant Blue (R250).
BP-RNase HI, have the same absorption coefficient,
A0.1%280 = 2.0, as that
of the wild-type protein (31). The absorption coefficient of
BP-RNase HI was estimated as 1.65 by using
= 1576 M
1 cm
1 for Tyr (× 5) and 5225 M
1 cm
1 for Trp (× 4) at 280 nm
(32), and a molecular mass of 15,860 Da.
-end. Streptavidin was covalently linked to the
dextran on the surface of research-grade CM5 sensor chips via primary
amino groups using the amine coupling kit from Pharmacia. Carboxylate
groups on the dextran were activated by injecting a mixture of
N-hydroxysuccinimide and
N-ethyl-N
-(3-dimethylaminopropyl)carbodiimide followed by 20 µl of 25 µg/ml streptavidin in 10 mM
sodium acetate, pH 5.7. Ethanolamine hydrochloride, pH 8.5, was
injected to block unreacted N-hydroxysuccinimide esters.
Typically, the resonance unit (RU) value increase following this
procedure was 500-1000. Twenty microliters of RNA-DNA hybrid (50-100
nM) in TBS buffer (10 mM Tris-HCl, 50 mM NaCl, 1 mM EDTA, 1 mM
-mercaptoethanol, 0.005% Tween P20, pH 8.0) were injected onto the
streptavidin-modified sensor chip giving an increase of 50-100 RU.
Unoccupied streptavidin was blocked by biotin.
koff values obtained from Equation 1 are
not very reliable for low ranges (34). Therefore,
koff values were calculated from the
dissociation phases of the sensorgrams using Equation 2.
(Eq. 1)
Association constants (KA) were calculated
using Equation 3.
(Eq. 2)
Association constants (KAeq) can
also be obtained from equilibrium levels of the analyte binding to the
surface (RUeq) using Equation 4.
(Eq. 3)
(Eq. 4)
Three versions of E. coli RNase HI were examined by plasmon resonance studies as follows: wild-type enzyme, a mutant (D134A) with very low RNase H activity, and a protein whose handle or basic protrusion had been removed, presumably altering its ability to bind to RNA-DNA hybrids. This latter protein is described here for the first time.
Preparation of RNase HI Variant Missing the Basic ProtrusionBy studying the binding kinetics of enzymes one can
distinguish whether the lower activity of the mutant form of a protein is due to poor binding to the substrate or to a defect in catalytic activity. Previously, it has been shown that the inactivity of a
protein derived from the RNase H domain of HIV-1 reverse transcriptase can be overcome by inserting handle region of E. coli RNase
HI into the corresponding position of the HIV-1 RNase H protein. These
results support the role of the handle in binding to the RNA-DNA
substrate (18, 19). To further substantiate this function of the handle
region, we made a deletion of amino acid residues 83-100 and replaced
it with four randomly generated amino acid residues. It has previously
been shown that E. coli strain MIC3001, with the
rnhA-339::cat and recB270(TS) mutation,
can effectively screen for genes encoding a functional RNase H molecule
(35). However, no clone was obtained after deletion-substitution of the
handle region of E. coli RNase HI that could support growth of MIC3001 at the restrictive temperature, suggesting that the basic
protrusion of E. coli RNase HI is critical for enzyme
function in vivo. We screened the clones for an RNase HI
variant that can be overproduced in a soluble form. When 22 E. coli HB101 transformants bearing pJAL600 derivatives were analyzed
by SDS-polyacrylamide gel electrophoresis for overproduction, only one
was able to produce soluble RNase HI in amounts suitable for subsequent
purification (data not shown). Plasmid DNA was isolated from this
transformant, and the DNA sequence of the mutated rnhA gene
was determined. The deduced amino acid sequence connecting
Ile82 and Val101 was Arg-Thr-Asn-Ser. This
mutant protein, BP-RNase HI, was recovered in a soluble form after
sonication lysis, and the lysate was subjected to DE52 chromatography
using a column equilibrated with 10 mM Tris-HCl, pH 7.5, containing 1 mM EDTA. The protein was eluted from the
column by linearly increasing the NaCl concentration from 0 to 0.5 M. Fractions containing the protein were combined and
applied to a Sephacryl-S300 (Pharmacia) (2.2 × 90 cm) equilibrated with 10 mM Tris-HCl, pH 7.5, containing 1 mM
EDTA and 0.1 M NaCl. Fractions containing the pure protein
were combined and used for further analyses. The amount of the mutant
protein
BP-RNase HI purified from a 1-liter culture was about 0.64 mg.
No enzymatic activity was
detected for the mutant protein BP-RNase HI either in the presence
of the Mg2+ or Mn2+ ion, indicating that this
mutant is completely inactive. CD spectra of the wild-type protein and
of the mutant protein
BP-RNase HI are shown in Fig.
2. Few spectral changes were detected
between these proteins in the far ultraviolet region, in which the
spectra reflect the content of the secondary structure of the protein (Fig. 2a). This result suggests that the mutant protein
BP-RNase HI folds correctly. On the other hand, in the near
ultraviolet region, in which the spectra reveal environments of
aromatic amino acid residues, the spectra were clearly distinguishable
(Fig. 2b). This change in the spectrum may be due to a local
conformational change upon the deletion of the basic protrusion and/or
to the loss of two tryptophan residues.
Determination of Kinetic Parameters
For the binding of RNase
HI molecules to the 36-bp RNA-DNA hybrid, TBS containing RNase HI at
concentrations ranging from 10 nM to 1.5 µM
were passed, at a flow rate of 5 µl/min, over the surface of the
sensor chip, on which the substrate had been immobilized. The slope of
the ln(RU1/RUn) versus RU plots in each
sensorgram was linear (data not shown), indicating that dissociation of
the enzyme from the substrate occurs in an apparently first-order
reaction. The sensorgrams are shown in Fig.
3. When the RNase HI molecules bound to
the sensor chip were dissociated by running the RNase-free buffer at a
flow rate of 5 µl/min, the dissociation rate constants
(koff) decreased as the initial amount of the
RNase HI molecules bound to the sensor chip decreased. This change is
probably due to rebinding of the protein to the RNA-DNA hybrid on the
sensor chip when the hybrid is present at high densities on the
surface. Increasing the flow rate to 100 µl/min increased the
dissociation rate constant nearly 5-fold, but the
koff value was still dependent on the initial
number of RNase HI molecules bound to the sensor chip (Fig.
4). Addition of the poly(rA)·poly(dT)
competing substrate to the buffer for dissociation at 39 µg/ml
resulted in a 10-20-fold increase in the dissociation rate constant
(Fig. 4). The koff values were nearly constant
when the initial amount of the protein bound to the sensor chip
increased, indicating that the competition completely abolishes the
rebinding. In the presence of this competing substrate, binding
of RNase HI to the immobilized hybrid was completely inhibited (data
not shown).
kon values were calculated using Equation 1.
Association profiles fit satisfactorily with a mono-exponential
equation for protein concentrations up to 100 nM.
ks versus C plots for 10-80
nM protein concentrations were linear (Fig.
5), indicating that, unlike the
koff values, the kon
values are not affected by RNase H concentrations in the binding
buffer.
Similar sensorgrams were obtained when the 30-, 24-, or 18-bp RNA-DNA hybrids were used as immobilized substrates (data not shown). Kinetic parameters of the interaction between RNase HI and these substrates were similar with, at most, a 2-fold variation between them (Table I).
|
When the
concentration of RNase HI in the buffer was greater than 0.1 µM, RU values showed a slow increase followed by a fast increase, never reaching a plateau (Fig.
6). This suggests that two binding phases
exist, a slow phase and a fast phase. The RUeq/C versus RUeq plots for the fast binding phase were
obtained over the range of 25 nM to 1.5 µM by
using the end points of the fast binding mode as the RUeq
values (Fig. 7). These plots were
biphasic with two different kinetic association constants.
KA values derived from linearization of the plots
obtained at low RNase HI concentrations ranged from 25 to 50 nM (KAeq1) and at higher
RNase HI concentrations from 0.4 to 1.5 µM
(KAeq2) (Table I). At higher RNase HI
concentrations, KAeq values were
approximately one-fourth to one-seventh those calculated from
data collected at lower RNase HI concentrations.
Stoichiometry of RNase HI Bound to RNA-DNA Hybrids
An
increase in RU values upon binding of analyte on the surface of the
chip is proportional to the mass of analyte. Therefore, using the
relationship RUnucleic acids = 0.8 ×
RUprotein (36), one can calculate the ratio of RNase HI
and RNA-DNA hybrid from the equation R =
RURNase HI/
RUhybrid × MWhybrid/MWRNase HI × 0.8; where
RUhybrid is the increase in RU value upon binding of the
hybrid to the streptavidin surface,
RURNase HI is the
increase in RU value upon binding of RNase HI to the hybrid, and
MWhybrid and MWRNase HI are the molecular
weights of the hybrid and RNase HI, respectively. The RUmax
values used for the determination of stoichiometry were obtained from
the x intercepts of the RUeq/C versus
RUeq plots. The stoichiometry obtained by linear
transformation of the plots at lower RNase HI concentrations (25-50
nM (n1)) was 1.19 ± 0.07, 1.34 ± 0.10, 2.17 ± 0.04, and 3.18 ± 0.05 for the
18-, 24-, 30-, and 36-bp RNA-DNA hybrids, respectively. The
stoichiometry obtained by linear transformation of the plots at higher
RNase HI concentrations of (0.4-1.5 µM (ntotal)) was 1.56, 1.72, 2.62, and 4.11 ± 0.14 for the 18-, 24-, 30-, and 36-bp RNA-DNA hybrids, respectively.
The values ntotal
n1
(n2), which correspond to the number of the
RNase HI molecules bound to the hybrids with the smaller
KA values, were 0.37, 0.38, 0.45, and 0.92 for 18-, 24-, 30-, and 36-bp RNA-DNA hybrid, respectively. The numbers of RNase
HI molecules binding to the hybrids in the slow binding phase
(nslow) was estimated from the increase in RU
values in the slow binding phase at 1.5 µM, because, even
at this concentration, the slow binding phase was far from equilibrium.
The nslow values were 0.19, 0.20, 0.31, and 0.56 for 18-, 24-, 30-, and 36-bp RNA-DNA hybrid, respectively.
Kinetic measurements were also carried out using the inverse
experimental system, in which RNase HI molecules are immobilized on the
surface. A histidine tag fused to the N terminus of RNase HI allows the
protein to be immobilized on the chip in a unique homogeneous
orientation. This enables one to analyze the interactions between a
single RNase HI molecule and its RNA-DNA hybrid substrate. For binding
of the 36-bp RNA-DNA hybrid to RNase HI, TBS buffer (without EDTA)
containing various concentrations (from 50 nM to 1.5 µM) of the 36-bp RNA-DNA hybrid were injected onto the
surface of the chip, on which RNase HI had been immobilized.
Immediately following injection of the hybrid, RNase HI (1 µM) was injected as a competitor to avoid rebinding of
the hybrid to other RNase HI molecules on the surface of the sensor
chip. Binding of the 36-bp hybrid to immobilized RNase HI gave
sensorgrams similar to those obtained for the binding of RNase HI to
the immobilized hybrid (Fig.
8a). The RU versus
RU/C plot was comprised of a dominant binding phase with
high affinity and an additional binding phase with low affinity (Fig.
8b), similar to those obtained with immobilized substrate.
Kinetic parameters were also similar to those determined when the
substrate was immobilized (Table II).
|
When binding of RNase HI to the 36-bp
2-O-methylated substrate was analyzed by BIAcoreTM, a
rapid increase of the RU value was observed, which was followed by a
slow increase in RU value (Fig.
9a). Such biphasic binding is
similar to that observed for the binding to unmodified substrate.
However, the slow phase observed in binding to the modified substrate
was more pronounced than that observed for the normal substrate. For
the rapid binding phase, the kon value for the
modified substrate was one-seventh of that for the unmodified
substrate, whereas the koff value for the
modified substrate was 4-fold higher than that for the unmodified substrate (Table III). These results
indicate that 2
-O-methylation of the substrate considerably
impairs interaction between enzyme and substrate. When the
RUeq/C values were plotted versus
RUeq over a range of 0.1-1.5 µM for the
rapid binding phase, a linear relationship was observed. This indicates
that the binding is monophasic (Fig. 9b). The
KA value obtained from the plot concurred with the
kon/koff values. The
stoichiometry calculated from the RUeq value was 3.7 ± 0.38. The number of RNase HI molecules bound to the 36-bp modified
substrate in the slow binding phase (nslow) was
estimated from the increase in the RU value at 1.5 µM to
be 0.88 ± 0.05.
|
Because the 2-O-methylated substrate cannot be cleaved by
RNase HI, it was possible to determine kinetic parameters for binding of RNase HI to the modified substrate (36-bp) in the presence of 10 mM MgCl2. Slow monophasic binding was observed
in the presence of 10 mM MgCl2 (Fig.
10). The kon
value was 3.5% that obtained for the rapid binding phase in the
absence of MgCl2 (Table III). The
koff value was also lower than that obtained in
the absence of MgCl2 (Table III). Binding of the enzyme to
the modified substrate in the presence of 10 mM
MgCl2 was so weak that the association constant and the
stoichiometry could not be determined from the RUeq/C
versus RUeq plots. The number of the RNase HI
molecules bound to the modified substrate in the presence of 10 mM MgCl2 was estimated from the increase in the
RU value at 1.5 µM to be 2.92 ± 0.11.
Effect of Asp134 to Ala Mutation on the Binding of RNase HI to an RNA-DNA Hybrid
Kinetic parameters for binding of the mutant protein D134A to the 36-bp unmodified substrate are shown in Table IV. The kon value was similar to that obtained for the wild-type enzyme, whereas the koff value was about half that of the wild-type enzyme. Consequently, the KA value for the mutant protein D134A is slightly higher than that of the wild-type protein. This result indicates that the Asp134 to Ala mutation does not seriously affect the binding of RNase HI to the substrate. Kinetic analysis of the enzymatic activity has shown that the Asp134 to Glu, Gln, Ser, and Thr mutations, which greatly reduce the enzymatic activity, do not seriously affect the Km value (5). However, this finding could not completely exclude the possibility that the Asp134 to Ala mutation almost fully inactivates the enzyme by reducing the binding affinity of the enzyme to the substrate, because no kinetic data were available for the mutant protein D134A. The current study excludes such a possibility.
|
When the binding
of BP-RNase HI (9.2 µM) to the 36-bp RNA-DNA hybrid
was analyzed by BIAcoreTM, a slight increase in RU value was observed,
which corresponds to that observed for 2.5 nM wild-type protein (Fig. 11). From the increase of
the RU level, the affinity of the
BP-RNase HI for the 36-bp RNA-DNA
hybrid was estimated to be 0.025% that of the wild-type protein for
the same substrate. This result suggests that the basic protrusion is
the major contributor to the interaction between enzyme and
substrate.
In
this study, association and dissociation of RNase HI with RNA-DNA
hybrid substrates were monitored in real time using the BIAcoreTM
system. Kinetic parameters for binding of RNase HI to hybrids were
constant for concentrations up to 0.1 µM. When RNaseHI was injected at this concentration, we calculated that 0.9 and 2.8 RNase HI molecules were bound to the 18- and 36-bp hybrids, respectively. Similar kinetic parameters were also obtained for binding
of the 36-bp hybrid to immobilized RNase HI which binds only a single
substrate molecule. These results suggest that multiple RNase HI
molecules bind to the substrate simultaneously and independently with
little or no cooperativity. KD values obtained in this study (~2 × 108 M) agree well
with those obtained from the titration of RNA-DNA hybrids with RNase HI
using data collected from changes in CD spectra upon binding to the
substrate (10
8-10
9 M) (37).
We observed two phases of binding (with fast and slow rates) when the concentration of RNase HI was greater than 0.1 µM (Fig. 6). Of the total protein bound to the substrate at 1.5 µM, ~10% bound at a slower "on" rate and 90% bound at a faster rate. The phase with the faster on rate consisted of two parts, 70% with high and 20% low association constants, as indicated by the biphasic RU versus RU/C plot (Fig. 5). It is likely that the binding with the higher association constant is predominant under physiological conditions under which the concentration of RNase HI is expected to be low. Therefore, it also seems likely that binding with higher affinity is productive binding, with additional, nonspecific binding. One explanation for the secondary interactions is that free RNase HI may interact with RNase HI molecules already bound to the substrate. However, multiphasic binding was also observed when RNase HI was immobilized, and the 36-bp RNA-DNA hybrid was in the mobile phase. Taken together, these results suggest an additional weaker interaction between RNase HI and the substrate that is separate from the normal binding site.
Length of the RNA-DNA Hybrid Interacts with RNase HIThe
numbers of RNase HI molecules bound to the hybrids with high
association constants are ~1.2 for the 18-bp hybrid, ~1.3 for the
24-bp hybrid, ~2.2 for the 30-bp hybrid, and ~3.2 for the 36-bp
hybrid. Kinetic analyses, under conditions in which the RNA-DNA hybrid
was cleaved at a unique site (22), or using RNA-DNA hybrids with
2-O-methylnucleosides, which limits the cleavage at a
single site (38), suggest that RNase HI interacts with 9-10 bp of
RNA-DNA hybrid. Using these values, one can expect the maximum number
of RNase HI molecules able to bind to the 18-, 24-, 30-, and 36-bp
hybrid to be 1-2, 2, 3, and 3-4, respectively. However, to completely
saturate these relatively short RNA-DNA hybrids, each RNase HI
molecule(s) would need to bind in such a manner as to permit the
maximum number of other RNase HI molecules to bind. For example, if one
RNase HI molecule were to bind to the middle of the 18-bp RNA-DNA
hybrid, there would be only 4 bp at either end of the hybrid accessible
to other RNase HI molecules. Thus, the observed values will always be
lower than the maximum values. Therefore, our current results are
consistent with a 9-10-bp binding size (22, 38).
It has been reported that the conformations of
2-O-methyl RNA-DNA hybrids are similar to those of normal
RNA-DNA hybrids (39). Therefore, a 2
-O-methyl RNA-DNA
hybrid can provide a good model with which to investigate the
interaction between RNase HI and the substrate. Our data show that the
association constant between RNase HI and the
2
-O-methylated substrate was one-thirtieth that between
enzyme and the substrate. Methylation of 2
-OH group at the cleavage
site, which greatly reduces catalytic efficiency, also slightly reduces
the affinity between the enzyme and the substrate (a 5-fold difference
in the Km value), possibly due to steric hindrance
(9). It is reported that the 2
-OH group of the nucleoside adjacent to
the 3
-side of the cleaved phosphodiester bond acts as a proton donor
and acceptor and that the second nucleoside from the 5
-side of the
cleaved phosphodiester bond acts as a proton acceptor (20). Therefore,
the reduced affinity consequent to 2
-O-methylation may be
due to steric hindrance by the methyl groups and/or loss of the
hydrogen bonds. These results support an interaction of 2
-OH group and
RNase HI. However, the current results cannot exclude the possibility
that methylation of the 2
-OH group of the hybrid creates a substrate
that binds to the enzyme in a different manner (e.g. in the
opposite orientation or with different spacing).
The presence of Mg2+ may
affect the RNase HI/RNA-DNA interaction in two ways as follows: first,
Mg2+ may neutralize the charge of the phosphates and
decrease the ionic interaction between the basic protrusion and the
RNA-DNA hybrid; and second, Mg2+ binds to the acidic amino
acid residues at the catalytic center of the enzyme, a process
necessary for cleavage. We find that, in the presence of
Mg2+, the affinity between RNase HI and
2-O-methyl RNA-DNA hybrid was reduced by 83% (Fig. 9 and
Table III). A large reduction in affinity (3-4 orders of magnitude) in
the presence of Mg2+ has been reported for the nonspecific
interaction between EcoRV and DNA (24), a reduction that was
interpreted as due to the displacement by the enzyme of
Mg2+ bound to DNA. Using
BP-RNase HI, a protein lacking
the basic protrusion, we found a drastic reduction in substrate binding (Fig. 11). Therefore, we believe the reduction in affinity in the presence of Mg2+ results from the displacement of
Mg2+ bound to the RNA-DNA hybrid by the basic protrusion.
In contrast, the specific interaction between EcoRV and DNA
is not affected by Mg2+ ions (24). Thus, the effect of the
Mg2+ ions seems to depend on its mode of interaction. Such
a difference in the effect of the Mg2+ ions could account
for a moderate reduction in affinity of RNase HI for its substrate as
compared with that of EcoRV for DNA. However, a 20%
reduction in the "off" rate counteracted, in part, this reduction
in affinity. In the absence of the Mg2+ ion, electrostatic
repulsion between the RNA-DNA hybrid and the acidic amino acid residues
in the catalytic center of RNase HI (Asp10,
Glu48, Asp70, and Asp134) would be
expected to decrease affinity. The catalytic Mg2+ ion would
eliminate the negative charge repulsion at the active site of the
enzyme, resulting in less release of the enzyme from the substrate.
Keck and Marqusee (40) have recently shown that the basic
protrusion of E. coli RNase HI is not essential for
activity. They reported that the mutant protein, in which the amino
acid residues 83-95 are replaced by six glycine residues, exhibited RNase H activity in the presence of the Mn2+ ion but not in
the presence of the Mg2+ ion. In contrast, we showed that
BP-RNase HI, in which residues 83-100 are replaced by
Arg-Thr-Asn-Ser, exhibited little RNase H activity either in the
presence of the Mg2+ or Mn2+ ion. This apparent
discrepancy in the enzymatic activity of the mutant protein, which
lacks the basic protrusion, may result from the difference in the size
of the deleted region or in the flexibility of the linker which is
substituted for the basic protrusion.
In this report, we discussed the importance of the basic protrusion for
the interaction with the substrate, on an assumption that BP-RNase
HI has basically the same folding topology as that of the wild-type
protein. The possibility that the deletion of the basic protrusion
causes a gross structural change may not be excluded, because
BP-RNase HI has little enzymatic activity. However, the similarity
in the far UV CD spectra between
BP-RNase HI and the wild-type
protein and the fact that both the RNase H domain of HIV-1 reverse
transcriptase and the E. coli RNase HI variant with the
deletion of the basic protrusion (40) fold correctly strongly suggest
that
BP-RNase HI also folds correctly.
We thank Drs. Susana M. Cerritelli for technical assistance; Wu-Po Ma for providing the E. coli RNase HI; Xinyi Zhan for providing His-tagged E. coli RNase HI; S. Khilko for providing the NTA ligand; and H. Yamada, L. Stolz, E. Kanaya, K. Morikawa, S. Iwai, T. Tanaka, and M. Wakasa for encouragement and helpful discussions.