| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, August 5, 1994; and in revised form, December 18, 1994) From the
2`-5` oligoadenylate (2-5(A)) synthetase and protein
kinase, RNA activated (PKR) are the only two known enzymes that bind
double-stranded RNA (dsRNA) and get activated by it. We have previously
identified their dsRNA binding domains, which do not have any sequence
homology. Here, we report a profound difference between the two enzymes
with respect to the structural features of the dsRNA that are required
for their activation. The adenoviral virus-associated type I (VAI) RNA
cannot activate PKR, although it binds to the protein and thereby
prevents its activation by authentic dsRNA. In contrast, we observed
that VAI RNA can both bind and activate 2-5(A) synthetase.
Mutations in VAI RNA, which removed occasional mismatches present in
its double-stranded stems, markedly enhanced its 2-5(A)
synthetase-activating capacity. These mutants, however, are incapable
of activating PKR. Other mutations, which disrupted the structure of
the central stem-loop region of the VAI RNA, reduced its ability to
activate 2-5(A) synthetase. These debilitated mutants could bind
to the synthetase protein, although they fail to bind to PKR. Interferons (IFNs) ( Although both PKR and
2-5(A) synthetase are activated by dsRNA, their activation
characteristics are quite different. PKR binds dsRNA with a much higher
affinity than 2-5(A) synthetase. ( We are interested in identifying the specific structural
features of an RNA molecule that are recognized by PKR and 2-5(A)
synthetase for binding and activation. Many RNA molecules, which are
not perfectly double stranded, can bind to these enzymes. Several
viruses encode RNAs that are partially double-stranded and are known to
interact with PKR (10) . Most of these viral RNAs inhibit the
action of PKR, although reovirus mRNA can activate it(11) . It
is not known if and how these RNAs affect the action of 2-5(A)
synthetase. In the current study, we have addressed this question by
studying the effects of adenoviral virus-associated (VA) RNAs on
synthetase activity. Adenoviruses, which are relatively insensitive to
IFN, encode two RNA polymerase III-directed small RNAs (VAI and VAII
RNAs) that accumulate to high levels at late stages of infection. Both
RNAs are about 160 nucleotides long and are highly structured. In virus
infections, VAI RNA is obligatory for efficient translation of viral
and cellular mRNAs at late times. It binds to and blocks the activation
of PKR produced by the cell, thereby enabling protein synthesis to
proceed at normal
levels(12, 13, 14, 15) . The
structure of VAI RNA consists of two long imperfectly base-paired stems
of 20-22 base pairs (bp) joined at the center by a domain that is
structurally complex and that contains a short stem loop and two
adjacent loops. This complex short stem loop and the two adjacent loops
together are referred to as the central
domain(16, 17) . Although initially it was thought
that the long duplex regions might be important in blocking the
activation of the PKR, recent mutational analysis showed that it is the
central domain that is critical for function(16, 17) . In this paper, we report an unexpected observation that VAI RNA can
bind to and activate 2-5(A) synthetase in a
concentration-dependent manner. VAII RNA is less effective in this
regard. We also show that the activation potential of the VAI RNA is
primarily due to the long duplex regions of the molecule because the
activation of 2-5(A) synthetase by VAI RNA increased dramatically
and reached to the levels of activation mediated by dsRNA when the
imperfectly base-paired duplex regions of VAI RNA were converted to
perfectly base-paired regions.
The mutants 741, 745, 746, and 748 are linkerscan substitution
mutants in which HindIII linker sequences are substituted in
the following locations: 741, between nucleotides 76 and 90; 745,
between nucleotides 105 and 117; 746, between nucleotides 116 and 126;
and 748, between nucleotides 134 and 143(21) . Mutant VAI-CB
was constructed by substituting DNA sequences between Csp45I
(+61) and BstEII (+99) sites with a synthetic
double-stranded oligonucleotide in which sequences have been altered to
make the stem III of VAI RNA perfectly double stranded. Similarly, VAI
BR was constructed by substituting DNA sequence between BstEII
(+99) and EcoRI (+160) to make stem I perfectly
double stranded(20, 21) .
Figure 1:
Activation of bacterially expressed
2-5(A) synthetase by various dsRNAs. Synthetase activity was
measured as described under ``Experimental Procedures.''
Parts of the autoradiogram are shown. Positions of 2-5(A)
molecules and inorganic phosphate (P
For
studying 2-5(A) synthetase-RNA interactions, we developed an
electrophoretic mobility shift assay. The synthetic 82-bp dsRNA was
radiolabeled and used as the probe. As shown in Fig. 2A, increasing amounts of the synthetase protein
shifted the mobility of increasing amounts of the probe. The shifted
band could be competed out by excess unlabeled probe but not by
single-stranded RNA (data not shown). To confirm that the observed
shift in the dsRNA mobility was due to its binding to the 2-5(A)
synthetase protein and not to any accompanying impurity, an antibody
supershift assay was performed (Fig. 2B). An antibody
raised against a synthetic peptide, whose sequence corresponds to a
region of the synthetase protein(22) , supershifted a portion
of the synthetase-dsRNA complex (lane4). Similar
supershifting was observed with an antibody produced against the
purified bacterially produced synthetase (data not shown), whereas the
preimmune serum did not cause any supershift (lane3). Once the specificity of the mobility shift assay was
established, it was used to examine the potential interaction between
VAI RNA and 2-5(A) synthetase. In the experiment shown in Fig. 3A, unlabeled VAI RNA could effectively compete
with the 82-bp dsRNA probe to prevent the complex formation (lane4), although a concentration of VAI RNA 10 times higher
than that of poly(I)
Figure 2:
Electrophoretic mobility shift assay for
measuring dsRNA-2-5(A) synthetase interactions. All lanes had 0.5 ng of labeled 82-bp dsRNA probe. A: lane1, no protein; lanes2-5,
increasing amounts of 2-5(A) synthetase added. The openarrow shows the position of the probe, and the solidarrow shows the position of the shifted complex. B: lane1, no protein added; lanes2-4, 500 ng of 2-5(A) synthetase; lane3, 5 µl of preimmune serum; lane4,
5 µl of antipeptide serum. The openarrow shows
the position of the probe, the lowersolidarrow shows the position of the shifted complex, and the uppersolidarrow shows the position of supershifted
complex.
Figure 3:
VAI
RNA-2-5(A) synthetase interactions. A, all lanes had 0.5
ng of labeled 82-bp dsRNA probe. Lane1, no protein; lanes2-4, 500 ng of 2-5(A) synthetase
added; lane3, 5 ng of poly(I)
Figure 4:
Activation of 2-5(A) synthetase by
VAI RNA. Lane1, no RNA activator; lane2, 25 µg/ml VAI RNA; lane3, 100
µg/ml VAI RNA; lane4, 100 µg/ml poly(A); lane5, 100 µg/ml poly(U); lane6, 100 µg/ml
poly(A)
To further characterize the requirements of activation of
2-5(A) synthetase by VAI RNA, its dose response (Fig. 5)
and kinetics of the reaction (Fig. 6) were measured. Increasing
concentrations of VAI RNA increasingly catalyzed the enzyme reaction (Fig. 5). The rate of reaction appeared to level off between 50
and 100 µg/ml VAI RNA. The enzyme kinetics was linear for at least
2 h (Fig. 6), irrespective of the activator RNA used. Note that
at every time point, the amount of 2-5(A) synthesized was about
10 times higher when an authentic dsRNA was used instead of VAI RNA (Fig. 6, A and B). The synthetic dsRNA used in
this experiment contained 164 nucleotides, whereas VAI RNA contains 157
nucleotides. Thus, on a weight basis, VAI RNA is about 10 times less
effective in activating 2-5(A) synthetase than a totally
double-stranded RNA.
Figure 5:
Effect of varying concentrations of VAI
RNA on synthetase activity. 2-5(A) synthetase activity was
measured in the presence of increasing concentrations of VAI RNA. The
amounts of 2-5(A) synthesized were quantitated by phosphorimager
analysis of the thin layer chromatogram; they are presented in
arbitrary units.
Figure 6:
Kinetics of 2-5(A) synthesis.
Amounts of 2-5(A) synthesized at various times in response to (A) 25 µg/ml VAI RNA and (B) 25 µg/ml
synthetic dsRNA are shown. The data are presented in arbitrary units
after subtracting the corresponding values without any activator RNA at
each time point.
Figure 7:
Secondary structures of wild type and
mutant VAI RNAs. A, experimentally derived secondary structure
of the VAI RNA. The structure is based on single-strand-specific
ribonuclease sensitivity analysis(16) . See text for details. B, experimentally derived secondary structure of various
mutant VAI RNAs based on single-strand-specific ribonuclease cleavage
analysis previously described(20, 21) . RNase
cleavages are shown by arrowheads. Pronounced cleavages are
shown by solidcircles next to the arrowheads. Weak cleavages are shown by opencircles next to the arrowheads. The mutated
nucleotides are boxed. All VAI RNAs derived from T7 constructs
contain 6 uridine residues followed by a glycine rather than the
1-4 uridine residues found in vivo. As no structural
changes are observed for the VAI-CB and VAI-BR RNAs, the cleavage maps
of these RNAs are shown on the wild type structure. The nucleotides
substituted in the wild type sequence are shown in parentheses. In VAI-CB, 2 bases that are inserted to pair with
2 stacked bases at 41 and 42 are shown by diamonds.
Since stems I and III are the
two longest duplex regions of the VAI RNA molecule, they are probably
the major determinants of its capacity to activate 2-5(A)
synthetase. These two stems, however, are not made of perfect
base-paired sequences. Both stems contain five G-U pairs and two
mismatches. In addition, stem III contains two stacked bases
(nucleotides 42 and 43) (Fig. 7A). VAI RNA mutants, in
which these pairing defects have been eliminated, were used for
evaluating the roles of stems I and III in 2-5(A) synthetase
activation. Two such mutants were used. In VAI-BR mutant, stem I is
perfectly base paired, whereas in VAI-CB mutant stem III is perfectly
base paired (Fig. 7B). Neither of these mutants can
activate PKR, although they bind to it(20) . When tested for
2-5(A) synthetase activation, both were more effective than wild
type VAI RNA (Fig. 8). There was, however, a quantitative
difference between the efficacies of the two mutants. VAI-BR was about
4 times more potent than wild type VAI RNA, whereas VAI-CB was more
than 10 times as effective. VAI-CB was almost as effective as the 82-bp
authentic dsRNA or poly(I)
Figure 8:
Activation of 2-5(A) synthetase by
VAI RNA mutants with perfectly base-paired duplex regions. Synthetase
activation capacities of three mutants of VAI RNA were compared with
those of VAI RNA, poly(I)
Figure 9:
Activation of 2-5(A) synthetase by
VAII RNA and central domain mutants of VAI RNA. Four mutants of VAI
RNA, in which different central domain regions have been disrupted, and
VAII RNA were compared with wild type VAI RNA for their abilities to
activate 2-5(A) synthetase. For mutant 746, the mean value of two
independent determinations is shown. Other conditions are the same as
in Fig. 8.
One reason for the central domain mutant inefficiency in
activating the synthetase could be due to their failure to bind to it
efficiently. This possibility is a strong one since these mutants
cannot bind PKR, although the wild type VAI RNA
can(16, 20) . We tested this possibility by using the
newly developed electrophoretic mobility shift assay (Fig. 2).
Three mutant RNAs and the wild type VAI RNA were radiolabeled at the
3`-ends and used as probes for 2-5(A) synthetase binding. Like
VAI RNA, all the mutants formed shifted complexes (Fig. 10).
Moreover, there was no discernible difference among their efficiencies
in complex formation. These results suggest that the central domain
mutant inefficiency in activating the synthetase cannot be attributed
to a poorer binding of these molecules to the protein.
Figure 10:
Binding of VAI RNA central domain mutants
to 2-5(A) synthetase. The odd-numbered lanes contained
no protein, and the even-numbered lanes contained 2-5(A)
synthetase. Lanes1 and 2, VAI RNA; lanes3 and 4, mutant 741 RNA; lanes5 and 6, mutant 745 RNA; lanes7 and 8, mutant 748 RNA. All RNAs were labeled
at the 3`-end. The openarrow shows the position of
the probe, and the solidarrows show the positions of
shifted complexes.
This paper reports development of new tools for studying the
mode of activation of 2-5(A) synthetase by dsRNA. A recently
cloned murine 2-5(A) synthetase cDNA (8) was expressed in
bacteria using an inducible expression vector. The protein was
expressed as a fusion protein containing a hexahistidine tag at the
amino terminus. This facilitated its purification using affinity
chromatography and the subsequent production of an antibody. The
histidine tag at the amino terminus did not affect the functioning of
the enzyme or its activation by dsRNA. This general procedure can
therefore be used for producing and purifying other isozymes of
2-5(A) synthetase and their mutants. We also developed an
electrophoretic mobility shift assay for monitoring the physical
interaction between 2-5(A) synthetase and dsRNA. A perfectly
base-paired synthetic dsRNA was used as the probe for this assay. This
assay will be useful for monitoring the relative affinities of
different 2-5(A) synthetases and their mutants for dsRNA. It
should also be very useful for quick and quantitative measurements of
the affinities of different RNA and DNA molecules for this protein by
using them as unlabeled competitors for binding. We are interested
in understanding how dsRNA interacts with specific cellular proteins
regulating their biochemical activities. There are two complementary
aspects of this interest: one is to delineate the physicochemical
characteristics of dsRNA-protein interaction, and the other is to
evaluate the regulatory roles of such interactions in cell physiology.
The current study sheds light mainly on the first aspect and
illuminates differences between two IFN-induced dsRNA-activable enzymes
with respect to the structural requirements of their activator RNAs.
The main observation reported here is that the partially
double-stranded adenoviral VAI RNA interacts with 2-5(A)
synthetase and activates it. Although this is the first report of VAI
RNA interactions with 2-5(A) synthetase, similar interactions
with PKR have been extensively studied by us and
others(17, 20, 21, 23) . Both in
vivo and in vitro studies have demonstrated that VAI RNA
blocks the activation of PKR and thereby maintains normal levels of
protein synthesis in adenovirus-infected cells. As a result, unlike the
wild type virus, mutants lacking VA RNAs replicate poorly, especially
in IFN-treated cells. Analysis of VAI RNA secondary structure (15) showed that the RNA exists in solution as a highly
base-paired molecule with two long imperfectly base-paired stems of
23-25 bp joined at the center by a short stem-loop structure (Fig. 7A). The base-paired duplex regions were thought
to be the crucial determinants of VAI RNA for its interaction with PKR.
Extensive mutational analyses, however, revealed that in addition to
the ds regions, other structural elements present in the complex
stem-loop structure are required for PKR-inhibitory activity of VAI
RNA. Our recent study has demonstrated the importance of the short
stem-loop structure in the central domain of VAI RNA for its
interaction with PKR (20) . VAI mutants with mutations in this
region, but not elsewhere in the central domain or in the long duplex
regions, are defective in binding and inhibiting PKR. These and other
results suggest that PKR recognizes a specific secondary or tertiary
structure in VAI RNA. This interaction is distinct from its recognition
and binding to authentic dsRNA, although the same domain of PKR may be
responsible for both interactions(24) . Our studies have
identified the dsRNA binding domain (DRBD) of 2-5(A)
synthetase(8) . The DRBD is located at the amino terminus of
the protein between residues 1 and 158. Similar studies with PKR
established that its DRBD is also located at the amino
terminus(9) . Although both DRBDs bind dsRNA, we could not
detect any homology in their primary and secondary
structures(9) . Thus, these two IFN-inducible enzymes
apparently belong to two different families of dsRNA binding proteins.
Many other members of the PKR family, which bear sequence homology in
their DRBDs, have been identified(25) . However, the
2-5(A) synthetase family has no other known members. As the noted
structural difference would suggest, the characteristics of dsRNA
interactions of the two enzymes are also different. PKR binds dsRNA
much more strongly than 2-5(A) synthetase; 0.3 M NaCl
dissociates the 2-5(A) synthetase-dsRNA complex, whereas the
PKR-dsRNA complex is not dissociated even in the presence of 1 M NaCl. The
above differences in the properties of dsRNA interactions of the two
enzymes led us to investigate the nature of interactions between
2-5(A) synthetase and adenoviral VA RNAs. Our EMSA results
clearly demonstrated that VAI RNA can bind to 2-5(A) synthetase.
Moreover, it could compete with authentic dsRNA for binding, thus
suggesting that they may be binding to the same site on the protein.
These results are consistent with the known properties of VAI RNA
interactions with PKR. However, unlike its action on PKR, VAI RNA
activated 2-5(A) synthetase. Like authentic dsRNA, this
activation was concentration dependent. On a weight basis, VAI RNA was
about 10-fold less effective than perfect dsRNA. This is not
unexpected, given the fact that VAI RNA is only partially double
stranded. Even for perfectly base-paired dsRNAs, there are quantitative
differences in their efficacies to activate 2-5(A) synthetase.
For example, in the results reported here, we observed that
poly(A) The results obtained with mutant
VAI RNAs are intriguing. Mutants 741, 745, 746, and 748, in which the
structure of the central domain of VAI RNA has been disrupted (Fig. 7), do not bind to or inhibit the action of
PKR(21) . Our results showed that they are also less effective
than wild type VAI RNA in activating 2-5(A) synthetase. They are,
however, not totally inert. 3-5-fold stimulation was observed
with these mutants. This level of activation is not due to nonspecific
interaction between 2-5(A) synthetase and any RNA molecule since
even a high concentration of perfectly single-stranded RNA, such as
poly(A) or poly(U), could not activate the enzyme (Fig. 4).
Thus, it is reasonable to speculate that the residual ds structure of
these mutants is probably responsible for activating the enzyme.
Consistent with the observed, albeit reduced activity of the mutant VAI
RNAs, they efficiently bound to 2-5(A) synthetase as monitored by
EMSA (Fig. 10). Thus, the observed differences in the ability to
activate the enzyme was not reflected at the level of binding to the
protein. It is interesting to note that VAII RNA was about 5-fold less
efficient than VAI RNA in activating 2-5(A) synthetase. The
reason for the inability of VAII RNA to efficiently activate
2-5(A) synthetase is not clear. Currently, the structure of the
VAII RNA is not known. Computer-generated structures indicate the
presence of ds regions; experimentally determined structure will be
necessary to evaluate these results. The second set of mutants,
VAI-BR and VAI-CB, were much better activators than the wild type VAI
RNA. In these mutants, the occasional mismatches present in stems I and
III have been repaired. Their increased efficacy suggests that these
stem structures are important determinants of the ability of VAI RNA to
activate 2-5(A) synthetase. Our results also demonstrate that
occasional mismatches in the ds region of an RNA can affect its
activating potency considerably. Surprisingly, there is a marked
difference between the efficacies of VAI-BR and VAI-CB mutants, which
indicates that stem III of VAI RNA is a more important element than
stem I. The VAI-CB mutant was almost as good as the 82-bp perfect dsRNA
and poly(I) The underlying mechanism
of differential response of the two enzymes to VAI RNA and its mutants
remains to be determined. For one set of mutants, those in the central
domain, the difference is clearly in their ability to bind to the two
proteins. The wild type and the stem mutants, however, can bind to both
the proteins. One can speculate that specific conformational changes in
the two proteins are required for activating them, and VAI RNA or the
BR and CB mutants can promote this change in 2-5(A) synthetase
but not in PKR. Alternatively, it is possible that binding of at least
two molecules of PKR to the same RNA molecule is necessary for its
intermolecular autophosphorylation, and VAI RNA may not be able to bind
two PKR molecules. The specific conformational change model is favored
by the fact that wild type PKR and mutants of PKR, which do not bind
dsRNA, can be activated by heparin(27) . Small molecules, such
as heparin, are not likely to bind to two large PKR molecules
simultaneously, although such a possibility has not been ruled out
experimentally.
Volume 270,
Number 7,
Issue of February 17, 1995 pp. 3454-3461
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
5` Oligoadenylate Synthetase by
Adenoviral VAI RNA (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
)induce the synthesis of two
enzymes that require double-stranded RNA (dsRNA) as their
activators(1) . One of these enzymes, a dsRNA-dependent protein
kinase (PKR), phosphorylates the 
subunit of the translation
initiation factor, eIF-2, and thus inhibits the initiation of protein
synthesis(2, 3) . The other enzyme, 2`-5`
oligoadenylate (2-5(A)) synthetase, polymerizes ATP into
2-5(A), which in turn activates a latent ribonuclease leading to
the degradation of mRNAs(4) . There are multiple isozymes of
2-5(A) synthetase, all of which are induced by IFN and activated
by dsRNA. The 2-5(A) synthetase pathway is responsible for
inhibiting the replication of picornaviruses in IFN-treated
cells(5, 6, 7) .
)This dissimilarity
in the dsRNA binding property is reflected in the absence of any
structural homology between the dsRNA binding domains of the two
enzymes(8, 9) . Unlike 2-5(A) synthetase, PKR
dose response to dsRNA is biphasic. Low concentrations of dsRNA
activate it, and high concentrations inhibit it (2) . This
phenomenon is probably due to the need for two PKR molecules to bind to
the same dsRNA molecule so that intermolecular autophosphorylation is
possible.
Materials
Restriction enzymes,
poly(I)
poly(C), RNasein, and other fine chemicals were from
Boehringer Mannheim. Plasmid pGEM3 was from Promega. Plasmids pET3a and
pET15b and the His.Bind resin were from Novagen. Polyethyleneimine
cellulose plates were from EM Scientific.
[-
P]ATP (specific activity, 800 Ci/mmol)
was from DuPont NEN.2-5(A) Synthetase Subcloning
The
coding region of 2-5(A) synthetase 9-2 cDNA (8) was
subcloned into plasmid pET3a at NdeI and BamHI sites.
The plasmid 2-5(A) synthetase/pGEM3 (8) was digested with XhoII and BamHI, and the fragment containing the
coding region was purified and ligated with NdeI/BamHI-digested pET3a with a NdeI-XhoII linker adapter. The insert from
2-5(A) synthetase/pET3a was subcloned into the histidine tag
vector plasmid pET15b in NdeI and BamHI restrictions
sites after releasing the 1491-bp insert from pET3a with the same
enzymes.Bacterial Expression and Purification of 2-5(A)
Synthetase
Initial transformation and screening was done in Escherichia coli (DH5). For production of the protein,
appropriate clones were introduced into E. coli BL21 (DE3).
Unless specified, all purification steps were carried out at 4 °C.
An overnight culture of the colony in LB broth (Difco) containing 100
µg/ml ampicillin was diluted ten times and was grown and induced at
0.6 optical density units with 2 mM isopropyl-1-thio-
-D-galactopyranoside for 2 h. The
induced cells (500 ml) were chilled immediately and harvested at 10,000
g for 15 min, washed twice with ice-cold 20 mM Tris-HCl, pH 7.5, resuspended in 20 mM Tris-HCl, pH 8.00,
300 mM NaCl, 5 mM imidazole, 0.5% Triton X-100, and 1
mM phenylmethylsulfonyl fluoride (binding buffer) to give a
20% (w/v) suspension (approximately 12.5 ml), and sonicated on ice in
5-ml portions for a total of 1.8 min with each pulse of 30 s. The
resulting suspension was then spun at 40,000 g for 60
min to remove the cell debris, and the supernatant (approximately 12
ml) was then applied to a 1-ml bed volume of His.Bind resin column,
which had been successively washed with the binding buffer and 50
mM NiSO
, followed by equilibration with the
binding buffer. The column was then washed with 20 volumes of binding
buffer followed by 20 volumes of washing buffer, which had the same
ingredients as the binding buffer but with 60 mM imidazole.
The enzyme was then eluted with two bed volumes of elution buffer
containing 20 mM Tris-HCl, pH 8.00, 500 mM NaCl, 500
mM imidazole, and 1 mM phenylmethylsulfonyl fluoride.
The eluate was dialyzed against 20 mM Tris-HCl, pH 7.5, and
10% glycerol and was stored at -70 °C till further use.2-5(A) Synthetase Assay
In-solution
synthetase assay was done according to the method of Mory et
al.(18) with minor modifications. Unless otherwise
specified, 10 µl of reaction mixture containing the enzyme, 20
mM Tris-HCl, pH 7.5, 20 mM magnesium acetate, 2.5
mM dithiothreitol, 5 mM ATP, 5 µCi of
[-P]ATP (specific activity, 800 Ci/mmol),
and the activator RNA was incubated for 2 h at 30 °C. The reaction
was stopped by boiling the sample for 3 min and was then centrifuged at
14,000 g for 10 min; 8 µl of the supernatant was
incubated for 3 h at 37 °C with 3 µl of 1 unit/µl of calf
intestine alkaline phosphatase. 2 µl of the sample was then spotted
on a polyethyleneimine-cellulose thin layer chromatography plate and
was resolved in 750 mM KHPO, pH
3.5(19) . The 2-5(A) formed was then quantitated by
exposing the plate to storage phosphor screen and expressed as
arbitrary units.Preparation of dsRNA
The 82-bp synthetic
dsRNA was prepared by hybridizing sense and antisense transcripts
generated in vitro using the template plasmid
pGEM3-9T(9) . Complementary strands were generated by
digesting the plasmid with appropriate restriction enzymes followed by
transcription with SP6 and T7 polymerase. Hybridization was carried out
at 50 °C for 6 h in 40 mM PIPES, pH 6.4, 400 mM NaCl, 1 mM EDTA, and 80% deionized formamide. The sample
was then diluted with 10 volumes of RNase digestion buffer containing
300 mM NaCl, 10 mM Tris-HCl, pH 7.4, 5 mM EDTA, and 40 µg/ml RNase A, incubated at 20 °C for 1 h,
and extracted with phenol-chloroform. The RNA was recovered by
precipitation with ethanol.Preparation of VA RNAs
The VAI genes were
transcribed in vitro using T7 RNA polymerase and plasmids in
which the wild type VAI, VAII, or mutant VAI gene has been cloned
behind a T7 promoter. The transcribed RNA samples were purified by
electrophoresis on a 6% native polyacrylamide gel before use. None of
these RNA preparations could activate PKR either at low or high
concentrations. If needed, the RNAs were radiolabeled at the 5`-end
using polynucleotide kinase and [
-P]ATP or
at the 3`-end using P as described before(16) . Electrophoretic Mobility Shift Analysis
(EMSA)
Binding reactions were performed by mixing the
end-labeled dsRNA or VAI RNAs and 2-5(A) synthetase in binding
buffer (20 mM Tris-HCl, pH 7.5, 50 mM KCl, 2 mM MgCl, 5% glycerol) and incubating at 25 °C for 10
min. The amount of labeled RNA per lane was about 20,000 cpm. After
incubation, a dye-glycerol solution was added, and the samples were
loaded immediately on a 4% polyacrylamide gel
(acrylamide:bisacrylamide, 29.2:0.8). The gel was cast in 40 mM Tris-glycine buffer and was prerun in the same buffer for 45 min
at 150 V. Radioactivity was detected by autoradiography. The same
results were obtained by using 0.25 Tris borate buffer for
casting and running the gel. In competition experiments, cold
competitor was added 15 min prior to the addition of labeled RNA. Dried
gels were analyzed by autoradiography.2-5(A) Synthetase Antibody
The
antipeptide B antibody was a gift from Judith Chebath(22) .
Antibody against purified bacterially produced histidine-tagged
9-2 synthetase was raised in rabbits using standard procedures.
For supershifting EMSA assays, equal quantities of immune or preimmune
sera were included in the reaction mixture.
VAI RNA-2-5(A) Synthetase
Interactions
VAI RNA is known to inhibit the activation of
PKR by authentic dsRNA. This inhibition occurs because VAI RNA competes
with dsRNA for binding to PKR, but, unlike dsRNA, it cannot activate
the enzyme(15, 20) . To study the effect of VAI RNA on
2-5(A) synthetase, we first wanted to determine if it can bind to
the protein. The enzyme used for this purpose was the mouse 9-2
isozyme of 2-5(A) synthetase whose cDNA we have previously
cloned(8) . The enzyme was expressed in E. coli as a
hexahistidine-tagged protein and purified by affinity chromatography
following procedures described under ``Experimental
Procedures.'' The purified enzyme was incubated with
[-P]ATP and activator RNAs, and the
resultant 2-5(A) molecules were detected by a thin layer
chromatographic assay. As shown in Fig. 1, little 2-5(A)
was synthesized in the absence of activator RNA (lane1). Poly(I)poly(C) of heterogenous length (lane2) and a synthetic dsRNA of 82 bp (lane3) were highly efficient in activating the enzyme. These
results demonstrated that the bacterially produced hexahistidine-tagged
2-5(A) synthetase has enzymatic properties indistinguishable from
those of the native enzyme isolated from mammalian cells.
) are indicated. Lane1, no activator; lane2,
poly(I)poly(C); lane3, in vitro synthesized 82-bp dsRNA. All RNAs were used at 25
µg/ml.
poly(C) (lane3) was
required for observing the same level of competition. These results
suggested that VAI RNA can bind to 2-5(A) synthetase. This
suggestion was directly tested in the experiment shown in Fig. 3B. VAI RNA was radiolabeled at the 5`-end and
used as the probe to bind 2-5(A) synthetase. Several shifted
complexes were formed (lane2). Although the reason
for the appearance of multiple complexes is not known, they may
represent alternate conformational states of the VAI RNA. These results
demonstrated that VAI RNA can directly bind to the 2-5(A)
synthetase protein.
poly(C) added as
the competitor; lane4, 500 ng of VAI RNA added as
the competitor. B, both lanes had 0.5 ng of 5`-labeled VAI
RNA. Lane1, no protein; lane2,
500 ng of 2-5(A) synthetase added. The openarrow shows the positions of the probes, and the solidarrows show the positions of the shifted
complexes.
Activation of 2-5(A) Synthetase by VAI
RNA
In the next series of experiments, we examined the
potential of VAI RNA to activate 2-5(A) synthetase. The VAI RNA
preparation used for these studies was free of any dsRNA contamination,
as judged by gel electrophoresis, and it could not activate PKR either
at low or high concentration ( (20) and data not shown). VAI
RNA could, however, effectively activate 2-5(A) synthetase (Fig. 4, lanes2 and 3). At 100
µg/ml it was 5-fold less effective than poly(A)poly(U) (lanes3 and 6) and 10-fold less effective
than poly(I)poly(C) (data not shown). Two single-stranded RNAs,
poly(A) (lane4) and poly(U) (lane5) could not activate the enzyme. These results
demonstrated that VAI RNA is an effective activator of 2-5(A)
synthetase under conditions at which single-stranded RNAs have no
effects.
poly(U).
Role of ds Stems of VAI RNA in Activating
2-5(A) Synthetase
The structure of VAI RNA has been
studied
extensively(15, 16, 17, 20) .
Computer modeling has suggested several alternative secondary
structures that have been experimentally tested by using susceptibility
to limited RNase digestion as an assay. The experimentally derived
secondary structure of VAI RNA consists of two long duplex regions,
stem I and III, connected at the center by a short duplex region, stem
II (Fig. 7A). Stem I consists of a duplex in which
nucleotides 1-22 are base paired to nucleotides 134-155
(numbers here represent the position of the nucleotides from the 5`-end
with G start(16) ). Stem II consists of a short duplex in which
nucleotides 31-35 are base paired to nucleotides 128-132.
In stem III, the longest of these, nucleotides 37-62 are base
paired with nucleotides 71-94. As a result, nucleotides
63-71 exist as a loop (loop B). The structure in the central part
of the molecule that is referred to as the central domain is complex
and poorly defined. This part of the molecule contains a small loop in
the 5`-side (loop A) and a short stem loop (loop C) and an additional
minor loop (loop D) in the 3`-side.
poly (C). These results indicate that
stems I and III are the structural elements in VAI RNA, which are
important for its ability to activate 2-5(A) synthetase. Between
the two stems, stem III, consisting of 26 bp, appears to be more
important than stem I, which contains 22 bp.
poly(C), and in vitro transcribed 82-bp dsRNA. Occasional mismatches in stem I of VAI
RNA have been repaired in the mutant VAI-BR; similar restoration has
been done in stem III of the mutant VAI-CB. The results are presented
as -fold stimulation over the activity without any RNA activator. The numbers over each bar show the numbers of independent
measurements with the specific RNA. The corresponding errorbars are also shown. The incubations were for 2 h with 25
µg/ml RNA.
Role of the Central Domain
The central
domain of the VAI RNA molecule (Fig. 7A) is thought to
be necessary for maintaining the integrity of its structure. In the
next experiment, we examined the effects of disrupting this region on
VAI RNA ability to activate 2-5(A) synthetase. Four central
domain mutants, sub741, sub745, sub746, and sub748, were used for this
purpose. These mutants contain linker scan substitution mutations, and
their secondary structures have been experimentally determined (Fig. 7B, (20) ). Each of the mutations
disrupts the central domain and distorts the secondary structure of the
molecule in a distinct and different way. Like wild type VAI RNA, none
of these mutants can activate PKR, but unlike the parental molecule,
they also fail to bind to PKR(20, 21) . All of these
mutants were less efficient than wild type VAI RNA in activating
2-5(A) synthetase (Fig. 9). None of them, however, was
unable to activate the enzyme. These results indicate that the central
domain is not essential, although its disruption leads to a marked
reduction in the activity. This reduction may be caused by a
concomitant perturbation of the structures of stems I and III. Indeed,
among the four mutants, sub745 was the most effective one (Fig. 9). The two-stem structures are preserved in this mutant (Fig. 7B), although there is a minor perturbation in
the structure of stem III when compared with wild type VAI RNA. Another
VA RNA, VAII, was a poor activator of 2-5(A) synthetase (Fig. 9). Since the secondary structure of VAII RNA has not been
determined, the underlying structural defect cannot be identified at
this time.
Activation of PKR, but not of 2-5(A)
synthetase, by dsRNA has a biphasic dose response; low concentrations
activate, whereas high concentrations do not. Finally, modified dsRNAs
or dsRNAs with periodic mismatches have been shown to activate the two
enzymes with different efficiencies(5, 26) .poly(U) was less efficient than poly (I)poly(C). The
VAI RNA preparations used in our studies were free of any contaminating
dsRNA which, is sometimes generated as an aberrant product during in vitro transcription of RNAs. The strongest evidence for
such a conclusion is functional: these preparations of RNAs could not
activate PKR either at a low or a high concentration. Moreover, no
contaminating RNA could be detected in these preparations when analyzed
by denaturing gel electrophoresis.poly(C), thereby suggesting that the 26-bp perfectly
paired stem III could be an optimal inducer by itself. The VAI-BR and
VAI-CB RNAs, which could activate 2-5(A) synthetase so
efficiently, cannot activate PKR, although they can bind to it
efficiently and are more active in inhibiting the activity of PKR as
compared with the wild type VAI RNA(20) . This exemplifies a
major difference in the structural requirements of dsRNAs to be
effective as activators of the two enzymes.
))
We thank Judith Chebath for the 2-5(A)
synthetase antipeptide antibody and Karen Matthews for expert
secretarial assistance.
©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:
|
|
 |
|
  R. J. Molinaro, B. K. Jha, K. Malathi, S. Varambally, A. M. Chinnaiyan, and R. H. Silverman Selection and cloning of poly(rC)-binding protein 2 and Raf kinase inhibitor protein RNA activators of 2',5'-oligoadenylate synthetase from prostate cancer cells Nucleic Acids Res., December 2, 2006; 34(22): 6684 - 6695. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wang, S.-A. Xue, G. Hallden, J. Francis, M. Yuan, B. E. Griffin, and N. R. Lemoine Virus-Associated RNA I-Deleted Adenovirus, a Potential Oncolytic Agent Targeting EBV-Associated Tumors Cancer Res., February 15, 2005; 65(4): 1523 - 1531. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Whitmore, M. J. DeVeer, A. Edling, R. K. Oates, B. Simons, D. Lindner, and B. R. G. Williams Synergistic Activation of Innate Immunity by Double-Stranded RNA and CpG DNA Promotes Enhanced Antitumor Activity Cancer Res., August 15, 2004; 64(16): 5850 - 5860. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Samuel Antiviral Actions of Interferons Clin. Microbiol. Rev., October 1, 2001; 14(4): 778 - 809. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Goodbourn, L. Didcock, and R. E. Randall Interferons: cell signalling, immune modulation, antiviral response and virus countermeasures J. Gen. Virol., October 1, 2000; 81(10): 2341 - 2364. [Full Text]
|
|
|
|
 |
|
 A. Naganuma, A. Nozaki, T. Tanaka, K. Sugiyama, H. Takagi, M. Mori, K. Shimotohno, and N. Kato Activation of the Interferon-Inducible 2'-5'-Oligoadenylate Synthetase Gene by Hepatitis C Virus Core Protein J. Virol., September 15, 2000; 74(18): 8744 - 8750. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. N. Sarkar, S. Bandyopadhyay, A. Ghosh, and G. C. Sen Enzymatic Characteristics of Recombinant Medium Isozyme of 2'-5' Oligoadenylate Synthetase J. Biol. Chem., January 15, 1999; 274(3): 1848 - 1855. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kumar and G. G. Carmichael Antisense RNA: Function and Fate of Duplex RNA in Cells of Higher Eukaryotes Microbiol. Mol. Biol. Rev., December 1, 1998; 62(4): 1415 - 1434. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Hartmann, P. L. Norby, P. M. Martensen, P. Jorgensen, M. C. James, C. Jacobsen, S. K. Moestrup, M. J. Clemens, and J. Justesen Activation of 2'-5' Oligoadenylate Synthetase by Single-stranded and Double-stranded RNA Aptamers J. Biol. Chem., February 6, 1998; 273(6): 3236 - 3246. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ghosh, S. N. Sarkar, W. Guo, S. Bandyopadhyay, and G. C. Sen Enzymatic Activity of 2'-5'-Oligoadenylate Synthetase Is Impaired by Specific Mutations that Affect Oligomerization of the Protein J. Biol. Chem., December 26, 1997; 272(52): 33220 - 33226. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ghosh, S. Y. Desai, S. N. Sarkar, P. Ramaraj, S. K. Ghosh, S. Bandyopadhyay, and G. C. Sen Effects of Mutating Specific Residues Present Near the Amino Terminus of 2'-5'-Oligoadenylate Synthetase J. Biol. Chem., June 13, 1997; 272(24): 15452 - 15458. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
