Download Analysis of the stimulation of reporter gene expression by the ¢r3

Journal of General Virology (1993), 74, 1055 1062. Printedin Great Britain
1055
Analysis of the stimulation of reporter gene expression by the ¢r3 protein
of reovirus in co-transfected cells
P. E. M . M a r t i n and M . A. M c C r a e *
Department of Biological Sciences, University of Warwick, Coventry CV4 7AL, U.K.
The stimulation of reporter gene expression following
co-transfection with the $4 gene of mammalian reoviruses was analysed. The a3 protein of type 3 reovirus
gave a five- to eightfold increase in expression of
chloramphenicol acetyltransferase and fl-galactosidase
but this was found to be dependent upon the nature of
the promoter being used to drive reporter gene expression. The o-3 protein of reovirus type 1 failed to
stimulate reporter gene expression under any of the
conditions used. Hybrid constructs between the $4
genes of reoviruses type 1 and 3 were used to map the
stimulation characteristic to the carboxy-terminal third
of the gene. Analysis of the level of ~r3 protein
accumulation in transfected cells showed that the
reovirus type 3 protein accumulated to a much higher
level than that of reovirus type 1. Using the hybrid gene
constructs this higher level of protein accumulation was
shown to co-segregate with the ability to stimulate
reporter gene expression.
Introduction
being responsible for the serotype-specific modulation of
host macromolecular synthesis following infection
(Sharpe & Fields, 1982). The $4 gene encodes the protein
o-3, which is a major component of the outer shell of the
virus particle (McCrae & Joklik, 1978; Mustoe et al.,
1978). Sequence analysis of the $4 gene from the reovirus
type 1 and 3 serotypes (Giantini et aI., 1984; Atwater
et al., 1986) has shown that both are 1196 nucleotides in
length with a single long open reading frame encoding
the ~r3 protein of 365 amino acids. Despite the observed
differences in the phenotypes of these genes with respect
to host cell RNA and protein synthesis inhibition, the
level of sequence identity was found to be very high:
94 % at the nucleotide level and 96 % at the amino acid
level with the observed differences scattered throughout
the gene (Giantini et al., 1984; Atwater et al., 1986).
In an attempt to begin to deduce the molecular
mechanism by which o-3 protein exerts its inhibitory
effects on host macromolecular synthesis Giantini &
Shatkin (1989) have reported using a transient eukaryote
expression system to examine the effect(s) of o-3 protein
on expression of a co-transfected reporter gene. It had
been expected from its effects following infection with
type 3 virus that the transfected o-3 protein would inhibit
reporter gene expression. However, contrary to expectations co-transfection of reovirus type 3 o-3 protein
resulted in a marked stimulation of reporter gene
expression (Giantini & Shatkin, 1989). This report
describes work that extends these original observations
and using in vitro generated recombinants between
The pathogenic process associated with virus infection
can be conveniently divided into extracellular and
intracellular pathogenesis. Extracellular pathogenesis
can be defined as the process by which the infecting virus
enters the host and finds its way to the interior of the
target cell. Intracellular pathogenesis on the other hand
is the process by which the virus achieves productive
replication within the target cell. A major component of
intracellular pathogenesis is the steps taken by the virus
to redirect the metabolism of the cell to favour its growth
and this often involves changes at both the transcriptional and translational level. Amongst the more striking
examples of intracellular pathogenesis are the changes in
cellular RNA and protein synthesis induced by infection
of a variety of cell types with mammalian reoviruses
(Joklik, 1985). These examples have been of particular
interest because of the differences induced following
infection with each of the three reovirus serotypes. Thus
despite their structural similarity and ability to interact
genetically with each other the three serotypes of reovirus
have quite different effects on host cell RNA and protein
synthesis following infection. Reovirus type 1 has little or
no effect on host cell macromolecular synthesis, reovirus
type 2 causes a rapid almost complete inhibition whereas
infection with reovirus type 3 results in a moderate
inhibition at late times post-infection (Joklik, 1985).
Analysis of reassortants between the three serotypes
identified the smallest genomic RNA segment $4 as
0001-1517 © 1993SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Tue, 02 May 2017 11:16:23
1056
P. E. M. Martin and M. A. McCrae
Table 1. Construction details of plasmids used
Insert
(a) Hind III 5'
Description
BamHI
Full-length $4 cDNA cloned into the HindlIUBarnHl sites of the
Bluescribe polylinker.
$4 gene
(6)
EcoRI
HindIII
BamHI Sinai
HindlII/BamHI ligation of the $4 gene into the polylinker
downstream of the LTR of HIV-1 in pHXBH2. Final plasmids
HX-1 and HX-3.
HIV LTR
(c)
EcoRI
HindlII
SmaI
SalI
EcoRI/SmaI ligation of the LTR $4 region from HX plasmids into
i
the polylinker upstream of a polyadenylation site in a Bluescribe
plasmid. Final plasmids HPA-1 and HPA-3.
Poly(A)
(a3
EcoRI
HindIII
SalI/EcoRI SalI/HindllI
Insertion of the SV40 origin of replication downstream of the
polyadenylation site by blunt end ligation. Final plasmids HIV-1
and HIV-3.
SV40 origin
12
5t
10
C
)<
t
8
b,
<
0
0
ttIV-1
-
HIV-1
+
HIV-3
.
HIV-3
+
Con
Con
+
"
HeLa
+
' " ~'
~ '
Cos-I
+
Fig. 1. The effect of a3 protein of reovirus type 1 and type 3 on CAT reporter gene expression driven by the immediate-early promoter
of CMV. (a) COS-1 cells were transfected with 1 gg of the reporter plasmid CMV-CAT in the presence ( + ) or absence ( - ) of l gg
of the tat expression plasmid pSV40tat. For 'HIV-1 ' experiments, 10 gg of a plasmid carrying the reovirus type 1 S4 gene under the
control of the LTR of HIV was added to the transfection mixtures and for 'HIV-3' experiments, 10 gg of a plasmid carrying the
reovirus type 3 $4 gene under the control of the same HIV promoter was added. Control (Con) cultures were co-transfected with 10 gg
of the control plasmid pEXP6 to ensure that all cultures received the same amount of plasmid DNA. (b) HeLa or COS-I cells were
transfected with 1 gg of the plasmid H I V ~ A T in which the CAT reporter gene expression was being driven by the LTR of HIV in
the presence ( + ) or absence ( - ) of 1 ~tg of pSV40tat. All cultures were also co-transfected with the control plasmid pEXP6 to ensure
that results were comparable to those shown in (a).
cDNAs prepared from the $4 genes of reovirus types 1
and 3 begins the process of localizing the region(s) of the
gene responsible for the differences in phenotype found
between reovirus types 1 and 3 with respect to this
stimulatory effect.
Methods
Plasmid construction. All cDNA manipulations were performed as
described by Sambrook et al. (1989). The $4 cDNAs were generated in
this laboratory from total viral dsRNA using oligonucleotides specific
to the 5" and 3' ends of the $4 genes to prime cDNA synthesis. The
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Tue, 02 May 2017 11:16:23
Reporter gene stimulation by a3 protein
construction of the a3 protein expression plasmids used is outlined in
Table 1. Plasmids pSV40-1 and pSV40-3 were produced by inserting
the $4 cDNA into the HindlII/BamHI sites of the plasmid pEXP6.
This plasmid carries the simian virus 40 (SV40) origin, early promoter
and polyadenylation signals flanking either end of the M 13 polylinker
(K. Leppard, personal communication). The construction of plasmids
HYB1-3 and HYB3-1 is outlined in Fig. 4.
Maintenance of cells. COS cells were maintained in Dulbecco's
modified eagle's MEM supplemented with 10% fetal calf serum and
100 lag/ml penicillin/streptomycin. Cells were split 1:10 every 4 days.
Transfection and analysis o f reporter gene expression. Subconftuent
monolayers of COS cells in 60 mm tissue culture dishes were transfected
in triplicate with 10 lag of S4-containing DNA, 1 lag of a tat proteinexpressing plasmid [pSV40-tat which carries the tat protein of human
immunodeficiency virus type 1 (HIV-1) under the control of the early
promoter of SV40] and 1 lag of reporter gene plasmid using the calcium
phosphate technique (Gorman et al., 1982). Cells were harvested 48 h
post-transfection, centrifuged and cell pellets were stored at - 2 0 °C
until required. Cell lysates were prepared as described by Gorman et al.
(1982). The protein concentration of each lysate was determined using
the Bio-Rad protein assay kit and a standard amount of protein was
used in each chloramphenicol acetyltransferase (CAT) assay. CAT
activity was measured using the two ftuor diffusion assay as described
by Neuman et al. (1987). Briefly, cell extracts were increased to 50 lal
with 100 mM-Tris-HC1 pH 7.5 and heated at 70 °C for 15 min to reduce
background activity. The reaction mix consisted of 50 ~tl of l0 mMchloramphenicol, 25lal 1M-Tris HC1 pH7.5, 0.1 laCi [3H]acetyl
coenzyme A and 124 lal of distilled water. This was mixed with the heattreated cell extract, placed in a 5 rnl scintillation vial and overlaid with
non-aqueous scintillation fluid (toluene with 0.5% PPO). Samples
were incubated at room temperature in a scintillation counter
and radioactivity was counted every 15 min over a 3 h period, flGalactosidase (]?-gal) assays were performed as described by Sambrook
et al. (1989).
Pulse labelling o f cells and immunoprecipitation. At 48 h posttransfection, duplicate samples of COS-1 cells were starved for 1 h in
methionine-free medium and then pulse-labelled with 100 laCi of
[35S]methionine per dish in methionine-free medium for 6 h. One well
of each sample was then harvested into 0.25 ml of 50 mM-Tris-HC1
pH 8 and the other was chased by overlaying the cells with 1 ml of 100fold methionine-containing medium for a further 6 h. Equal amounts
of labelled proteins as determined by acid precipitation (Sambrook
et al., 1989) were immunoprecipitated.
For immunoprecipitation 50 lal samples were increased to 100 gl with
twofold concentrated IP3 (400 mM-TribHC1 pH 8, 100 mM-NaC1,
2 mM-EDTA, 2% NP40, 2% deoxycholate, 0.2% SDS) (Watson et al.,
1982), 10 lal of undiluted serum (polyclonal rabbit sera raised against
either purified reovirus type 1 or type 3 as appropriate) was added and
the mixture incubated for 1 h at 4 °C with gentle agitation. Control
experiments, in which these sera were used to immunoprecipitate viral
proteins from samples in which [aSS]methionine-labelled uninfected
and infected cells were mixed such that the viral proteins constituted
only a small percentage of those labelled, were carried out to show that
each serum was capable of efficiently precipitating reovirus type 1 or 3
viral proteins as appropriate (results not shown). Following incubation
the precipitation mixtures were added to pre-swollen Protein
~ S e p h a r o s e (5 mg in 100 lal IP3) and incubated for a further 1 h at
4 °C. The Sepharose was extracted by microcentrifugation (1 min) and
the supernatant discarded. The pelleted beads were then sequentially
washed with IP2 (IP3 with 20 mg/ml BSA), IP2 with 1 M-NaCI, twice
with IP3 and finally IP1 (20 mM-Tris-HC1 pH 8, 100 mM-NaC1, 1 mMEDTA, 1% NP40). Immune complexes were disrupted by boiling in
2 % SDS-5 % 2-mercaptoethanol and analysed on 10 % polyacylamide
gels (Laemmli, 1970).
1057
Results
Effect of promoter on reporter gene stimulation
The initial strategy chosen to analyse the effect of o-3
protein on reporter gene expression made use of two
types of plasmid. In the first a full-length cDNA copy of
the $4 gene from either reovirus type 1 or type 3 was
positioned downstream of the long terminal repeat
(LTR) sequence from HIV. The rationale for choosing
this promoter is that it is trans-activated by the tat gene
product of HIV offering the possibility of regulated
expression of o-3 protein in the system, in a manner
analogous to that employed by Sun & Baltimore (1989)
in their analysis of the role of the protein 2A of
poliovirus in the inhibition of host cell protein synthesis.
The second plasmid had the reporter gene for CAT
positioned downstream of the immediate early promoter
of cytomegalovirus (CMV). This promoter was chosen
as we wanted to examine the effect of 0-3 protein
expression in several different cell types and it has been
reported to be a strong constitutive promoter in a wide
~4
e~0
~3
-6
0
SV40-1
SV40-3
HIV-1
HIV-3
Con
Fig. 2. The effect of a3 protein of reovirus type 1 and type 3 on fl-gal
reporter gene expression. COS-1 cells were transfected with 1 lag of a
reporter gene plasmid in which the Eseherichia coli lacZ ~-gal) gene
was driven by the early promoter of SV40. For 'HIV-I' and 'HIV-3'
experiments the transfection mixture was supplemented with 10 gg of
the plasmids carrying the $4 gene of either reovirus type 1 or 3 under
the control of the LTR of HIV (see Fig. 1) and 1 lag of pSV40tat. For
'SV40-1' and 'SV40-3' experiments the only difference was that
expression of the appropriate reovirus or3 protein was driven by the
early promoter of SV40. The control (Con) culture was co-transfected
with 10 lag of pEXP6 to ensure all cultures received the same amount
of plasmid DNA. The level of reporter gene expression measured has
been normalized relative to that seen in the control culture.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Tue, 02 May 2017 11:16:23
1058
P. E. M. Martin and M. A. M c C r a e
3-5
(a)
3
"t
2-5
×
d
2
¢..)
i~ 1-5
~2
<
1
0.5
0
~
ABC ABC ABC
TK CMV RSV
Promoter
•
ABC
SV40
0
TK
RSV SV40CMV
Promoter
Fig. 3. The influence of promoter being used to drive reporter gene expression on the effects of reovirus type 1 and 3 ~r3 protein on
reporter gene expression. (a) COS-1 cells were co-transfected with 10 gg of SV40-1 (A), SV40-3 (B) (see Fig. 2) or the control plasmid
pEXP6 (C) and 1 gg of various CAT plasmids in which the CAT gene expression was being driven by different promoters: TK,
thymidine kinase promoter of herpes simplex virus type 1; CMV, immediate early promoter of CMV; RSV, LTR of Rous sarcoma
virus; SV40, early promoter of SV40. CAT expression has been normalized in each case to that seen in the control cells transfected with
pEXP6. (b) Absolute values of CAT reporter gene expression obtained using the various promoters in control cells co-transfected with
pEXP6 (i.e. a, C).
range of transfected cell lines. However, as shown in Fig.
1(a), when these plasmids were used to transfect COS-1
cells no stimulation of reporter gene expression was
obtained either with the reovirus type 1 or type 3 S4 gene
irrespective of the presence or absence of the tat gene
construct. This negative result was obtained despite the
fact that in the control experiments for these assays, cotransfection with a tat construct gave an approximately
40-fold stimulation of CAT gene expression when being
driven by the LTR from HIV (Fig. 1b). In parallel with
this work using CAT, experiments were also set up using
p-gal as the reporter gene. The objective of these
experiments was to assess how generally applicable was
the unexpected stimulation of CAT expression seen
following co-transfection with the $4 gene of reovirus
type 3. Fig. 2 shows that by using fl-gal as the marker of
host gene expression the results reported previously
(Giantini & Shatkin, 1989) could be replicated and
extended. Thus co-transfection with the $4 gene from
reovirus type 3 resulted in a greater than sixfold
stimulation of fl-gal activity when a3 protein expression
was being driven by the tat-activated HIV promoter and
more than fivefold stimulation when being driven by the
constitutive SV40 early promoter (Fig. 2). By contrast,
co-transfection with the $4 gene from reovirus type 1
resulted in no significant change in the level of reporter
gene expression compared to that seen in cells transfected
with the fl-gal construct alone. To exclude the possibility
that there was some defect in the reovirus type 1 cDNA
clone used in these experiments, in vitro transcription/
translation studies were done using both reovirus type 1
and type 3 cDNAs. The results (data not shown)
indicated that both types of cDNA directed production
of the same level of full-length o.3 protein.
It was surprising that o.3 protein appeared to have
different effects on the two reporter gene systems used.
This and the apparent failure to replicate the results of
Giantini & Shatkin (1989), who clearly demonstrated
that co-transfection with o.3 protein of type 3 virus
stimulated the CAT reporter gene expression led us to
seek explanations for these apparent anomalies. An
obvious difference between the results obtained using the
CAT and fl-gal reporter systems and between the CAT
expression results of Fig. 1 and those reported by
Giantini & Shatkin (1989) lay in the promoter being used
to drive reporter gene expression. To investigate whether
the stimulation of reporter gene expression by the o-3
protein of reovirus type 3 was dependent on the promoter
driving the reporter gene, co-transfection of reovirus
type 1 and type 3 $4 genes with the CAT gene under the
control of a number of constitutive promoters was
examined. The results (Fig. 3) showed that stimulation of
CAT gene expression by o.3 protein of reovirus type 3
was indeed dependent on the promoter being used to
drive the reporter gene expression, with ascending levels
of stimulation being observed using the thymidine kinase
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Tue, 02 May 2017 11:16:23
Reporter gene stimulation by o-3 protein
1059
KpnI
~ L
TR
~l~i~ $4-3
~
5300bp
5300 bp
..K--StyI824
~
~i--StyI824
/ Poly(A)
Poly(A)
DigestwithKpnI and StyI, isolatefragments
carrying LTR and 5' end of geneand cross-ligate
nI
KpnI
TR
~_~LTR
$4-I
HYB1-3
5300 b p ~ S t y I
~ _ . ~
824
Poly(A)
#
/
~
.... ~ "%'~,$4-3
HYB3-1 '~iiii!
5300bp
~ S ~ y I 824
~
'
Poly(A)
Fig. 4. Constructionof plasmidscarryinghybridversionsof the reovirus$4 gene.
promoter of herpes simplex virus, the LTR of Rous
sarcoma virus and the early promoter of SV40 (Fig. 3 a).
By contrast, confirming the result shown in Fig. 1 (a),
when CAT gene expression was being driven by the
immediate early promoter of CMV, then co-transfection
of the $4 gene of reovirus type 3 did not result in any
stimulation of reporter gene expression (Fig. 3 a). In the
case of the reovirus type 1 $4 gene, it failed to stimulate
expression of CAT from any of the promoters used (Fig.
3 a). The results shown in Fig. 3 (a) have been normalized
to show relative CAT activities in order to allow simple
cross-comparison; however, as shown in Fig. 3 (b) the
absolute levels of reporter gene expression from the
various promoters did differ, with the CMV promoter
giving over 10-fold more activity than any of the others.
Localization of stimulatory doma#7 of the type 3 o-3
protein
The clear difference in activity of the reovirus type 1 and
type 3 $4 genes observed in the co-transfection experiments prompted attempts to localize the region of $4
responsible for this difference in phenotype. To achieve
this, two reovirus type 1/type 3 hybrid $4 constructs
were generated as shown in Fig. 4. In these constructs
approximately one-third of the 3' part of the $4 gene was
switched between the two types of cDNA. When these
hybrid $4 genes were used in co-transfection experiments
with a fl-gal reporter gene construct, the hybrid
(HYB1-3) encoding the carboxy-terminal third of the
reovirus type 3 protein behaved like the native type 3
gene and the reciprocal hybrid (HYB3-1) failed to
stimulate reporter gene expression just like the native
type 1 gene (Fig. 5).
Level of o-3 protein expression in co-transfected cells
Despite the fact that in vitro transcription/translation
studies (data not shown) showed that both the reovirus
type 1 and type 3 $4 cDNAs were expressed with equal
efficiencies it remained possible that the different o-3
proteins had different stabilities in vivo and this accounted for their different effects on reporter gene expression.
To investigate this possibility duplicate monolayers of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Tue, 02 May 2017 11:16:23
1060
P. E. M. Martin and M. A. McCrae
(a)
1
2
3
4
5
6
7
8
9
10
XS--~
•"~
4
~S
crl
"-d 2
~3 ---~
0
HIV-1
HIV-1
HIV-3
Con
HYB1-3 HYB3-1
5'
--
HIV-3
HYB1-3 ~
HYB3 1
(b)
3'
1
2
3 4
5 ,.6
7 8 9 10 ......
Xs--~
~
'
~
:
~: : : , ~
~
Fig. 5, The effect of hybrid $4 genes on reporter gene expression. COS-1
cells were co-transfected with 10 lag of plasmids carrying the various
$4 gene constructs or control plasmid pEXP6, 1 lag of pSV40tat and
1 lag of the reporter gene plasmid carrying the E. coli lacZ (fl-gal) gene
under the control of the early promoter of SV40. Levels of reporter
gene expression have been normalized relative to that seen for the
control cells transfected with pEXP6.
sub-confluent COS-1 cells were transfected with 10 lag of
the various plasmids carrying reovirus $4 gene constructs
and where appropriate 1 lag of pSV40tat. The medium
was replaced with medium lacking methionine 48 h after
transfection, and after starving for 1 h cells were labelled
with a 6 h pulse of [a~S]methionine (100 laCi/ml). At the
end of the pulse period one set of cells was harvested
immediately and the second set chased for a further 6 h
in medium containing 100 times the normal methionine
concentration before being harvested. Immunoprecipitation and subsequent polyacrylamide gel analysis of
these samples were carried out as detailed in Methods.
Samples of labelled reovirus-infected cells for use as
markers on the gels were generated by pulse labelling
(100 ~tCi/ml [aSS]methionine for 20 min) virus-infected
(50 p.f.u, cell) cells at 15 h post-infection. Chase samples
in this case were prepared as for the transfected cells but
with a 2 h chase period.
The results (Fig. 6) show that the a3 protein of type 1
virus is very unstable in transfected cells and consequently accumulates to only very low levels whereas the
a3 protein of reovirus type 3 is stable and therefore
accumulates to much higher levels in the transfected cells
(Fig. 6 a). Using the hybrid $4 constructs it was possible
to show that the stability phenotype of the a3 protein in
]As
~3
Fig. 6. Analysis of the level of a3 protein expression in transfected cells.
(a) Marker lanes are shown for pulse-labelled cells infected with
reovirus type 1 (lane 1), and reovirus type 3 (lane 10) and chase samples
of pulse-labelled cells infected with reovirus type 1 (lane 2) and reovirus
type 3 (lane 9). Immunoprecipitates from pulse-labelled cells transfected
with SV40-1 (lane 3), SV40-3 (lane 5) and HIV-1 (lane 7) (see Fig. 2)
and respective immunoprecipitate of chase samples (lanes 4, 6 and 8).
(b) Lanes 1, 10, 2 and 9 as in (a). Immunoprecipitates from pulselabelled cells transfected with HYB3-1 (lane 3), HYB1-3 (lane 5) (see
Fig. 4) and SV40-1 (lane 7) and SV40-3 (lane 8) (see Fig. 2) and chase
samples from pulse-labelled cells transfected with HYB3-1 (lane 4) and
HYB1-3 (lane 6). The positions of various reovirus proteins are
indicated down each side of the gels.
transfected cells was controlled by the carboxy-terminal
third of the protein (Fig. 6b).
Discussion
This study has shown that the use of reporter genes to
monitor the effects of transfected viral genes on cellular
gene activity can be influenced by the choice of promoter
system used to drive its expression. Thus the initial
attempts to replicate the results of Giantini & Shatkin
(1989) showing that co-transfection with the $4 gene of
reovirus type 3 stimulates the expression of a CAT
reporter gene, driven by the immediate early promoter of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Tue, 02 May 2017 11:16:23
Reporter gene stimulation by o3 protein
CMV, in co-transfected cells were unsuccessful. This
contrasted with a good stimulation by co-transfected
reovirus type 3 e3 gene of the expression of a second
reporter gene, fl-gal. The only difference between the two
reporter gene systems, apart from the reporters themselves, was in the promoter used to drive expression. This
led to a study of the role of the promoter chosen to drive
reporter gene expression on effects seen when cotransfected with reovirus $4 gene constructs. This
revealed that if CAT expression was under the control of
the herpes simplex virus thymidine kinase promoter, the
LTR of Rous sarcoma virus or the early promoter of
SV40 then in each case expression of reovirus type 3 03
protein in co-transfected cells did result in an increase in
reporter gene activity similar to that previously reported
(Giantini & Shatkin, 1989). The only significant difference between these promoter systems and that from
CMV lay in the amount of reporter gene expression, with
the CMV-driven expression being over 10-fold higher
than that given by the other promoters. It is possible that
at these much higher levels of reporter gene expression
the stage in the transcription/translation process affected
by o-3 protein has ceased to be the rate-limiting one and
hence any change that it may induce is masked.
In line with their differential effects on the inhibition of
host macromolecular synthesis in virus-infected cells the
$4 genes of reovirus type 1 and type 3 were found to have
markedly different effects in this transfection system.
Thus no stimulation of either reporter gene expression by
co-transfection with the reovirus type 1 $4 gene was
observed with any of the promoter systems examined. To
begin mapping the amino acids responsible for this
differential phenotype of the two reovirus $4 genes, two
hybrid $4 constructs were generated in which the last
approximately one-third of the o3 protein coding
sequence was exchanged between the two genes. Analysis
of the effects of these hybrid constructs on reporter gene
expression demonstrated that the type 3 virus phenotype
segregated with the 3' third of the $4 gene. To examine
the degree to which the difference in o.3 protein
phenotype resulted from the amount of viral protein
accumulated in transfected cells, a series of immunoprecipitation experiments was done. These showed that
the amount of reovirus type 1 o3 protein accumulating
was much lower than that for reovirus type 3 and that
this difference again co-segregated with the carboxyterminal third of the reovirus type 3 protein. The region
of the $4 gene exchange in the hybrid gene experiments
differs by 27 nucleotides between reovirus types 1 and 3
but these changes result in only six amino acid changes
most of which are conservative in nature. This is
nevertheless a functionally important region of the
protein, as the present study has demonstrated that it
affects the level of ~73 protein accumulated in transfected
1061
cells and hence presumably the different phenotypes
observed with respect to stimulation of co-transfected
reporter genes. It is also of interest that studies using
protease digestion have also located the dsRNA-binding
activity of a3 in this region of the protein (Huismans &
Joklik, 1976; Schiff et al., 1988).
One possible mechanism by which o3 protein has been
suggested to exert its effect on host cell macromolecular
synthesis is by blocking the activation of the dsRNAactivated eIF-2e protein kinase involved in protein
synthesis (Imani & Jacobs, 1988; Giantini & Shatkin,
1989; Bischoff & Samuel, 1989; Samuel & Brody, 1990).
In this hypothesis the known dsRNA-binding activity of
o3 protein (Huismans & Joklik, 1976) is postulated to
sequester dsRNAs produced during viral replication,
thereby preventing them from activating the eIF-2~
protein kinase and hence allowing the preferential
translation of viral mRNAs which predominate at late
times post-infection. In transfected cells a similar
mechanism may operate with the expressed o3 protein
complexing with naturally occurring dsRNA structures
and hence shifting the efficiency of protein translation
so that more of the co-transfected reporter gene is
expressed.
As this work was being completed Seliger et al. (1992)
reported an extension of the earlier studies by this group
on o3 protein-mediated stimulation of reporter gene
expression. This study also examined the effect of o3
protein from different virus serotypes. In common with
the present study they found that reovirus type 1 03
protein did not stimulate reporter gene expression to the
same degree as that from reovirus type 3. However in
marked contrast to the present study their analysis of the
amount of o3 protein made in transfected cells showed it
to be approximately the same for the two viruses (Seliger
et al., 1992). Given the clear differences between the
levels of the two proteins seen in the present study, and
that by hybrid gene analysis we found this to be a
property of the carboxy-terminal third of the protein,
it is difficult to resolve this anomaly. One possible
explanation may lie in the period over which o3 protein
synthesis was analysed. In our study protein labelling
was done over a 6 h period and samples were also
analysed after a 6 h chase period to ascertain both the
steady-state levels of the protein and obtain an indication
of its stability. In the study of Seliger et al. (1992) protein
labelling was done for only 15 min and hence their
analysis may have been biased towards examining rates
of o-3 protein synthesis from the transfected genes rather
than its steady-state level.
The modulation of protein synthesis following infection is obviously of profound importance in the
overall intracellular pathogenesis caused by a virus. This
study has suggested that a particular region of the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Tue, 02 May 2017 11:16:23
1062
P. E. M. Martin and M. A. McCrae
reovirus $4 gene is involved in this regulation. If the
potential for reverse genetics opened up by the observation that reovirus RNA can be made to be infectious
(Roner et al., 1990) can be realized then it should be
possible to dissect this phenomenon in much greater
detail.
This work was supported by grants from the MRC. During the
course of this work P.E.M.M. was a recipient of an MRC Studentship
for training in research methods.
References
ATWATER, J.A., MUNEMITSU, S.M. & SAMUEL, C.E. (1986).
Biosynthesis of reovirus polypeptides. Molecular cDNA cloning and
nucleotide sequence of the reovirus serotype 1 Lang strain s4 mRNA
which encodes the major capsid surface protein g3. Biochemical and
Biophysical Research Communications 136, 183-192.
BISCHOFr, J.R. & SA~rCEL, C.E. (1989). Mechanism of interferon
action: activation of the human P1/eIF-2~ protein kinase by
individual reovirus s-class mRNAs. Virology 172, 106-115.
GIANTINI, M. & SHATKIN,A. J. (1989). Stimulation of chloramphenicol
acetyltransferase mRNA translation by reovirus capsid polypeptide
a3 in co-transfected COS cells. Journal of Virology 63, 2415-2421.
GIANTINI, M., SELIGER, L. S., FURUICHI, Y. & SHATKIN,A. J. (1984).
Reovirus type 3 genome segment $4. Nucleotide sequence of the gene
encoding a major surface protein. Journal of Virology 52, 984-987.
GORMAN, C.M., MOFFAT, L. F. & HOWARD, B. H. (1982). Recombinant genomes which express chloramphenicol acetyltransferase in
mammalian cells. Molecular and Cellular Biology 2, 1044-1051.
H~SMANS, H. & JOKLm, W. K. (1976). Reovirus coded polypeptides in
infected cells: isolation of two native monomeric polypeptides with
affinity for single-stranded and double-stranded RNA respectively.
Virology 70, 411-424.
IMANI, F. & JACOBS,L. J. (1988). Inhibitory activity for the interferoninduced protein kinase is associated with the reovirus serotype 1 a3
protein. Proceedingsof the National Academy of Sciences, U.S.A. 85,
7887-7891.
JOKLIK, W. K. (1985). Recent progress in reovirus research. Annual
Review of Genetics 19, 537-575.
LAEMYILI, U.K. (1970). Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature, London 227,
680-683.
McCgAE, M. A. & JOKLIK, W. K. (1978). The nature of the polypeptide
encoded by each of the 10 double stranded RNA segments of
reovirus type 3. Virology 89, 578-593.
MUSTOE, T. A., RAMIG, R. F., SnARer, A. H. & FIELDS, B. N. (1978).
Genetics ofreovirus. Identification of the dsRNA segments encoding
the polypeptides of the g and a size classes. Virology 89, 594-604.
NEUMANN, J. R., MORENCY, C. A. & RUSSIAN, K. O. (1987). A novel
rapid assay for chloramphenicol acetyltransferase gene expression.
Biotechniques 5, 444-447.
RONER, M. R., SUTPHIN, L. A. & JOKLIK, W. K. (1990). Reovirus RNA
is infectious. Virology 179, 84~852.
SAMBROOK, J., FRITSCH, E.F. & MANIATIS, T. (1989). Molecular
Cloning." A Laboratory Manual, 2nd edn. New York: Cold Spring
Harbor Laboratory.
SAM~L, C.E. & BRODY, M.S. (1990). Biosynthesis of reovirusspecified polypeptides. 2-Aminopurine increases the efficiency of
translation of reovirus sl mRNA but not s4 mRNA in transfected
cells. Virology 176, 106-113.
SCHrVF, L.A., N1BERT, M.L., CO, M. S., BROWN, E.G. & FIELDS,
B. N. (1988). Distinct binding sites for zinc and double-stranded
RNA in the reovirus outer capsid protein a3. Molecular and Cellular
Biology 8, 273-283.
SELIGER, L. S., GIANTINI, M. & SHATKIN, A.J. (1992). Translational
effects and sequence comparisons of the three serotypes of the
reovirus $4 gene. Virology 187, 20~210.
SHARPE, A. H. & FIELDS, I . N. (1982). Reovirus inhibition of cellular
RNA and protein synthesis. Role of the $4 gene. Virology 122,
381-391.
SUN, X. & BALTIMORE,D. (1989). Human immunodeficiency virus tatactivated expression of poliovirus protein 2A inhibits mRNA
translation. Proceedingsof the National Academy of Sciences, U.S.A.
86, 2143-2146.
WATSON, R. J., WEIS, J. H., SALSTROM,J. S. & ENQUIST, L. W. (1982).
Herpes simplex virus type 1 glycoprotein D gene: nucleotide sequence
and expression in E. coll. Science 218, 381-384.
(Received 10 December 1992; Accepted 8 February 1993)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Tue, 02 May 2017 11:16:23