J°urna!.°l~General.Vir°!°gY.
(! 996).'..7.7t..2~.29T2~.36. P[!nted..in.Great.B.[!!a!n
..........................................................................................................................
Characterization of RNA synthesis during a one-step growth
curve and of the replication mechanism of bovine viral
diarrhoea virus
Y u n h a o G o n g , 1 Rachel T r o w b r i d g e , 1 T h o m a s B. M a c n a u g h t o n , ~ E d w i n G. W e s t a w a y , 1
Anthony
D. S h a n n o n 2 a n d Eric J. G o w a n s ~
Sir Albert Sakzewski Virus Research Centre, Royal Children's Hospital, Brisbane 4029, Australia
2Elizabeth Macarthur Agricultural Institute, Camden, New South Wales 2570, Australia
The noncytopathic Australian bovine viral diarrhoea
virus (BVDV) Trangie isolate was used to establish a
one-step growth curve and to investigate previously
uncharacterized aspects of pestivirus replication.
The eclipse phase was found to be approximately 8
to 10 h and the first appearance of viral antigen
assayed by immunofiuorescence occurred around
6 h post-infection (p.i.). Both positive- and negative-sense virus RNAs were first detected at 4 h p.i.
by Northern blot hybridization using strand-specific
RNA probes generated by in vitro transcription from
cDNA cloned from the NS3 region. The ratio of
positive- to negative-sense virus RNA changed from
2:1 at 4 h p.i. to 10:1 at 12 h p.i. and thereafter.
The kinetics of synthesis showed that the rate of
synthesis of positive-strand viral RNA increased
rapidly from 6 h p.i., whereas the rate of synthesis of
Introduction
Bovine viral diarrhoea virus (BVDV) is a member of the
genus Pestivirus in the family Flaviviridae (Francki eta/., I991).
Pestiviruses are a group of small, enveloped, positive-stranded
RNA viruses which are major pathogens of domestic animals
worldwide (reviewed by Baker, 1987; Brownlie, 1991; Collett,
1992; Moennig & Plagemann, 1992; Thiel et al., 1996). BVDV
is divided into cytopathic (cp) and noncytopathic (ncp)
biotypes depending on the production of a cytopathic effect
(CPE) in cell culture (Donis & Dubovi, 1987). The genome is
approximately 12"5 kb and varies in size between cp and ncp
Authorfor correspondence:EricJ. Gowans.
Fax +61 7 3253 1401. [email protected]
The nucleotidesequencepresentedin this paper has been submitted
to GenBankand assignedaccessionnumber U53408.
OOO1-4063 © 1996 SGM
the negative-strand remained constant. The copy
number of genomic RNA determined by Northern
blot hybridization analysis was estimated to be
1.5 x 104 copies per cell, 16 to 24 h p.i. Viral RNA
species that were thought to represent replicative
intermediate (RI) and replicative forms (RF) were
detected after electrophoretic separation by ureaPAGE. Confirmation of the identity of the RI and RF
was obtained using LiCI precipitation and RNase A
digestion of [3H]uridine-labelled RNA. Pulse-chase
labelling of BVDV-infected cells was consistent with
synthesis of nascent BVDV RNA through an RI
derived by strand displacement from an RF template
and thus the synthesis of BVDV RNA is likely to be
similar to the model proposed for fiavivirus replication.
isolates due to the insertion of cellular sequences or gene
rearrangements (Meyers et al., 1991, 1992; Qi et al., 1992;
Greiser-Wilke et aI., 1993; Ridpath & Bolin, 1995). The
genome contains one large open reading frame (ORF) which
encodes a polyprotein of about 4000 amino acids (Collett et al.,
1988a; Renard et al., 1987; Deng & Brock, I992; Ridpath &
Bolin, I995). A 5' untranslated region (5'UTR) of about 385
nucleotides (nt), containing an internal ribosome entry site
(IRES) (Deng & Brock, 1993 ; Poole et al., 1995), and a 3'UTR
of about 220 nt lacking a poly(A)+ tail (Collett et al., 1988a)
flank the long ORF. Translation of pestiviruses is considered to
occur in a cap-independent manner and is initiated by binding
of the ribosomes to the IRES (Poole et al., 1995).
Entry of BVDV into permissive cells is believed to occur by
receptor-mediated endocytosis and is pH-dependent (Boulanger et aI., 1992; Xue et aL, 1993; Flores & Donis, 1995).
Although the cellular receptor for BVDV has not been
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 22:34:52
_~72c.
characterized, virus attachment is assumed to be mediated b y
the envelope glycoproteins E2 (gp53) a n d / o r E0 (gp48) (Donis
et al., 1988). The molecular features of BVDV R N A replication
have n o t b e e n described (reviewed b y Collett et al., 1989;
Collett 1992; M o e n n i g & Plagemann, 1992). Nevertheless, the
replication complex probably comprises an R N A - d e p e n d e n t
R N A polymerase (RdRP), a helicase, virus-encoded cofactors
and cellular proteins. NS3 possesses helicase, serine protease
and NTPase activities (Wiskerchen & Collett, 1991; Tamura et
at., 1993; W a r r e n e r & Collett, 1995) and NS5 is a candidate for
the RdRp because it contains a characteristic G D D motif
(Collett et al., 1991).
Hepatitis C virus (HCV) is an important h u m a n p a t h o g e n
currently classified in a separate genus in the family Flaviviridae
(Francki et at., 199I). It has a similar genomic organization and
polyprotein processing strategy to that of the pestiviruses and
shares h o m o l o g y and conserved motifs with the 5'UTR, NS3
and NS5 (Miller & Purcell, 1990; H o u g h t o n et al., 1991;
Howard, 1992). Both BVDV and H C V have the ability to cause
chronic and persistent infections. It is possible that H C V and
BVDV m a y have a similar replication mechanism. However,
difficulties arise in s t u d y i n g H C V replication due to the low
level viraemia in infected individuals and because the virus
cannot be g r o w n reproducibly in cell culture. W e believe that
the s t u d y of the BVDV replication mechanism m a y provide a
g o o d model for H C V replication and persistent infection.
Thus, the aim of this study was to develop a one-step
g r o w t h curve for BVDV and to examine the appearance of
different forms of BVDV R N A as a first step in the construction
of a model for BVDV R N A replication.
Methods
• Cells and virus. Pestivirus-negative bovine turbinate (BTu) cells
and lamb testis (LT) cells were obtained from the Elizabeth Macarthur
Agricultural Institute (EMAI), Sydney and were cultured in Dulbecco's
modified Eagle's medium (DMEM), supplemented with 10% pestivirusflee bovine serum (DMEM-10) raised from pestivirus-free animals at
EMAI, 100 U/ml penicillin and 100 ~tg/ml streptomycin (Gibco-BRL).
For cell maintenance, DMEM containing 2% pestivirus-free bovine
serum (DMEM-2) was used. The ncp BVDV Trangie isolate was also
obtained from EMAI and was biologically cloned three times by limiting
dilution.
• Virus propagation and titration. The virus was propagated at
37 °C in either BTu or LT cells at an m.o.i, of 0"1 to 0"2 and harvested at
day 6 post-infection (p.i.). The culture fluid was removed, the infected
cells were frozen and thawed once and, after low speed centrifugation,
the clarified supematants were pooled and stored at -- 70 °C.
Virus was titrated by immunoperoxidase staining of infected cells.
Basically, 10-fold serial virus dilutions were added to confluent BTu cells
in a 96-well plate (Nundon). After i h virus adsorption at 37 °C, the
medium was removed and the cells were covered with 1'5 % hydroxylpropyl methylcellulose (HPMC) in DMEM-2, and then incubated at
37 °C for 3 or 4 days. The HPMC was then removed and the cells were
fixed with 5% formalin for 10 min. Immunoperoxidase staining for
plaques was then performed using monoclonal antibodies (MAbs) to the
~73C
NS3 antigen of BVDV as described by Richards (1991). Virus stocks
prepared and titrated in this way were shown to have a titre of
3-2 x 107 p.f.u./ml from a static culture in a I50cm ~ flask and
1"1 x 108 p.f.u./ml from a 900 cm2 roller bottle.
• Immunofluorescence staining (IF). BTu cells were grown on
coverslips in a 24-well plate (Nunclon) prior to infection with BVDV. At
selected time-points, the coverslips were air dried and the cells fixed with
5 % formalin as above. The coverslips were incubated overnight with the
anti-NS3 MAbs at 4 °C, washed, then incubated with FITC-conjugated
rabbit anti-mouse antibody (Dako) at 37 °C for 90 min. After a final wash,
the coverslips were mounted in Tris-PBS--glycerol pH 8"6 and examined
for specific staining using a fluorescence microscope.
• One-step growth curve. Confluent BTu cells in a 24-weU plate
were washed with PBS, BVDV Trangie at an m.o.i, of 20 was added and
the plates were then incubated at 4 °C for I h. After washing, i ml
DMEM-2 at 37 °C was added and samples were harvested at various
time-points as described in Results. At each time-point, the supematant
fluid was stored in aliquots at - 70 °C and used to titrate the extracellular
virus; the infected cells were washed, trypsinized, resuspended in i ml of
DMEM-2, frozen and thawed once, pelleted to remove the cell debris,
and the supernatant was used to titrate the cell-associated virus. At each
time-point, immunofluorescence staining was performed as above to
define the first appearance of viral antigen.
• RNA extraction. Total RNA was extracted from BVDV-infected
cells and mock-infected cells as described by Chomczynski & Sacchi
(1987). The RNA pellet was dissolved in DEPC-treated water, quantified
by UV spectrophotometry and stored at --70 °C. The quality of the
RNA was determined by electrophoresis of I I~gRNA on a formaldehyde
agarose gel (Sambrook et al., 1989).
• Generation of strand-specific RNA probes. Plasmid pYG80-7
was linearized by HindIII and RNA transcribed in the presence of
[c¢-a2P]UTP(Bresatec) by T7 RNA polymerase to generate the negativesense RNA probe. Similarly, pYGS0-7 was linearized with XbaI and the
positive-sense RNA probe synthesized by T3 RNA polymerase. The
transcription reactions were performed using the Riboprobe kit (Promega)
and the DNA template then digested with RQ1 DNase. The length of the
probe was checked by agarose gel electrophoresis. Only probes which
were full length, i.e. I-2 kb, were used in Northern blot hybridization
reactions which were performed essentially as described (Sambrook et al.,
1989).
• Radiolabelling of viral RNA. BVDV-infected (m.o.i. of 20) and
mock-infected BTu cells were incubated in medium containing 10 lig/ml
actinomycin D (AMD) for 2 h and the virus RNA was radiolabelled by
the addition of 20 ~tCi/ml [5,6-aH]uridine (Amersham) in DMEM-2 at
37 °C for 5 h. Total RNA was then extracted by the guanidinium
isothiocyanate method and analysed by PAGE in a 7 M-urea--3% gel (see
below). The LiC1fractionation and RNase A digestion were carried out as
described by Cleaves et al. (I98I).
• Gel electrophoresis of RNA. PAGE of the RNA was performed in
a 7 M-urea--3% gel essentially as described by Chu & Westaway (1985).
Briefly, the gel comprised 7 M-urea, 3 % acrylamide-bisacrylamide (I9:1),
I x TBE (89 mM-Tris, 89 mM-boric acid, 2"5 mM-EDTA), 0"I% ammonium persulphate and 0"063 % TEMED. Prior to electrophoresis, the gel
was rinsed with DEPC-treated water and pre-run at 100 V for 15 min.
The RNA samples were mixed with 2 vols of 1'5 x loading buffer
(10"5 M-urea, 1"5 × TBE) and loaded onto the gel. Electrophoresis was
carried out in 1 x TBE buffer at 80 V for approximately 3 h. The gel was
then fixed in 10 % methanol-10 % acetic acid and fluorography performed
using ENaHANCE (Du Pont).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 22:34:52
•
Pulse-chase labelling of viral RNA. BTu cells were infected with
IF (percentage positive cells)
BVDV at an m.o.i, of 20. At 11 h p.i. the infected cells were treated with
10 p.g/ml AMD for 1 h. The medium was then removed and the cells
were incubated in DMEM-2 with 100 ,Ci/ml [5,6-3H]uridine and
1 ~tg/ml AMD for 30 min. Fresh DMEM-2 containing 1 p.g/ml AMD
was then added and the cells were incubated at 37 °C and sampled at
intervals up to a maximum of 6 h.
100
0 0 40 100100100 100 100100
' ' ' ' ' ' ' ' '
75
,q..
Results
x
9
One-step growth curve
To establish a one-step growth curve for ncp BVDV in BTu
cells, the cells were infected with BVDV Trangie at an m.o.i, of
20 p.f.u, per cell. Virus adsorption was performed at 4 °C to
permit virus binding, but not internalization. At various times
p.i., the extracellular virus and cell-associated virus titres were
determined by titration. Virus antigen expression was monitored by IF staining in situ and the proportion of infected cells
was confirmed by an infectious centre assay performed 2 h p.i.
(data not shown).
The results (Fig. 1) showed that little infectious extracellular
and cell-associated virus was detected between 0 to 10 h;
thereafter the virus titre showed a rapid increase between 10
and 16 h and then reached a plateau between 16 and 18 h.
Further time-points were not examined. Thus, in this experiment, the eclipse phase was approximately 10 h and the
complete replication cycle appeared to be 12-14 h. The level
of extracellular virus exceeded that of cell-associated virus by
a factor of 6 at 16 h p.i.; this is consistent with a previous study
by Nuttall (1980) and with data derived for enveloped RNA
viruses in general (Burleson et al., 1992).
The coverslips harvested at each time-point were assayed
by IF to detect the first appearance of viral antigen. The results
showed that approximately 40 % of infected cells were positive
at 6 h p.i. and 100 % were positive at 8 h p.i. (data not shown).
It is probable that the vast majority of cells were infected at
time O, but the level of viral antigen failed to reach detectable
levels in all cells until 8 h p.i. This interpretation was confirmed
by an infectious centre assay which showed, on two different
occasions, that 70% and 100% of cells, respectively, were
infected 2 h p.i. (data not shown).
The effect of AMD on BVDV replication was also
examined. After adsorption the virus inoculum was removed
and the medium was replaced with DMEM-2 containing
different concentrations of AMD. The results (data not shown)
showed that I ~tg/ml and 10 I~g/ml AMD reduced the virus
titre to 64% and 50%, respectively, whereas cellular RNA
synthesis was inhibited by 97"8% and 99% with I ~tg/ml and
10 ,g/ml, respectively. Thus AMD had no significant effect on
BVDV replication.
Characterization of viral RNA synthesis
Although the synthesis of flavivirus RNA is well characterized, the same is not true for the pestiviruses. Thus, in order
to examine the kinetics of BVDV RNA synthesis, we employed
50
25 ~ . . - . . 0 . . - . . 0 ,
2 4 6
,
8 10 12 14 16 18 20
Time (h)
Fig. 4. One-step growth curve of BVDVTrangie isolate in BTu cells at an
m.o.i, of 20. The extracellular (17) and cell-associatedvirus (0) were
titrated by immunoperoxidasestaining and viralantigen was detected by
IF.
Northern blot hybridization using strand-specific RNA probes
prepared by in vitro transcription.
The template used for in vitro transcription was cloned from
the Trangie isolate of BVDV after RT-PCR amplification. The
primers [sense, 5' CTGGAGACTGGCTGGGCTTACACAC
(position 55 76-5600); antisense, 5' AGCATTTGTAGCCACGATTA (position 6762-6781)] were designed to target the
conserved NS3 region according to the published sequence of
the NADL cp isolate (Collett et al., 1988a). As the exact
boundaries of NS3 have not been determined, the region was
estimated based on published data and hydrophilicity profiles
(Collett et al., 1988b, 1991; Rumenapf et al., 1993). The
amplified fragment (1206 nt) was confirmed by agarose gel
electrophoresis, cloned into pBluescript and the specificity
confirmed by mapping internal restriction sites and nucleotide
sequencing. The nuceotide sequence was obtained from at least
two clones and sequenced from both strands. The consensus
sequence (GenBank accession number: U53408) is 84%
identical to that of the NADL cp isolate and 83 % to that of the
SD-1 ncp isolate, whereas the predicted amino acid sequence is
97% similar to both. This indicates that NS3 is highly
conserved within BVDV isolates. One of these clones,
pYG80-7, was used to generate RNA probes.
The strand specificity of the probes was confirmed by
hybridization of positive- and negative-sense RNA probes
with RNA extracted from concentrated virus and infected cells,
respectively (data not shown). Only the negative-sense probe
reacted with the RNA from the concentrated virus whereas
both positive- and negative-sense probes hybridized with
RNA from the infected (but not uninfected) cells. To examine
BVDV RNA in infected cells, the total RNA harvested from
each time-point during a one-step growth curve was denatured
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 22:34:52
!7!
M
0
2
4
6
Time (h)
8 10 12 14 16
5000
18
(a)
4oo0
3000
(b)
L) 2000
1000
Fig. 2. Detection of positive- (a) and negative- (b) sense BVDV RNA from
infected cells during the course of the one-step growth curve by Northern
blot hybridization using 32p-labelled strand-specific RNA probes generated
from the NS3 clone pY(380-7. The total RNA extracted at different timepoints was denatured by glyoxaI-DMSO (Sambrook et al., 1989) : 2 I~g
RNA was loaded in each lane in (a) and 5 Ilg RNA per lane in (b). The
exposure period was 1 9 h in (a) and 50 h in (b). Lane M, mock-infected
cellular RNA.
by glyoxal-DMSO, separated by agarose gel electrophoresis,
then transferred to a nylon membrane. The results of
hybridization are shown in Fig. 2, and indicate that both
positive- and negative-sense BVDV RNA were first detected
at 4 h p.i. The earlier time-points were negative, showing that
the levels of input virus RNA were undetected. Maximum
levels of virus RNA were detected at 16 h p.i. Furthermore,
only one band was detected, indicating that no subgenomic
RNA encoding NS3 was produced during the replication cycle.
Subgenomic RNA species were also undetected with RNA
probes (a gift from Marc Collett) designed to detect the
envelope gene region (data not shown).
To examine the kinetics of positive- and negative-strand
RNA synthesis after hybridization, the region of the membrane
containing the BVDV RNA band and the bound probes was
excised and counted by liquid scintillation. This experiment
was possible because the probes were derived from the same
plasmid and were radiolabelled to the same specific activity;
the RNA targeted by the positive- and negative-probes was
also derived from the same preparation. The results (Fig. 3)
show that positive-strand RNA synthesis increased rapidly
from 6 h p.i. whereas negative-strand RNA synthesis failed to
show any significant increase. These data were reproducible.
Since negative-strand viral RNA is thought to function as a
template for positive-strand RNA synthesis, these results
imply a recycling role for negative-strand RNA in BVDV RNA
replication.
Quantification of BVDV RNA in infected cells and the
ratio of positive- to negative-strand BVDV RNA
It has often been suggested that members of the F/aviviridae
replicate poorly and that the level of virus in infected cells is
low. However, few data are available to confirm this.
Consequently, we examined the level of BVDV RNA in
infected cells. To do this, the intensity of the genomic RNA
_~73;
0
0
5
10
Time (h)
15
20
Fig. 3. Kinetics of positive- (R) and negative-strand ( ~ ) viral RNA
synthesis in BVDV-infected cells detected by Northern blot hybridization.
The membrane containing the BVDV RNA band was excised and the
bound 32p-labelled probe was counted by liquid scintillation.
signal generated from Northern blot hybridization analysis of
total cellular RNA, as determined by the level of 32p bound to
the membrane, was compared with the signal generated after
hybridization of the negative-sense probe to a known amount
of the DNA standard (pYGS0-7). Since, by calculation, 10 ng
of pYGS0-7 DNA contains 2"16 x 103 molecules, it was
estimated that BVDV-infected cells contained approximately
1"5 x i04 copies of genomic RNA per cell, 16 to 24 h p.i. (data
not shown). This reflects the low level of replication in infected
cells when compared with poliovirus, which produces 4 x i0 s
genome copies per cell (reviewed by Rueckert, 1990).
We also examined the ratio of positive- to negative-strand
BVDV RNA at each time-point in the one-step growth curve
by comparing the level of a2p bound after Northern blot
hybridization (Fig. 3). The probe was shown to be in excess in
similar experiments as, after hybridization, the probe mixture
was re-used successfully on a second blot. The ratio increased
rapidly from around 2 : 1 at 4 h p.i. to around 10:1 at 12 h p.i.
and thereafter remained relatively stable until 18 h p.i. This
result is consistent with that of flavivirus RNA synthesis in
which the ratio of positive- to negative-sense RNA was 10:1
at the peak of dengue 2 virus infection (Cleaves et al., I981).
Detection and characterization of replicative
intermediate and replicative forms of BVDV RNA
Replicative intermediate (RI) and replicative forms (RF) of
picornavirus and flavivirus RNA are produced during virus
replication (Richards & Ehrenfeld, 1990; Cleaves et al., 1981;
Chu & Westaway, 1985). However, only an RNA species
which was resistant to RNase digestion in BVDV-infected cells
was detected in one study (Purchio et al., 1983) and the
putative RF and RI RNA species likely to be present in
pestivirus-infected cells have remained undetected (reviewed
by Collett et al., I989; Collett, 1992). To detect the putative RI
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 22:34:52
M
UD
D
LiC!
RNaseA
P
HS LS
S
Chase time
M
UD M
0'
15'
30'
lh
2h
4h
6h
RI
SS
RF --~
Fig. 4
Fig. 5
Fig. 4. Gel electrophoresis and fluorography of [3H]uridine-labelled intracellular RNA purified from BVDV-infected and mockinfected BTu cells, 48 h p.i. In some experiments the RNA was then treated with LiCI or RNase A. RI, RF and ssRNA were
separated by PAGE in a 7 M-urea3% gel. Lane M, mock-infected BTu RNA; lane UD, undenatured BVDV-infected BTu total cell
RNA; lane D, as lane UD except that the sample was denatured in the urea loading buffer at 75 °C before loading; lane S, LiCIsoluble fraction (mainly dsRNA) ; lane P, LiCI-insoluble fraction (mainly ssRNA and partial dsRNA) ; lane HS, RNase A digestion
of total cell RNA in high salt (2 x SSC) ; lane LS, RNase A digestion in low salt (0-1 x SSC). The equivalent of 5 I~g total RNA
was loaded in each lane.
Fig. 5. Pulse-chase labelling of BVDV-infected BTu cells. Cells were infected with an m.o.i, of 20 and, at 11 h p.i., the cells
were labelled with 100 I~Ci/ml [SH]uridine for 30 min after AND treatment for 1 h. The medium was then replaced with DMEM2 + 1 Mg/ml AMD and the [3H]uridine chased for the indicated times. The RNA was extracted and electrophoresed in a 7 Murea-3% gel. RNA (5 l~g) was loaded in each lane. Lane UD, an undenatured RNA aliquot from BVDV-infected cells (same RNA
aliquot as in lane UD in Fig. 4) ; lane M, mock-infected BTu cellular RNA at the end of pulse period (chase 0 min), lanes 0 - 6 h
represent the duration of the chase. The exposure period was 2 months.
and RF of BVDV RNA and to investigate the repIication
mechanism, we employed the strategies described by Chu &
Westaway (1985) for flaviviruses. BVDV-infected BTu cells
were treated with I0 gg/ml AMD for 2 h at 48 h p.i., and then
metabolically labelled with [3H]uridine. The total RNA from
infected cells and mock-infected cells was then extracted and
examined by PAGE in a 7 M-urea-3 % gel. This system only
denatures ssRNA (Chu & Westaway, 1985).
The results (Fig. 4) showed that synthesis of the ribosomal
RNA species was completely inhibited by AMD as shown in
the mock-infected cellular RNA (lane M). Three virus-specific
RNA species were readily resolved (lane UD). Although two
bands were shadowed by a low level of label incorporated into
cellular RNA that was resistant to AMD treatment (presumably
45S pre-ribosomal RNA and aggregated cell RNA), the gel
profile was clear and quite different to the profile of the mockinfected cells which were treated in exactly the same way as
the infected cells. The profile was similar to that of Kunjin virus
RNA (Chu & Westaway, i985) in that a proportion of the viral
RNA failed to enter the gel. This band was shown to contain
virus-specific RNA following denaturation in situ after gel
electrophoresis and subsequent Northern blot hybridization
(data not shown). Consequently, the bands were interpreted to
represent RI (upper band), the intermediate band to represent
genomic length ssRNA and the rapidly migrating band (lower
band) to represent RF. This latter band was not detected after
the RNA was denatured at 75 °C in 7 M-urea loading buffer for
3 min prior to electrophoresis (lane D), concomitantly with an
increase in the level of ssRNA. Thus, the lower band comprises
dsRNA and is Iikely to represent RF. The increase in the
intensity of the ssRNA band may also have resulted from
conversion of the RI species, although we are unable to explain
why the RI was not fully denatured in this experiment. We
assume that the 7 M-urea treatment is inadequate for complete
denaturation. Overall, the increase in the ssRNA band (lane D)
appears greater than the total labelled RNA prior to denaturation (lane UD) and we assume that this results from
quenching of the counts in the RI band of the UD lane.
The likely identity of the RI and RF species in infected cells
was confirmed by 2 M-LiC1 fractionation and RNase A
digestion. When analysed by urea-PAGE, the LiCl-soluble
fraction, predicted to contain predominantly dsRNA, showed
a similar electrophoretic mobility to the rapidly migrating band
described above as RF (Fig. 4, lane S). This species was also
found to be resistant to digestion with I gg/ml RNase A in
high salt (2 x SSC, lane HS), but sensitive to digestion in low
salt (0"1 x SSC, lane LS). Thus, this band shows the characteristics of dsRNA and can be identified as RF. The LiC1
precipitate was found to contain ssRNA and partially dsRNA
(RI) (lane P) that were sensitive to RNase A digestion in high
and low salt (lanes HS and LS, respectively), as expected for
ssRNA or partially dsRNA. This presents additional evidence
that the upper and intermediate bands are most likely to
represent RI and genomic ssRNA, respectively.
The role of RF and RI in virus replication
A s described above, RI and RF can be d e t e c t e d in B V D V infected
cells, b u t
their
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 22:34:52
precise roles in
pestivirus
RNA
'.73!
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii iii ii ii iiiiiiiiiiiiii!i¸
replication remain unknown. The RI and RF species of
picornaviruses and flaviviruses were previously thought to
serve different functions during virus replication. The RI
species in poliovirus was considered to function as a precursor
to ssRNA and contains four to eight nascent strands per
molecule (Baltimore & Girard, 1966); poliovirus RF was
thought to represent a by-product or 'dead end' product
unable to support further replication (reviewed by Richards &
Ehrenfeld, 1990). In contrast, the RI in Kunjin virus contains
only one nascent strand per template (Westaway, 1987) and
RF is thought to be the template for the production of nascent
RNA which is synthesized by strand displacement to form an
RI with a single growing strand (Chu & Westaway, I985).
Recently, however, the replication of poliovirus RNA was
proposed to be initiated at the end of the dsRNA (RF) (Andino
et al., 1993; Harris eta[., I994). Thus, the template for the
production of nascent poliovirus and flavivirus RNA is thought
to be the RF as proposed for flaviviruses by Chu & Westaway
(1985). To investigate how the RI and RF species of BVDV
may contribute to virus replication, pulse-chase labelling of
infected cells with [aHluridine was performed 11 h p.i.. After
separation of extracted RNA by PAGE in a 7 M-urea--3% gel,
a virus-specific band migrating faster than 28S ribosomal RNA
was detected that represents RF (Fig. 5). A second band which
failed to enter the gel was also detected that represents the RI
and possibly aggregated radiolabelled cell RNA (despite the
presence of AMD). Thus, only RI and RF of virus RNA were
detected at the end of the 30 min pulse labelling period (Fig. 5,
chase 0 rain). However, due to the inability to remove all
[dH]uridine from residual intracellular pools after the pulse, the
incorporation of label into the RI and RF species increased
slowly during the chase period. Nevertheless, the data clearly
show a slow accumulation of genomic length virus ssRNA
from i h onwards. This is in stark contrast to the rapid
appearance of ssRNA in a similar study of Kunjin virus RNA
synthesis that showed the appearance of ssRNA after only a
10 min chase period (Chu & Westaway, 1985). These data
suggest that BVDV RNA replication proceeds much more
slowly than that of Kunjin virus RNA. Clearly, incorporation
of [dH]uridine into the RF during the pulse period suggests that
the BVDV RF is involved in viral RNA replication. However,
the exact role of the RF and the RI in BVDV RNA replication
is still unclear, as the high stability of the RF, although
consistent with the model for flavivirus replication proposed
by Chu & Westaway (1985), is also consistent with other
interpretations (see below).
Discussion
The main aim of this study was to examine the previously
uncharacterized replication mechanism of BVDV as a model
for HCV replication. We have chosen a ncp BVDV isolate for
this study because it may resemble HCV more closely.
Consequently, we have developed a one-step growth curve for
_~73L
BVDV in BTu cells using an m.o.i, of 20 and shown that viral
RNA and antigen were first detected at 4 h and 6 h p.i.,
respectively. The eclipse phase was found to be approximately
10 h. Our data also showed that the titre of cell-flee virus
exceeded that of cell-associated virus by a factor of 6. A
previously reported one-step growth curve for ncp BVDV
used a lower m.o.i. (0"2 TCIDs0 per cell; Nuttall, 1980) and
showed the eclipse phase to be 6 to 8 h p.i. However, all cells
may not be infected simultaneously at the low m.o.i. The
eclipse phase was reported to be 8 to 12 h in a further set of
experiments (Johnson et aI., 1990) using an m.o.i, of 5"2 TCIDs0
per cell, but no cell-associated virus titre was reported. The
results of Northern blot hybridization (Fig. 2) showed that the
input RNA was undetected and thus the signal represented
progeny viral RNA following infection. Virus was released at
10 h p.i., presumably when positive-strand virus RNA and
proteins had accumulated sufficiently for virus packaging to
OCCUr.
Examination of BVDV RNA within infected cells showed
that each cell contained approximately I"5 x I 0 4 genome
copies per cell and the ratio of positive to negative strands was
10:1. This latter figure is in close agreement with the ratio
reported for flavivirus-infected cells (Cleaves et al., 1981).
However, the ratio of positive to negative strand RNA may
differ in different cells or even between different BVDV
isolates (Deng & Brock, 1993).
This study has also described, for the first time, the
appearance of RI and RF of BVDV RNA. Based on studies of
Kunjin virus, Chu & Westaway (1985) proposed a replication
model for flaviviruses in which the RF functions as a recycling
template to permit asymmetric replication of virus RNA in a
semiconservative manner. In this model, incoming viral RNA
is copied into negative strand RNA to form RF. Subsequent
RNA synthesis using the RF as template is accompanied by
strand displacement (forming RI), and the displaced positivestrand RNA is released upon completion of the nascent strand.
The completed nascent positive-strand and the negativestrand in the template constitute the RF which is recycled.
Consequently, the recycling of the RF mainly involves
recycling of the negative strand. Our data shown in Figs 2, 3,
4 and 5 are consistent with the Kunjin virus model. As shown
in Fig. 3, the levels of positive-strand RNA increased rapidly in
infected cells from 6 h p.i., whereas the levels of negative
strand RNA remained virtually constant through the course of
the one-step growth curve. These relative rates of increase in
accumulation of positive- and negative-strand RNA are
consistent with a recycling role for negative strand in the RF
during semiconservative virus RNA replication and our data in
Fig. 5 are also consistent with this model. However, additional
studies are necessary to determine unequivocally the exact
roles of BVDV RI and RF, as the high stability of the
radiolabelled RF is also consistent with a conservative model of
virus RNA replication. Although the nature of the RI has not
been examined in detail in this study, our preliminary data of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 22:34:52
iiiii!!iiiiiiiiii iiii iii iii i iii
RNase resistance (data not shown) suggest that the BVDV RI
may contain more than one nascent strand. O n g o i n g studies
are designed to clarify these aspects of the model.
To our knowledge, this is the first description of several
molecular features of pestivirus replication; we have examined
the kinetics of R N A synthesis, the ratio of positive- to
negative-strand RNA, and the appearance of viral R N A species
in infected cells that showed characteristic properties of RI and
RF. RI and RF have long been suspected to exist in pestivirusinfected cells; Purchio eta]. (1983) identified a virus-specific
R N A species in BVDV-infected bovine cells that was resistant
to RNase A treatment and showed similar features to RF with
an apparent size of 8 kb. Furthermore, the published N A D L
BVDV sequence was based on c D N A synthesized from the
LiCl-soluble fraction which mainly comprised dsRNA (Collett
et al., I988a). W e have now provided evidence of RI and RF in
BVDV-infected cells that can be separated b y gel electrophoresis.
This work was supported by a grant from the National Health & Medical
Research Council of Australia. Mr Yunhao Gong is supported by an
Overseas Postgraduate Research Scholarship, University of Queensland
Postgraduate Research Scholarship and a Fellowship from the Royal
Children's Hospital Foundation. We thank Dr Marc Collett for the gift of
the structural gene probes.
References
Andino, R., Rieckhof, G. E., Achacoso, P. L. & Baltimore, D. (1993).
Poliovirus RNA synthesis utilizes an RNP complex formed around the 5'end of viral RNA. EMBO Journal 12, 3587-3598.
Baker, J. C. (1987). Bovine viral diarrhoea virus: a review. Journal of the
American Veterinary Medical Association I90, 1449-1458.
Baltimore, D. & Girard, M. (1966). An intermediate in the synthesis of
poliovirus RNA. Proceedings of the National Academy of Sciences, USA 56,
741-748.
Boulanger, D., Dubuisson, J. & Pastoret, P.-P. (1992)o Studies on the
neutralization mechanism of bovine viral diarrhoea virus. Proceedings of
the Second Symposium on Pestiviruses, pp. 157-161. Veyrier-du-Lac:
European Society for Veterinary Virology.
Brownlie, J. (1991). The pathways for bovine virus diarrhoea virus
biotypes in the pathogenesis of disease. Archives of Virology, Supplementum 3, 79-96.
Burleson, F. G, Chambers, T. M. & Wiedbrauk, D. k. (1992). Virology:
A Laboratory Manual, pp. 100-103. San Diego: Academic Press.
Chomczynski, P. & Sacchi, N. (1987). Single-step method of RNA
isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.
Analytical Biochemistry 162, I56--159.
Chu, P. W. G. & Westaway, E. G. (1985). Replication strategy of Kunjin
virus : evidence for recycling role of replicative form RNA as template in
semiconservative and asymmetric replication. Virology 140, 68-79.
Cleaves, G., Ryan, T. E. & Schlesinger, R. W. (1981). Identification and
characterization of type 2 dengue virus replicative intermediate and
rep[icative form RNAs. Virology 111, 73-85.
Collett, M. S. (1992). Molecular genetics of pestiviruses. Comparative
Immunology, Microbiology and Infectious Diseases 15, 145-I54.
iiiii
i i i i i i i i i!i i i i i
Collett, M.S., Larson, R., Gold, C., Strick, D., Anderson, D.K. &
Purchio, A. F. (1988 a). Molecular cloning and nucleotide sequence of
the pestivirus bovine viral diarrhoea virus. Virology 165, I91-199.
Collett, M. S., Anderson, D. K. & Retzel, E. (1988b). Comparisons of
the pestivirus bovine viral diarrhoea virus with members of the
Flaviviridae. Journal of General Virology 69, 2637-2643.
Collett, M. S., Moennig, V. & Horzinek, M. C. (1989). Recent advances
in pestivirus research. ]ournaI of General Virology 70, 253-266.
Collett, M.S., Wiskerchen, M., Welniak, E. & Belzer, S. K. (1991).
Bovine viral diarrhoea virus genomic organization, Archives of Virology,
Supplementum 3, 19-Z7.
Deng, R. & Brock, K.V. (1992). Molecular cloning and nucleotide
sequence of a pestivirus genome, noncytopathic bovine viral diarrhoea
virus strain SD-1. Virology 191, 867-879.
Deng, R. & Brock, K.V. (1993). 5' and 3' untranslated regions of
pestivirus genome: primary and secondary structure analyses. Nucleic
Acids Research 21, 1949-1957.
Donis, R.O. & Dubovi, E.J. (1987). Differences in virus-induced
polypeptides in cells infected by cytopathic and noncytopathic biotypes
of bovine virus diarrhoea-mucosal disease virus. Virology 158, 168-I73.
Donis, R. 0., Corapi, W. & Dubovi, E. J. (1988). Neutralizing monoclonal antibodies to bovine viral diarrhoea virus bind to the 56K to 58K
glycoprotein. Journal of General Virology 69, 77-86.
Flores, E. F. & Denis, R. O. (1995). Isolation of a mutant MDBK cell line
resistant to bovine viral diarrhoea virus infection due to a block in viral
entry. Virology 208, 565-575.
Francki, R. I. B., Fauquet, C. M., Knudson, D. L. & Brown, F. (editors)
(1991 ). Classification and Nomenclature of Viruses. Fifth Report of the
International Committee on the Taxonomy of Viruses. Archives of
Virology, Supplementum 2, 223-233,
Greiser-Wilke, I., Haas, L., Dittmar, K., Liess, B. & Hoennig, V. (1993).
RNA insertions and gene duplications in the nonstructura] protein p125
region of pestivirus strains and isolates in vitro and in vivo. Virology 193,
977-980.
Harris, K.S., Xiang, W., Alexander, L., Lane, W.S., Paul, A.V. &
Wimmer, E. (1994). Interaction of poliovirus polypeptide 3CD with the
5' and 3' termini of the poliovirus genome. Journal of Biological Chemistry
269, 27004-27014.
Houghton, M., Weiner, A., Han, J., Kuo, G. & Choo, Q. L. (1991).
Molecular biology of the hepatitis C viruses: implications for diagnosis,
development and control of viral disease. Hepatology 14, 381-388.
Howard, C. R. (1992). Hepatitis C virus: a comparison with pestiviruses
and other flaviviruses. Proceedings of the Second Symposium on Pestiviruses,
pp. 1~6. Veyrier-du-Lac: European Society for Veterinary Virology.
Johnson, J. C. & Rosenbusch, R. F. (1990). Effect of host cell on the in
vitro characteristics expressed by two bovine viral diarrhoea virus strains.
Veterinary Microbiology 21, 319-328.
Meyers, G., Tautz, N., Dubovi, E.J. & Thiel, H.-J. (1991). Viral
cytopathogenicity correlated with integration of ubiquitin-coding sequences. Virology 180, 602--616.
Meyers, G., Tautz, N., Stark, R., Brownlie, J., Dubovi, E. J., Collett, M.
S. & Thiel, H.-J. (1992). Rearrangement of viral sequences in
cytopathogenic pestiviruses. Virology 191, 368-386.
Miller, R. H. & Purcell, R. H. (1990). Hepatitis C shares amino acid
sequence similarity with pestiviruses and flaviviruses as well as members
of two plant virus supergroups. Proceedings of the National Academy of
Sciences, USA 87, 2057-206I.
Moennig, V. & Plagemann, P. G. W. (1992). The pestiviruses. Advances
in Virus Research 41, 53-98.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 22:34:52
!¸73!
Poole, T. L., Wang, C., Popp, R. A., Potgieter, L. N. D., Siddiqui, A. &
Rumenapf, T., Unger, G., Strauss, J. H. & Thiel, H.-J. (1993). Processing
of the envelope glycoproteins of pestiviruses. Journal of Virology 67,
3288-3294.
Collett, M. S. (1995). Pestivirus translation initiation occurs by internal
ribosome entry. Virology 206, 750-754.
Sambrook, J., Fritsch, E. g. & Maniatis, T. (1989). Molecular Cloning: A
Laboratory Manual, 2nd edn. New York: Cold Spring Harbor Laboratory.
Purchio, A. F., Larson, R. & Collett, M. S. (1983). Characterization of
virus-specific RNA synthesized in bovine cells infected with bovine viral
diarrhoea virus. Journal of Virology 48, 320-324.
Tamura, J. K., Warrener, P. & Collett, M. S. (1993). RNA-stimulated
NTPase activity associated with the pS0 protein of the pestivirus bovine
viral diarrhoea virus. Virology 193, 1-10.
Qi, F., Ridpath, J. F., Levis, T., Bolin, S. R. & Berry, E.S. (1992).
Analysis of the bovine viral diarrhoea virus genome for possible cellular
insertions. Virology 189, 285-292.
Thiel, H.-J., Plagemann, P. & Moennig, V. (1996). Pestiviruses. In Fields
Virology 3rd edn, pp. 1059-1073. Edited by B. N. Fields, D. M. Knipe &
P. M. Howley. Philadelphia: Lippincott-Raven.
Renard, A., Dina, D. & Martial, J. (1987). Vaccines and diagnostics
derived from bovine diarrhoea virus. European Patent, no, 0208672 .
Warrener, P. & Collett, M.S. (1995). Pestivirus NS3 (PS0) protein
possesses RNA heIicase activity. Journal of Virology 69, 1720-1726.
Westaway, E.G. (1987). Flavivirus replication strategy. Advances in
Virus Research 33, 45-90.
Wiskerchen, M. & Collett, M.S. (1991). Pestivirus gene expression:
Nuttall, P.A. (1980). Growth characteristics of two strains of bovine
virus diarrhoea virus. Archives of Virology 66, 365-369.
Richards, S. G. (1991). The rapid diagnosis of bovine pestivirus infections,
Chapter 4, pp. 36-43. PhD thesis, University of Sydney, Australia.
Richards, O, C. & Ehrenfeld, E. (1990). Poliovirus RNA replication.
Current Topics in Microbiology and Immunology 161, 89-119.
protein PS0 of bovine viral diarrhoea virus is a proteinase involved in
polyprotein processing. ViroIogy 184, 341-350.
Ridpath, J. F. & Bolin, S. R. (1995). The genomic sequence of a virulent
bovine viral diarrhoea virus (BVDV) from the type 2 genotype: detection
of a large genomic insertion in a noncytopathic BVDV. Virology 212,
39-46.
Xue, W. & Minocha, H., C. (1993). Identification of the cell surface
receptor for bovine viral diarrhoea virus by using anti-idiotypic
antibodies. Journal of General Virology 74, 73-79.
Rueckert, R. R. (1996), Picomaviridae : the viruses and their replication.
In Fields Virology, 3rd edn, pp. 609--654. Edited by B. N. Fields, D. M.
Knipe & P. M. Howley. Philadelphia: Lippincott-Raven.
Received 9 April 1996; Accepted 16 July 1996
~_73f
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 22:34:52