Journal
of General Virology (2001), 82, 93–101. Printed in Great Britain
...................................................................................................................................................................................................................................................................................
Deletion or substitution of the aphthovirus 3h NCR abrogates
infectivity and virus replication
M. Sa! iz,1,2 S. Go! mez,1,2 E. Martı! nez-Salas2 and F. Sobrino1,2
1
2
Centro de Investigacio! n en Sanidad Animal, INIA, 28130 Valdeolmos, Madrid, Spain
Centro de Biologı! a Molecular ‘ Severo Ochoa ’ (CSIC-UAM), Universidad Auto! noma de Madrid, Cantoblanco, 28049 Madrid, Spain
The 3h noncoding region (NCR) of the genomic picornaviral RNA is believed to contain major cisacting signals required for negative-strand RNA synthesis. The 3h NCR of foot-and-mouth disease
virus (FMDV) was studied in the context of a full-length infectious clone in which the genetic
element was deleted or exchanged for the equivalent region of a distantly related swine
picornavirus, swine vesicular disease virus (SVDV). Deletion of the 3h NCR, while maintaining the
intact poly(A) tail as well as its replacement for the SVDV counterpart, abrogated virus replication
in susceptible cells as determined by infectivity and Northern blot assays. Nevertheless, the
presence of the SVDV sequence allowed the synthesis of low amounts of chimeric viral RNA at
extended times post-transfection as compared to RNAs harbouring the 3h NCR deletion. The failure
to recover viable viruses or revertants after several passages on susceptible cells suggests that the
presence of specific sequences contained within the FMDV 3h NCR is essential to complete a full
replication cycle and that FMDV and SVDV 3h NCRs are not functionally interchangeable.
Introduction
Foot-and-mouth disease virus (FMDV) is the causative
agent of an acute systemic disease of cloven-hooved animals
considered to be a major animal health problem world-wide
(Kitching, 1999). FMDV is the prototypic member of the
aphthoviruses, which belong to the family Picornaviridae
(Rueckert, 1990). Swine vesicular disease is a highly contagious
disease caused by the enterovirus swine vesicular disease virus
(SVDV), which also belongs to the family Picornaviridae
(Sutmoller, 1992). The clinical signs produced by SVDV are
indistinguishable from those caused by FMDV in pigs. Both
viruses share the ability to infect in vitro cultured IBRS-2 cells
whereas only FMDV infects BHK-21 cells.
The FMDV genome consists of a single molecule of
messenger-sense RNA of about 8500 nt containing a single
ORF encoding the viral polyprotein, which is subsequently
processed into the viral products (Ryan et al., 1989). The 5h end
of the RNA is covalently linked to a small viral protein (VPg)
(Sangar et al., 1977). The 5h noncoding region (NCR), about
1200 nt in length, is highly structured (Clarke et al., 1987) and
contains several genetic elements necessary to control essential
Author for correspondence : Encarnacio! n Martı! nez-Salas.
Fax j34 91 3974799. e-mail emartinez!cbm.uam.es
0001-7315 # 2001 SGM
functions in the replication cycle. This region harbours an
internal ribosome entry site (IRES) element that promotes the
cap-independent translation initiation of the viral genome
(Ku$ hn et al., 1990 ; Lo! pez de Quinto & Martı! nez-Salas, 1997).
Immediately downstream of the stop codon at the end of the
polyprotein ORF is a 3h NCR of about 90 nt followed by a
genetically encoded poly(A) tract (Chatterjee et al., 1976).
The picornavirus 3h NCR has been predicted to contain
stable secondary and tertiary structures by computer algorithms and biochemical structure probing (Pilipenko et al., 1992 ;
Todd & Semler, 1996). A cis-acting determinant for replication
is believed to reside in the 3h NCR where primary sequence
and\or higher order motifs are presumably recognized by the
replication complex in order to initiate the synthesis of the
negative-strand RNA (Richards & Ehrenfeld, 1990 ; Rohll et al.,
1995). Engineered disruption of the relevant structures predicted in this region results in an impairment of the replicative
capacity of the viral genome (Rohll et al., 1995 ; Pierangeli et al.,
1995).
Binding of viral proteins, or their precursors, to the 3h NCR
has been reported for encephalomyocarditis virus (Cui et al.,
1993) and poliovirus (PV) (Harris et al., 1994). In addition,
picornavirus 3h NCR interacts with host proteins. A complex
of proteins binds to human rhinovirus type 14 (HRV14) and
PV1 3h NCRs and a deletion mutant in this region exhibited
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:41:10
JD
M. Sa! iz and others
reduced binding capacity and a defective replication phenotype
(Todd et al., 1995). Likewise, cellular factors interact with the
3h NCR of PV3, HRV14 and coxsackievirus B4 (CB4) and
mutations reducing binding affinity had a deleterious effect on
virus replication (Mellits et al., 1998). A nucleolar protein,
nucleolin, has been shown to bind poliovirus 3h NCR.
Relocalization of nucleolin induced by virus infection seems to
play a role in the early stages of the virus replication cycle
(Waggoner & Sarnow, 1998). The functional significance of
those interactions remains to be determined, although some of
them may play an important role in regulating the initiation of
the negative-strand once the cis-elements have mediated the
first recognition event or in modulating the specificity of this
process.
Moreover, internal cis-acting replication elements (CREs)
located in the picornavirus genome coding region are now
being identified. This is the case with a hairpin structure in the
HRV14 capsid protein precursor (McKnight & Lemon, 1998),
in the VP2 coding sequence of two cardioviruses (Lobert et al.,
1999) or a similar element located in the PV 2C coding region
(Goodfellow et al., 2000). These findings contribute to
highlight the complexity of a process in which different viral
and cellular factors contribute to promote efficient initiation of
negative-strand RNA synthesis (Agol et al., 1999).
Little is known about the role of the 3h NCR in
aphthoviruses although some experimental evidence supports
its relevance in the FMDV replication cycle. RNA transcripts
spanning the 3h-terminal region of the FMDV genome, in both
sense and antisense orientations, have been reported to inhibit
infective particle formation following co-microinjection or cotransfection with viral RNA in BHK-21 cells (Gutie! rrez et al.,
1994). Inhibition of virus yield was also observed in FMDVinfected cells transiently expressing the interfering RNAs
(Bigeriego et al., 1999).
In this report, we have examined the role of the 3h NCR of
FMDV in the virus replication cycle by the deletion of this
element and by its substitution with its counterpart of another
swine picornavirus, SVDV, in a full-length infectious cDNA
clone of FMDV. The effects of those alterations on replication
and translation have been followed by infectivity and RNA
analysis.
Methods
Construction of plasmids. Plasmids are based on an FMDV O1K
infectious full-length cDNA clone, pSP65FMDVpolyC (Falk et al.,
1992 ; Zibert et al., 1990), which has an average poly(C) tract of 30 nt.
Plasmid pDM, which contains unique StuI and EcoRV sites flanking the 3h
NCR, was made by mutagenic PCR (Kunkel, 1985). In this clone, three
nucleotide mismatches were artificially introduced to create the StuI
and EcoRV sites at positions 9 and 86 from the stop codon, respectively
(Fig. 1).
To construct plasmid p∆74, pDM was digested with StuI and EcoRV
and the large fragment released was religated using T4 DNA ligase. The
resulting clone has a deletion of most of the 3h NCR element (74 nt).
JE
Fig. 1. Schematic representation of different constructs based on an
FMDV full-length cDNA clone. The restriction sites used for cloning are
indicated. The sequence of the 3h region is shown above each plasmid.
Artificially engineered residues are shown in bold letters. The stop codon
is underlined and SVDV sequences are boxed.
Thus, only 15 virus residues remain between the stop codon and the
poly(A) tract.
Prior to the construction of plasmid pCHIM, construct pTAG was
made by recombinant PCR (Kunkel, 1985) using pDM as a template and
primer pairs OL-1 (5h CGGACCGCAAATTGGCTCAGCGG 3h) with
ILAV (5h CCTCTGAGGGCCTAGGCGTCA 3h) and IRAV (5h TGACGCCTAGGCCCTCAGAGG 3h) with ILE-1 (5h TTTTTGGATATCAAGGAAGCGGGAAAAACCC 3h), respectively. The bold letters indicate nucleotides that were substituted in order to create the underlined
AvrII site at nt k2 from the stop codon. A unique CelII site at position
635, relative to the first nucleotide of the 3D gene sequence (Fig. 1), was
used for the 5h insertion of the PCR products. The final PCR product was
digested with CelII and EcoRV (sites underlined in external primers OL1 and ILE-1, respectively) and ligated into the pDM clone in which the
fragment CelII–EcoRV had been previously excised. Thus, plasmid pTAG
contains a unique AvrII site spanning the stop codon, which changes
TAA in the original construct to TAG.
cDNA containing the 3h NCR of SVDV was amplified by RT–PCR
from viral RNA of strain UK (Seechurn et al., 1990) using primers
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:41:10
FMDV 3h NCR is essential for replication
SVDCHIM-R (5h TTTTTTTTTTTTTTTGATATCCGCACCGAATACGG 3h) and SVDCHIM-L (5h GGACTCCTTTTAACCTAGGAATTAGAGCACAATTAG 3h), which introduce the underlined EcoRV and
AvrII sites, respectively. The RT–PCR product was then digested with
AvrII and EcoRV and cloned into pTAG in which the fragment
AvrII–EcoRV had been excised. The resulting plasmid, pCHIM, has the 3h
NCR of SVDV inserted into the FMDV infectious clone and only a G
residue remains between the stop codon and the aphthoviral sequence. In
the 3h end, 6 nt corresponding to the FMDV 3h NCR remain preceding
the poly(A) tract.
All plasmids were transformed into E. coli strain DH5α cells and
sequenced along the 3h modified region using the fmol DNA cycle
sequencing system (Promega). Prediction of secondary structures in the
3h NCR of the different constructs was performed using the M-Fold
program of the GCG package. The average length of the poly(A) tract in
all the constructs was 58 nt.
In vitro RNA transcription. Synthetic RNAs were obtained by
transcribing HpaI-linearized plasmids with SP6 RNA polymerase (Promega), according to the manufacturer’s instructions. After transcription,
the reaction mixture was incubated with RQ1 DNase (1 U\µg RNA)
(Promega). RNAs were then phenol–chloroform-extracted and ethanolprecipitated. RNA integrity and concentration were determined by
native agarose gel electrophoresis.
Cells and transfections. Cells were cultured in Dulbecco’s
modified Eagle’s medium supplemented with 2 to 5 % foetal bovine
serum and 1 mM HEPES pH 7n4. Semiconfluent (80 %) BHK-21 or IBRS2 (a swine-derived cell line) cell monolayers, split 1 day prior to
transfection, were transfected with 3 to 5 µg of transcripts. Liposomemediated transfection (Rose et al., 1991) was performed as described
(Lo! pez de Quinto & Martı! nez-Salas, 1997) and transfected cells were
incubated at 37 mC or 34 mC until cytopathic effect (CPE) was observed.
For those mutants that did not induce a visible CPE, new rounds of
amplification were performed collecting separately the transfection
medium and the monolayers lysed by three cycles of freeze–thawing.
The lysate and the transfection medium were transferred independently
onto fresh cell monolayers.
Infection of suckling mice. Virus used as the inoculum was
produced by transfection of RNA onto BHK-21 cells followed by two
rounds of amplification. Groups of 12 mice, 7-day-old, were inoculated
intraperitoneally with 100 µl of a virus dilution in PBS. Dead animals
were scored from 2 to 7 days after inoculation. LD corresponds to virus
&!
titre required to kill 50 % of the inoculated animals.
In vitro translation. About 200 and 400 ng of each transcript was
translated in 10 µl of rabbit reticulocyte lysate (Promega) in the presence
of 10 µCi of [$&S]methionine (10 mCi\ml). Reaction mixtures were
incubated at 30 mC for 1 h. Aliquots of the translation products were
loaded onto a 12 % SDS–polyacrylamide gel. The gels were then dried
and exposed to X-ray films (Agfa Curix RP2).
Northern blot. Extracts from transfected cells (about 1i10'),
frozen at different times post-transfection, were prepared in 260 µl of
0n5 % NP-40, 120 mM NaCl and 50 mM Tris–HCl, pH 7n8. The lysate
was then centrifuged at 12 000 r.p.m. for 5 min and total cytoplasmic
RNA was purified from the supernatant with TRI Reagent (Sigma).
Aliquots of RNAs (12n5 % of each sample) were resolved through a
formaldehyde gel and blotted onto a nylon Zeta probe membrane (BioRad) using the vacublot XL system (Pharmacia). The probe was prepared
by PCR using the 932 bp CelII–HpaI fragment from pSP65FMDVpolyC
as a template (about 200 ng) and the antisense primer I-18 (5h
GCCACCACGATGTCGTCTCC 3h) in the presence of 50 µCi of [α-
$#P]dCTP (6000 Ci\mmol ; Amersham). The probe was subsequently
purified using a Nick column (Pharmacia). The resulting probe is a singlestranded DNA fragment of 392 nt complementary to nt 817–1207 of the
3Dpol gene. About 5i10' c.p.m. of the probe was added to the
hybridization solution (50 % formamide in 0n12 M Na HPO , pH 7n2,
#
%
0n25 M NaCl, 7 % SDS) and incubated overnight at 42 mC. The membrane
was then washed three times for 5 min at room temperature with 2i
SSC\0n1 % SDS ; 0n5i SSC\0n1 % SDS ; 0n1i SSC\0n1 % SDS, respectively, and exposed to X-ray film.
RT–PCR analysis of viral RNA. Serial dilutions of the total
cytoplasmic RNA extracted as above were analysed by RT–PCR to
determine the best signal to background ratio (data not shown). Basically,
RNAs (1 µl of a 1\10 dilution) were heated at 93 mC for 2 min with 10
pmol of the antisense primer I-18 in a final volume of 10 µl, then cooled
down at room temperature for 10 min prior to RT for 1 h in a 20 µl
reaction containing 50 mM Tris–HCl, pH 8n3, 75 mM KCl, 3 mM
MgCl , 10 mM DTT, 1 mM each dNTP, 40 U of RNasin RNase inhibitor
#
(Promega) and 20 U of SuperScript II RNase H− reverse transcriptase
(Gibco BRL).
For PCR amplification, 1\5 to 1\10 of the RT reaction was used in the
50 µl PCR mixture containing 10 mM Tris–HCl, pH 8n3, 50 mM KCl,
1n5 mM MgCl , 0n2 mM each dNTP, 20 pmol each of the respective
#
primers and 5 U of AmpliTaq DNA polymerase (Perkin Elmer). Samples
were amplified in a UNO-thermoblock (Biometra) using a program that
included an initial incubation at 94 mC for 10 min and 72 mC for 2 min
(the enzyme was added at this point), followed by 30 cycles of
denaturation at 94 mC for 1n5 min, annealing at 50 mC for 1 min and
extension at 72 mC for 1n5 min. All the RT–PCR products were dependent
on the presence of reverse transcriptase in the reaction mixture (data not
shown), proving the specificity for RNA templates.
The primers used for the PCR analysis of viral RNA were D3-I(k) (5h
GTACTGTGTGTAGTACTG 3h), complementary to nt 286–303 of
the VP3 gene, and D3-V(j) (5h GACCCGAAGACGGCTGACC 3h),
corresponding to nt 61–80 of the VP3 gene. The expected amplification
product is a DNA fragment of 252 bp.
For positive-strand RNA amplification experiments D3-I was used as
the antisense primer in the RT reaction at 42 mC.
Negative-strand RNA amplification was carried out using an external
primer in the RT reaction, VP3NEST(j) (5h CCAACGTGCACGTCGCGGGT 3h), corresponding to the sequence identical to nt 617–637 of
the VP2 gene, and cDNA synthesis was performed at 50 mC. Prior to PCR
amplification with primers D3-I and D3-V, primer VP3NEST was
removed by filtering the RT reaction through a Centricon YM-30 column
(Millipore). Under these conditions, specific amplification dependent on
the negative-strand was achieved for up to 1 ng of input RNA.
As a control for RNA extraction, an RT–PCR reaction using primers
designed for amplification of cellular mRNA was performed using
oligonucleotides g3pdh 3h(k) (5h AAGTTGTCATGGATGACCTTGGCCA 3h) and g3pdh 5h(j) (5h CCATCACCATCTTCCAGGAGCGAG
3h), which amplify a 287 bp DNA fragment corresponding to nt 387–673
of human glyceraldehyde-3-phosphate dehydrogenase gene (accession
no. X01111).
Results
RNAs harbouring a 3h NCR deletion are not infectious
To examine the role of the 3h NCR in the FMDV infectious
cycle we constructed a derivative plasmid, pDM, from the
FMDV O1K infectious full-length clone in which two unique
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:41:10
JF
M. Sa! iz and others
Table 1. Infectivity of the different FMDV transcripts on BHK-21 and IBRS-2 cells and
virulence of the recovered viruses in suckling mice
Infectivity was estimated as the ability to induce CPE in primary transfections on BHK-21 and IBRS-2 cells at
37 mC or on IBRS-2 cells at 34 mC. Transfected cells were incubated at 37 mC under 0n6 % agar overlay and
monolayers were fixed and stained at 44 h post-transfection. The infectivity of the virus preparation in BHK21 cells was 1i10( p.f.u.\ml and this preparation was used as the inoculum for suckling mice.
Primary transfections
Infectivity
RNA
SP65FMDVpolyC
DM
∆74
TAG
CHIM
37 mC
34 mC
p.f.u./µg RNA
(BHK-21)
j
j
k
j
k
j
j
k
j
k
10%
10%
1
10%*
1
p.f.u./µg RNA
(IBRS-2)
Virulence in
suckling mice
(LD50/ml)
4i10%
3i10%
1
5i10%*
1
10$n&
10$n#
k

k
* Small plaque phenotype.
, Not determined.
restriction sites flanking the 3h NCR (StuI and EcoRV) were
introduced (Fig. 1). RNAs derived from plasmid pDM are
infectious at a level similar to that of the RNA derived from the
wt construct (pSP65FMDVpolyC) (Table 1).
The viability of the viruses recovered from transfections
with SP65FMDVpolyC and DM RNAs was analysed in
suckling mice infected with second-passage viruses. The
virulence observed for pDM-derived virus was similar to that
obtained for the virus stock derived from the original clone
(Table 1). This result proves the infection ability of viruses
derived from pDM in an animal system.
To study the biological relevance of FMDV 3h NCR, we
constructed the p∆74 derivative (Fig. 1), harbouring a deletion
of most of the 3h NCR. p∆74 retains 15 virus residues between
the stop codon and the poly(A) tract, which remains intact in
all constructs. In vitro-transcribed RNAs harbouring the ∆74
mutation were transfected onto BHK-21 and IBRS-2 monolayers and cells were kept at 37 mC and monitored for CPE
development. No CPE could be observed in these monolayers
after 3 to 4 days while the positive control, DM RNA,
consistently induced CPE about 19 h post-transfection.
In order to preserve cell viability for extended times
following transfection and to give a better chance to slowgrowing viruses to complete their life-cycle, transfected cells
were incubated at 34 mC for up to 9 to 11 days. As BHK-21
cells exhibited an impaired ability to grow at 34 mC, these
experiments were performed using IBRS-2 cells. During this
time no CPE could be observed in cells transfected with ∆74
RNAs whereas transcripts derived from the positive control,
pDM, induced CPE about 48 h post-transfection (Table 1).
No infectious virus could be recovered from cells transfected with ∆74 RNA, as indicated by the negative results of
JG
two additional rounds of amplification performed by transferring the transfection medium and the cell lysate onto fresh
monolayers. Therefore, we conclude that deletion of the 3h
NCR was deleterious for virus multiplication.
Exchange of the FMDV 3h NCR for its homologue from
SVDV fails to produce infectious RNA
The use of chimeric viruses provides a useful tool to study
the conservation of functional signals to initiate replication
between related viruses. In order to study the possibility of a
functional substitution of the 3h NCR of FMDV for that
element of another picornavirus, we chose SVDV, a distantly
related swine picornavirus, which has a similar growth capacity
on IBRS-2 cells as FMDV and shares a 45 % primary sequence
homology in the 3h NCR. Prior to the construction of the
FMDV–SVDV chimera, a new plasmid, pTAG, containing a
unique AvrII site spanning the stop codon of the polyprotein
which changes the codon from TAA to TAG was made (Fig. 1).
Secondary structure prediction of the 3h NCR region in pTAG
is identical to that of pDM, consisting of two main stem–loops
that are highly conserved among all aphthovirus strains (data
not shown). The intermediate construct pTAG allows the
precise excision of the 3h NCR as an AvrII–EcoRV cassette and
its replacement with a heterologous sequence bearing those
restriction sites while keeping the 3Dpol sequence and the
poly(A) tract intact. The infectivity on BHK-21 and IBRS-2
cells shown by transcripts produced from pTAG was similar to
that obtained with pDM transcripts (Table 1), but the RNAs
derived from the pTAG mutant caused smaller plaques. To
construct pCHIM, the 3h NCR of SVDV RNA was amplified
by RT–PCR and then ligated into pTAG (Fig. 1). Thus, pCHIM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:41:10
FMDV 3h NCR is essential for replication
Fig. 3. Northern blot analysis of positive-strand viral RNA in IBRS-2 cells
transfected with the transcripts indicated on the top of the panel or mocktransfected. Cells were incubated at 34 mC and harvested at the indicated
times (in h) following transfection. The blot was probed with a α-32Plabelled single-stranded DNA fragment complementary to residues 817 to
1207 of the 3Dpol gene. The unspecific band (28S RNA), present also in
mock-transfected cells, serves as a sample loading control.
Fig. 2. Translation efficiency of transcripts in the rabbit reticulocyte lysate
system. Lanes 1 and 2, 3 and 4, and 5 and 6 correspond to 200 ng and
400 ng of ∆74, CHIM and TAG RNAs, respectively. (a) SDS–PAGE and
autoradiography of the in vitro translation products derived from RNA
templates shown in (b). Protein molecular mass markers are indicated in
kDa.
is a chimera of the FMDV full-length infectious clone in which
the 3h NCR has been exchanged for that of SVDV. pCHIM
contains a synonymous nucleotide substitution in the last
codon of the 3D protein (Fig. 1). The predicted secondary
structure for pCHIM shows no disruption of the secondary
structure predicted for the SVDV RNA in that region (data not
shown).
RNAs transcribed from pCHIM failed to induce visible CPE
when transfected onto BHK-21 or IBRS-2 monolayers incubated at 37 mC (Table 1). Nevertheless, when cells transfected
with CHIM RNA were incubated at 34 mC a diffuse cellular
damage as detected by cell detachment was observed in IBRS2 cells at extended times post-transfection (about 4 days). This
effect was not visible in cells transfected with ∆74 RNA or in
mock-transfected cells (data not shown). In no case were we
able to recover infectious particles from cells transfected with
CHIM RNAs, incubated at either 34 mC or 37 mC, after two
additional passages on IBRS-2 cells using both the supernatant
and the lysate from transfected cells harvested at different
times post-transfection (from 3 to 9 days). These results
indicate that the 3h NCRs of FMDV and SVDV are not
functionally interchangeable.
Mutations in the 3h NCR have no effect on the in vitro
translation efficiency of the FMDV RNA
To assess that the phenotype observed for mutant RNAs
bearing extensive deletion or heterologous sequences in the 3h
NCR was due to a major replication defect, we assayed the in
vitro translation efficiency of the transcripts to rule out the
possibility of significant translation dysfunction in these
mutants (Fig. 2). Equal amounts of RNAs derived from
constructs p∆74, pCHIM and pTAG (Fig. 2 b) were used to
program in vitro translation reactions in a rabbit reticulocyte
lysate system. The translation products are shown in Fig. 2 (a).
No significant differences in the viral protein patterns could be
detected, suggesting that the multiplication defect associated
with mutant RNAs was independent of the efficiency of
translation.
Analysis of RNA derived from primary transfections
The 3h NCR of different picornavirus RNAs has been
shown to be involved in replicative events (Pierangeli et al.,
1995 ; Richards & Ehrenfeld, 1990 ; Rohll et al., 1995).
Accordingly, the lack of infectivity of ∆74 and CHIM RNAs
could be due to a replication defect. In order to explore this
possibility, the accumulation of positive-strand viral RNA in
cells transfected with transcripts prepared from p∆74, pCHIM
and pTAG was assessed (Fig. 3). Full-length positive-strand
RNA could be detected only in cells transfected with TAG
RNA. Overexposure of the film did not yield full-length viral
RNA bands in lanes corresponding to ∆74 and CHIM RNAs.
Our failure to detect viral RNA in cells transfected with 3h
mutant RNAs by Northern blot suggested major dysfunction
defects for those mutants at the replication level.
In a further attempt to detect low levels of virus replication
derived from those RNAs, we carried out an RT–PCR analysis
of the RNAs extracted from transfected cells in order to
specifically amplify viral positive- and negative-strand RNAs.
Fig. 4 (a) shows the results of the negative-strand amplification.
Comparison of lanes corresponding to the three different
transcripts indicates that neither ∆74 nor CHIM RNAs were
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:41:10
JH
M. Sa! iz and others
Fig. 4. RT–PCR analysis of transfected cells. IBRS-2 cells were incubated
at 34 mC and harvested at the indicated times (in h) following transfection
with the transcript indicated on the top of each panel. (a) Detection of
negative-strand RNA following cDNA synthesis with primer VP3NEST. The
PCR reaction was carried out using primers D3-I and D3-V as described in
Methods. (b) Positive-strand RNA RT–PCR with primers D3-I and D3-V. (c)
RT–PCR of cellular RNA with primers g3pdh3h and g3pdh5h. The expected
sizes of the products are indicated with an arrow. M, size of markers (bp).
able to reach replication levels comparable to the levels of
infectious RNAs derived from pTAG, even at late times posttransfection (119 h). In contrast, a significant difference was
observed between ∆74 and CHIM RNAs after 42 h posttransfection. In cells transfected with chimeric RNAs harbouring the 3h NCR of SVDV negative-strand RNA was
detected at the 42 h time-point. However, at early times posttransfection (0n5 h), positive RT–PCR amplification was
detected in both ∆74 and CHIM RNA lanes, presumably
arising from remnants of transfection input RNA. The
differences observed in synthesis of negative-strand RNA at
42 h post-transfection are unlikely to be due to incomplete
elongation by aberrant initiation events of the negative-strand
RNA as the primers used in the assay amplify an internal
fragment of the genome (VP3).
The results of the RT–PCR amplification of positive-strand
RNA are shown in Fig. 4 (b). The fact that positive-strand RNA
remains detectable at times post-transfection as late as 119 h,
even for genomes unable to replicate (as suggested in Fig. 4 a)
could be explained by the presence of low amounts of input
RNA undetectable in the Northern blot assay. Nevertheless,
slight differences in the band intensities were observed among
the transcripts despite of the lack of linearity in the assay. The
JI
intensity of the bands from monolayers transfected with
CHIM RNA was higher when compared to those transfected
with ∆74 RNA, particularly at 42 h post-transfection. This is
consistent with the detection of small amounts of negativesense viral RNA in cells transfected with pCHIM transcripts
(Fig. 4 a).
Fig. 4 (c) shows the results of the RT–PCR reactions with
g3pdh primers as a control of sample RNA loading. The
intensity of the bands with these primers did not correlate with
the intensities of bands shown in Fig. 4 (a) and (b), indicating
that the differences observed in those reactions were not due
to significant differences in RNA load.
Sequencing of the PCR products corresponding to positivestrand RNA amplification confirmed the presence of the ∆74-,
SVDV 3h NCR- and TAG-specific sequences in viral RNAs
from the respective transfections at 42, 90 and 119 h posttransfection.
The results obtained from cells transfected with mutant
RNAs by RT–PCR analysis suggest that a certain advantage
was conferred by the presence of the heterologous element
from a distant picornavirus over the deletion of this region,
even though none of the FMDV RNAs bearing the ∆74
deletion or the SVDV sequence in the 3h NCR were replicationcompetent at levels sufficient to make them viable and produce
infectious viruses. However, we failed to detect the presence of
viral proteins in cells transfected with ∆74 and CHIM
transcripts in Western blot experiments (data not shown).
Thus, CHIM RNA might replicate to a small extent but this is
still below the threshold required for further rounds of virus
multiplication.
Discussion
The precise mechanisms involved in initiating and regulating replicative events in infected cells remain elusive in
picornaviruses. However, the relevance of the 3h NCR in
initiation of negative-strand RNA synthesis has been well
established as an important cis-acting signal mediating the
process. Extensive mutagenesis analysis of this region has
shown primary sequence and secondary and tertiary order
motifs essential for preserving a wt replication phenotype.
However, the extent of conservation of those signals among
different picornaviruses is difficult to predict, as results inferred
for different viruses may be in apparent contradiction. Binding
of several viral and host proteins to the 3h NCR has also been
reported. Some of these interactions may play a role in
replication.
Previous work on picornaviruses has shown that PV and
HRV14 RNAs harbouring complete deletions of the 3h NCR
are infectious. The recovered viruses demonstrated a defective
growth phenotype and an RNA replication defect, suggesting
that the presence of the 3h NCR may not be an absolute
requirement for genome replication in these viruses (Todd et
al., 1997). In contrast, individual excision of the two stem–
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:41:10
FMDV 3h NCR is essential for replication
loops of PV3 3h NCR had been previously reported to abolish
replication (Rohll et al., 1995).
We have examined the role of the 3h NCR in FMDV, an
aphthovirus for which there is no available information
regarding replication signals. Our results with the 3h NCR
mutants show that ∆74 RNAs were not infectious when
transfected into susceptible cells as no CPE could be observed
and we were unable to recover infectious virus after two
rounds of amplification.
The functional inter-exchange of PV 3h NCR for that of
CB4 or HRV14 has been reported (Rohll et al., 1995). When the
donor sequences came from hepatitis A virus (HAV) or bovine
enterovirus, the replication levels were significantly lower than
those for the PV3, CB4 or HRV14 recombinants exhibiting the
wt replication phenotype. In contrast, Pierangeli et al. (1995)
reported the deleterious effect on replication of the substitution
of PV 3h NCR with the equivalent sequences of HAV RNA.
Since the ∆74 deletion, spanning most of the FMDV 3h
NCR and including the two predicted stem–loops, resulted in
an impairment of infectivity, we studied the functional
substitution of the element with that of the distantly related
enterovirus SVDV. A chimeric construct, pCHIM (Fig. 1), did
not induce visible CPE in BHK-21 or IBRS-2 cells at 3 to 4 days
post-transfection, as compared to DM RNAs, which induced
CPE about 19 h post-transfection. We were not able to recover
infectious virus in this case, similar to the situation seen with
the ∆74 mutant.
In order to hypothesize replication defects in ∆74 and
CHIM RNAs, it was necessary to rule out the possibility of
major translation defects. In recent reports the highly conserved 98 nt X region of the 3h NCR of hepatitis C virus has
been shown to enhance translation (Ito et al., 1998 ; Ito & Lai,
1999). When the in vitro translation efficiency of the different
FMDV transcripts in a rabbit reticulocyte lysate system was
compared, no relevant differences could be detected between
infectious and uninfectious RNAs (Fig. 2). This result suggests
that the inability of ∆74 and CHIM RNAs to direct RNA
replication efficiently was unrelated to their translation initiation efficiency.
In further experiments we used IBRS-2 cells that were able
to grow at lower temperatures (34 mC) and may have provided
trans-acting factors specifically required for the replication
competence of the FMDV–SVDV chimera.
The accumulation of positive-sense FMDV RNA, monitored by Northern blot, indicated that only TAG transcripts
were able to replicate successfully, while RNAs derived from
p∆74 and pCHIM did not accumulate at levels detectable by
Northern blot.
In our search for RNA synthesized de novo in cells
transfected with ∆74 and CHIM RNAs, we carried out an
RT–PCR analysis to detect lower amounts of RNA. In the
negative-strand amplification experiments we detected viral
RNA at 42 h post-transfection in cells transfected with CHIM
RNA, but not in monolayers transfected with ∆74 RNA (Fig.
4 a). Positive-strand amplification revealed slight differences in
the intensity of the bands corresponding to ∆74 and CHIM
RNA-transfected cells respectively, at different times posttransfection.
Therefore, we conclude that the deletion of the 3h NCR had
a deleterious effect on FMDV replication. Its presence seems to
be a strict requirement for the initial rounds of replication as no
viable viruses, including revertants or pseudorevertants containing additional mutations, were recovered. The SVDV 3h
NCR, replacing the FMDV homologue, was unable to provide
the same function in cis. The extent of identity of both
elements at the primary sequence level is 45 % and the
secondary structure predicted for SVDV 3h NCR does not
resemble that of FMDV 3h NCR. The secondary structure was
maintained intact in the pCHIM construct (data not shown).
FMDV and SVDV cause indistinguishable symptomatic diseases in pigs and productively infect IBRS-2 cultured cells.
However, their replicative signals, at least at the 3h NCR level,
seem to be different. The lack of infectivity of CHIM transcripts
on IBRS-2 cells is unlikely to be due to the absence of specific
host factors for SVDV replication. Interestingly though, the
presence of the SVDV 3h NCR conferred to the RNA a slight
advantage in early replication events as compared to RNAs
harbouring complete 3h NCR deletions.
Our results for FMDV may be in apparent contradiction to
data obtained using entero- and rhinoviruses. Although the
main replication mechanisms have been assumed to be common
for all picornaviruses, some group-dependent strategies cannot
be ruled out.
Novel internal domains essential for replication are being
discovered in the coding regions of the picornaviral genome.
All these CREs resemble each other functionally but differ in
sequence, structure and location. The CREs in rhino- and
cardioviruses are not functionally exchangeable while the CRE
of Mengo virus can replace that of Theiler’s virus (Lobert et al.,
1999). Interestingly, a lethal 3h NCR mutation could be partly
rescued by a compensating mutation within the CRE of
HRV14 (McKnight & Lemon, 1998).
The ability of an FMDV replicon lacking the Lb and most
of the P1 (capsid protein precursor) coding sequences to
replicate suggests that no CRE is contained in that region of
the genome (McInerney et al., 2000). An additional distinguishing feature among enteroviruses, rhinoviruses and aphthoviruses is the different type of IRES contained within their
5h end regions differing in sequence and structure.
The possible interaction of the 5h and 3h NCRs during
translation and replication of picornavirus RNA is currently
being investigated. Inter-regulation of both processes seems to
be mediated by the formation of specific RNA–protein
complexes in the 5h end clover leaf structure (Gamarnik &
Andino, 1998). Additionally, recent data favour a model where
binding of cellular proteins may establish a physical link between the 5h and 3h NCRs (Herold & Andino, 2000 a). Several
IRESs, including those of different virus families, have been
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:41:10
JJ
M. Sa! iz and others
exchanged without drastic consequences for virus replication
(Alexander et al., 1994 ; Rohll et al., 1994 ; Lu & Wimmer, 1996).
However, some requirements in terms of functional compatibility of sequences present in the 5h end, 3h end and CREs
may exist and therefore make it difficult to predict the viability
of modified RNAs bearing engineered mutations and replacements in those regions. Thus, disruption of signals strictly
required for efficient replication, present in the 3h NCR, may
cause major growing defects leading to non-viable phenotypes
as well as revertants and pseudorevertants carrying compensatory mutations in the 3h NCR or other regions involved
in the regulation of the process.
The essential role played by internal sequences of the 3h
NCR has been recently highlighted in a model of initiation of
negative-strand synthesis (Paul et al., 2000), in which uridylylated VPg was transferred to the 3h NCR and then transferred
back to the poly(A) tract. The fact that replicative forms of PV
RNA can be isolated by poly(A) tract selection (Herold &
Andino, 2000 b) strongly supports this model. Consistent with
the biological relevance of signals within the 3h NCR, the data
shown here for FMDV indicate that specific sequences within
this region are absolutely required to initiate replication, since
neither deletion nor SVDV-substituted 3h NCR mutants could
replicate efficiently in transfected cells. What is noteworthy is
that all the mutants shared the same poly(A) tails and the last
four residues, irrespective of their replication competence,
indicating that either a structural element essential for FMDV
RNA synthesis is missing or disrupted in p∆74 and pCHIM or
that more than four residues in the FMDV 3h NCR are required
to initiate replication.
Our data on FMDV 3h NCR add new information to the
poorly characterized regulation of aphthoviruses replication.
The experimental approach using derivatives from an infectious FMDV cDNA clone allows the study of infectivity
both in vitro on cultured cell lines and in vivo in suckling mice.
We are currently focussed on the study of the minimal
requirements needed at the level of primary and secondary
structures for replication competence in FMDV RNA.
We are grateful to Ewald Beck who kindly provided plasmid
pSP65FMDVpolyC containing the full-length sequence of FMDV O1K.
We thank J. I. Nu! n4 ez for his experimental help with mice. This work was
supported by CICYT (grant BIO 99-0833-C02-01), Comunidad de
Madrid (grant 08.2\0024\1997) and DGES (grant PM 98.0122). M. Sa! iz
was a postdoctoral fellow from the Spanish Ministry of Education and
Culture.
References
Agol, V. I., Paul, A. V. & Wimmer, E. (1999). Paradoxes of the replication
of picornaviral genomes. Virus Research 62, 129–147.
Alexander, L., Lu, H. H. & Wimmer, E. (1994). Polioviruses containing
picornavirus type 1 and\or type 2 internal ribosomal entry site elements :
genetic hybrids and the expression of a foreign gene. Proceedings of the
National Academy of Sciences, USA 91, 1406–1410.
BAA
Bigeriego, P., Rosas, M. F., Zamora, E., Martı! nez-Salas, E. & Sobrino,
F. (1999). Heterotypic inhibition of foot-and-mouth disease virus
infection by combinations of RNA transcripts corresponding to the 5h
and 3h regions. Antiviral Research 44, 133–141.
Chatterjee, N. K., Polatnick, J. & Bachrach, H. L. (1976). Cell-free
translation of foot-and-mouth disease virus RNA into identifiable noncapsid and capsid proteins. Journal of General Virology 32, 383–394.
Clarke, B. E., Brown, A. L., Currey, K. M., Newton, S. E., Rowlands,
D. J. & Carroll, A. R. (1987). Potential secondary and tertiary structure in
the genomic RNA of foot-and-mouth disease virus. Nucleic Acids Research
15, 7067–7079.
Cui, T., Sankar, S. & Porter, A. G. (1993). Binding of encephalomyocarditis virus RNA polymerase to the 3h-noncoding region of the viral
RNA is specific and requires the 3h-poly(A) tail. Journal of Biological
Chemistry 268, 26093–26098.
Falk, M. M., Sobrino, F. & Beck, E. (1992). VPg gene amplification
correlates with infective particle formation in foot-and-mouth disease
virus. Journal of Virology 66, 2251–2260.
Gamarnik, A. V. & Andino, R. (1998). Switch from translation to RNA
replication in a positive-stranded RNA virus. Genes & Development 12,
2293–2304.
Goodfellow, I., Chaudhry, Y., Richardson, A., Meredith, J., Almond,
J. W., Barclay, W. & Evans, D. J. (2000). Identification of a cis-acting
replication element within the poliovirus coding region. Journal of
Virology 74, 4590–4600.
Gutie! rrez, A., Martı! nez-Salas, E., Pintado, B. & Sobrino, F. (1994).
Specific inhibition of aphthovirus infection by RNAs transcribed from
both 5h and 3h noncoding region. Journal of Virology 68, 7426–7432.
Harris, K. S., Xiang, W., Alexander, L., Lane, W. S., Paul, A. V. &
Wimmer, E. (1994). Interaction of poliovirus polypeptide 3CDpro with
the 5h and 3h termini of the poliovirus genome. Identification of viral and
cellular co-factors needed for efficient binding. Journal of Biological
Chemistry 269, 27004–27014.
Herold, J. & Andino, R. (2000 a). Functional circularization of a positivestranded RNA virus : a new model for the initiation of negative strand
RNA synthesis in poliovirus replication. Europic 2000 (Zagare Bay,
Gargano, Italy, 25–31 May 2000). Abstract G09.
Herold, J. & Andino, R. (2000 b). Poliovirus requires a precise 5h end for
efficient positive-strand RNA synthesis. Journal of Virology 74, 6394–
6400.
Ito, T. & Lai, M. M. C. (1999). An internal polypyrimidine-tract-binding
protein-binding site in the hepatitis C virus RNA attenuates translation,
which is relieved by the 3h-untranslated sequence. Virology 254, 288–296.
Ito, T., Tahara, S. M. & Lai, M. M. C. (1998). The 3h-untranslated region
of hepatitis C virus RNA enhances translation from an internal ribosomal
entry site. Journal of Virology 72, 8789–8796.
Kitching, R. P. (1999). Foot-and-mouth disease : current world situation.
Vaccine 26, 1772–1774.
Ku$ hn, R., Luz, N. & Beck, E. (1990). Functional analysis of the internal
translation initiation site of foot-and-mouth disease virus. Journal of
Virology 64, 4625–4631.
Kunkel, T. (1985). Rapid and efficient site specific mutagenesis without
phenotypic selection. Proceedings of the National Academy of Sciences, USA
82, 488–492.
Lobert, P. E., Escriou, N., Ruelle, J. & Michiels, T. (1999). A coding
RNA sequence acts as a replication signal in cardioviruses. Proceedings of
the National Academy of Sciences, USA 96, 11560–11565.
Lo! pez de Quinto, S. & Martı! nez-Salas, E. (1997). Conserved structural
motifs located in distal loops of aphthovirus internal ribosome entry site
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:41:10
FMDV 3h NCR is essential for replication
domain 3 are required for internal initiation of translation. Journal of
Virology 71, 4174–4175.
Lu, H. H. & Wimmer, E. (1996). Poliovirus chimeras replicating under
the translational control of genetic elements of hepatitis C virus reveal
unusual properties of the internal ribosomal entry site of hepatitis C virus.
Proceedings of the National Academy of Sciences, USA 93, 1412–1417.
McInerney, G. M., King, A. M., Ross-Smith, N. & Belsham, G. J. (2000).
Replication-competent foot-and-mouth disease virus RNAs lacking
capsid coding sequences. Journal of General Virology 81, 1699–1702.
McKnight, K. L. & Lemon, S. M. (1998). The rhinovirus type 14 genome
contains an internally located RNA structure that is required for viral
replication. RNA 4, 1569–1584.
Mellits, K. H., Meredith, J. M., Rohll, J. B., Evans, D. J. & Almond, J. W.
(1998). Binding of a cellular factor to the 3h untranslated region of the
RNA genomes of entero- and rhinoviruses plays a role in virus replication.
Journal of General Virology 79, 1715–1723.
Paul, A. V., Rieder, E., Kim, D. W. & Wimmer, E. (2000). Identification
of an RNA hairpin in poliovirus RNA that serves as a specific template for
the uridylylation of VPg in vitro. Europic 2000 (Zagare Bay, Gargano,
Italy, 25–31 May 2000). Abstract E04.
Pierangeli, A., Bucci, M., Pagnotti, P., Degener, A. M. & Perez Bercoff,
R. (1995). Mutational analysis of the 3’-terminal extra-cistronic region of
poliovirus RNA : secondary structure is not the only requirement for
minus-strand RNA replication. FEBS Letters 374, 327–332.
Pilipenko, E. V., Maslova, S. V., Sinyakov, A. N. & Agol, V. I. (1992).
Towards identification of cis-acting elements involved in the replication
of enterovirus and rhinovirus RNAs : a proposal for the existence of
tRNA-like terminal structures. Nucleic Acids Research 20, 1739–1745.
Richards, O. C. & Ehrenfeld, E. (1990). Poliovirus RNA replication.
Current Topics in Microbiology and Immunology 161, 89–119.
Rohll, J. B., Percy, N., Ley, R., Evans, D. J., Almond, J. W. & Barclay,
W. S. (1994). The 5h-untranslated regions of picornavirus RNAs contain
independent functional domains essential for RNA replication and
translation. Journal of Virology 68, 4384–4391.
Rohll, J. B., Moon, D. H., Evans, D. J. & Almond, J. W. (1995). The 3h
untranslated region of picornavirus RNA : features required for efficient
genome replication. Journal of Virology 69, 7835–7844.
Rose, J. K., Buonocore, L. & Whitt, M. A. (1991). A new cationic
liposome reagent mediating nearly quantitative transfection of animal
cells. Biotechniques 10, 520–525.
Rueckert, R. R. (1990). Picornaviridae and their replication. In Fields
Virology, 2nd edn, pp. 507–548. Edited by B. N. Fields & D. M. Knipe.
New York : Raven Press.
Ryan, M. D., Belsham, G. J. & King, A. M. (1989). Specificity of enzymesubstrate interactions in foot-and-mouth disease virus polyprotein
processing. Virology 173, 35–45.
Sangar, D. V., Rowlands, D. J., Harris, T. J. R. & Brown, F. (1977). A
protein covalently linked to foot-and-mouth disease virus RNA. Nature
268, 648–650.
Seechurn, P., Knowles, N. J. & McCauley, J. W. (1990). The nucleotide
sequence of a pathogenic swine vesicular disease virus. Virus Research 16,
255–274.
Sutmoller, P. (1992). Vesicular diseases, foreign animal diseases. In
Committee Report on Foreign Animal Diseases of the United States Animal
Health Association, pp. 368–382. Richmond, VA, USA.
Todd, S. & Semler, B. L. (1996). Structure-infectivity analysis of the
human rhinovirus genomic RNA 3h noncoding region. Nucleic Acids
Research 24, 2133–2142.
Todd, S., Nguyen, J. H. & Semler, B. L. (1995). RNA-protein
interactions directed by the 3h end of human rhinovirus genomic RNA.
Journal of Virology 69, 3605–3614.
Todd, S., Towner, J. S., Brown, D. M. & Semler, B. L. (1997).
Replication-competent picornavirus with complete genomic RNA 3h
noncoding region deletions. Journal of Virology 71, 8868–8874.
Waggoner, S. & Sarnow, P. (1998). Viral ribonucleoprotein complex
formation and nucleolar-cytoplasmic relocalization of nucleolin in
poliovirus-infected cells. Journal of Virology 72, 6699–6709.
Zibert, A., Maas, G., Strebe, l. K., Falk, M. M. & Beck, E. (1990).
Infectious foot-and-mouth disease virus derived from a cloned full-length
cDNA. Journal of Virology 64, 2467–2473.
Received 27 July 2000 ; Accepted 11 September 2000
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:41:10
BAB