lsevier.com/locate/yviro

Virology 343 (20

Role of double-stranded RNA and Npro of classical swine fever

virus in the activation of monocyte-derived dendritic cells

Oliver Bauhofer, Artur Summerfield, Kenneth C. McCullough, Nicolas Ruggli *

Institute of Virology and Immunoprophylaxis (IVI), Sensemattstrasse 293, CH-3147 Mittelhausern, Switzerland

Received 13 May 2005; returned to author for revision 25 June 2005; accepted 12 August 2005

Available online 8 September 2005

Abstract

Classical swine fever virus (CSFV) is a noncytopathogenic (ncp) positive-sense RNA virus that replicates in myeloid cells including

macrophages and dendritic cells (DC). The virus does not induce type I interferon (IFN-a/h), which in macrophages has been related to the

presence of the viral Npro gene. In the present work, the role of viral double-stranded (ds)RNA and Npro in the virus–host cell interaction has been

analyzed. Higher levels of detectable dsRNAwere produced by a genetically engineered cytopathogenic (cp) CSFV compared with ncp CSFV, and

cp CSFV induced IFN-a/h in PK-15 cells. With DC, there was only a small difference in the levels of dsRNA between the cp and ncp viruses, and

no IFN-a/h was produced. However, the cp virus induced a higher degree of DC maturation, in terms of CD80/86 and MHC II expression. Npro

deletion mutants induced an increase in DC maturation and IFN-a/h production–for both ncp and cp viruses–despite reduced replication

efficiency in the DC. Deletion of Npro did not influence dsRNA levels, indicating that the interference was downstream of dsRNA turnover

regulation. In conclusion, the capacity of CSFV to replicate in myeloid DC, and prevent IFN-a/h induction and DC maturation, requires both

regulated dsRNA levels and the presence of viral Npro.

D 2005 Elsevier Inc. All rights reserved.

Keywords: Pestivirus; Classical swine fever virus; Npro; Double-stranded RNA; Dendritic cell; Interferon-a/h; CD80/86; MHC II; Innate immunity

Introduction

Classical swine fever virus (CSFV) is an enveloped positive-

sense RNAvirus that belongs to the genus Pestivirus within the

Flaviviridae family (Fauquet et al., 2005). The virus is highly

contagious for pigs and wild boar and has a worldwide

distribution. Depending on the CSFV strain, the outcome of

an infection can vary from severe hemorrhagic symptoms with

high mortality rate to a mild and almost unapparent disease. The

other members of the Pestivirus genus are the bovine viral

diarrhea virus (BVDV) and the border disease virus (BDV) that

cause disease in cattle and sheep, respectively. Pestiviruses are

closely related to the flaviviruses and hepaciviruses, two

important groups of human pathogens of the Flaviviridae

family (Fauquet et al., 2005). The genome of the pestiviruses

0042-6822/$ - see front matter D 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.virol.2005.08.016

* Corresponding author. Fax: +41 31 848 9222.

E-mail addresses: oliver.bauhofer@ivi.admin.ch (O. Bauhofer),

artur.summerfield@ivi.admin.ch (A. Summerfield),

kenneth.mccullough@ivi.admin.ch (K.C. McCullough),

nicolas.ruggli@ivi.admin.ch (N. Ruggli).

consists of a single RNA molecule of positive polarity. It

contains one open reading frame (ORF) flanked by two non-

translated regions (NTR). An internal ribosome entry site

(IRES) located in the 5V NTR mediates cap-independent

translation initiation of a polyprotein. Co- and post-translational

processing by cellular and viral proteases leads to the proteins

Npro, C, Erns, E1, E2, p7, NS2, NS3, NS4A, NS4B, NS5A, and

NS5B (for review, see Lindenbach and Rice, 2001). Npro is

unique to pestiviruses within the Flaviviridae and exhibits

autoprotease activity resulting in rapid cleavage from the

nascent polyprotein (Rumenapf et al., 1998; Stark et al.,

1993; Wiskerchen et al., 1991). So far, Npro is the only gene

known to be non-essential for the pestivirus life cycle (Lai et al.,

2000; Tratschin et al., 1998). While mutant CSFV lacking the

Npro gene (DNpro CSFV) are replication-competent in cell

culture, they have lost their capacity to interfere with cellular

antiviral defense mechanisms, i.e. double-stranded (ds) RNA-

mediated apoptosis and interferon (IFN)-a/h induction (Ruggli

et al., 2003). In fact, DNpro CSFV, unlike wild-type virus,

induce IFN-a/h in macrophages and PK-15 cells (Ruggli et al.,

2003). DNpro CSFV are also avirulent in pigs but can induce a

05) 93 – 105

www.e

O. Bauhofer et al. / Virology 343 (2005) 93–10594

protective immune response (Mayer et al., 2004). Npro of CSFV

is an antagonist of dsRNA-mediated IFN-a/h induction

independently of the CSFV context (La Rocca et al., 2005;

Ruggli et al., in press). A recent report suggests that Npro of

BVDV is also involved in suppressing interferon-dependent

antiviral responses (Horscroft et al., 2005).

CSFV is a monocytotropic virus that replicates in dendritic

cells (DC) without inducing cell activation (Carrasco et al.,

2004). This is important considering that DC are central players

in the induction of early immune responses against pathogens.

One process employed by DC is activation triggered by

pathogen associated molecular patterns (PAMP). For RNA

viruses, an important PAMP inducing host response is the

dsRNA of replicative intermediates and replicative forms

produced during genome replication (for review, see Good-

bourn et al., 2000; Jacobs and Langland, 1996). Such dsRNA is

recognized in the cell by receptors such as toll-like receptor

(TLR)3 (Alexopoulou et al., 2001; Matsumoto et al., 2004), the

retinoic acid inducible gene I (RIG-I) product (Yoneyama et al.,

2004), the melanoma differentiation-associated gene 5 (mda-5)

product (Andrejeva et al., 2004; Kang et al., 2002), the

dsRNA-dependent protein kinase (PKR) and by the coupled

2V–5V oligoadenylate synthetase (2–5 OAS)/RNaseL system

(Gil and Esteban, 2000; Player and Torrence, 1998).

Although CSFV is generally noncytopathogenic (ncp) in cell

culture and in DC, cytopathogenic (cp) biotypes have been

isolated from infected animals (Aoki et al., 2001; Kosmidou et

al., 1998; Meyers and Thiel, 1995) or arose spontaneously after

multiple passages in cell culture (Meyers and Thiel, 1995;

Mittelholzer et al., 1997). With few exceptions (Aoki et al.,

2004; Gallei et al., 2005; Hulst et al., 1998), the cytopathic

effect (cpe) observed so far with cp CSFV correlated with the

presence of defective interfering particles (DI) associated with

the wild-type helper virus. This DI consists of the 5V NTR, thecodon for methionine, the genes NS3 to NS5B and the 3V NTR.The CSFV DI was shown to function as cp replicon

independently of the helper virus, establishing that the genes

Npro to NS2 were non-essential for pestivirus RNA replication

in cell culture (Behrens et al., 1998; Moser et al., 1999; Tautz et

al., 1999). In addition, cp CSFV could be rescued in cell culture

by transfection of ncp CSFV-infected cells with DI RNA

(Meyers et al., 1996) or by co-transfection of in vitro transcripts

of the cp replicon and the helper virus (Moser et al., 1999). For

CSFV, the significance of the cp biotype in pathogenesis is

probably minor (Aoki et al., 2003; Kosmidou et al., 1998), in

contrast to BVDV for which the cp biotypes are associated with

mucosal disease (Kummerer et al., 2000; Tautz et al., 1998). An

interesting observation is that cp pestiviruses produce a

significantly higher amount of total viral RNA compared with

the corresponding ncp biotype (Kupfermann et al., 1996;

Lackner et al., 2004; Mendez et al., 1998; Mittelholzer et al.,

1997; Vassilev and Donis, 2000). However, the double-stranded

form of the viral RNA has never been specifically analyzed.

The present study aimed to characterize the role of viral

dsRNA and Npro in CSFV-mediated innate immune activation

in vitro. Induction of IFN-a/h was analyzed by comparing the

IFN-a/h-competent PK-15 cell line with the SK-6 cell line that

cannot be stimulated to produce IFN-a/h (Ruggli et al., 2003).

Porcine monocyte-derived DC were employed to analyze the

role of viral dsRNA and Npro in DC activation and maturation.

Both ncp and cp CSFV biotypes were used as models of low

and high dsRNA-producing viruses, and the role played by

Npro was tested with DNpro mutants of ncp and cp CSFV.

Results

Cp CSFV produces higher amounts of dsRNA than ncp CSFV

Viral dsRNA is a well-known trigger of innate immune

activation. Previous studies demonstrated that cp pestiviruses

produced higher amounts of total viral RNAwhen compared to

the corresponding ncp pestivirus. However, the double-stranded

fraction of RNA occurring during replication of ncp and cp

CSFV has never been identified and quantified so far. Therefore,

we analyzed dsRNA in CSFV-infected cells by using the

monoclonal antibody (mAb) J2. Detection of dsRNA in

CSFV-infected SK-6 cells by immunofluorescence provided

first evidence that cp CSFV produced more dsRNA than its ncp

parent (Fig. 1A). While the ncp CSFV-infected cells exhibited

faint dsRNA dots (Fig. 1A, panel 4), the cp CSFV-infected cells

contained a high density of large bright fluorescent spots (Fig.

1A, panel 6). Immunostaining of the viral glycoprotein E2

resulted in a comparable signal for both ncp and cp CSFV (Fig.

1A, panels 3 and 5). The specificity of the mAb J2 for CSFV

dsRNA was assessed in a dot blot assay using in vitro

synthesized nucleic acids. For this purpose, serial dilutions of

plasmid DNA encoding the full-length genome (pA187-1) and

of in vitro transcribed CSFV single-stranded (ss)RNA and

dsRNA were dot-blotted onto a nitrocellulose membrane and

subjected to immunodetection as described. With identical

amounts of nucleic acids, a 400-fold stronger signal was

detected for CSFV dsRNA than for CSFV ssRNA, whereas

plasmid DNA was barely detectable even at the highest

concentration (Fig. 1B). The standard curve obtained with serial

dilutions of in vitro synthesized CSFV dsRNAwas then used to

quantify dsRNA in extracts from mock, ncp and cp CSFV-

infected SK-6 cells. Nucleic acid extracts were serially diluted

and dot-blotted onto a nitrocellulose membrane (Fig. 1C).

Quantification revealed a 5- to 10-fold higher amount of dsRNA

synthesized in cp CSFV-infected cells when compared to the ncp

CSFV. This represents the first demonstration that cp CSFV

produces higher levels of dsRNA than ncp CSFV, which may

result in differential activation of the innate immune response.

The amount of dsRNA produced by cp CSFV is dependent on

the cell type

We showed previously that the porcine cell lines SK-6 and

PK-15 display a different response to stimulation by dsRNA

(Ruggli et al., 2003, in press). Therefore, we analyzed dsRNA

in these two cell lines in comparison with primary porcine

monocyte-derived DC. For this purpose, flow cytometry

(FCM) was evaluated as an alternative to the dot blot assay

for the quantification of dsRNA levels produced upon CSFV

Fig. 1. Detection of dsRNA in mock- and CSFV-infected cells. (A) SK-6 cells were mock-treated (1, 2) or infected with ncp CSFV (3, 4) or with cp CSFV (5, 6) at an

m.o.i. of 2 TCID50/cell. At 16 h p.i., cells were fixed and analyzed by confocal microscopy using the E2-specific mAb HC/TC26 (1, 3, 5) or the dsRNA-specific

mAb J2 (2, 4, 6). The cytoskeleton was detected with fluorescent-conjugated phalloidin. (B) The specificity of the mAb J2 for dsRNAwas tested in a dot blot assay.

10-fold serial dilutions of dsRNA, ssRNA and plasmid DNAwere dot-blotted onto a nitrocellulose membrane. Detection was performed according to Materials and

methods. (C) dsRNA from mock-, ncp CSFV- or cp CSFV-infected SK-6 cells was quantified in a dot blot assay. Nucleic acid extracts obtained 16 h p.i. were diluted

as indicated and blotted onto a nitrocellulose membrane. dsRNA was detected according to Materials and methods and quantified by comparison with a standard

curve based on in vitro synthesized dsRNA. Amounts of dsRNA are indicated in nanograms for each dot.

O. Bauhofer et al. / Virology 343 (2005) 93–105 95

infection. First, SK-6 cells were either mock-infected or

infected with ncp or cp CSFV and monitored for dsRNA

levels by FCM. The difference of dsRNA fluorescence

intensity between ncp and cp CSFV was comparable to that

observed with the dot blot assay (Fig. 2A). Therefore, all

following analyses were carried out by FCM. When SK-6 cells,

PK-15 cells and DC were compared, the differences in dsRNA

levels between ncp and cp CSFV varied considerably. In SK-6

cells that do not produce IFN-a/h, infection with cp CSFV

resulted in the highest dsRNA production. In PK-15 cells,

which are capable of producing IFN-h, the amount of viral

dsRNA was lower compared to SK-6 cells. In DC, the

difference in dsRNA synthesis between ncp and cp CSFV

was the smallest (Fig. 2A). A statistical analysis of the

geometric mean fluorescence intensity (MFI) of two indepen-

dent experiments using ANOVA revealed a significant differ-

ence for the dsRNA levels between mock-treated cells and cells

infected with either ncp or cp CSFV for all the cell types (P <

0.004). The difference between ncp and cp CSFV was also

significant in SK-6 and PK-15 cells (P < 0.004) and in DC

(P < 0.005) (Fig. 2B). Interestingly, the cpe caused by cp CSFV

was strong in SK-6 cells, less pronounced in PK-15 cells (Fig.

2C) and absent in DC (data not shown). The cpe appeared in

both SK-6 and PK-15 cells 16 h post-infection (p.i.) and

resulted in complete lysis of the SK-6 monolayer 40 h p.i.,

whereas the PK-15 cells recovered (data not shown). No

evidence of cell death could be detected in DC at any time (data

not shown), despite virus replication detected by NS3 expres-

sion (Fig. 2C). Taken together, these observations show cell-

dependent differences in the control of virus replication and of

viral dsRNA levels.

Cp CSFV induces IFN-a/b in PK-15 cells but not in DC

As dsRNA is known to induce synthesis of IFN-a/h, weasked whether the level of dsRNA was related to the degree of

IFN-a/h induction. These experiments were performed with

PK-15 and DC cultures since it was observed previously that

no IFN-a/h could be induced in SK-6 cells (Ruggli et al.,

2003). Supernatants of cells infected with cp CSFV, ncp CSFV

and mock were tested for bioactive IFN-a/h after virus

neutralization, using the MxCAT reporter gene assay (Fig. 3).

O. Bauhofer et al. / Virology 343 (2005) 93–10596

Fig. 3. Analysis of IFN-a/h secretion induced by cp CSFV. PK-15 cells and DC

were treated with mock SK-6 cell extract, ncp CSFV and with 5-fold serial

dilutions of cp CSFV. The undiluted virus and the ncp CSFV corresponded to

an m.o.i. of 2 TCID50/cell for PK-15 and 5 TCID50/cell for DC. Supernatants

were collected 16 h p.i. for analysis of IFN-a/h bioactivity. The error bars

represent the 95% confidence interval (n = 5).

O. Bauhofer et al. / Virology 343 (2005) 93–105 97

In accordance with previous data (Carrasco et al., 2004; Ruggli

et al., 2003), ncp CSFV did not induce any detectable IFN-a/h,neither in PK-15 cells nor in DC. In PK-15 cells, cp CSFV

induced IFN-a/h in a non-linear, multiplicity of infection

(m.o.i.)-dependent manner. Interestingly, the infection of DC

with cp CSFV did not induce any detectable interferon

synthesis, which may relate to the only slight increase of

dsRNA when compared with ncp CSFV (see Fig. 2).

Cp CSFV induces DC maturation

Next, we investigated the maturation level of ncp and cp

CSFV-infected DC in order to determine whether these cells

respond differently to the two biotypes. For this purpose, the

level of CD80/86 and MHC II molecules on the cell surface

was monitored by FCM. Cells infected with cp CSFV

expressed higher levels of MHC II and CD80/86 molecules

(Fig. 4A). The differences of the geometrical MFI values

analyzed using ANOVA were statistically significant (P <

0.004 for CD80/86, P < 0.03 for MHC II). The infection with

ncp CSFV did not result in a statistically significant upregula-

tion of MHC II and CD80/86 when compared to mock,

although the raw data indicated a slight increase of the

expression levels. In conclusion, while viral protein expression

in DC did not differ between cp and ncp CSFV as measured by

the levels of Npro and NS3 (Fig. 4B), cp CSFV induced

maturation of DC without detectable IFN-a/h production.

Maturation of DC and IFN-a/b secretion can be triggered by in

vitro synthesized CSFV dsRNA

Since we demonstrated activation of DC by cp CSFV with a

small increase of dsRNA levels and without detectable

Fig. 2. Quantification of dsRNA in mock-, ncp and cp CSFV-infected SK-6 cells, PK

mock, ncp CSFVor cp CSFVat anm.o.i. of 2 (SK-6 and PK-15) respectively 5 TCID50

by FCM. (B) The geometric mean fluorescence intensity (MFI) values obtained in t

submitted to statistical analysis usingANOVAon ranks. TheP values obtained for the

parallel cultures of SK-6 and PK-15 cells were subjected to immunoperoxidase stain

IFN-a/h secretion, the next experiments were aimed at

determining the activation potential of in vitro synthesized

CSFV dsRNA. CSFV dsRNA obtained in vitro was not

replication-competent (data not shown) and was compared

with in vitro transcribed non-replicating CSFV ssRNA and

with plasmid DNA carrying the full-length CSFV genome

sequence as control. DC were lipofected with CSFV DNA,

ssRNA or dsRNA and monitored by FCM for CD80/86 and

MHC II upregulation. Transfection of DC with dsRNA resulted

in a significant increase in MHC II levels (P < 0.005) (Fig. 4C).

CD80/86 expression levels were not increased with dsRNA

and only slightly increased with plasmid DNA (Fig. 4C). The

survival rate of DC after lipofection of dsRNA was reduced

when compared to transfection with plasmid DNA and with

ssRNA (data not shown). Interestingly, transfection of DC with

synthetic CSFV dsRNA resulted in secretion of IFN-a/h(Fig. 4D), in contrast to the infection with cp CSFV (Fig. 3).

The induction was specific for dsRNA and was not due to the

transfection process since ssRNA and plasmid DNA did not

induce IFN-a/h secretion.

DNpro CSFV induces IFN-a/b secretion and DC maturation

Similar to cp CSFV in PK-15 cells (Fig. 3), DNpro CSFV is

an inducer of IFN-a/h in these cells and in macrophages

(Ruggli et al., 2003). The fact that DNpro CSFV is ncp suggests

that there is no direct link between cytopathogenicity and IFN-

a/h induction. To elaborate on this, we analyzed IFN-a/hsecreted by DC in response to infection with DNpro CSFV and

monitored CD80/86 and MHC class II expression. DC

produced IFN-a/h in response to infection with DNpro CSFV

(Fig. 5A) in absence of any detectable cpe. Furthermore, DNpro

CSFV induced an upregulation of CD80/86 and of MHC class

II molecules on DC (Fig. 5B). The difference in the CD80/86

upregulation between DNpro CSFV and ncp CSFV analyzed

using ANOVA was statistically significant (P < 0.008). The

increased activation in absence of Npro occurred despite slight

reduced levels of viral protein expression as measured by NS3

(Fig. 5C). This confirms that cytopathogenicity of CSFV is not

a prerequisite for DC activation and suggests that Npro is

required for the prevention of such activation.

DNpro CSFV- and ncp CSFV-infected cells produce compara-

ble amounts of dsRNA

We have shown above that IFN-a/h induction in PK-15

cells and activation of DC correlated with increased levels of

dsRNA observed with cp CSFV but that cytopathogenicity was

not a prerequisite for DC activation. Since DNpro CSFV

induced IFN-a/h and activated DC without being cytopatho-

genic, we analyzed whether Npro was involved in the regulation

-15 cells and DC. (A) SK-6 cells, PK-15 cells and DC were infected either with

/cell (DC). At 16 h (SK-6, PK-15) and 48 h (DC) p.i., dsRNA levels were analyzed

wo independent experiments (n = 10 for SK-6 and PK-15, n = 6 for DC) were

differences between each group are indicated. (C) At the time of dsRNAdetection,

ing for viral NS3. For DC, analysis of NS3 expression was performed by FCM.

Fig. 4. Role of cp CSFVand viral dsRNA in DC activation and IFN-a/h secretion. (A) CD80/86 and MHC II upregulation was measured in DC after infection with

ncp and cp CSFV at an m.o.i. of 5 TCID50/cell. (B) The Npro and NS3 expression levels in mock (lane 1)-, ncp CSFV (lane 2)- and cp CSFV-infected DC (lane 3)

were analyzed by SDS-PAGE and Western blotting. (C) CD80/86 and MHC II upregulation was measured in DC and after transfection with in vitro synthesized ss

and dsRNA of CSFV. Transfection of plasmid DNA coding for the full-length CSFV genome, mock transfection and mock-transfected DC supplemented with 1000

U/ml porcine IFN-a and 20 ng/ml porcine TNF (mock + IFN/TNF) served as controls. (D) IFN-a/h secretion was analyzed 48 h post-transfection. In panel (A), the

asterisk represents a statistical significant upregulation of CD80/86 (P < 0.002, n = 6) and of MHC II (P < 0.035, n = 6). In panel (C), the upregulation of MHC II

expression in DC transfected with dsRNAwas statistically significant as shown by the asterisk (P < 0.005, n = 3) when compared with mock, DNA and ssRNA. For

all the graphs, mean values are depicted. The error bars represent the 95% confidence interval.

O. Bauhofer et al. / Virology 343 (2005) 93–10598

of dsRNA levels during virus infection. For this purpose, we

quantified dsRNA induced by DNpro CSFV in comparison with

ncp CSFV in the three different cell types. There was no

statistical difference in the amounts of dsRNA between ncp and

DNpro CSFV in SK-6 and PK-15 cells (Figs. 6A and 6B). In

DC infected with DNpro CSFV, one population of cells

displayed a dsRNA-dependent fluorescence intensity compa-

rable with ncp CSFV-infected cells, and a second population of

cells showed a decreased dsRNA fluorescence at the level of

non-infected cells (Fig. 6A). This could be explained by the

fact that the replication of DNpro CSFV was impaired in DC

when compared with ncp CSFV. At the same m.o.i., only 40%

of the DC infected with DNpro CSFV expressed NS3, whereas

with ncp CSFV approximately 90% of the cells were NS3

positive at the same time after infection (Fig. 6C). From these

experiments, we conclude that IFN-a/h induction in PK-15

cells and activation of DC by DNpro CSFV are not due to

increased dsRNA accumulation.

The lack of Npro in cp CSFV enhances the cp CSFV-mediated

IFN-a/b production and MHC II upregulation

The observation that DNpro-mediated IFN-a/h production

was apparently not induced by higher levels of viral dsRNA

indicates that Npro regulates IFN-a/h induction downstream

of dsRNA. According to this, cp CSFV lacking the Npro gene

should display enhanced IFN-a/h induction and DC activa-

tion. In order to test this hypothesis, we rescued a cp DNpro

CSFV and analyzed virus replication, DC activation and IFN-

a/h induction. SK-6 cells, PK-15 cells and DC were either

mock-infected or infected with ncp CSFV, cp CSFV, DNpro

CSFV or cp DNpro CSFV. All the viruses replicated with

comparable efficiency in SK-6 cells but differed in IFN-a/h-competent cells such as PK-15 cells and DC. The strongest

reduction of virus replication was observed with cp DNpro

CSFV in both, PK-15 cells and DC (Fig. 7A). When IFN-a/hbioactivity was measured, cp DNpro CSFV infection resulted

in a 1.5-to 5-fold increase in IFN-a/h synthesis in PK-15

cells and in DC respectively, when compared with cp CSFV

(Fig. 7B). This was particularly remarkable with DC, in

which cp DNpro CSFV replication was not detectable when

measured by NS3 expression (Fig. 7A). Analysis of DC

activation did not reveal any significant difference in the

levels of CD80/86 with either virus tested (Fig. 7C).

However, for cp DNpro CSFV, expression of MHC II

molecules was increased by two-fold when compared to

either cp CSFV or DNpro CSFV (Fig. 7C). Taken together,

these experiments show that increased amounts of dsRNA

Fig. 5. IFN-a/h secretion and DC activation after infection with DNpro CSFV.

DC were treated with mock SK-6 cell extract, ncp CSFV and with 5-fold serial

dilutions of DNpro CSFV. The undiluted DNpro CSFV and the ncp CSFV

corresponded to an m.o.i. of 5 TCID50/cell. (A) At 48 h p.i., supernatants were

collected for analysis of IFN-a/h bioactivity. (B) The respective cells infected

with undiluted virus stocks or mock were analyzed for CD80/86 and MHC

class II expression by FCM. The difference between CD80/86 levels induced

by ncp CSFV and DNpro CSFV was statistically significant as shown by the

asterisk (P < 0.008, n = 6). The error bars represent the 95% confidence

interval. (C) Expression of the viral proteins Npro and NS3 in DC 48 h after

infection with mock (line 1), ncp CSFV (line 2) and DNpro CSFV (line 3) was

analyzed by SDS-PAGE and Western blotting.

O. Bauhofer et al. / Virology 343 (2005) 93–105 99

and the lack of Npro synergistically contribute to the

activation of DC and IFN-a/h induction.

Discussion

For immunological sensing of virus infections, DC play a

major role through PAMP receptors such as the TLRs. The

development of memory responses against antigen has been

demonstrated to depend on TLR stimulation (Iwasaki and

Medzhitov, 2004). Many viruses have successfully evolved

mechanisms to evade detection by the immune system,

including the capacity to prevent immune activation and

permit their replication in IFN-a/h-competent cells (for

selected reviews, see Basler and Garcia-Sastre, 2002; Good-

bourn et al., 2000; Tortorella et al., 2000; Weber et al.,

2004). Recently, we showed that CSFV has the capacity to

replicate in DC without inducing their activation (Carrasco et

al., 2004). In order to further elaborate on this observation,

the present study analyzed the interaction of CSFV with

monocyte-derived DC. Of particular interest were the dsRNA

intermediates formed by RNA viruses during their replica-

tion, clear candidates to trigger IFN-a/h production (for

review, see Goodbourn et al., 2000; Jacobs and Langland,

1996). In this same vein, it is known that cp pestiviruses

induce higher RNA levels than their ncp counterparts

(Kupfermann et al., 1996; Lackner et al., 2004; Mendez et

al., 1998; Mittelholzer et al., 1997; Vassilev and Donis,

2000), but it was not known whether this is also true for the

double-stranded forms of the RNA.

The present work demonstrates that cp CSFV produces

elevated amounts of dsRNA, compared with the ncp biotype,

but with variations dependent on the cell type infected. SK-6

cells accumulated more dsRNA than PK-15 cells and DC. In

fact, the latter showed only a minor difference in dsRNA levels

between the cp and ncp biotypes. With the PK-15 cells, this

might be reflecting the induction of IFN-a/h by the cp CSFV.

In DC, although there was no IFN-a/h detectable in the

supernatant, the increased expression of CD80/86 and MHC

class II after cp CSFV infection demonstrated DC activation,

suggesting that also in these cells an antiviral state controlling

dsRNA accumulation would be induced. Interestingly, cyto-

pathogenicity was observed only in cell cultures where high

amounts of dsRNA were produced, the cp CSFV being

cytopathic for SK-6 and PK-15 cells, but not for DC. Similar

results have been reported in terms of cp BVDV being

noncytopathogenic in bovine DC (Glew et al., 2003), although

these were induced to produce IFN-a/h.Honda et al. demonstrated that IFN-a/h signaling is required

for dsRNA-induced DCmaturation in vitro (Honda et al., 2003).

This would appear to be in contradiction with the observation

that cp CSFV upregulates CD80/86 and MHC II molecules

without detectable IFN-a/h bioactivity. Nevertheless, low levels

of IFN-a/h would quickly become assimilated by the DC and

thus be undetectable. Certainly, dsRNA can induce IFN-a/h in

DC as we demonstrate that in vitro synthesized CSFV-derived

dsRNA transfected into DC induced IFN-a/h. Along this line isour recent report that the activation of DC by transfected

synthetic mRNA molecules was dependent on secondary

dsRNA structures (Ceppi et al., 2005). Furthermore, a recent

study demonstrated that the addition of cationic lipids prolon-

gates the retention time of nucleic acids in the endosomal

compartment in DC and thus enhances the IFN response (Honda

et al., 2005).

DC activation by dsRNA can be mediated via TLR3 (for

review, see Matsumoto et al., 2004), but, with RNA virus

infections, it is necessary to consider TLR7 and TLR8 as

receptors for ssRNA (Barchet et al., 2005; Diebold et al.,

2004; Heil et al., 2004; Lund et al., 2004). However, the

monocyte-derived DC used in the present study do not

respond to in vitro synthesized CSFV ssRNA, supporting that

CSFV dsRNA is the main trigger for myeloid DC activation

O. Bauhofer et al. / Virology 343 (2005) 93–105100

Fig. 7. Virus replication, IFN-a/h induction and activation of DC infected with cp DNpro CSFV. SK-6, PK-15 and DC were infected with mock, ncp CSFV, DNpro

CSFV, cp CSFVor cp DNpro CSFVat an m.o.i. of 1 (SK-6 and PK-15) respectively 0.5 TCID50/cell (DC) and analyzed 16 h (SK-6 and PK-15) and 48 h (DC) p.i. (A)

The percentage of infected cells was assessed for the different viruses by measuring the percentage of cells expressing the viral protease NS3 by FCM. (B) The IFN-

a/h bioactivity was measured in the supernatant of infected cells 16 h p.i. For all the graphs (except for PK-15 in panel B, single values), mean values obtained from

2 independent experiments (n = 6) are shown, with error bars representing the 95% confidence interval. (C) At 48 h p.i., DC were harvested and analyzed for CD80/

86 and MHC class II expression.

O. Bauhofer et al. / Virology 343 (2005) 93–105 101

observed with cp CSFV. Human monocyte-derived DC also

lack responsiveness to ssRNA, in contrast to plasmacytoid

DC (reviewed in Liu, 2005).

Having identified the importance of dsRNA in triggering

myeloid DC, the question remained concerning the role of the

viral Npro. Previous work has shown Npro to be important for

counteracting cellular antiviral responses such as IFN-a/hsecretion (Horscroft et al., 2005; La Rocca et al., 2005; Ruggli

et al., 2003, in press). Although the absence of Npro did not

impair viral replication in SK-6 cells, DNpro CSFV was

incapable of efficient replication in DC. These mutants induced

IFN-a/h production by the DC, which would explain the

limitation of the virus replication, similar to that reported with

infected macrophages and PK-15 cells (Ruggli et al., 2003).

Horscroft and co-workers also reported that BVDV replicons

Fig. 6. Quantification of dsRNA in DNpro CSFV-infected cells. (A) SK-6 cells, PK-1

of 2 (SK-6 and PK-15) respectively 5 TCID50/cell (DC). At 16 h (SK-6, PK-15) and

MFI obtained in two independent experiments (n = 6) was submitted to statistica

between each group are indicated. (C) At the time of dsRNA detection, parallel cult

viral NS3. For DC, analysis of NS3 expression was performed by FCM.

lacking the Npro gene failed to establish stable replication in

Huh-7 cells (Horscroft et al., 2005). It is still unclear if this

antiviral state is due to IFN-a/h activity. What is known is that

CSFV replication in DC is impaired by IFN-a treatment (data

not shown).

Recently, Lackner and co-workers showed that the temporal

modulation of NS2–3 processing by the NS2 autoprotease is

crucial in RNA replication control and that the intracellular

level of NS3 strictly correlates with the efficiency of RNA

replication (Lackner et al., 2004). Whether these proteins

regulate the dsRNA levels remains to be established. Although

Npro can be implicated in controlling IFN-a/h induction, our

results show that Npro is not regulating the dsRNA turnover. It

is more likely that Npro is inhibiting the cellular response to

dsRNA. Using a virus engineered to combine elevated dsRNA

5 cells and DC were infected with mock, ncp CSFVor DNpro CSFVat an m.o.i.

48 h (DC) p.i., cells were fixed and dsRNA levels analyzed by FCM. (B) The

l analysis using ANOVA on ranks. The P values obtained for the differences

ures of SK-6 and PK-15 cells were subjected to immunoperoxidase staining for

O. Bauhofer et al. / Virology 343 (2005) 93–105102

levels in absence of Npro (cp DNpro CSFV), an additive effect

on IFN-a/h induction and DC activation was observed.

The present work has given rise to a model of CSFV–host

interaction governing virus evasion of host cell responses.

Regulation of CSFV RNA turnover with minimal accumulation

of dsRNA on one hand and Npro as a viral inhibitor of innate

immune activation on the other hand are key elements. They

account for a successful initiation of virus replication in the

primary target cells (monocytic cells), allowing dissemination

in the host. This is a particularly critical event for a virus

targeting cells of the immune system, especially highly

endocytic cells such as macrophages and dendritic cells. By

such means, CSFV not only replicates in IFN-producing cells,

but also controls IFN induction and DC maturation. Another

viral protein – Erns – may well be involved due to its proposed

control of dsRNA-mediated induction of IFN-h (Iqbal et al.,

2004). Further studies are required to understand how the

dsRNA turnover is regulated and what mechanisms are

involved in Npro-induced control of innate immune activation.

Such a detailed characterization will permit the understanding

of the early phases of CSFV pathogenesis.

Materials and methods

Cells

The swine kidney cell line SK-6 (Kasza et al., 1972) was

propagated in Earl’s minimal essential medium containing 7%

horse serum (GibcoBRL). The porcine kidney cell line PK-15

(American Type Culture Collection, Manassas, VA, USA) was

maintained in Dulbecco’s modified eagle medium (DMEM)

supplemented with non-essential amino acids, 1 mM Na-

Pyruvat and 5% horse serum (GibcoBRL). Porcine DC were

prepared as described elsewhere (Carrasco et al., 2001).

Briefly, peripheral blood mononuclear cells were isolated from

specific pathogen-free pigs bred at our institute. SWC3 positive

monocytes were sorted using the MACS system (Miltenyi

Biotec) and differentiated to DC by incubation for 4 days at 39

-C in DMEM with 10% porcine serum (Sigma) in the presence

of 100 U/ml recombinant porcine IL-4 and 150 ng/ml

recombinant porcine GM-CSF (kindly provided by Dr. Shigeki

Inumaru, Institute for Animal Health, Ibaraki, Japan). The IL-4

was prepared in our laboratory as described previously

(Carrasco et al., 2001).

Viruses

The viruses vA187-1 and vA187-DNpro were rescued by

electroporation of SK-6 cells with in vitro transcribed RNA

from the plasmid cDNA clones pA187-1 (Ruggli et al., 1996)

and pA187-DNpro (Ruggli et al., 2003) as previously described

(Moser et al., 1999). Cp CSFV was obtained by co-transfection

of SK-6 cells with in vitro transcribed RNA from plasmids

pA187-1 and pA187-D4764iv (Moser et al., 1999) at a ratio of

1:4. For the recovery of cp DNpro CSFV, SK-6 cells were co-

transfected with in vitro transcribed RNA from plasmids

pA187-DNpro and pA187-D4764iv at a ratio of 1:10. All the

virus titers were determined on SK-6 cells by standard end-

point dilution technique and were expressed as 50% tissue

culture infectious doses (TCID50)/ml. The m.o.i. used with SK-

6 and PK-15 cells was calculated from the titer obtained with

the respective cell line. Titration on DC was not performed, and

the m.o.i. was calculated based on the SK-6 titer.

In vitro synthesis of CSFV dsRNA

CSFV dsRNA was obtained by annealing of in vitro

transcribed full-length positive and negative-sense viral

RNA. For this purpose, plasmid pA187-1 was modified to

allow in vitro run-off transcription of negative-sense CSFV

RNA. A phosphorylated DNA cassette obtained by annealing

of primers T7-XhoI-L (5V-TATAGTGAGTCGTATTAC) and

T7-XhoI-R (5V-TCGAGTAATACGACTCACTATA) was ligat-ed between the SrfI and XhoI sites of pA187-1, resulting in

plasmid pA187-2xT7. Then, the SalI to ClaI fragment of

plasmid pA187-2xT7 containing the 5V-terminal T7 promoter

and the 5V end of the genome was replaced by the SalI to ClaI

fragment of plasmid pA187T3-SnaBI (Moser C., unpublished)

containing a unique SnaBI restriction site for linearization at

the precise 5V end of the genome. The new plasmid was called

pA187-5V-SnaBI. In vitro transcription of RNA was performed

with the MEGAscript T7 kit (Ambion) as described elsewhere

(Moser et al., 1999). Full-length run-off transcripts were

synthesized in independent reactions from SrfI-linearized

plasmid pA187-1 for positive-sense RNA and from SnaBI-

linearized plasmid pA187-5V-SnaBI for negative-sense RNA.

Subsequently, equimolar amounts of both RNA strands were

mixed and denatured at 75 -C for 5 min prior to annealing at

room temperature to obtain CSFV dsRNA. The integrity of the

dsRNA was confirmed by digestion experiments with RNase

A1 and T in 10 mM Tris–HCl pH 7.5, 300 mM NaCl and 5

mM EDTA. A non-replicating ssRNA control transcript was

synthesized from a pA187-1-derived construct carrying muta-

tions in the NS5B gene resulting in a Gly–Asp–Asp to Ala–

Ala–Ala substitution at amino acid positions 3627 to 3629 of

the polyprotein (details of the construction can be obtained on

request).

Transfection of DC

DC were transfected as previously described (Ceppi et al.,

2005). Briefly, cells were harvested after 4 days of culture and

washed twice with Opti-MEM (GibcoBRL). Nucleic acids

were diluted in Opti-MEM, and Transfast lipofection agent

(Invitrogen) was added at a ratio of 3 Al Transfast per 1 Agnucleic acid. The mixture was incubated at room temperature

for 15 min prior to addition to the cells. 4 Ag of nucleic acids

was transfected per 106 cells. The lipofection mix was

incubated with the cell suspension at 39 -C for 1 h. The cells

were then washed twice in culture medium supplemented with

recombinant porcine GM-CSF/IL4 (concentrations as indicated

above) and plated in 6-well plates. Two days post-transfection,

the cells were harvested and acquired by FCM for CD80/86

and MHC class II expression. Supernatants were subjected to

O. Bauhofer et al. / Virology 343 (2005) 93–105 103

analysis of IFN-a/h bioactivity. Lipofection efficacy was

monitored with an mRNA encoding EGFP (Ceppi et al., 2005).

Antibodies and cytokines

Monoclonal antibodies directed against the viral protein

NS3 (C16) (Greiser-Wilke et al., 1992) and the E2 glycoprotein

(HC/TC26) (Greiser-Wilke et al., 1990) (kindly provided by

Dr. Irene Greiser-Wilke, Hannover Veterinary School, Hann-

over, Germany) were used for virus detection by FCM and

immunohistochemistry. For the detection of viral proteins by

Western blotting, polyclonal rabbit sera against Npro (Ruggli

et al., in press) and NS3 (Ruggli N., unpublished) were utilized.

dsRNA was visualized with the dsRNA-specific mAb J2

(English and Scientific Consulting Bt., Szirak, Hungary)

described to detect the A-helix of double-stranded polyribo-

nucleotide complexes larger than 11 base pairs (Schonborn

et al., 1991). Secondary rabbit anti-mouse biotinylated anti-

body (DAKO) was used in combination with IR Dye-labeled

streptavidin (Rockland). The level of CD80/86 molecules

expressed on the cell surface was measured using human

CTLA4 coupled to murine immunoglobulin (Ancell). MHC II

levels were determined with mAb MSA3 (Hammerberg and

Schurig, 1986). Recombinant porcine IFN-a was produced in

HEK293EBNA cells as described elsewhere (Balmelli et al.,

2005). Porcine tumor necrosis factor a (TNF) was prepared

from L929 cells expressing porcine TNF, kindly provided by

Dr. Giuseppe Bertoni, Vetsuisse Faculty, University of Berne,

Switzerland (Von Niederhausern et al., 1993).

Immunostaining and confocal microscopy

SK-6 cells were cultured and infected with CSFV in eight-

well culture slides (Becton Dickinson). Prior to immunostain-

ing, the cells were washed with PBS, fixed with 4% (w/v)

paraformaldehyde for 15 min and washed again with PBS.

The cells were then incubated for 20 min at 4 -C with the

respective monoclonal antibody diluted in PBS/0.3% (w/v)

saponin (Sigma) and washed again three times with PBS

containing 0.1% (w/v) saponin. Detection was performed with

Alexa-488 fluorochrome-labeled anti-mouse secondary anti-

body (Molecular Probes). The cytoskeleton was stained with

Alexa-546-conjugated phalloidin (Molecular Probes). Analysis

was performed with a Leica TCS-SL spectral confocal

microscope and Leica LCS software (Leica).

dsRNA dot blot

A dot blot assay was established for the quantification of

dsRNA extracted from CSFV-infected cells. Eight million cells

were harvested 16 h p.i. and washed twice with PBS. The cells

were resuspended in 900 Al lysis buffer (200 mMNaCl, 100 mM

Tris–HCl pH 8, 2% sodium dodecyl sulfate and 2% h-mercaptoethanol). Nucleic acids were extracted and purified

from the lysate with phenol:chloroform:isoamylalcohol

(25:24:1) and precipitated with isopropanol. The pellet was

washedwith 70% ethanol, dried and resuspended in water. Serial

dilutions of the extracts were dot-blotted with a Minifold II dot

blot device (Schleicher & Schuell) onto a Protran nitrocellulose

membrane (Schleicher & Schuell). Nucleic acids were fixed by

baking the membrane at 80 -C for 1 h. The membrane was then

blocked for 1 h at room temperature in a blocking solution

consisting of Odyssey Blocking Buffer (LI-COR) diluted 1:1

with PBS (80 mM Na2HPO4, 20 mM NaH2PO4, 100 mM NaCl,

pH 7.3). The primary antibody J2 was diluted to a concentration

of 0.5 Ag/ml (1:2000) in the blocking solution. The secondary

biotinylated rabbit F(abV)2 anti-mouse IgG (DAKO) and the IR

Dye-labeled streptavidin were diluted 1:2000 and 1:2500

respectively in the blocking solution containing 0.1% Tween

20. Between incubation steps, the membrane was washed 4

times for 5 min each with PBS supplemented with 0.1% Tween

20. Two wash steps with PBS lacking detergent were carried out

prior to image acquisition and quantification using the Odyssey

Infrared Imaging System (LI-COR). Quantification of the

nucleic acids was based on a standard curve obtained with in

vitro synthesized CSFV dsRNA.

Western blotting

DC were lysed 48 h p.i. with a hypotonic buffer (20 mM

morpholinepropanesulfonic acid, 10 mM NaCl, 1.5 mM

MgCl2, 1% Triton X-100, pH 6.5). Viral proteins were

separated by sodium dodecyl sulfate polyacrylamide gel

electrophoresis (SDS-PAGE) and analyzed by Western blotting

as previously described (Ruggli et al., in press).

Flow cytometry

FCM was performed as described elsewhere (Summer-

field and McCullough, 1997). For staining of cell surface

molecules, unfixed cells were incubated for 15 min on ice with

the respective primary and secondary antibodies diluted in

CellWash (Becton Dickinson). After each antibody incubation,

cells were washed once with CellWash. For the detection of the

intracellular antigens NS3 and dsRNA, cells were fixed and

permeabilized using a cell fixation and permeabilization kit

(Harlan Sera-Lab Ltd). The primary mAb C16 and J2

respectively were diluted in the permeabilization agent. mAb

J2 was used at a working concentration of 10 Ag/ml (1:100

dilution). The acquisition was done by FCM (FACSCalibur,

Becton Dickinson).

Measurement of IFN-a/b bioactivity

Bioactive IFN-a/h in cell culture supernatant was measured

by using an Mx/CAT reporter gene assay initially developed for

the quantification of bovine IFN-a/h (Fray et al., 2001). This

assay was validated for porcine IFN-a/h and was performed as

previously described (Ruggli et al., 2003). Infectious CSFV

present in the samples to be analyzed was neutralized by

incubation for 1 h at 4 -C with polyclonal pig anti-CSFV serum

600–88 (Ruggli et al., 1995) diluted 1:100. RNAwas removed

by treating the samples for 1 h at 37 -C with a cocktail of 5 U/

ml RNaseA and 200 U/ml RNase T1 (Ambion).

O. Bauhofer et al. / Virology 343 (2005) 93–105104

Statistical analysis

Statistical and graphical analyses were carried out with the

SigmaStat and SigmaPlot software package (SPSS). If appli-

cable, parametric analysis of variance (ANOVA) combined

with the Student’s t test was used. If the samples of the

populations did not satisfy the normal distribution or the equal

variance conditions, ANOVA on ranks combined with a rank

sum test was used. The P values were adjusted with Bonferroni

correction.

Acknowledgments

This work was funded by the Swiss Federal Veterinary

Office (grant # 1.03.04) and in part by the Swiss National

Science Foundation (grant # 31-68237.02). We thank Christian

Griot and Martin A. Hofmann for continuous support and Jon-

Duri Tratschin for critical reading of the manuscript. We are

also grateful to Markus Gerber, Luzia Liu, Viviane Neuhaus,

Valerie Tache and Heidi Gerber for excellent technical

assistance. The SK-6 cell line was kindly provided by M.

Pensaert. We are also grateful to Martin D. Fray for the Mx/

CAT reporter gene assay, to Shigeki Inumaru for the porcine

GM-CSF and to Giuseppe Bertoni for the porcine TNF. We

thank Irene Greiser-Wilke for providing us with mAbs HC/

TC26 and C16.

References

Alexopoulou, L., Holt, A.C., Medzhitov, R., Flavell, R.A., 2001. Recognition

of double-stranded RNA and activation of NF-kappaB by Toll-like receptor

3. Nature 413, 732–738.

Andrejeva, J., Childs, K.S., Young, D.F., Carlos, T.S., Stock, N., Goodbourn,

S., Randall, R.E., 2004. The V proteins of paramyxoviruses bind the IFN-

inducible RNA helicase, mda-5, and inhibit its activation of the IFN-beta

promoter. Proc. Natl. Acad. Sci. U.S.A. 101, 17264–17269.

Aoki, H., Ishikawa, K., Sakoda, Y., Sekiguchi, H., Kodama, M., Suzuki, S.,

Fukusho, A., 2001. Characterization of classical swine fever virus

associated with defective interfering particles containing a cytopatho-

genic subgenomic RNA isolated from wild boar. J. Vet. Med. Sci. 63,

751–758.

Aoki, H., Ishikawa, K., Sekiguchi, H., Suzuki, S., Fukusho, A., 2003.

Pathogenicity and kinetics of virus propagation in swine infected with the

cytopathogenic classical swine fever virus containing defective interfering

particles. Arch. Virol. 148, 297–310.

Aoki, H., Sakoda, Y., Nakamura, S., Suzuki, S., Fukusho, A., 2004.

Cytopathogenicity of classical swine fever viruses that do not show the

exaltation of newcastle disease virus is associated with accumulation of

NS3 in serum-free cultured cell lines. J. Vet. Med. Sci. 66, 161–167.

Balmelli, C., Ruggli, N., McCullough, K., Summerfield, A., 2005. Fibrocytes

are potent stimulators of anti-virus cytotoxic T cells. J. Leukocyte Biol. 77,

923–933.

Barchet, W., Krug, A., Cella, M., Newby, C., Fischer, J.A., Dzionek, A., Pekosz,

A., Colonna, M., 2005. Dendritic cells respond to influenza virus through

TLR7- and PKR-independent pathways. Eur. J. Immunol. 35, 236–242.

Basler, C.F., Garcia-Sastre, A., 2002. Viruses and the type I interferon antiviral

system: induction and evasion. Int. Rev. Immunol. 21, 305–337.

Behrens, S.E., Grassmann, C.W., Thiel, H.J., Meyers, G., Tautz, N., 1998.

Characterization of an autonomous subgenomic pestivirus RNA replicon.

J. Virol. 72, 2364–2372.

Carrasco, C.P., Rigden, R.C., Schaffner, R., Gerber, H., Neuhaus, V., Inumaru,

S., Takamatsu, H., Bertoni, G., McCullough, K.C., Summerfield, A., 2001.

Porcine dendritic cells generated in vitro: morphological, phenotypic and

functional properties. Immunology 104, 175–184.

Carrasco, C.P., Rigden, R.C., Vincent, I.E., Balmelli, C., Ceppi, M., Bauhofer,

O., Tache, V., Hjertner, B., McNeilly, F., van Gennip, H.G., McCullough,

K.C., Summerfield, A., 2004. Interaction of classical swine fever virus with

dendritic cells. J. Gen. Virol. 85, 1633–1641.

Ceppi, M., Ruggli, N., Tache, V., Gerber, H., McCullough, K.C., Summerfield,

A., 2005. Double-stranded secondary structures on mRNA induce type I

interferon (IFN alpha/beta) production and maturation of mRNA-trans-

fected monocyte-derived dendritic cells. J. Gene Med. 7, 452–465.

Diebold, S.S., Kaisho, T., Hemmi, H., Akira, S., Sousa, C.R.E., 2004. Innate

antiviral responses by means of TLR7-mediated recognition of single-

stranded RNA. Science 303, 1529–1531.

Fauquet, C.M., Mayo, M.A., Maniloff, J., Desselberger, U., Ball, L.A., 2005.

Virus Taxonomy. Eighth Report of the International Committee on

Taxonomy of Viruses. Academic Press, SanDiego.

Fray, M.D., Mann, G.E., Charleston, B., 2001. Validation of an Mx/CAT

reporter gene assay for the quantification of bovine type-I interferon.

J. Immunol. Methods 249, 235–244.

Gallei, A., Rumenapf, T., Thiel, H.J., Becher, P., 2005. Characterization of

helper virus-independent cytopathogenic classical Swine Fever virus

generated by an in vivo RNA recombination system. J. Virol. 79,

2440–2448.

Gil, J., Esteban, M., 2000. Induction of apoptosis by the dsRNA-dependent

protein kinase (PKR): mechanism of action. Apoptosis 5, 107–114.

Glew, E.J., Carr, B.V., Brackenbury, L.S., Hope, J.C., Charleston, B., Howard,

C.J., 2003. Differential effects of bovine viral diarrhoea virus on monocytes

and dendritic cells. J. Gen. Virol. 84, 1771–1780.

Goodbourn, S., Didcock, L., Randall, R.E., 2000. Interferons: cell signalling,

immune modulation, antiviral response and virus countermeasures. J. Gen.

Virol. 81, 2341–2364.

Greiser-Wilke, I., Moennig, V., Coulibaly, C.O., Dahle, J., Leder, L., Liess, B.,

1990. Identification of conserved epitopes on a hog cholera virus protein.

Arch. Virol. 111, 213–225.

Greiser-Wilke, I., Dittmar, K.E., Liess, B., Moennig, V., 1992. Heterogeneous

expression of the non-structural protein p80/p125 in cells infected with

different pestiviruses. J. Gen. Virol. 73, 47–52.

Hammerberg, C., Schurig, G.G., 1986. Characterization of monoclonal

antibodies directed against swine leukocytes. Vet. Immunol. Immunopathol.

11, 107–121.

Heil, F., Hemmi, H., Hochrein, H., Ampenberger, F., Kirschning, C., Akira,

S., Lipford, G., Wagner, H., Bauer, S., 2004. Species-specific recognition

of single-stranded RNA via toll-like receptor 7 and 8. Science 303,

1526–1529.

Honda, K., Sakaguchi, S., Nakajima, C., Watanabe, A., Yanai, H., Matsumoto,

M., Ohteki, T., Kaisho, T., Takaoka, A., Akira, S., Seya, T., Taniguchi, T.,

2003. Selective contribution of IFN-alpha/beta signaling to the maturation

of dendritic cells induced by double-stranded RNA or viral infection. Proc.

Natl. Acad. Sci. U.S.A. 100, 10872–10877.

Honda, K., Ohba, Y., Yanai, H., Negishi, H., Mizutani, T., Takaoka, A., Taya,

C., Taniguchi, T., 2005. Spatiotemporal regulation of MyD88-IRF-7

signalling for robust type-I interferon induction. Nature 434, 1035–1040.

Horscroft, N., Bellows, D., Ansari, I., Lai, V.C., Dempsey, S., Liang, D.,

Donis, R., Zhong, W., Hong, Z., 2005. Establishment of a subgenomic

replicon for bovine viral diarrhea virus in Huh-7 cells and modulation of

interferon-regulated factor 3-mediated antiviral response. J. Virol. 79,

2788–2796.

Hulst, M.M., Panoto, F.E., Hoekman, A., van Gennip, H.G., Moormann, R.J.,

1998. Inactivation of the RNase activity of glycoprotein E(rns) of classical

swine fever virus results in a cytopathogenic virus. J. Virol. 72, 151–157.

Iqbal, M., Poole, E., Goodbourn, S., McCauley, J.W., 2004. Role for bovine

viral diarrhea virus Erns glycoprotein in the control of activation of beta

interferon by double-stranded RNA. J. Virol. 78, 136–145.

Iwasaki, A., Medzhitov, R., 2004. Toll-like receptor control of the adaptive

immune responses. Nat. Immunol. 5, 987–995.

Jacobs, B.L., Langland, J.O., 1996. When two strands are better than one: the

mediators and modulators of the cellular responses to double-stranded

RNA. Virology 219, 339–349.

O. Bauhofer et al. / Virology 343 (2005) 93–105 105

Kang, D.C., Gopalkrishnan, R.V., Wu, Q., Jankowsky, E., Pyle, A.M., Fisher,

P.B., 2002. mda-5: An interferon-inducible putative RNA helicase with

double-stranded RNA-dependent ATPase activity and melanoma growth-

suppressive properties. Proc. Natl. Acad. Sci. U.S.A. 99, 637–642.

Kasza, L., Shadduck, J.A., Christofinis, G.J., 1972. Establishment, viral

susceptibility and biological characteristics of a swine kidney cell line

SK-6. Res. Vet. Sci. 13, 46–51.

Kosmidou, A., Buttner, M., Meyers, G., 1998. Isolation and characterization of

cytopathogenic classical swine fever virus (CSFV). Arch. Virol. 143,

1295–1309.

Kummerer, B.M., Tautz, N., Becher, P., Thiel, H., Meyers, G., 2000. The

genetic basis for cytopathogenicity of pestiviruses. Vet. Microbiol. 77,

117–128.

Kupfermann, H., Thiel, H.J., Dubovi, E.J., Meyers, G., 1996. Bovine viral

diarrhea virus: characterization of a cytopathogenic defective interfering

particle with two internal deletions. J. Virol. 70, 8175–8181.

Lackner, T., Muller, A., Pankraz, A., Becher, P., Thiel, H.J., Gorbalenya, A.E.,

Tautz, N., 2004. Temporal modulation of an autoprotease is crucial for

replication and pathogenicity of an RNA virus. J. Virol. 78, 10765–10775.

Lai, V.C.H., Zhong, W.D., Skelton, A., Ingravallo, P., Vassilev, V., Donis, R.O.,

Hong, Z., Lau, J.Y.N., 2000. Generation and characterization of a hepatitis

C virus NS3 protease-dependent bovine viral diarrhea virus. J. Virol. 74,

6339–6347.

La Rocca, S.A., Herbert, R.J., Crooke, H., Drew, T.W., Wileman, T.E., Powell,

P.P., 2005. Loss of interferon regulatory factor 3 in cells infected with

classical swine fever virus involves the N-terminal protease, Npro. J. Virol.

79, 7239–7247.

Lindenbach, B.D., Rice, C.M., 2001. Flaviviridae: the viruses and their

replication. In: Knipe, D.M., Howley, P.M., Griffin, D.E., Lamb, R.A.,

Martin, M.A., Roizman, B., Straus, S.E. (Eds.), Fields Virology. Lippincott

Williams and Wilkins, Philadelphia, pp. 991–1041.

Liu, Y.J., 2005. IPC: professional type 1 interferon-producing cells and

plasmacytoid dendritic cell precursors. Annu. Rev. Immunol. 23, 275–306.

Lund, J.M., Alexopoulou, L., Sato, A., Karow, M., Adams, N.C., Gale,

N.W., Iwasaki, A., Flavell, R.A., 2004. Recognition of single-stranded

RNA viruses by Toll-like receptor 7. Proc. Natl. Acad. Sci. U.S.A. 101,

5598–5603.

Matsumoto, M., Funami, K., Oshiumi, H., Seya, T., 2004. Toll-like receptor 3:

a link between toll-like receptor, interferon and viruses. Microbiol.

Immunol. 48, 147–154.

Mayer, D., Hofmann, M.A., Tratschin, J.D., 2004. Attenuation of classical

swine fever virus by deletion of the viral Npro gene. Vaccine 22,

317–328.

Mendez, E., Ruggli, N., Collett, M.S., Rice, C.M., 1998. Infectious bovine viral

diarrhea virus (strain NADL) RNA from stable cDNA clones: a cellular

insert determines NS3 production and viral cytopathogenicity. J. Virol. 72,

4737–4745.

Meyers, G., Thiel, H.J., 1995. Cytopathogenicity of classical swine fever virus

caused by defective interfering particles. J. Virol. 69, 3683–3689.

Meyers, G., Thiel, H.J., Rumenapf, T., 1996. Classical swine fever virus:

recovery of infectious viruses from cDNA constructs and generation of

recombinant cytopathogenic defective interfering particles. J. Virol. 70,

1588–1595.

Mittelholzer, C., Moser, C., Tratschin, J.D., Hofmann, M.A., 1997. Generation

of cytopathogenic subgenomic RNA of classical swine fever virus in

persistently infected porcine cell lines. Virus Res. 51, 125–137.

Moser, C., Stettler, P., Tratschin, J.D., Hofmann, M.A., 1999. Cytopathogenic

and noncytopathogenic RNA replicons of classical swine fever virus.

J. Virol. 73, 7787–7794.

Player, M.R., Torrence, P.F., 1998. The 2–5A system: modulation of viral and

cellular processes through acceleration of RNA degradation. Pharmacol.

Ther. 78, 55–113.

Ruggli, N., Moser, C., Mitchell, D., Hofmann, M., Tratschin, J.D., 1995.

Baculovirus expression and affinity purification of protein E2 of classical

swine fever virus strain Alfort/187. Virus Genes 10, 115–126.

Ruggli, N., Tratschin, J.D., Mittelholzer, C., Hofmann, M.A., 1996.

Nucleotide sequence of classical swine fever virus strain Alfort/187 and

transcription of infectious RNA from stably cloned full-length cDNA.

J. Virol. 70, 3478–3487.

Ruggli, N., Tratschin, J.D., Schweizer, M., McCullough, K.C., Hofmann,

M.A., Summerfield, A., 2003. Classical swine fever virus interferes with

cellular antiviral defense: evidence for a novel function of Npro. J. Virol.

77, 7645–7654.

Ruggli, N., Bird, B.H., Liu, L., Bauhofer, O., Tratschin, J.D., Hofmann, M.A.,

in press. Npro of classical swine fever virus is an antagonist of double-

stranded RNA-mediated apoptosis and IFN-a/h induction. Virology.

Rumenapf, T., Stark, R., Heimann, M., Thiel, H.J., 1998. N-terminal protease

of pestiviruses: identification of putative catalytic residues by site-directed

mutagenesis. J. Virol. 72, 2544–2547.

Schonborn, J., Oberstrass, J., Breyel, E., Tittgen, J., Schumacher, J., Lukacs,

N., 1991. Monoclonal antibodies to double-stranded RNA as probes of

RNA structure in crude nucleic acid extracts. Nucleic Acids Res. 19,

2993–3000.

Stark, R., Meyers, G., Rumenapf, T., Thiel, H.J., 1993. Processing of pestivirus

polyprotein: cleavage site between autoprotease and nucleocapsid protein of

classical swine fever virus. J. Virol. 67, 7088–7095.

Summerfield, A., McCullough, K.C., 1997. Porcine bone marrow myeloid

cells: phenotype and adhesion molecule expression. J. Leukocyte Biol. 62,

176–185.

Tautz, N., Meyers, G., Thiel, H.J., 1998. Pathogenesis of mucosal disease, a

deadly disease of cattle caused by a pestivirus. Clin. Diagn. Virol. 10,

121–127.

Tautz, N., Harada, T., Kaiser, A., Rinck, G., Behrens, S., Thiel, H.J., 1999.

Establishment and characterization of cytopathogenic and noncytopatho-

genic pestivirus replicons. J. Virol. 73, 9422–9432.

Tortorella, D., Gewurz, B.E., Furman, M.H., Schust, D.J., Ploegh, H.L.,

2000. Viral subversion of the immune system. Annu. Rev. Immunol. 18,

861–926.

Tratschin, J.D., Moser, C., Ruggli, N., Hofmann, M.A., 1998. Classical swine

fever virus leader proteinase Npro is not required for viral replication in cell

culture. J. Virol. 72, 7681–7684.

Vassilev, V.B., Donis, R.O., 2000. Bovine viral diarrhea virus induced

apoptosis correlates with increased intracellular viral RNA accumulation.

Virus Res. 69, 95–107.

Von Niederhausern, B., Bertoni, G., Hertig, C., Pfister, H., Peterhans, E., Pauli,

U., 1993. Cloning and expression in mammalian cells of porcine tumor

necrosis factor alpha: examination of biological properties. Vet. Immunol.

Immunopathol. 38, 57–74.

Weber, F., Kochs, G., Haller, O., 2004. Inverse interference: how viruses fight

the interferon system. Viral. Immunol. 17, 498–515.

Wiskerchen, M., Belzer, S.K., Collett, M.S., 1991. Pestivirus gene

expression: the first protein product of the bovine viral diarrhea virus

large open reading frame, p20, possesses proteolytic activity. J. Virol. 65,

4508–4514.

Yoneyama, M., Kikuchi, M., Natsukawa, T., Shinobu, N., Imaizumi, T.,

Miyagishi, M., Taira, K., Akira, S., Fujita, T., 2004. The RNA helicase

RIG-I has an essential function in double-stranded RNA-induced innate

antiviral responses. Nat. Immunol. 5, 730–737.