Skip to main content
Journal of Virology logoLink to Journal of Virology
. 1999 Dec;73(12):10224–10235. doi: 10.1128/jvi.73.12.10224-10235.1999

Mutations Abrogating the RNase Activity in Glycoprotein Erns of the Pestivirus Classical Swine Fever Virus Lead to Virus Attenuation

Gregor Meyers 1,*, Armin Saalmüller 1, Mathias Büttner 1
PMCID: PMC113076  PMID: 10559339

Abstract

Classical swine fever (CSF) is a severe hemorrhagic disease of swine caused by the pestivirus CSF virus (CSFV). Amino acid exchanges or deletions introduced by site-directed mutagenesis into the putative active site of the RNase residing in the glycoprotein Erns of CSFV abolished the enzymatic activity of this protein, as demonstrated with an RNase test suitable for detection of the enzymatic activity in crude cell extracts. Incorporation of the altered sequences into an infectious CSFV clone resulted in recovery of viable viruses upon RNA transfection, except for a variant displaying a deletion of the histidine codon at position 297 of the long open reading frame. These RNase-negative virus mutants displayed growth characteristics in tissue culture that were undistinguishable from wild-type virus and were stable for at least seven passages. In contrast to animals inoculated with an RNase-positive control virus, infection of piglets with an RNase-negative mutant containing a deletion of the histidine codon 346 of the open reading frame did not lead to CSF. Neither fever nor extended viremia could be detected. Animals infected with this mutant did not show decrease of peripheral B cells, a characteristic feature of CSF in swine. Animal experiments with four other mutants with either exchanges of codons 297 or 346 or double exchanges of both codons 297 and 346 showed that all these RNase-negative mutants were attenuated. All viruses with mutations affecting codon 346 were completely apathogenic, whereas those containing only changes of codon 297 consistently induced clinical symptoms for several days, followed by sudden recovery. Analyses of reisolated viruses gave no indication for the presence of revertants in the infected animals.


Pestiviruses are the etiologic agents of economically important diseases of animals in many countries worldwide. According to the host animals from which the viruses originate, presently known pestivirus isolates have been grouped into three different species which together form one genus within the family Flaviviridae. A new taxonomy with four virus species has been proposed that is based on the results of sequence comparison studies and takes into account that pestiviruses infect different host species. Pestiviruses predominantly found in ruminants are two types of bovine viral diarrhea virus and border disease virus. These viruses are responsible for a variety of syndromes characterized by different clinical symptoms (1, 18, 32). Classical swine fever virus (CSFV), formerly named hog cholera virus, represents the fourth species and is causative for classical swine fever (CSF), a severe disease that results in high morbidity and mortality of infected swine (18, 32). Acute CSF is characterized by pyrexia and leukopenia. Diseased animals show anorexia and diarrhea and in late stage central nervous disorders, hemorrhages in the skin, mucosa, and inner organs. Like all pestiviruses, CSFV is immunosuppressive during acute infection. A characteristic feature detected early after infection of swine is a dramatic decrease of peripheral B cells (30). Infection with CSFV variants of high virulence mostly leads to death of the affected animal. However, recent CSFV outbreaks in Europe resulted predominantly from viruses apparently inducing milder and chronic forms of the disease (18, 32).

Like other members of the family Flaviviridae, pestiviruses are small enveloped viruses with a single-stranded RNA genome of positive polarity. The pestivirus RNA lacks both 5′ cap and 3′ poly(A) sequences and contains one long open reading frame (ORF) coding for a polyprotein of about 4,000 amino acids, which encompasses all viral proteins arranged in the order NH2-Npro-C-Erns-E1-E2-p7-NS2-NS3-NS4A-NS4B-NS5A-NS5B-COOH. The polyprotein gives rise to 11 to 12 final cleavage products by co- and posttranslational processing involving cellular and viral proteases (21). Protein C and the glycoproteins Erns, E1, and E2 represent structural components of the pestivirus virion (33). E2 and, to a lesser extent, Erns were found to be targets for antibody neutralization (7, 19, 36, 37, 38). Erns lacks a typical membrane anchor and is secreted in considerable amounts from the infected cells (23). A highly unusual feature of this protein is its RNase activity that was first identified by characteristic sequence motifs and then proven by enzymatic tests with the purified protein (9, 28, 40). The function of this enzymatic activity for the viral life cycle is presently unknown. Experimental destruction of the RNase by site-directed mutagenesis of the respective sequence in the genome of a CSFV vaccine strain has been reported to result in a cytopathogenic virus that has growth characteristics in cell culture equivalent to wild-type virus (10). We report here on the establishment of mutants of the virulent CSFV strain Alfort/Tübingen and the effects of the introduced mutations with regard to the pathogenicity of the virus in its natural host.

MATERIALS AND METHODS

Cells and viruses.

PK15 cells were obtained from the American Type Culture Collection (Rockville, Md.). CSFV strain Alfort/Tübingen was reisolated from organs of an experimentally infected moribund animal (22). CSFV Eystrup originated from Behring, where it served as a challenge virus; the virus was obtained from the CSFV reference strain collection of the Federal Research Centre for Virus Diseases of Animals (kindly provided by R. Ahl). Cells were grown in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal calf serum (FCS; tested for the absence of pestiviruses and antibodies against pestiviruses) and nonessential amino acids. Cells and virus stocks were tested regularly for the absence of mycoplasma contamination.

The BSR clone of BHK-21 cells was kindly provided by J. Cox (Department of Immunology, Federal Research Centre for Virus Diseases of Animals, Tübingen, Germany).

Infection of cells and immunofluorescence assay.

Since pestiviruses are known to be associated with the host cells, lysates of infected cells were used for reinfection of culture cells. Lysates were prepared by freezing and thawing cells at 48 h postinfection and were stored at −70°C. If not indicated differently in the text, a multiplicity of infection (MOI) of about 0.01 was used for infections.

For detection of infected cells in immunofluorescence assays, cells were fixed with ice-cold methanol-acetone (1:1) for 15 min, air dried, rehydrated with phosphate-buffered saline (PBS) and then incubated with a mixture of anti-CSFV monoclonal antibodies (MAbs) a18 and 24/16 (38, 39). Bound antibodies were detected with a fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse serum (Dianova, Hamburg, Germany).

Generation of cDNA clones with mutations in the Erns gene.

Restriction, cloning, and other standard procedures were done essentially as described previously (27). Restriction and modifying enzymes were obtained from New England BioLabs (Schwalbach, Germany), Pharmacia (Freiburg, Germany), Gibco-BRL (Eggenstein, Germany), and Boehringer Mannheim GmbH (Mannheim, Germany).

Starting with the full-length cDNA clones pA/CSFV (14), from which infectious cRNA can be obtained by in vitro transcription, subclones were generated. A XhoI/SspI fragment of pA/CSFV was cloned into pBluescript SK(+) and cut with XhoI and SmaI, leading to plasmid p666. Single-stranded plasmid DNA was produced according to the method of Kunkel et al. (13) by using Escherichia coli CJ236 cells (Bio-Rad, Munich, Germany) and the VCMS single-strand phage (Stratagene, Heidelberg, Germany). The single-stranded DNA was converted to double strands by using the Phagemid In Vitro Mutagenesis Kit (Bio-Rad) and the following synthetic oligonucleotides as primers: Ol/C-297-L, AGGAGCTTACTTGGGATCTG; Ol/C-346-L, GGAACAAACTTGGATGGTGT; Ol/C-297-K, ACAGGAGCTTAAAAGGGATCTGGC; Ol/C-346-K, ATGGAACAAAAAGGGATGGTGTAA; Ol/C-297-d, AACAGGAGCTTAGGGATCTGGCCC; and Ol/C-346-d, GAATGGAACAAAGGATGGTGTAAC.

The double-stranded plasmid DNA was used for transformation of E. coli XL1-Blue cells (Stratagene). Bacterial colonies harboring plasmids were isolated via ampicillin selection, and plasmid DNA was prepared and further analyzed by nucleotide sequencing by using the T7 polymerase sequencing kit (Pharmacia, Freiburg, Germany).

Plasmids containing the desired mutations and no second site changes were used for construction of full-length cDNA clones. An XhoI/NdeI fragment from the mutagenized plasmid was inserted together with an NdeI/BglII fragment derived from plasmid 578 (pCITE 2a, containing the XhoI/BglII fragment from pA/CSFV) into pA/CSFV cut with XhoI and BglII. To restore the wild-type sequence in full-length clones containing mutated Erns genes, a 1.7-kb XhoI/NdeI fragment was exchanged for the corresponding fragment from p666.

A second series of constructs was established for expression in the vaccinia virus-T7 system. To do so, plasmid p666 was mutagenized with oligonucleotide Cs/NcoI as described above to yield a construct with an NcoI site at the ATG of the CSFV ORF. The plasmid with the mutation (p666/Mut-NcoI) was cut with NcoI and NdeI, and the cDNA fragment was cloned into pCITE 2a (AGS, Heidelberg, Germany), restricted with the same enzymes, leading to plasmid p704. Plasmid p705 was constructed by inserting an NcoI/HindIII fragment from p704 together with a HindIII/NdeI fragment from p666/297-L into pCITE 2a, cut with NcoI and NdeI. The presence of the desired mutations in the resulting constructs was verified by DNA sequencing. To establish p706, a plasmid equivalent to p704 with a deletion covering the complete Erns coding region, plasmid p578 was cut with BsaBI and XbaI, ligated with a BanI/XbaI fragment of the same plasmid together with oligonucleotides Cs/C-E1+ and Cs/C−E1−. From the resulting construct, a HindIII/NdeI fragment was inserted together with an NcoI/HindIII fragment from p666/Mut-NcoI into pCITE 2a cut with NcoI and NdeI.

The oligonucleotides used were as follows: Cs/NcoI, TGTACATGGCCCATGGAGTTG; Cs/C−E1+, CAATCTTGCTGTACCAGCCTGTAGCAGCCG; and Cs/C−E1−, GCACCGGCTGCTACAGGCTGGTACAGCAAGATTG.

In vitro transcription and RNA transfection.

Amounts (2 μg) of the respective cDNA construct were linearized with the restriction enzyme SrfI and purified by phenol extraction and ethanol precipitation. Transcription with T7 RNA polymerase (NEB, Schwalbach, Germany) was carried out in a total volume of 50 μl of transcription mix (40 mM Tris-HCl, pH 7.5; 6 mM MgCl2; 2 mM spermidine; 10 mM NaCl; 0.5 mM concentrations of each ATP, GTP, CTP, and UTP; 10 mM dithiothreitol; 100 μg of bovine serum albumin per ml) with 50 U of T7 RNA polymerase in the presence of 15 U of RNAguard (Pharmacia). After incubation at 37°C for 1 h the reaction mixture was passed through a Sephadex G-50 spin column (27) and further purified by phenol extraction and ethanol precipitation.

Transfection was done with a suspension of 3 × 106 PK15 cells and ca. 0.5 μg of in vitro-transcribed RNA bound to DEAE-dextran (Pharmacia). For positive controls, usually 5 μg of total RNA from PK15 cells infected with the respective CSFV isolate was used for transfection. The RNA–DEAE-dextran complex was established by mixing RNA dissolved in 100 μl of Hanks balanced salt solution (HBSS) (35) with 100 μl of DEAE-dextran (1 mg/ml in HBSS) and incubation for 30 min on ice. Pelleted cells were washed once with DMEM without FCS, centrifuged, and then resuspended in the RNA–DEAE-dextran mixture. After 30 min of incubation at 37°C, 20 μl of dimethyl sulfoxide was added, and the mixture was incubated for 2 min at room temperature. After the addition of 2 ml of HBSS, cells were pelleted and washed once with HBSS and once with medium without FCS. Cells were resuspended in DMEM with FCS and seeded in a 10.0-cm-diameter dish. At 48 to 72 h posttransfection cells were split and seeded as appropriate for subsequent analyses.

Northern (RNA) hybridization.

RNA was prepared 48 h after infection by using either the Trizol reagent as recommended by the supplier (Gibco-BRL) or cesium chloride density gradient centrifugation as described before (22). Gel electrophoresis, radioactive labeling of the probe, hybridization, and posthybridization washes were done as described before (22). A 2.2-kb SalI fragment from CSFV Alfort cDNA clone 4.2 (14) was used as a probe.

RT-PCR.

Reverse transcription of 2 μg of heat-denatured RNA (2 min at 92°C, 5 min on ice in 11.5 μl of water in the presence of 30 pmol of reverse primer) was done after the addition of 8 μl of reverse transcriptase (RT) mix (125 mM Tris-HCl, pH 8.3; 182.5 mM KCl; 7.5 mM MgCl2; 25 mM dithiothreitol; 1.25 mM concentrations of dATP, dTTP, dCTP, and dGTP), 15 U of RNAguard, and 50 U of Superscript (Gibco-BRL) for 45 min at 37°C. After addition of paraffin (Paraplast; melting point, 55°C) and 2 min at 80°C, the tubes were placed on ice and 30 μl of PCR mix (8.3 mM Tris-HCl, pH 8.3; 33.3 mM KCl; 2.2 mM MgCl2; 0.42 mM each of dATP, dTTP, dCTP, and dGTP; 0.17% Triton X-100; 0.03% bovine serum albumin; 5 U of Taq polymerase [Appligene, Heidelberg, Germany]) was added. Amplification was carried out in 30 cycles (30 s at 94°C, 30 s at 54°C, and 60 s at 72°C).

The primers for RT-PCR were as follows: upstream, CATGCCATGGCCCTGTTGGCTTGGGCGGTGATA; and downstream, GGAATTCTCAGGCATAGGCACCAAACCAGG. The amplified cDNA fragments were purified by preparative agarose gel electrophoresis. Elution of DNA from the agarose gel was done with the Qiaex Kit (Qiagen, Hilden, Germany). For sequencing with the upstream primer, the Big Dye Terminator Cycle Sequencing Kit (Perkin-Elmer/Applied Biosystems, Weiterstadt, Germany) was used. Analysis of the sequencing products was done with an ABI Prism 377 DNA Sequencer (Perkin-Elmer/Applied Biosystems).

Transient expression, metabolic labeling, radioimmunoprecipitation, and SDS-PAGE.

Transient expression of transfected plasmids with vaccinia virus vTF7-3 (kindly provided by B. Moss) (8) was done as described before (31), except that the labeling time was reduced to 5 h. CSFV-infected PK15 cells (1.5 × 106 per 3.5-cm dish) were labeled for 8 h with 0.5 mCi of [35S]methionine-[35S]cysteine (Promix; Amersham, Braunschweig, Germany) per ml. The labeling medium contained no cysteine and no methionine. Cell extracts were prepared under denaturing conditions. Extracts were incubated with 5 μl of undiluted serum. Precipitates were formed with cross-linked Staphylococcus aureus (11), analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and processed by fluorography by using En3Hance (New England Nuclear, Boston, Mass.).

Determination of RNase activity.

For analysis of RNase activity after CSFV infection, PK15 cells were infected with the mutant viruses at an MOI of 0.01. Infection with wild-type virus served as a positive control, whereas noninfected cells were used as a negative control. At 48 h postinfection cells were washed twice with PBS and then lysed in 0.4 ml of lysis buffer (20 mM Tris-HCl, 100 mM NaCl, 1 mM EDTA, 2 mg of bovine serum albumin per ml, 1% Triton X-100, 0.1% deoxycholic acid, 0.1% SDS). The lysate was added to 1.5-ml reaction tubes, sonified (Branson sonifier B12; 120 W for 20 s in a cup horn water bath), and cleared by centrifugation (5 min, 14,000 rpm, Eppendorf centrifuge, 4°C), and the supernatant was subjected to ultracentrifugation (Beckman tabletop ultracentifuge, 60 min at 4°C and 45,000 rpm in a TLA 45 rotor).

To determine the RNase activity of transiently expressed proteins, BHK21 cells (clone Bsr) were infected with vTF7-3 and transfected with the different constructs as described above. About 10 h posttransfection, the cells were lysed and further processed as described for CSFV infected cells. For these analyses, cells infected with vTF7-3 but not transfected served as a control.

Determination of RNase activity was done essentially as described before (28). If not otherwise specified, assays were conducted in a total volume of 200 μl containing 5 or 50 μl of supernatant of the second centrifugation step and 80 μg of Poly(U) (Pharmacia) in RNase assay buffer (40 mM Tris-acetate [pH 6.5], 0.5 mM EDTA, 5 mM dithiothreitol). After incubation of the reaction mixture at 37°C for 1 h, 200 μl of 1.2 M perchloric acid–20 mM lanthanum sulfate was added. After 15 min of incubation on ice the mixture was centrifuged for 15 min at 4°C and 14,000 rpm in an Eppendorf centrifuge. To the supernatant, 3 volumes of water were added and an aliquot of the mixture was analyzed by measuring the optical density at 260 nm (OD260) by using an Ultrospec 3000 spectrophotometer (Pharmacia). As a positive control, RNase A from bovine pancreas (Serva, Heidelberg, Germany) with an activity of 85 Kunitz units per mg of protein was used instead of the cell extract in some experiments.

For time course experiments, the assay was scaled up to a total volume of 1.5 ml. At each time point, 160 μl of the reaction mixture was removed and mixed with 160 μl of the acid solution. All further steps were the same as those described above.

To determine the influence of antibodies on the RNase activity Erns, 0.1 μl of cell extract was incubated in a total volume of 160 μl containing 20 μl 10× RNase assay buffer and 5 μl of rabbit antiserum against Erns (kindly provided by R. Stark and H.-J. Thiel, Institut für Virologie, FB Veterinärmedizin, Justus Liebig Universität Giessen) or a rabbit preimmune serum for 1 h at 37°C, followed by further incubation for 16 h at 4°C. RNase activity was measured after the addition of 40 μl of Poly(U) solution (containing 80 μg of RNA) and incubation at 37°C for 45 min. Acid precipitation and further analysis were done as described above.

Animal experiments.

For each mutant three piglets (German Landrace; 20 to 25 kg) were used. If not specified, the infection dose was 0.5 × 105 to 1 × 105 50% tissue culture infective doses (TCID50) per animal, depending on the size of the animals; two-thirds of the inoculate was administered intranasally (one-third in each nostril), and one-third was given intramuscularly. The different groups were housed in separate isolation units. Blood was taken from the vena jugularis at the time points indicated in the Results section. Coagulation was inhibited with heparin (20 IU/ml of blood) or sodium citrate (3.8% [wt/vol]).

In challenge experiments, the animals were inoculated with 2 × 105 TCID50 of CSFV strain Eystrup.

To test the animals for the presence of virus in the blood, PK15 cells seeded in a 24-well plate were incubated with 100 μl of citrate-treated or heparinized blood and 100 μl of medium (animal experiment 1 or 3, respectively) or with 50 μl of isolated peripheral blood leukocytes (prepared as described previously [25]) and 150 μl of medium (animal experiment 2). After 1 h at 37°C, the mixture was removed and the cells were washed twice with medium and incubated for 48 to 72 h at 37°C. Infection of cells was demonstrated by immunofluorescence (see above).

To look for virus-neutralizing antibodies in blood serum or plasma, samples were diluted 1:2 and then in steps of 1:3 with medium. Next, 50 μl of the diluted samples were mixed with 50 μl of medium containing 30 TCID50 of virus (CSFV Alfort/Tübingen). After 90 min of incubation at 37°C, 100 μl of cell suspension (1.5 × 104 cells) was added, and the mixture was seeded in 96-well plates. After 72 h the cells were fixed with ice-cold acetone-methanol and analyzed for infection by immunofluorescence with MAb a18 (38).

Isolation of PBMC and FCM.

PBMC were isolated from heparinized blood samples by use of Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) centrifugation (1,100 × g, 30 min) as described earlier (25).

Staining of PBMC for two-color flow cytometric (FCM) analyses was performed in a two-step procedure: (i) incubation with MAb B-ly4 [αhuman CD21, immunoglobulin G1 [IgG1]; Pharmingen, San Diego, Calif. [4, 26, 29]) and MAb 74-22-15A [αSWC3, IgG2b, Dr. J. K. Lunney, USDA ARS, Beltsville, MD (20)] and (ii) incubation with isotype-specific conjugates (αIgG2b-FITC and αIgG1-PE; Southern Biotechnology, Birmingham, Ala.). Between each of the incubation steps on ice, cells were washed with PBS–2% FCS (vol/vol). The analyses were performed on a FACStarplus (Becton Dickinson, Mountain View, Calif.) as described earlier (24).

The percentages of CD21-positive B lymphocytes and SWC3-positive myeloid cells were calculated by using WinMDI or Cell Quest software.

RESULTS

Introduction of mutations into Erns and RNase test.

The pestivirus Erns sequence contains two short stretches of amino acids that show limited but significant homology to sequences found in different known RNases of plant and fungal origin. The respective regions of the pestivirus proteins correspond to residues 295 to 307 and residues 338 to 357 in the polyprotein of the CSFV strain Alfort/Tübingen and are regarded as essential for enzymatic activity (9, 28). Both of these stretches of amino acids are conserved among pestiviruses and contain a histidine residue (positions 297 and 346, respectively); these two residues are believed to be involved in the catalytic reaction (28). Substitution of each of these residues for lysine in the Erns protein of a CSFV vaccine strain resulted in the destruction of RNase activity (10). To be able to test the influence of mutations affecting the RNase activity of Erns on the pathogenicity of a virulent CSFV strain, such mutants were established on the basis of the CSFV strain Alfort/Tübingen. An XhoI/SspI fragment of the infectious clone pA/CSFV was subcloned into pBluescript SK(+), and a variety of mutants was established with the oligonucleotides Ol/C-297-L, Ol/C-346-L, Ol/C-297-K, Ol/C-346-K, Ol/C-297-d, and Ol/C-346-d. Mutagenesis with the first four oligonucleotides led to substitution of the histidine residues at positions 297 or 346 of the polyprotein for lysine (oligonucleotides with names ending with “-K”) or leucine (oligonucleotides with names ending with “-L”), respectively (Fig. 1). In addition to plasmids coding for proteins with one of these mutations, double mutants were also established with both of the histidine codons at positions 297 and 346 exchanged for triplets coding for leucine or lysine, respectively. Another double mutant encoded a protein with lysine at position 297 and leucine at position 346. Oligonucleotides Ol/C-297-d and Ol/C-346-d were used to delete the histidine codons at positions 297 and 346, respectively (Fig. 1). Introduction of the desired mutations and the absence of unwanted second site mutations was verified by nucleotide sequencing.

FIG. 1.

FIG. 1

Sequences encoded by CSFV Alfort/Tübingen (CSFV) and the different Erns mutants. Only those parts of the Erns sequence containing the conserved regions with the putative active-site residues of the RNase are shown. Residues conserved with regard to known RNases of fungal origin are indicated in boldface. Residues 297 and 346 of the polyprotein are shaded and underlined.

To analyze the effect of the introduced mutations on the RNase function of Erns, it was necessary to measure the enzymatic activity of the altered proteins. Published assay systems rely on the purification of Erns with specific MAbs and determination of the degradation of high-molecular-weight RNA by measuring the generation of acid-soluble nucleic acid (9, 10, 28). As a quick and easy alternative, we established an assay to measure RNase activity in crude cell extracts. Initial tests were conducted with proteins expressed in the vaccinia virus-T7 system (8). Different expression constructs were established that start with the original translational initiation codon of the CSFV ORF and end with an NdeI site located within the region coding for glycoprotein E1. Initially, three constructs were established that contained either the wild-type sequence, a mutant with leucine replacing the histidine codon at position 297 of the ORF, or a mutant from which the complete Erns gene was deleted (plasmids p704, p705, or p706, respectively). After transient expression of these constructs, cell extracts were prepared and aliquots thereof were tested for the ability to degrade Poly(U). As a control, the extract of nontransfected, vTF7-3 infected cells was used. After 60 min of incubation, the residual high-molecular-weight RNA was precipitated with perchloric acid and the acid-soluble reaction products were quantified by determination of the OD260. Extracts of cells transfected with plasmid p704 showed significant RNase activity, whereas the values determined after transfection of p705 or p706 were in the same range as the negative control (Fig. 2). This result was not due to major differences in the transfection or expression efficiencies since accompanying analyses of radioactively labeled proteins from equivalently transfected dishes by using immunoprecipitation and phophorimager-based quantification showed similar expression levels (Fig. 2). To test the influence of protein concentration on the results, the assay with the extract of p704- or p705-transfected cells was repeated with different protein concentrations. The undiluted extracts or dilutions of 1:5 to 1:1,000 were used. For p704, OD260 values of >2 were recorded for dilutions up to 1:100. At higher dilutions, significantly reduced RNase activity was observed. In contrast, no concentration dependency was detected for the assays with extract from p705-transfected cells (not shown). These analyses indicated that the amount of extract used for the experiment summarized in Fig. 2 resulted in saturation of the assay with enzyme. Thus, small differences in protein amounts cannot be responsible for our results.

FIG. 2.

FIG. 2

Expression of Erns in the vaccinia virus vTF7-3 system. The different cDNA constructs were transfected into BHK21 cells infected with vTF7-3. The cells in one set of dishes were labeled with radioactive amino acids. The labeled proteins were precipitated with antisera specific for Npro or Erns, and equal volumes of the precipitates were separated by PAGE. The gels were analyzed by fluorography (upper part) or by measuring the amount of radioactivity contained in the specific bands by using a phosphorimager (middle). The protein extracts of a second set of dishes were used for the determination of RNase activity (lower part). Cells infected with vTF7-3 but not transfected with a plasmid were used as negative controls. For the RNase test, 100 ng of RNase A from bovine pancreas (0.0085 Kunitz units) served as a positive control. RNase activity was determined by measuring the OD260 due to the release of acid-soluble RNA. OD values of ca. 0.4 to 08 as measured for the negative control are not due to enzymatic degradation of RNA, since no decrease was observed when less of the cell extracts was used for the test.

In a further assay, the time course of the enzymatic degradation of the RNA was determined. RNA degradation was induced by the extract of p704-transfected cells and stopped at eight different time points ranging from 0 to 55 min. Determination of the amount of acid-precipitable RNA showed a clear time dependency of the reaction, reaching its maximum between 35 and 45 min (not shown).

It has been shown before that the RNase activity of Erns can be blocked by antibodies specific for this protein (40). To demonstrate that the degradation of RNA observed in the above-described assay indeed is due to the enzymatic activity of Erns, an equivalent assay was conducted after incubation of cell extracts with either a monospecific antiserum against Erns or a preimmune serum. When extracts of a nontransfected vaccinia virus control or cells transfected with p705 were tested, similar values were determined regardless of whether the antiserum or the preimmune serum was used (Fig. 3). However, the analysis with extracts of cells transfected with p704 showed a significantly reduced OD value after incubation with the antiserum in comparison to that after incubation with the preimmune serum. Taken together, these results show that the described assay allows the detection of Erns RNase activity and the discrimination between wild-type Erns and mutants with changes abrogating the enzymatic activity of the protein.

FIG. 3.

FIG. 3

Influence of antibodies on the RNase activity. Diluted extracts of cells transfected with p704 or p705 were first incubated with a polyclonal antiserum directed against Erns (anti-Erns) or a rabbit preimmune serum (NS) and then tested for RNase activity. An extract from cells infected with vTF7-3 (control) or 10 ng of RNase A from bovine pancreas (0.00085 Kunitz units) served as controls.

In order to analyze the effect of the different mutations on the RNase activity of Erns, expression constructs equivalent to p704 were established for all mutants. After expression in the vaccinia virus vTF7-3 system, the analysis of the enzymatic activity in the cell extracts showed that the RNase was inactivated in all mutants (not shown). Thus, as expected, all the exchanges and deletions affecting the putative active center of the RNase destroyed its enzymatic activity.

Generation of RNase-negative mutants of CSFV Alfort/Tübingen.

Substitution of histidine 297 or 346 in the polyprotein of the CSFV vaccine strain for lysine did not influence viral viability or growth properties (10). To test the effects of the different mutations introduced into the sequence of CSFV Alfort/Tübingen on virus viability, fragments containing the altered sequences were inserted into the infectious cDNA clone pA/CSFV replacing the wild-type sequence. RNA transcribed from the resulting constructs, termed pC-297-L, pC-297-K, pC-297-d, pC-346-L, pC-346-K, pC-346-d, pC-297-L/346-L, and pC-297-K/346-L, was used for transfection of porcine kidney (PK15) cells. The cells were analyzed 3 days posttransfection by immunofluorescence for the expression of CSFV proteins. For all constructs except for pC-297-d, positive signals were obtained. Analysis of RNA prepared at 4 days posttransfection by Northern blot with a CSFV-specific probe allowed the detection of viral RNA for all these mutants, whereas RNA prepared from noninfected cells or cells transfected with RNA transcribed from pC-297-d showed no signal (not shown). Freeze-thaw extracts of the transfected cells found positive in the first tests were prepared and used for infection of fresh cells with an MOI of about 0.01. Again, the infected cells showed positive immunofluorescence signals, proving that viable viruses had been recovered after transfection of the RNAs with the mutations in Erns. RNA was prepared 48 h postinfection and analyzed by Northern blot. Virus genome could again be detected in all RNA samples except for the noninfected control (Fig. 4). When analyzing RNA yields with a phosphorimager we observed differences of up to a factor of 3 between individual samples in one experiment (data not shown). These differences seemed to be not significant when several independent experiments were compared. Since, however, RNA yield was only analyzed rather late after infection, an influence of the mutations on RNA synthesis cannot be excluded completely and has to be analyzed in further experiments.

FIG. 4.

FIG. 4

Northern blot with RNA from cells infected with cell extracts prepared from cells transfected with in vitro-transcribed RNA containing the mutations resulting in the indicated changes in the Erns sequence. Total RNA of the infected cells was separated in an agarose gel under denaturing conditions, transferred to a nylon membrane, and hybridized with a CSFV-specific probe. On the left side of the gel, the bands of an RNA ladder are indicated.

In the plasmid pC-297-d, from which infectious virus could not be recovered, the wild-type sequence was restored. The exchanged cDNA fragment was sequenced and found to be free of unwanted mutations. From the restored plasmid, infectious virus could be recovered, showing that the failure to obtain viable virus from pC-297-d was indeed due to the deletion of the histidine codon.

To test whether the virus mutants showed an RNase-negative phenotype, we tried to measure RNase activity in extracts of the infected cells. For all viruses, two dishes were infected with an MOI of 0.01, and the cell extracts were prepared 2 days postinfection, a time point at which all cells were found to be infected according to immunofluorescence analyses (not shown). To determine the amount of Erns generated in the infected cells, a second set of two dishes with infected cells was labeled with 35S-amino acids and analyzed by immunoprecipitation and phosphorimager quantification of Erns. The amount of Erns protein present in the different extracts was found to be very similar (data not shown). Noninfected cells and cells infected with the virus recovered from an infectious cDNA clone without mutation served as negative and positive controls, respectively. Infection with the virus without mutation resulted in measurable RNase activity, whereas all mutants were in the same range as the negative control (Fig. 5). Thus, the different exchanges or the deletion of codon 346 in the Erns gene did not interfere with viability of the virus but resulted in an RNase-negative phenotype.

FIG. 5.

FIG. 5

Analysis of RNase activity in extracts of cells infected with CSFV Alfort/Tübingen (Alfort), the virus recovered from the infectious clone pA/CSFV [V(pA/CSFV)], or the indicated mutants thereof at 1 passage postinfection. The given OD values represent the averages of two independent experiments.

The growth characteristics of the recovered virus mutants were analyzed by recording the growth curves. In different experiments with samples obtained from independent transfections no significant differences were observed between the mutants and the virus recovered from pA/CSFV, the construct containing the wild-type sequence (data not shown). Since RNA viruses are known to exhibit a high rate of spontaneous mutation, revertants or pseudorevertants might arise during virus propagation. Therefore, the virus mutants C-346-L, C-346-d, C-297-L/346-L, and C-297-K/346-L were passaged seven times and then analyzed for RNase activity. A virus recovered from pA/CSFV, displaying no mutation in Erns, and noninfected cells served as controls. All mutants still showed an RNase-negative phenotype, indicating that at least no significant portion of the virus population present after seven passages had restored the enzymatic activity (not shown). As a further test, RNA was prepared from cells infected with the different viruses of the seventh passage, and the Erns-coding region was amplified by RT-PCR. The RT-PCR products were isolated, and the region harboring the mutations was sequenced. No indication for reversion was found (not shown). Thus, it can be concluded that the mutations abrogating RNase activity are stable over several passages, indicating that the loss of RNase function does not represent a considerable disadvantage for virus replication in vitro.

Animal experiments with RNase mutant C-346-d.

The function of the RNase activity of Erns for pestiviruses is not known. Our data, as well as those published by Hulst et al. (10), show that this enzymatic activity is dispensable for growth of CSFV in tissue culture cells. It therefore can be speculated that the RNase plays a role during virus replication in the natural hosts of pestiviruses. To analyze the effects of RNase mutations on CSFV propagation in pigs, animal experiments were conducted. To minimize the danger of the generation of revertants possibly leading to unclear results, the deletion mutant C-346-d was used for the first experiment. Three pigs of about 25 kg (body weight) were each infected with 105 TCID50 of virus. As a control, a second group (n = 3) was inoculated with the same amount of V(pA/CSFV), the virus recovered from the infectious clone without RNase mutation. Each group of animals was housed separately in isolation units, and the body temperatures of all animals were recorded daily (Fig. 6). Blood was taken from the animals two times before infection and on days 3, 5, 7, 10, 12, and 14 postinfection. The animals infected with the wild-type virus showed typical symptoms of CSF, such as fever, ataxia, anorexia, diarrhea, central nervous disorders, and hemorrhages in the skin. Virus could be recovered from the blood on days 3 (animal 68), 5, 7, 10, and 14 (animals 68, 78, and 121) postinfection (Table 1). The animals were killed in a moribund stage at day 14 postinfection. At this time, no virus-neutralizing antibodies could be detected (data not shown).

FIG. 6.

FIG. 6

Body temperature curves of animals infected with V(pA/CSFV), the virus recovered from the infectious cDNA clone pA/CSFV (solid lines), or C-346-d, the mutant thereof that contains a deletion of codon 346 (broken lines).

TABLE 1.

Detection of CSFV in the blood of animals infected with V(pA/CSFV) (animals 68, 78, and 121) or C-346-d (animals 70, 72, and 74) determined at the indicated time pointsa

Animal no. Detection of CSFV at (days postinfection):
3 5 7 10 14
68 + + + + +
78 + + + +
121 + + + +
70
72
74
a

PK15 cells were incubated for 1 h with samples of citrate-treated blood obtained from the animals at the indicated time points postinfection. After 72 h the cells were analyzed by immunofluorescence with a CSFV-specific MAb. 

In contrast, the animals infected with the mutant did not develop clinical symptoms. The temperature remained normal over the whole experimental period (Fig. 6). Food consumption remained normal, and no reduction in average body weight gain was observed. At no time could virus be recovered from the blood. Nevertheless, the animals became infected, since all of them developed neutralizing antibodies that were already detectable by day 17 postinfection. The neutralization titers increased during the observation period until day 69 postinfection (Table 2).

TABLE 2.

Determination of neutralization titers of serum samples obtained from animals infected with C-346-d at the indicated time pointsa

Animal no. Neutralization titer at (days postinfection):
−3 0 17 25 69 76 79 87
70 1:18 1:162 1:162 1:162 1:486 1:1,458
72 1:18 1:54 1:486 1:1,458 1:1,458 1:4,374
74 1:6 1:54 1:162 1:162 1:486 1:1,458
a

Animals were challenged with CSFV Eystrup on day 69 postinfection. 

To analyze whether the infection with the mutant virus had induced protective immunity, a challenge experiment was conducted 10 weeks (day 69) after the infection with the CSFV mutant by using the highly pathogenic heterologous CSFV strain Eystrup. A 2 × 105 TCID50 of virus was used for the infection. This amount of virus had been found to be sufficient to induce lethal disease in several earlier studies (12, 17). However, the animals previously infected with the CSFV RNase mutant did not show symptoms of disease after challenge infection. Neither fever nor viremia could be detected (not shown), but an increase in neutralizing antibodies indicated productive infection and replication of the challenge virus (Table 2, values for days 76, 79, and 87 postinfection).

To show that the observed attenuation of the mutant virus is indeed due to the deletion of the histidine at position 346 of the polyprotein and not a consequence of an unidentified second site mutation, the wild-type sequence was restored by substitution of a 1.7-kb XhoI/NdeI fragment of the full-length clone pC-346-d for the corresponding fragment of pA/CSFV displaying the wild-type sequence. The fragment excised from pC-346-d was analyzed by nucleotide sequencing for mutations; except for the deletion of the triplet coding for histidine 346 of the polyprotein no difference with regard to the wild-type sequence was found. From the cDNA construct with the rescued mutant, virus C-346-d/Rs could be recovered that grew as well as wild-type virus and showed equivalent RNase activity (not shown). In a second animal experiment, the rescued virus was used for the infection of pigs. As a control, the deletion mutant C-346-d recovered after an independent RNA transfection experiment from the cDNA construct isolated from a second bacterial clone was used. Again, two groups consisting of three animals were used. Since the animals were younger (ca. 20 kg [body weight]) than those in the first experiment, 5 × 104 TCID50 of virus were used for infection this time. Again, the animals infected with the mutant showed no clinical signs. Two animals had fever for one day (pigs 43 and 47; Fig. 7). One of these animals also showed viremia after 1 day, whereas in the blood of the other two animals virus could not be detected (Table 3). Nevertheless, these animals developed neutralizing antibodies (Table 4) and were protected against a lethal CSFV challenge. Challenge was again performed by infection with a 2 × 105 TCID50 of challenge strain Eystrup. The animals did not show clinical signs after challenge, and the temperature remained normal (not shown). In contrast to the pigs infected with the deletion mutant, the animals inoculated with the rescued wild-type virus had fever (Fig. 7) and developed fatal CSF. One animal had to be killed 11 days after infection; the other two were killed 3 days later. All animals showed typical pathological signs of CSF, such as hemorrhages in different organs (not shown). The blood of the animals was found to contain virus on all days tested (Table 3).

FIG. 7.

FIG. 7

Body temperature curves of animals infected with C-346-d/Rs, the virus recovered from the infectious cDNA clone in which the deletion present in C-346-d had been removed (solid lines) or the original mutant C-346-d (broken lines).

TABLE 3.

Detection of CSFV in the blood of animals infected with C-346-d/Rs (animals 27, 28, and 30) or C-346-d (animals 43, 47, and 87) at the indicated time pointsa

Animal no. Detection of CSFV at (days postinfection):
4 7 11 14
27 + + + +
28 + + + +
30 + + + D
43 +
47
87
a

PK15 cells were incubated for 1 h with peripheral blood lymphocytes purified from blood obtained from the animals at the indicated time points postinfection. After 48 h the cells were analyzed by immunofluorescence with a CSFV-specific MAb. D, animal already dead. 

TABLE 4.

Determination of neutralization titers of serum samples obtained from animals infected with C-346-d/Rs (animals 27, 28, and 30) or C-346-d (animals 43, 47, and 87) at the indicated time pointsa

Animal no. Neutralization titer at (days postinfection):
0 7 11 14 18 33 40 56 78
27 1:6 1:6
28 1:6 1:6
30 D
43 1:6 1:54 1:162 1:162 1:486 1:1,458
47 1:2 1:54 1:162 1:162 1:1,458 1:1,458
87 1:6 1:18 1:162 1:162 1:162 1:486
a

Animals 43, 47, and 87 were challenged with CSFV Eystrup on day 36 postinfection. D, animal already dead. 

Animal experiments with viruses exhibiting point mutations.

The results of the animal experiments showed that the deletion of codon 346 in mutant C-346-d apparently does not prevent virus replication in its natural host but has a major attenuating effect on CSFV. It was tempting to speculate that the drastically reduced virulence of the mutant could be due to the destruction of the RNase activity residing in Erns. However, other effects of the mutation on functions of Erns, i.e., by its probable influence on protein structure, could not be excluded. We therefore performed similar animal experiments with four other RNase-negative viruses, namely, the point mutants C-297-L, C-297-K, and C-346-L and the double mutant C-297-L/346-L. The animals infected with C-346-L or C-297-L/346-L did not show clinical signs of CSF. They had no fever, except for one animal of the group infected with the double mutant that developed fever due to injury on the hind leg; the body temperature of this animal returned to normal values after treatment with an antibiotic (Fig. 8A and B). All animals infected with the double mutant and one animal inoculated with C-346-L showed transient viremia for 1 day. Productive infection of all animals could be demonstrated by the development of neutralizing antibodies (not shown).

FIG. 8.

FIG. 8

FIG. 8

FIG. 8

FIG. 8

Body temperature curves of animals infected with CSFV mutants C-346-L (A), C-297-L/346-L (B), C-297-L (C), or C-297-K (D).

The members of the other two groups infected with viruses containing only mutations in their genomes that affected codon 297 all developed symptoms of disease. The pigs showed fever over a period of up to 5 days (Fig. 8C and D) and also anorexia, apathia, and diarrhea for 1 to 3 days. However, in contrast to the pathology after wild-type virus infection, these animals completely recovered. Their body temperature returned to normal values at around days 9 to 11, and also the other clinical symptoms vanished during the observation period.

Immunological analyses.

The data presented above strongly indicate that the RNase activity residing in glycoprotein Erns of CSFV plays a central role in the pathogenesis of CSF in swine. After identification of the enzymatic activity of Erns, it has been speculated that the RNase might be important for the virus in its interaction with the immune system of the host (3, 9, 10, 28). A characteristic feature of CSF in swine is a dramatic decrease of B cells early after infection (30). To look at whether destruction of RNase activity has an influence on the reduction of the number of B lymphocytes, the percentage of B lymphocytes compared to the percentage of myeloid cells in peripheral blood mononuclear cells (PBMC) was studied during the course of infection by using two-color FCM analyses.

Parallel to the increase of the body temperature (Fig. 7), animals infected with the wild-type-like virus mutant C-346-d/Rs (animals 27, 28, and 30) showed a severe decrease of CD21-positive B lymphocytes. For 7 days after infection the percentages of CD21-positive B lymphocytes dropped to less than 10% of the PBMC population (pig 27; 34% prior infection to 8%; pig 28, 31 to 7%; pig 30, 20 to 5%) (Fig. 9).

FIG. 9.

FIG. 9

Percentage of CD21-positive B lymphocytes in the PBMC of animals infected with virus mutants. Swine infected with the original mutant (C-346-d) are indicated by broken lines. The group of animals infected with the virus recovered from the infectious cDNA clone, in which the deletion had been removed (C-346-d/Rs), is indicated by solid lines. The percentage of CD21-positive B lymphocytes was determined as described in Materials and Methods.

In contrast, animals infected with the RNase-inactivated virus mutant C-346-d (animals 43, 47, and 87) showed no influence of the virus infection on the composition of their PBMC population (Fig. 9). Neither a reduction in the absolute number of PBMC (data not shown) nor in the percentage of CD21-positive B lymphocytes could be detected for up to 46 days after infection.

DISCUSSION

Among members of the family Flaviviridae, pestiviruses are unique because of the expression of two peculiar proteins with enzymatic activity. One of these proteins is the protease Npro. Although Npro is somewhat reminiscent of the leader proteases found in several other viruses, there is still no indication for a function of this polypeptide. Studies with an infectious cDNA of CSFV Alfort187 showed that the Npro gene can be deleted without major effects on virus viability (34). The second unique pestivirus protein is Erns, a glycoprotein exhibiting RNase activity. Erns is essential for virus propagation; attempts to grow a CSFV mutant, from which the complete Erns gene had been deleted, failed (17). This result was not surprising since Erns represents a structural component of the pestivirus particle and a target for virus-neutralizing antibodies. However, destruction of the enzymatic activity of Erns in a CSFV vaccine strain was found to have no impact on virus viability in tissue culture (10). To analyze the effects of mutations abrogating RNase activity after infection of the natural host, we introduced a variety of mutations into the Erns gene of CSFV Alfort/Tübingen. All of these mutations affected either codon 297 or codon 346 of the long ORF in the viral genome that both are conserved between the pestivirus proteins and several other known RNases and have been proposed to be involved in the catalytic reaction (28). Analyses of the functional effects of the different mutations are dependent on a test system for the enzymatic activity of Erns. Such tests have either been conducted with Erns purified after CSFV infection or baculovirus-based expression or protein enriched by specific interaction with a MAb directed against Erns (9, 10, 28, 40). We present here a new test that allows determination of Erns activity in crude cell extracts without further purification of the protein. Since RNases are common cellular enzymes, such a test could not be expected to work without any background problems. However, all control experiments clearly proved the specificity of the assay. This conclusion is based on the facts that RNA degradation occurs in a time- and concentration-dependent manner, that positive results are dependent on the presence of Erns within the cells, and that RNA degradation can be specifically blocked with antibodies directed against Erns. The observed specificity of the test may be explained by the large quantity of Erns in comparison to cellular RNases. However, time course experiments did not indicate the presence of low levels of RNase activity in cell extracts that did not contain functional Erns. It might be that the use of Poly(U) as a substrate is at least in part responsible for the specificity since this homopolymeric RNA could be an inappropriate substrate for cellular RNases, even though bovine RNase A proved to be able to degrade Poly(U) in the control experiments.

As expected, the RNase activity was completely destroyed by all the changes tested in the present study. A critical question was whether all mutations would allow the recovery of viable viruses after introduction into the infectious CSFV cDNA clone. The sequences flanking the two histidine residues in Erns are conserved among pestiviruses, and it is of course possible that this conservation is not only due to the preservation of the active center of the RNase but is also essential for other important functions of this protein. Therefore, Hulst et al. (10) decided to substitute in their experiments the histidine residues for lysine to conserve the basic character at the respective positions in order to minimize putative effects on the protein structure, and this resulted in viable viruses growing as well as wild-type virus. We show here that both histidine residues can also be replaced by the hydrophobic residue leucine without obvious effects on virus propagation in tissue culture. Even after deletion of codon 346, a virus was recovered that grew as well as the wild-type virus. No indication for changes restoring the RNase activity or altering the introduced mutations or flanking sequences could be identified. Thus, the inactivation of the RNase and the putative changes with regard to other hypothetical vital Erns functions are obviously without significant effects on virus replication in tissue culture.

In contrast, a viable virus could not be recovered after deletion of codon 297 showing that residue 297 of the polyprotein is important for an essential Erns function that is independent of the RNase activity.

It has been reported that after inactivation of the RNase of a CSFV vaccine strain by exchanging one of the putative active-site histidine residues against lysine the recovered virus mutants exhibited a cytopathogenic phenotype in tissue culture (10). We did not find any indication for cytopathogenicity after infection with one of our CSFV mutants. Since we also analyzed mutants containing lysine instead of histidine, the different results cannot be due to the nature of the introduced mutations. However, both the parental viruses and the cells used for virus propagation were different, and thus the observed discrepancy can be due to either one of these parameters or the cell/virus interaction as a complex system. Further analyses are necessary to find out the reason(s) for these different observations.

The main purpose of our study was to investigate the effects of mutations abrogating the RNase activity of Erns after infection of the natural host. To get the most clearcut results, we first analyzed the mutant with the deletion of codon 346, for which the danger of obtaining revertants in the animals is low. This virus mutant was found to be completely apathogenic, whereas the pigs infected with the wild-type virus developed fatal CSF. Further experiments with several other virus variants showed that all of the tested mutations inactivating the RNase residing in Erns had an attenuating effect. However, the degree of attenuation varied with the residue affected by the mutations. All viruses showing changes at position 346 did not induce clinical symptoms. In contrast, piglets infected with mutants with changes affecting only codon 297 showed typical symptoms of disease. However, the animals recovered very quickly by around day 9 to 11, the time point at which animals infected with the wild-type virus usually enter an irreversible critical stage of the disease. The fact that the mutants with changes of codon 346 exhibit a higher degree of attenuation indicates that the respective changes do not only affect the RNase activity of Erns but have additional effects on a function of this protein that is important for pathogenesis but not for replication in tissue culture. It is difficult to speculate about the reason for this finding. There seems to be a tendency toward reduced spread in the animals of viruses containing changes at codon 346 of the genome since for these mutants it was difficult or even impossible to reisolate virus from the blood. This might simply point at less-effective replication in the animal, resulting in a lower virus load in the circulation, or it could be the result of important differences such as an interference of the mutation with the ability to infect specific cell types involved in virus distribution. It is, however, very likely that infection of cells and virus replication have occurred also with these mutants, since all of the infected animals developed protective immunity, as demonstrated by the challenge experiments. In conclusion, the results of the animal experiments indicate that inactivation of the RNase activity of Erns leads to attenuation of CSFV and, due to additional effects of mutations at position 346, the degree of attenuation is higher for variants containing changes at this position. There is still a theoretical chance that it is not the inactivation of the RNase but unidentified side effects of the mutations that are responsible for the attenuation of the tested virus mutants. The sequences flanking the putative active side residues of the enzyme are highly conserved among pestiviruses, and it is rather clear that this conservation is not only necessary to preserve the RNase activity but also important for additional functions of the structural protein Erns. It therefore should be kept in mind that the causative relationship between inactivation of RNase activity and the observed attenuation of the viruses has not been formally proven and will be very difficult to prove. However, it seems very unlikely that a variety of different mutations at different positions of the Erns protein that destroy the RNase activity all result in attenuation and that this effect should be independent of the inactivation of the enzymatic activity.

The mutations introduced into the Erns gene seem to be stable not only in tissue culture but also in animus. Analyses of viruses reisolated from the blood of the infected animals did not result in any indications for the presence of revertants. It therefore can be concluded that RNase-negative mutants of CSFV are able to replicate and spread in the natural host. Based on these results, it seems unlikely that the temporary symptoms of disease observed after infection with mutants C-297-L or C-297-K are due to revertants. This conclusion is also based on the fact that, in comparison to infections with wild-type virus, there was no delay in the time point at which fever and clinical symptoms were first observed. If these symptoms were dependent on the generation of virus revertants, a delay would be expected, since some time has to pass to allow the introduction of the necessary mutations. The observation that fever and other clinical symptoms start at the usual time point after infection with C-297-L or C-297-K can also shed some light on the putative mechanism of attenuation of these mutants. Attenuation is due to a faster recovery of the animals rather than to a delayed or strongly ameliorated development of pathological signs. The beginning of the recovery was obvious by days 9 to 11. It has been reported that neutralizing antibodies can first be demonstrated at this time point (days 10 to 12 [32]). It has already been speculated that the RNase function of Erns could be involved in the strategy of pestiviruses to interact with the host immune system (3, 9, 10, 28). Erns is secreted from the infected cells and can be found in the serum of infected animals in considerable amounts. It could be that this protein represents at least one of the factors responsible for the severe reduction or even depletion of lymphocytes, especially B cells, observed during CSFV infection, which hardly can be solely due to the destruction of virus-infected cells. This putative role of Erns could be facilitated by its nature as a secreted protein circulating in the blood of the infected animal. Other RNases are known to exert cytotoxic activities and can have antitumor and immunomodulatory effects (2, 4, 5, 41). Experiments with purified Erns of CSFV indicated that it can induce apoptosis upon addition to lymphocytes of different species (3). Thus, it might be that the RNase activity of the secreted Erns is responsible for the specific decrease of B-cell numbers in the infected animals that probably represents one of the factors leading to virus-induced immunosuppression. The proposed attenuating effect resulting from inactivation of the RNase would then be due to abrogation of B-cell decrease and a faster development of a protective immune response. Our experiments with the mutant C-346-d indeed show that the reduction of B-cell numbers does not occur in animals infected with this variant. Of course, this result does not prove that inactivation of the RNase caused the observed effect, but it fits very well with the working hypothesis outlined above. Further experiments with the other RNase-negative mutants will be conducted in order to analyze this question in more detail.

ACKNOWLEDGMENTS

We thank Silke Esslinger, Petra Wulle, Angelika Braun, and Gaby Kuebart for excellent technical assistance.

REFERENCES

  • 1.Baker J C. Bovine viral diarrhea virus: a review. J Am Vet Med Assoc. 1987;190:1449–1458. [PubMed] [Google Scholar]
  • 2.Benner S A, Allemann R K. The return of pancreatic ribonucleases. Trends Biochem Sci. 1989;14:396–397. doi: 10.1016/0968-0004(89)90282-x. [DOI] [PubMed] [Google Scholar]
  • 3.Bruschke C J M, Hulst M M, Moormann R J M, van Rijn P A, van Oirschot J T. Glycoprotein Erns of pestiviruses induces apoptosis in lymphocytes of several species. J Virol. 1997;71:6692–6696. doi: 10.1128/jvi.71.9.6692-6696.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.D’Alessio G. New and cryptic biological messages from RNases. Trends Cell Biol. 1993;3:106–109. doi: 10.1016/0962-8924(93)90166-x. [DOI] [PubMed] [Google Scholar]
  • 5.D’Alessio G, Di Donato A, Parente A, Picoli R. Seminal RNase: a unique member of the ribonuclease superfamily. Trends Biochem Sci. 1991;16:104–106. doi: 10.1016/0968-0004(91)90042-t. [DOI] [PubMed] [Google Scholar]
  • 6.Denham S, Zwart R J, Whittall J T D, Pambusch M, Ccorteyn A H, Bianchi A T J, Murtaugh M P, Parkhouse R M E, Tlaskalova H, Sinkora J, Sinkora M, Rehakova Z. Monoclonal antibodies putatively identifying porcine B cells. Vet Immunol Immunopathol. 1998;60:317–328. doi: 10.1016/s0165-2427(97)00108-6. [DOI] [PubMed] [Google Scholar]
  • 7.Donis R O, Corapi W, Dubovi E J. Neutralizing monoclonal antibodies to bovine viral diarrhea virus bind to the 56K to 58K glycoprotein. J Gen Virol. 1988;69:77–86. doi: 10.1099/0022-1317-69-1-77. [DOI] [PubMed] [Google Scholar]
  • 8.Fuerst T R, Niles E G, Studier F W, Moss B. Eukaryotic transient expression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase. Proc Natl Acad Sci USA. 1986;83:8122–8126. doi: 10.1073/pnas.83.21.8122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Hulst M M, Himes G, Newbigin E, Moormann R J M. Glycoprotein E2 of classical swine fever virus: expression in insect cells and identification as a ribonuclease. Virology. 1994;200:558–565. doi: 10.1006/viro.1994.1218. [DOI] [PubMed] [Google Scholar]
  • 10.Hulst M M, Panoto F E, Hooekmann A, van Gennip H G P, Moormann R J M. Inactivation of the RNase activity of glycoprotein Erns of classical swine fever virus results in a cytopathogenic virus. J Virol. 1998;72:151–157. doi: 10.1128/jvi.72.1.151-157.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Kessler S W. Use of protein A-bearing staphylococci for the immunoprecipitation and isolation of antigens from cells. Methods Enzymol. 1981;73:442–459. doi: 10.1016/0076-6879(81)73084-2. [DOI] [PubMed] [Google Scholar]
  • 12.König M. Virus der klassischen Schweinepest: Untersuchungen zur Pathogenese und zur Induktion einer protektiven Immunantwort. Ph.D. thesis. Hannover, Germany: Veterinary School of Hannover; 1994. [Google Scholar]
  • 13.Kunkel T A, Roberts J D, Zakour R A. Rapid and efficient site-specific mutagenesis without phenotypic selection. Methods Enzymol. 1987;154:367–392. doi: 10.1016/0076-6879(87)54085-x. [DOI] [PubMed] [Google Scholar]
  • 14.Meyers G, Rümenapf T, Thiel H-J. Molecular cloning and nucleotide sequence of the genome of hog cholera virus. Virology. 1989;171:555–567. doi: 10.1016/0042-6822(89)90625-9. [DOI] [PubMed] [Google Scholar]
  • 15.Meyers G, Tautz N, Becher P, Thiel H-J, Kümmerer B M. Recovery of cytopathogenic and noncytopathogenic bovine viral diarrhea viruses from cDNA constructs. J Virol. 1996;70:8606–8613. doi: 10.1128/jvi.70.12.8606-8613.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Meyers G, Thiel H-J, Rümenapf T. Classical swine fever virus: recovery of infectious viruses from cDNA constructs and generation of recombinant cytopathogenic swine fever virus. J Virol. 1996;67:7088–7095. doi: 10.1128/jvi.70.3.1588-1595.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Meyers, G. Unpublished results.
  • 18.Moennig V, Plagemann J. The pestiviruses. Adv Virus Res. 1992;41:53–91. doi: 10.1016/s0065-3527(08)60035-4. [DOI] [PubMed] [Google Scholar]
  • 19.Paton D J, Lowings J P, Barrett A D. Epitope mapping of the gp53 envelope protein of bovine viral diarrhea virus. Virology. 1992;190:763–772. doi: 10.1016/0042-6822(92)90914-b. [DOI] [PubMed] [Google Scholar]
  • 20.Pesvovitz M D, Lunney J K, Sachs D H. Preparation and characterization of monoclonal antibodies reactive with porcine PBL. J Immunol. 1984;133:368–375. [PubMed] [Google Scholar]
  • 21.Rice C M. Flaviviridae: the viruses and their replication. In: Fields B N, Knipe D M, Howley P M, editors. Fields virology. Philadelphia, Pa: Lippincott-Raven; 1996. pp. 931–959. [Google Scholar]
  • 22.Rümenapf T, Meyers G, Stark R, Thiel H-J. Hog cholera virus characterization of specific antiserum and identification of cDNA clones. Virology. 1989;171:18–27. doi: 10.1016/0042-6822(89)90506-0. [DOI] [PubMed] [Google Scholar]
  • 23.Rümenapf T, Unger G, Strauss J H, Thiel H-J. Processing of the envelope glycoproteins of pestiviruses. J Virol. 1993;67:3288–3294. doi: 10.1128/jvi.67.6.3288-3294.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Saalmüller A, Hirt W, Maurer S, Weiland E. Discrimination between two subsets of porcine CD8+ cytolytic T lymphocytes by the expression of CD5 antigen. Immunology. 1994;81:578–583. [PMC free article] [PubMed] [Google Scholar]
  • 25.Saalmüller A, Reddehase M J, Bühring H-J, Jonjic S, Kozinowski U H. Simultaneous expression of CD4 and CD8 antigens by a substantial proportion of resting porcine T lymphocytes. Eur J Immunol. 1987;17:1297–1301. doi: 10.1002/eji.1830170912. [DOI] [PubMed] [Google Scholar]
  • 26.Saalmüller, A. Unpublished results.
  • 27.Sambrook S, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1989. [Google Scholar]
  • 28.Schneider R, Unger G, Stark R, Schneider-Scherzer E, Thiel H-J. Identification of a structural glycoprotein of an RNA virus as a ribonuclease. Science. 1993;261:1169–1171. doi: 10.1126/science.8356450. [DOI] [PubMed] [Google Scholar]
  • 29.Sopp P, Redknap L, Howard C. Cross-reactivity of human leucocyte differentiation antigen monoclonal antibodies on porcine cells. Vet Immunol Immunopathol. 1998;60:403–408. doi: 10.1016/s0165-2427(97)00114-1. [DOI] [PubMed] [Google Scholar]
  • 30.Susa M, König M, Saalmüller A, Reddehase M J, Thiel H J. Pathogenesis of classical swine fever: B-lymphocyte deficiency caused by hog cholera virus. J Virol. 1992;66:1171–1175. doi: 10.1128/jvi.66.2.1171-1175.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Tautz N, Thiel H-J, Dubovi E, Meyers G. Pathogenesis of mucosal disease: a cytopathogenic pestivirus generated by an internal deletion. J Virol. 1994;68:3289–3297. doi: 10.1128/jvi.68.5.3289-3297.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Thiel H-J, Plagemann G W, Moennig V. The pestiviruses. In: Fields B N, Knipe D M, Howley P M, editors. Fields virology. Philadelphia, Pa: Lippincott-Raven; 1996. pp. 1059–1073. [Google Scholar]
  • 33.Thiel H-J, Stark R, Weiland E, Rümenapf T, Meyers G. Hog cholera virus: molecular composition of virions from a pestivirus. J Virol. 1991;65:4705–4712. doi: 10.1128/jvi.65.9.4705-4712.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Tratschin J D, Moser C, Ruggli N, Hofmann M A. Classical swine fever virus leader proteinase Npro is not required for viral replication in cell culture. J Virol. 1998;72:7681–7684. doi: 10.1128/jvi.72.9.7681-7684.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Van der Werf S, Bradley J, Wimmer E, Studier F W, Dunn J J. Synthesis of infectious poliovirus RNA by purified T7 RNA polymerase. Proc Natl Acad Sci USA. 1986;83:2330–2334. doi: 10.1073/pnas.83.8.2330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.van Rijn P A, van Gennip H G, de Meijer E J, Moormann R J. Epitope mapping of envelope glycoprotein E1 of hog cholera virus strain Brescia. J Gen Virol. 1993;74:2053–2060. doi: 10.1099/0022-1317-74-10-2053. [DOI] [PubMed] [Google Scholar]
  • 37.Weiland E, Ahl R, Stark R, Weiland F, Thiel H-J. A second envelope glycoprotein mediates neutralization of a pestivirus, hog cholera virus. J Virol. 1992;66:3677–3682. doi: 10.1128/jvi.66.6.3677-3682.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Weiland E, Stark R, Haas B, Rümenapf T, Meyers G, Thiel H-J. Pestivirus glycoprotein which induces neutralizing antibodies forms part of a disulfide-linked heterodimer. J Virol. 1990;64:3563–3569. doi: 10.1128/jvi.64.8.3563-3569.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Weiland E, Thiel H-J, Hess G, Weiland F. Development of monoclonal neutralizing antibodies against bovine viral diarrhea virus after pretreatment of mice with normal bovine cells and cyclophosphamide. J Virol Methods. 1989;24:237–244. doi: 10.1016/0166-0934(89)90026-8. [DOI] [PubMed] [Google Scholar]
  • 40.Windisch J M, Schneider R, Stark R, Weiland E, Meyers G, Thiel H-J. RNase of classical swine fever virus: biochemical characterization and inhibition by virus-neutralizing monoclonal antibodies. J Virol. 1996;70:352–358. doi: 10.1128/jvi.70.1.352-358.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Wu Y N, Mikulski S M, Ardelt W, Rybak S M, Youle R J. A cytotoxic ribonuclease. J Biol Chem. 1993;268:10686–10693. [PubMed] [Google Scholar]

Articles from Journal of Virology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES