IMPORTANT LEGAL NOTICE - The information on this site is subject to a disclaimer and a copyright notice.
European Flag    Europa The European Commission Research Agriculture FAIR
index by Area Animal and plant health, animal welfare
index by Country United Kingdom

Control of Classical Swine Fever by Molecular Diagnosis and Epidemiology

Contract nr: FAIR-CT95-0707
Project nr: 707
Project type: SC
Starting date: 01/03/1996
Duration: 39 months
Total cost: 1,533,359 EUR
EC Contribution: 944,139 EUR
Scientific Officer: Isabel MINGUEZ-TUDELA
Research topic: Animal health

Despite intensive efforts, it has not proved possible to eradicate classical swine fever(CSF) in Europe and new outbreaks of the disease continue to occur within EU Member States and Eastern Europe. The direct costs - but also the indirect costs, relating to international trade of both live pigs and pig products - of these outbreaks are enormous. In order to deal with outbreaks, there is a short-term need for improved diagnosis and a longer-term need for a better understanding of why the disease is reappearing.

The aims are to improve the diagnosis of classical swine fever and our ability to trace the spread of the disease. The former will be dealt with through the development, optimisation and harmonised validation of new diagnostic methods based on detection of CSF virus antigens and RNA; the latter through the development of molecular epizootiological methods to permit the origin and spread of outbreaks to be more precisely determined.
The objectives are:

1). to develop novel techniques for swine fever diagnosis that will be superior to existing methods because of enhanced specificity and sensitivity, and the ability to handle a large numbers of tests rapidly;
2). to establish molecular epidemiology as a forensic tool for tracing the origin and spread of new outbreaks of CSF;
3). to speed up the introduction of molecular techniques into routine use by coordination of effort and joint validation of methods.

As regards diagnostic tests, particular emphasis will be placed on establishing tests based on reverse transcription - polymerase chain reaction (RT-PCR) for screening blood samples of live pigs and tissue from abattoir carcasses. Molecular epidemiology will be based on the establishment of a specific fingerprint for each outbreak that will be compared to others in order to determine the likely patterns of the spread.

The principal approaches are:
Task 1. Establish a large and representative panel of CSF viruses
Task 2. Development of specific RT-PCR techniques
Task 3. Strain discrimination by genetic sequencing
Task 4. Comparison of sequence discrimination with other methods
Task 5. Biological characterisation of viruses
Task 6. Obtain organ and blood specimens from infected and uninfected pigs
Task 7. Examine mutation rates of CSF viruses over time

Current situation/results:
Task 1. Establish a large and representative panel of CSF viruses:
The total virus collection now includes more than 600 CSF viruses with associated histories and archived, readily available stocks, compared to around 150 at the start of the project. Information about the isolates was stored on a specially established computer programme accessible via the World Wide Web homepage of the Community Reference Laboratory for swine fever. The database also contains genetic sequence data from the 5'NTR, E2 and/or NS5B regions of the genome (see Task 3 below).

Task 2. Development of specific RT-PCR techniques:
To choose an optimal storage method for blood samples, EDTA-, heparin-, citrate- as well as coagulated blood samples and serum collected at various time points after experimental infection of pigs were compared with respect to their suitability for virus isolation by cell culture and for virus detection by RT-PCR. No significant differences were found between the different blood samples. In general virus was detected in samples between day 4 and day 8 after infection. Samples from day 10 were only positive in RT-PCR but constantly negative in cell culture isolation. Blood taken at day 14 post infection was virus-negative both by RT-PCR as well as by cell culture isolation.
Next, the suitability of commercial kits was assessed for extracting viral RNA to be subsequently detected by RT-PCR. The comparison was carried out on a set of 31 blood samples taken at various times after CSFV infection from three different animals. The Trizol RNA extraction procedure performed equally as well as the Qiagen method but was distinctly less expensive, and was therefore chosen for all future work. Towards the end of the project, a new method - Flowgen Pure Script was also found to be highly suitable and convenient when used to analyse blood samples.
In the earlier part of the project, prototype reverse transcription-polymerase chain reaction (RT-PCR) methods were developed utilising fluorogenic probes to improve specificity and to automate reading. A second generation of improved tests has been developed which significantly reduce the risks of false positive results.
Two large ring tests have been conducted to compare the performance of RT-PCR testing in different laboratories. These ring tests have shown that whereas many different tests are available that provide an adequate level of test sensitivity and specificity, test performance is highly variable between laboratories. There is a clear need for greater quality control in all aspects of RT-PCR testing.

Task 3. Strain discrimination by genetic sequencing:
Viral sequencing efforts were concentrated on three areas of the CSFV genome within the 5'NTR, the E2 gene and the NS5B gene. A large number of viruses have been sequenced using this approach. This data is a very valuable resource for future studies and for tracing back the likely sources of new outbreaks. In order to make the data available to other laboratories through the World Wide Web, a searchable web interface was programmed, which can be accessed using a browser such as Netscape or the Internet Explorer.

The value of genetic typing to the tracing of disease outbreaks was shown repeatedly, and has demonstrated:

1) virus dissemination from a point source of introduction;
2) transmission between domestic pigs and wild boar;
3) transmission across national frontiers;
4) outbreaks of differing virulence associated with very closely related viruses;
5) local persistence of particular variants, most likely in infected wild boar;
6) differentiation between field and vaccine viruses.

Task 4. Comparison of sequence discrimination with other methods:
A study was conducted in the first year of the project to compare antigenic and genetic methods. The results indicated that the correlation between genetic and antigenic typing methods was poor for CSF viruses of genetic group 1, but was better for those of genetic group 2. Since most recent isolates have been of group 2 (especially in Europe), a further study has been undertaken in the final year, utilising German CSF viruses of subgroup 2.3, and characterisation by means of a panel of monoclonal antibodies from Tubingen. The new results indicate that at least for viruses of this genetic subgroup, there is a correlation between the genetic and antigenic typing methods. Monoclonal antibody typing of additional subgroup 2 viruses that have already been typed genetically, is underway.
Methods were developed to utilise amplicons from RT-PCR of the 5'NTR to distinguish between CSFV and other pestiviruses, and between CSFV of groups 1 and 2. The latter can be useful in discriminating between vaccine viruses (all group 1) and field viruses (group 1 does not occur in Europe at present). Further work on use of restriction fragment length polymorphism (RFLP) typing was abandoned, since it seems that the method cannot match the discrimination that is possible from sequencing. Sequencing techniques have now advanced to the stage that they are not only more precise, but also generally quicker and simpler to perform than typing with either monoclonal antibodies or by RFLP.

Task 5. Biological characterisation of viruses and Task 6. Obtain organ and blood specimens from infected and uninfected pigs:
These are considered together due to the close relationship between the tasks.
During the first two years of the project, six animal experiments were performed to characterise different CSFV isolates and to compare the effect of dose and breed susceptibility on the course of infection. In the first three experiments, different Sardinian viruses were compared, all of which were found to be of low-moderate virulence and to induce a sub-acute to chronic form of CSF.
Towards the end of the project three animal experiments have been conducted under standardised conditions to compare virus distribution in pigs infected with either a highly virulent, moderately virulent or an avirulent strain of CSFV. Whereas there was a marked difference in virus distribution, for the avirulent virus, this was not apparent comparing the moderately and highly virulent viruses. In the latter case, the main differences were that the highly virulent strain was more rapidly disseminated throughout the body, and could be found earlier and at a higher titre in most tissues.

Task 7. Examine mutation rates of CSF viruses over time:
Phylogenetic analyses of large numbers of classical swine fever viruses from outbreaks in many different countries have revealed that for several different parts of the genome, the nucleotide sequences of the viral RNA do not change rapidly during transmission between farms, even when followed over a period of several years. This low variability is in contrast to findings with other RNA viruses.
The genetic stability of CSFV has important consequences for molecular epidemiology. It means that discrimination between viruses from closely related outbreaks is not easy, and it may only be possible to say that certain outbreaks are linked rather than to trace the direction of spread. On the other hand, it has also been shown that quite small changes in sequence can be epidemiologically significant.

The aim of improving the diagnosis of classical swine fever and our ability to trace the spread of the disease has been achieved through the following:

  • improved RT-PCR diagnosis - optimised sample collection and processing methods; new sample storage methods; simplified methodology with validated results confirming high sensitivity, tailored specificity, and automated read-out; requirements for quality control;
  • improved virus detection in fixed tissues - new methods for immunohistochemistry and in situ hybridisation;
  • improved genetic typing - a large virus archive and sequence database linked to the internet; an improved knowledge of the extent and distribution of the diversity of CSF; a standardised approach to genetic typing using phylogenetics; a simple alternative to phylogenetics for genetic sequence comparison; a demonstration of the utility (and limitations) of genetic typing to tracing the spread of CSF; and a comparison of genetic typing to other typing methods.

In addition, it has been shown that the CSFV genome is rather stable; an animal model for the study of virulence has been established; but genetic markers of virulence have not been identified.

David James Paton
Veterinary Laboratories Agency
Woodham Lane
UK-KT15 3NB Surrey
Tel: +44 1932 35 72 85
Fax: +44 1932 35 72 39


  • Frank KOENEN
    Nationaal Instituut voor Diergeneeskundig Onderzoek
    Groeselenberg 99
    B-1180 Bruxelles
    Tel.: +32 2 375 44 55
    Fax: +32 2 375 09 79

    Hannover Veterinary School
    Tierarztliche Hochschule Hannover
    Buenteweg 17
    D-30559 Hannover
    Tel.: +49 511 953 88 47
    Fax: +49 511 953 88 98

  • Sandor BELAK
    National Veterinary Institute
    Husargatan 3
    S-751 23 Uppsala
    Tel.: +46 18 67 40 00
    Fax: +46 18 17.45 17

  • Martin A. HOFMANN
    Institute of Virology and Immunoprophylaxis
    CH-3147 Mittelhausern
    Tel.: +41 31 848 92 11
    Fax: +41 31 848 92 22

  • Zygmunt PEJSAK
    The National Veterinary Institute
    Al. Partyzantow 57
    PL-24 100 Pulawy
    Tel.: +48 81 86 30 51
    Fax: +48 81 86 25 95

  • Domenico RUTILI
    Istituto Zooprofilattico Sperimentale dell'Umbria e delle Marche -Perugia
    Via G. Salvemini 1
    I-06126 Perugia
    Tel.: +39 075 94 32 39
    Fax: +39 075 350 47

  Search Top