Comparative sequence analysis of classical swine fever virus isolates from the epizootic in The Netherlands in 1997-1998.

Sixteen classical swine fever virus (CSFV) field isolates from outbreaks of classical swine fever from the period between February 1997 and March 1998 in the Netherlands were sequence analysed. Parts of the 5' noncoding region (5'NCR) and the E1/E2 gene were sequenced after RT-PCR. The obtained sequences were compared with isolates of recent outbreaks in Europe and those of former outbreaks in the Netherlands. Sequence alignment of the 5'NCR region (321 bp) revealed that the isolates of the Dutch outbreak of 1997-1998 were closely linked to an isolate of the CSF outbreak that started in Paderborn, Germany in 1996. A relatively large fragment of the E1/E2 gene of 850 bp, including the antigenic region of E2, which is one of the most variable regions of the CSFV genome, was sequenced to determine whether this region can be used for epidemiology within an epizootic. Epidemiological tracing of transmission of virus was followed, starting from the first isolate and a line of five generations of viruses was analysed. Besides this, new isolates which could not be epidemiologically linked to preceding ones were also characterised. Differences between the isolates of the Dutch outbreak were minor both for the linked as well as for the non-linked isolates, indicating that all isolates have a common origin. Furthermore, our data show for the first time the genetic stability of CSFV even in the highly variable antigenic region of the E2 gene during a major epidemic lasting more than 1 year.     

Molecular epidemiology of classical swine fever in the Russian Federation

The ability to discriminate between various classical swine fever virus (CSFV) strains and isolates is a prerequisite for following the spread of the virus after an outbreak. To determine the relatedness between Russian CSFV isolates from different geographical regions, three fragments of the viral genome (5' NTR, the variable region of the E2 gene and a fragment of the NS5B gene) were sequenced and used for genetic typing. Thirty-one field isolates were obtained from CSF outbreaks which occurred between 1994 and 1999. In addition, three attenuated strains were included in the study, namely the LK and CS vaccine strains, and the moderately virulent 238H isolate. The vaccine strains have been used in Russia for more than 30 years. Our results showed that all field isolates are in subgroup 1.1 together with Alfort 187 and with the highly virulent strain Shimen. In contrast, the CS and LK vaccine strains belong to subgroup 1.2. While there is no evidence for the reversion of the two vaccine strains to wild type, it is feasible that the highly virulent Shimen strain, which has been used as a challenge strain for many years, contributed to field strain generation. The Russian field isolates from the 1990s can be distinguished from the CSF virus isolates which occurred in the EU Member States in the same decade, as here all outbreaks were caused by CSF viruses belonging to subgroup 2.     

Improved strategy for phylogenetic analysis of classical swine fever virus based on full-length E2 encoding sequences

ABSTRACT: Molecular epidemiology has proven to be an essential tool in the control of classical swine fever (CSF) and its use has significantly increased during the past two decades. Phylogenetic analysis is a prerequisite for virus tracing and thus allows implementing more effective control measures. So far, fragments of the 5′NTR (150 nucleotides, nt) and the E2 gene (190 nt) have frequently been used for phylogenetic analyses. The short sequence lengths represent a limiting factor for differentiation of closely related isolates and also for confidence levels of proposed CSFV groups and subgroups. In this study, we used a set of 33 CSFV isolates in order to determine the nucleotide sequences of a 3508-3510 nt region within the 5′ terminal third of the viral genome. Including 22 additional sequences from GenBank database different regions of the genome, comprising the formerly used short 5′NTR and E2 fragments as well as the genomic regions encoding the individual viral proteins Npro, C, Erns, E1, and E2, were compared with respect to variability and suitability for phylogenetic analysis. Full-length E2 encoding sequences (1119 nt) proved to be most suitable for reliable and statistically significant phylogeny and analyses revealed results as good as obtained with the much longer entire 5′NTR-E2 sequences. This strategy is therefore recommended by the EU and OIE Reference Laboratory for CSF as it provides a solid and improved basis for CSFV molecular epidemiology. Finally, the power of this method is illustrated by the phylogenetic analysis of closely related CSFV isolates from a recent outbreak in Lithuania.     

The complete nucleotide sequence of segment L2 of Ibaraki virus encoding for the antigen recognized by neutralizing antibodies

The complete nucleotide sequence of cDNA clones representing the L2 dsRNA from Japan isolate of epizootic hemorrhagic disease serotype 2 (EHDV-2JPN) was determined. The EHDV-2JPN L2 gene is 3002 base pairs long with a single open reading frame of 2949 bp which predicts a polypeptide of 982 amino acid residues. Comparison of VP2 sequence between Japan and North American Isolates of EHDV-2 showed a 72% homology in spite of the same serotype, although those among the North American isolates showed a high genetic identity (>97%).     

Comparison of viraemia- and clinical-based estimates of within- and between-pen transmission of classical swine fever virus from three transmission experiments

Analyses of recent classical swine fever (CSF) epidemics in the European Union have shown that silent circulation of CSF virus (CSFV) occurs before the first outbreak is detected and this may lead to a large epidemic. However, severity of CSF disease signs may be linked with efficacy of disease transmission, the most severely affected animals having a higher infectivity than the less affected ones. The purpose of this study was to combine disease transmission quantification methods with CSF clinical signs quantification tools to investigate whether clinical signs, considered as infectivity markers, may allow us to calculate reliable estimates for disease transmission parameters. Data from three transmission experiments were used, varying according to the viral strain (Eystrup or Paderborn) and to the contact structure between experimentally inoculated and contact animals (direct or indirect contact). Within- and between-pen basic reproduction ratios (R0) were compared using viraemia data or clinical data. Between-pen R0 estimates were close and not significantly >1, with either strain or computation mode (using viraemia or clinical data). Conversely, within-pen R0s (Paderborn strain) computed using clinical data appeared higher than the estimates obtained using viraemia data. A models comparison (Bayes information criterion) showed a better fit of the clinical-based models, for both strains. This suggests that, in affected herds, the most severely affected animals could play a prominent role in CSFV transmission.     

Genetic typing of recent classical swine fever virus isolates from Croatia

During a period of 5 years (1997-2001) several outbreaks of classical swine fever (CSF) were recorded in Croatia. For genetic typing, fragments of 150 nucleotides within the 5'-non-translated region (5'-NTR) and 190 nucleotides within the E2 glycoprotein coding gene of nine field isolates that were derived from domestic pigs and wild boars were used. For better epizootiological understanding, isolates from other European countries were included in the study. The results show that the isolates belong to subgroups 2.1 and 2.3 of CSF virus. Isolates from subgroup 2.1 were collected from domestic pigs during sporadic outbreaks in June 1997 and are genetically closely related. A genomic similarity between these isolates and CSF virus isolates from pigs in other European countries from the same year could also be confirmed. In contrast, the isolate from October 1997 was found to be a member of subgroup 2.3, and is closely related to European CSF virus isolates from outbreaks in the last decade in Western and Central European countries. These results show that two different sources of CSF virus caused outbreaks in Croatia during the same year. Furthermore, a close relationship was found between an isolate from a domestic pig in 1999 and isolates of subgroup 2.3 that originated from Croatian wild boars.     

The leukocyte count is a valuable parameter for detecting classical swine fever

In this paper we describe a study of the use of the white blood cell count (wbcc) as a parameter for detecting outbreaks of Classical Swine Fever (CSF). Meta-analysis of the results of challenge experiments revealed that oronasal infection of SPF-pigs with the virulent CSF virus (CSFV) strains Brescia or NL9201 resulted in a significant decrease in the average white blood cell count during the first week after inoculation of the virus. Challenge of conventional finishing pigs and sows with the moderately virulent strain Paderborn also resulted in a significant decrease in the average wbcc. However, this decrease was not observed after inoculation of SPF pigs with the mildly virulent CSFV strains Henken, Zoelen, or Bergen. The usefulness of clinical inspection in combination with wbcc to detect CSF outbreaks in the field was examined using the results of 214 EDTA blood specimens collected from 22 infected herds and 7250 EDTA blood specimens collected from 1450 non-infected herds. Half of the infected herds had been infected with the moderately virulent CSFV strain Venhorst (closely related to strain Paderborn) during the 1997-98 epidemic in the Netherlands. The other half had been infected with the moderately virulent CSFV strain Loraine. Using these data as a starting point, 1000 samples of one to ten specimens were generated by Monte Carlo simulation. These simulated samples and the samples of the non-infected herds were analysed by use of Receiver Operating Characteristic curves. On the basis of that analysis, the optimal number of animals whose wbcc needed to be determined to detect a CSF outbreak was five. With this number of animals, in conjunction with the threshold of 8000 white blood cells per mm3 (meaning that a herd is designated as CSF suspect if one or more of the five specimens has a white blood cell count of 8000 leukocytes/mm3 or less), the test procedure had a herd sensitivity (HSE) of 94.5% and a herd specificity (HSP) of 97.2%). The HSE is defined as the percentage of samples of infected herds with a positive result of the test procedure; HSP is defined as the percentage of uninfected herds with a negative result of the test procedure. We conclude that the wbcc can help the veterinary practitioner to detect outbreaks of CSF caused by (moderately) virulent CSFV strains. However, for the detection of outbreaks caused by mildly virulent CSFV strains, the contribution of the wbcc is doubtful. Development of additional tools that can improve the clinical diagnosis of the veterinary practitioner remains desirable.     

Studies on the virulence of two field isolates of the classical Swine Fever virus genotype 2.3 rostock in wild boars of different age groups

The virulence of two isolates of the classical swine fever virus (CSFV) was studied in experimentally infected wild boars of different ages. The isolates, originating from wild boars shot in Mecklenburg-Western Pomerania (isolate '1829-NVP') and in Rhineland-Palatinate (isolate '11722-WIL'), belong to the genetic subgroup 2.3 Rostock. Clinical picture, transient viraemia, virus excretion and gross lesions at necropsy as well as a failure of virus detection at the end of the experiment revealed that this virus subtype was only moderately virulent. Whereas one subadult wild boar and both 7-week-old wild boar piglets infected intranasally became sick and died, only one of three 8-week-old animals which survived after contact infection remained CSFV positive until the end of the experiment [34 days post infection (dpi)], although neutralizing antibodies were present. This underlines the role of young boars in CSF epidemics. The isolate '11722-WIL' was shed by an infected adult wild boar and was transmitted to susceptible piglets. Interestingly, all animals which became sick and died also were found to be infected with a secondary pathogen. Therefore, we assume that after infection with moderately virulent CSFV simultaneous infections with other pathogens may be important for the clinical course and the outcome of the disease as well as for a spread of the virus in field.     

Virulence of recent and former classical swine fever virus isolates evaluated by their clinical and pathological signs

The clinical diagnosis of classical swine fever (CSF) still caused problems to the veterinarians during the last decade. The primary CSF outbreak was often detected too late and, meanwhile, the virus had spread. Consequently, the recent classical swine fever virus isolates (CSFV) were suspected to be of low virulence. The purpose of the study was to quantify the virulence of four recent CSFV by evaluating the clinical and pathological signs caused by different CSFV. Pigs of the same breed and age group were inoculated intranasally with CSFV from recent epidemics in European Union (EU) member states. The CSFV used are registered in the data base of the EU Reference Laboratory for CSF and belong to different genotypes: 2.1, 2.2 and 2.3 respectively. Clinical signs of CSF were evaluated by using a score system suggested previously (Mittelholzer et al., 2000: Vet. Microbiol. 74, 293). For the evaluation of pathological lesions, a new pathological score was introduced. The four CSFV tested here were classified as moderately virulent in general, although one CSFV may cause different clinical courses, ranging from highly virulent to avirulent. This indicates the importance of additional factors in the host animal for virulence. Differences in the clinical and pathological signs between these four recent CSFV were rather minor, emphasizing that the genetic typing of CSFV is absolutely essential. Differences towards former CSFV (e.g. reference virus strain Alfort 187) were more pronounced, especially regarding the onset and duration of the disease, the occurrence of skin haemorrhages and pathological lesions of kidney, subcutis and serosae. It is concluded that clinical diagnosis of CSF is rather difficult in pigs up to 14 days post-CSFV infection using these four CSFV, emphasizing the need for careful differential diagnosis and the laboratory investigation for CSF at an early stage.     

Nucleotide sequence and genetic analysis of the leader region of Canadian, American and European equine arteritis virus isolates.

The extreme 5' end, the entire leader sequence of the Arvac vaccine strain, and 10 equine arteritis virus (EAV) isolates, including the ATCC Bucyrus reference strain and 5 Canadian field isolates, were determined and compared at the primary nucleotide and secondary structure levels. The leader sequence of eight EAV isolates, including the Bucyrus reference strain, and the leader sequence of the Arvac vaccine strain was determined to be 206 nt in length (not including the putative 5' cap structure-associated nucleotide) whereas those of the 86AB-A1 and 86NY-A1 isolates were found to be 205 and 207 nt in length, respectively. The sequence identity of the leader sequences, between the different isolates and the Bucyrus reference strain, ranged from 94.2 to 98.5%. Phylogenetic analysis and estimation of genetic distances, based on the leader nucleic acid sequences, showed that all EAV isolates/strains are likely to represent a large phylogenetically-related group. An AUG start codon found at position 14 in all EAV isolates/strains could initiate an open reading frame (ORF) that could produce a polypeptide of 37 amino acids, except for the 86NY-A1 isolate where the intraleader polypeptide would contain 54 amino acids. Computer-predicted RNA secondary structures were identified in the 11 EAV leader regions analyzed. All EAV isolates/strains showed 3 conserved stem-loops (designated A, B and C). An additional conserved stem-loop (D) was observed in 7 EAV isolates, including the Bucyrus reference strain. The leader region distal to stem-loop D did not contain conserved sequences or stem-loop structures common to the EAV isolates/strains.     

Molecular epidemiology of classical swine fever in Italy

To gain an insight into the molecular epidemiology of classical swine fever (CSF) in Italy, virus isolates originating from outbreaks that occurred between 1985 and 2000 in wild boar or in domestic pigs in mainland Italy and in Sardinia were analysed by genetic typing. For this, a fragment (190 nucleotides) of the E2 glycoprotein gene was sequenced and phylogenetic analyses were performed, including older Italian isolates and isolates from recent outbreaks in Europe for comparison. The results show that in mainland Italy, several independent epidemiological events occurred in the last decade. In the north of the country, viruses of genotype 2.2 have persisted in wild boar, causing sporadic outbreaks in domestic pigs. In contrast, viruses of subgroups 2.1 and 2.3 appeared only intermittently in different regions of the mainland. In 1997, classical swine fever virus (CSFV) isolates belonging to the subgroup 2.1 and genetically and epidemiologically related to the Dutch isolate in Venhorst, affected domestic pigs exclusively. The isolates of subgroup 2.3, derived from wild boar as well as from domestic pigs were closely related to isolates from Germany and Poland. In Sardinia, CSF is an endemic in wild boar and affects domestic pigs also. Genetic typing showed that viruses of subgroups 1.1 and 2.3 have been present, the last ones being unrelated to the mainland viruses of the same subgroup. Due to the large quantities of pig and wild boar meat imported in some parts of Italy, it cannot be established if these viruses were always present in either the mainland or Sardinia, or if they represent recent introductions.     

禽流感病毒分离株A/Chicken/Xinjiang/1/96(H14N5)NP基因序列分析

对ＲＴＰＣＲ扩增的禽流感病毒Ａ／Ｃｈｉｃｋｅｎ／Ｘｉｎｊｉａｎｇ／１／９６（Ｈ１４Ｎ５）分离株核蛋白基因进行了序列分析。结果表明：所克隆的基因包含了全部核蛋白阅读框架,编码区为１４９４个核苷酸。比较性研究表明,该毒株与Ａ／Ｍａｌａｒｄ／Ａｓｔｒａｋｈａｎ（Ｇｕｒｊｅｖ）／２６３／８２（Ｈ１４Ｎ５）有很高的同源性,而与人流感病毒株有较大,显示出明显的禽流感病毒特征。     

猪瘟兔化弱毒疫苗株基因组的遗传变异分析

参照已发表的猪瘟病毒基因组序列设计了9对引物，用RT-PCR从猪瘟兔化弱毒疫苗株细胞培养物中扩增得到了覆盖猪瘟病毒基因组全长的9个cDNA片段，将所得cDNA片段分别克隆至pMD18-T载体中，经测序和拼接后，获得了猪瘟兔化弱毒疫苗株基因组全序列。序列分析表明，猪瘟兔化弱毒疫苗株基因组全长12310个碱基，其5’非编码区（5＇-NCR）和3＇-NCR分别由373和239个碱基组成，在3’末端有富含T的碱基插入，其间为1个大的开放阅读框架，编码3898个氨基酸残基的多聚蛋白，与国内外已发表的另外7个猪瘟兔化弱毒疫苗株基因组全序列相比，核苷酸同源性为98．7％～99．9％，氨基酸同源性为98．6％～99．9％。基因组全序列比较显示，猪瘟兔化弱毒疫苗株基因组在遗传上相当稳定。     

Improved sero-monitoring assay for classical swine fever (CSF) using the recombinant E2 protein of a recent Korean isolate

Classical swine fever (CSF) is a highly contagious disease of pigs that causes fever, diarrhea and paralysis, often resulting in death. E2 is the major structural protein of the CSF virus (CSFV) and mediates the entrance of the virus, subsequently inducing a neutralizing immune response. In this study, the E2 gene of a recent Korean isolate of CSF, SW03, was cloned and the DNA sequence was compared to other strains via phylogenetic analysis. With the purified E2 protein, an enzyme-linked immunosorbent assay (ELISA) was developed for the serodiagnosis of CSFV infection. The sensitivity and specificity of the E2-ELISA were 96.1% and 94.8%, respectively. A total of 17 out of 485 field-collected pig sera tested demonstrated conflicting results between two ELISA methods, a commercial kit and the E2-ELISA. Of these sera, 60% were determined to be CSFV positive by a virus neutralization test (VNT), suggesting involvement of different immune responses in the cases of CSFV infection. As the E2-ELISA was developed using a recent Korean isolate, SW03, this assay is capable of rapidly identifying newly emerging CSFV strains.     

Emergence of 2.1. subgenotype of classical swine fever virus in pig population of India in 2011

Background: Limited studies are available on molecular epidemiology of classical swine fever virus (CSFV) in India and are restricted to domestic pigs. These studies show the presence of 1.1. genotype.

Hypothesis/objectives: The aim of the present study was to subgenotype four CSFV isolates, two each from the outbreaks of CSF in wild (Sus scrofa) and domestic pigs of Mizoram state, India, in 2011.

Animals and methods: CSFV isolates were subjected to nucleotide sequencing in E2 and NS5B genomic regions. Phylogenetic analysis of the isolates in both genomic regions was carried out with 39 Indian isolates (4 isolates from the present study of Mizoram state and 35 isolates from the other states of India) and 57 reference sequences retrieved from the GenBank database. Two of the 39 isolates from India were collected from wild boar and were subgenotyped as 2.1. Out of 37 isolates from domestic pigs, only two were subgenotyped as 2.1.

Results: The analysis revealed the emergence of 2.1. subgenotype of CSFV in both wild and domestic pigs in India.

Conclusions and clinical importance: The isolates from domestic pigs of Mizoram state (CSF/MZ/KOL/73 and CSF/MZ/AIZ/115) were grouped in genotype 1 and subgenotype 1.1., thus confirming that the source of CSF outbreaks in domesticated pigs in Mizoram was not from wild pigs. The current study forms an essential step for better understanding of the epidemiology of 2.1 subgroup as well as the movement and spread of the disease in India.     

Transmission of classical swine fever virus by artificial insemination.

Classical swine fever (CSF) virus was introduced into an artificial insemination centre during the CSF epizootic of 1997-1998 in the Netherlands. The risk of further spread of CSF virus via contaminated semen was recognised, but could not be assessed because scientific data on this issue were not available. An animal experiment was performed to determine whether CSF virus could be transmitted via artificial insemination with contaminated semen. Three boars were inoculated with a CSF virus field isolate and from Day 5 till Day 18 thereafter, ejaculates were collected and prepared for insemination. Ruttish sows were inseminated with the extended semen from Day 5 till Day 18 after inoculation of the boars. All the inoculated boars remained healthy throughout the experiment and developed CSF neutralising antibodies between 14 and 21 days after inoculation. Virus was isolated from several semen samples collected from 5 till 11 days after inoculation. Two out of six sows inseminated with CSF contaminated semen seroconverted after insemination. All the other sows remained seronegative. In the foetuses of both the seropositive sows, CSF virus was detected at approximately 35 days post insemination. These results demonstrate that adult boars infected with CSF virus can excrete virus with semen and can, subsequently, transmit the virus to sows and their foetuses via artificial insemination.     

Replication of classical swine fever virus strains and isolates in different porcine cell lines

Classical swine fever virus (CSFV) is an economically important pathogen of domestic pigs and wild boar. Due to the highly variable clinical picture of CSF, laboratory methods are essential for an unambiguous diagnosis. Virus isolation using cell culture is still considered the gold standard. It is based on the incubation of permissive cells with organ or leukocyte preparations followed by antigen detection. In the "EU Diagnostic Manual for CSF Diagnosis", the permanent cell line PK(15) (porcine kidney) is recommended. In the European Reference Laboratory (EURL) a clone of this cell line, PK(15)A, and the STE (swine testicular epitheloid) cell line are in use for propagation of CSFV. The aim of this work was to assess the relative ability of eleven permanent cell lines derived from various organs of wild boar and domestic pig, respectively, to support the replication of different strains and isolates in comparison to these cell lines. An avirulent and a highly virulent laboratory CSFV strain, and several recent field isolates from domestic pigs and wild boars were used. Titers were determined after one, two and three virus passages, and after 48 and 120 h of incubation. Of the eleven cell lines analyzed, two were found that replicated all the tested CSFV strains and field isolates. Those may be useful for improving diagnosis of CSFV and for preparing low-passaged virus stocks of new isolates.     

The classical swine fever epidemic 1997-1998 in The Netherlands: descriptive epidemiology

The objective of this paper is to describe the severe epidemic of classical swine fever (CSF) in The Netherlands in 1997–1998 under a policy of non-vaccination, intensive surveillance, pre-emptive slaughter and stamping out in an area which has one of the highest pig and herd densities in Europe.

The primary outbreak was detected on 4 February 1997 on a mixed sow and finishing pig herd. A total of 429 outbreaks was observed during the epidemic, and approximately 700 000 pigs from these herds were slaughtered. Among these outbreaks were two artificial insemination centres, which resulted in a CSF-suspect declaration of 1680 pig herds (mainly located in the southern part of The Netherlands). The time between introduction of CSF virus (CSFV) into the country and diagnosis of CSF in the primary outbreak was estimated to be approximately 6 weeks. It is presumed that CSFV was spread from The Netherlands to Italy and Spain via shipment of infected piglets in the beginning of February 1997, before the establishment of a total stand-still of transportation. In June 1997, CSFV is presumed to be introduced into Belgium from The Netherlands.

Pre-emptive slaughter of herds that had been in contact with infected herds or were located in close vicinity of infected herds, was carried out around the first two outbreaks. However, this policy was not further exercised till mid-April 1997, when pre-emptive slaughter became a standard operational procedure for the rest of the epidemic. In total, 1286 pig herds were pre-emptively slaughtered. (approximately 1.1 million pigs). A total of 44 outbreaks (10%) was detected via pre-emptive slaughter.

When there were clinical signs, the observed symptoms in infected herds were mainly atypical: fever, apathy, ataxia or a combination of these signs. In 322 out of 429 outbreaks (75%), detection was bases on clinical signs observed: 32% was detected by the farmer, 25% by the veterinary practitioner, 10% of the outbreaks by tracing teams and 8% by screening teams of the veterinary authorities. In 76% of the outbreaks detected by clinical signs, the farmer reported to have seen clinical symptoms for less than 1 week before diagnosis, in 22% for 1–4 weeks before diagnosis, and in 4 herds (1%) the farmer reported to have seen clinical symptoms for more than 4 weeks before diagnosis.

Transportation lorries played a major role in the transmission of CSFV before the primary outbreak was diagnosed. It is estimated that approximately 39 herds were already infected before the first measures of the eradication campaign came into force.

After the first measures to stop the spread of CSFV had been implemented, the distribution of the most likely routes of transmission markedly changed. In most outbreaks, a neighbourhood infection was indicated.

Basically, there were two reasons for this catastrophe. Firstly, there was the extent of the period between introduction of the virus in the region and detection of the first outbreak. As a result, CSFV had opportunities to spread from one herd to another during this period. Secondly, the measures initially taken did not prove sufficient in the swine- and herd-dense region involved.     

Genetic typing of classical swine fever virus

Three regions of the classical swine fever virus (CSFV) genome that have been widely sequenced were compared with respect to their ability to discriminate between isolates and to segregate viruses into genetic groups. Sequence data-sets were assembled for 55 CSFVs comprising 150 nucleotides of the 5' non-translated region, 190 nucleotides of the E2 envelope glycoprotein gene and 409 nucleotides of the NS5B polymerase gene. Phylogenetic analysis of each data-set revealed similar groups and subgroups. For closely related viruses, the more variable or larger data-sets gave better discrimination, and the most reliable classification was obtained with sequence data from the NS5B region. No evidence was found for intertypic recombination between CSFVs. A larger data-set was also analysed comprising 190 nucleotides of E2 sequence from 100 CSFVs from different parts of the world, in order to assess the extent and global distribution of CSFV diversity. Additional groups of CSFV are evident from Asia and the nomenclature of Lowings et al. (1996) [Lowings, P., Ibata, G., Needham, J., Paton, D., 1996. J. Gen. Virol. 77, 1311-1321] needs to be updated to accommodate these. A tentative assignment, adapting rather than overturning the previous nomenclature divides CSF viruses into three groups with three or four subgroups: 1.1, 1.2, 1.3; 2.1, 2.2, 2.3; 3.1, 3.2, 3.3, 3.4. The expanding data-base of CSFV sequences should improve the prospects of disease tracing in the future, and provide a basis for a standardised approach to ensure that results from different laboratories are comparable.