Bovine viral diarrhea virus (BVDV) has been classified within the genus Pestivirus of the family Flaviviridae (Wengler et al., 1995). It has a positive single-stranded RNA genome of approximately 12.3kb size, encoding one open reading frame (ORF) that is translated into a single polyprotein of about 4000 kDa (Collett et al., 1988; Meyers and Thiel, 1996). The polyprotein is subsequently cleaved into 4 structural and 6 to 7 non-structural proteins by viral and cellular proteases (Thiel et al., 1993). The ORF, which starts with the Npro viral autoprotease, is flanked at the 5’ and 3’ termini by untranslated regions (5’-UTR, 3’-UTR) (Collett et al., 1988; Meyers and Thiel, 1996).
Two species of BVDV, BVDV-1 and BVDV-2, have been described (Pellerin et al., 1994; Ridpath et al., 1994). The genotype is determined by genetic typing, usually using sequences from the 5’-UTR, Npro and E2 genetic regions. BVDV is genetically highly variable, particularly BVDV-1, where at least 11 genogroups have been identified (Vilcek et al., 2001). BVDV-1a and BVDV-1b seem to be the most prevalent genogroups worldwide (Fulton et al., 2003; Vilcek et al., 2004). In addition, due to their effects on permissive cells, two biotypes can be distinguished, the cytopathic (cp) and the noncytopathic (ncp) (Donis and Dubovi, 1987; Greiser-Wilke et al., 1992).
Animals which become infected intrauterinely with ncp-BVDV (50 to 150 days of gestation) are epidemiologically important, as they may become tolerant to the virus. These persistently infected (PI) animals are the major source of virus maintenance and shedding in dairy farms and readily infect all susceptible animals. Many persistently infected animals appear clinically normal, while others may be weak at birth (Kelling, 1996).
BVDV-1 infections involve mainly respiratory, reproductive and enteric diseases, causing considerable economic losses to the cattle farming industry worldwide (Baker et al., 1954; Brownlie et al., 1987; Nettleton and Entican, 1995). BVDV-2 causes similar clinical signs to BVDV-1, except that infection with a highly virulent isolate may lead to thrombocytopaenia and fatal haemorrhagic syndrome (Corapi et al., 1989; Carman et al., 1998; Liebler-Tenorio et al., 2002).
Genetic and antigenic diversity of BVDV is important to consider for the design of effective vaccination programs (Fulton et al., 2003; Vilcek et al., 2004). The objective of this study was to determine genetic diversity of BVD viruses present in dairy herds from Costa Rica, Central America, and to describe management routines carried out in these farms that are important for the epidemiology of the disease.
MATERIALS AND METHODS
A total of 16 BVDV isolates were analyzed in this study. Samples were collected in 2006 from dairy herds (Figure 1) or submitted to the Laboratory of Virology between 1987 and 2006. Seven positive animals were detected among the 1,980 serum samples collected and tested individually for BVDV antigen using a capture ELISA (HerdCheck, BVDV/Serum Plus, Idexx Laboratories, Österbybruk) during 2006 from 29 specialized dairy farms from Cartago, Heredia, Guanacaste, San José and Alajuela. The number of animals tested for BVDV antigen to determine the percentage of PI animals was determined based on Cannon and Roe (1982) (1% expected prevalence of PI animals and a 95% confidence level), taking into account a total of 60,000 dairy cattle and 1,600 dairy herds in Costa Rica (Anonymous, 2000). Surveys were carried out on these farms to obtain information about cattle management practices (purchase of animals, BVDV examination and vaccination practices, replacement strategies), handling of calves (elimination strategies and location of calves on the farm) and observation of BVDV associated diseases (retarded growth, abortion, mucosal disease, hemorrhagic syndrome) (Table 2). The additional 9 positive samples were clinical submissions to the Laboratory of Virology, Universidad Nacional in Costa Rica, over the last 20 years (1987-2006). All 9 samples were tested the day of submission for virus isolation and tested afterwards under direct immunofluorescence. Non-infected Madin-Darby Bovine Kidney (MDBK) cells were included as negative control. The isolates were kept at –70°C, until molecular testing was carried out in 2006. Data concerning the year of submission and the region of origin was available only for 5 samples.
All 16 BVDV positive samples were subject to 4 passages on MDBK cell cultures grown in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% BVDVfree fetal bovine serum (FBS) and antibiotics (200 IU of penicillin/ml, 200 µg of streptomycin/ml). Total RNA was extracted from cell culture supernatants using QIAamp Viral RNA Mini Kit® (Qiagen, Hilden) following manufacturer’s instructions. Non-inoculated MDBK cell cultures were included as negative controls.
Reverse Transcriptase PCR (RT-PCR) and Sequencing
Synthesis of cDNA was performed with random hexamers (Fermentas Inc, Glen Burnie, Maryland) in 40 μl total reaction volume. Reverse transcriptase polymerase chain reaction (RT-PCR) was carried out as previously described (Letellier et al., 1999). A 221 bp DNA product was amplified from the 5’-UTR for BVDV-1 and BVDV-2. Primers used were B3 (5’-GGT AGC AAC AGT GGT GAG-3’) and B4 (5’-GTA GCA ATA CAG TGG GCC-3’) to determine BVDV-1 and primers B5 (5’-ACT AGC GGT AGC AGT GAG-3’) and B6 (5’-CTA GCG GAA TAG CAG GTC-3’) to determine BVDV-2, respectively. RT-PCR products were visualized by 2% agarose gel electrophoresis and subsequent staining in ethidium bromide. Using primer OL100 (5’-CAT GCC CTT AGT AGG ACT AGC-3’) and primer 1400R (5’-ACC AGT TGC ACC AAC CAT G-3’), a 1341 bp product encompassing the Npro region was amplified (Becher et al., 1999). PCR amplicons were purified using QIAquick® PCR Purification Kit (Qiagen) and sequenced in both directions (MWG Biotech AG, Martinsried, Germany).
Nucleotide sequences were edited and analysed using the BioEdit software, version 18.104.22.168 (Hall, 1999). As with most BVDV isolates, only shorter fragments encompassing about 300 nu are available from GenBank. The corresponding fragments were used for calculation of the phylogenetic tree. Alignment and calculation of the unrooted phylogenetic neighbour-joining tree was performed in ClustalX 1.83 (Thompson et al., 1997) and bootstrapping was performed with 1,000 replicates. For the graphic output, Tree View software, version 1.6.6 (Page, 1996) was applied. The EMBL accession numbers of the sequences included as controls are listed in Table 1.
A total of 7 BVDV positive samples were detected in 4 out of 29 tested dairy farms in 2006, determining a percentage of 0.35% of PI animals. The survey yielded the following herd management practices: all farms had a veterinarian visit periodically (100%), 69% of the farms obtained their replacements completely within their own farm, whereas for the rest (31%) replacements were purchased from other farms. All the farms reported abortion and calves with retarded growth, and in 83% of the cases weak calves were eliminated from the farms. A total of 90% of the farms kept their calves at a short distance from adults, whereas 10% of the farms raised calves together with adult bovines. BVDV vaccines were used in 14% of the farms (Table 2).
The analysis by RT-PCR indicated that the 16 BVDV-positive samples for antigen capture ELISA and direct immunofluorescence assay belonged to BVDV species 1. BVDV-2 isolates were not detected. The determination of the nucleotide sequence of the Npro coding region was performed for 11 of these 16 BVDV field isolates. Nine field isolates (CR-01 to CR-06, CR11, CR-14 and CR-16) had 100% identical sequences in the fragment analyzed, whereas the other 2 field isolates (CR-12 and CR-15) had single base exchanges. The resulting phylogenetic tree indicated that the 11 field isolates from different geographic areas and collected at different dates belonged to the BVDV-1b subgroup (Figures 1 and 2). They formed a cluster that was separated by a high bootstrap value from other isolates in this subgroup.
In the present study, a low percentage of PI animals was detected, which could be consequence of good herd management strategies in most of the dairy farms surveyed. Another possible reason for a low percentage of PI animals could be the sampling method used, since approximately 25% of adult animals were tested in each farm. The PI animals detected in two farms with good herd management strategies were 2 heifers identified only by laboratory testing. The other two herds where 5 PI animals were detected had an open herd management system, did not eliminate weak or retarded calves and vaccinated against BVDV. The PI animals detected in these two farms were 4 calves (5 to 9 months of age) showing reduced growth and 1 heifer (Greiser-Wilke et al., 2003).
The phylogenetic tree calculated using the 300 nu fragment of the Npro clearly allowed for distinguishing between the individual BVDV subgroups, showing high bootstrap values (Nei and Kumar, 2000).
The results of the present study indicate that BVDV field viruses from Costa Rica could be genetically homogeneous with minimal variability over time (from 1987 to 2006) and geographical origin. These results are in accordance with reports from the United States of America, where BVDV-1b was recognized as the most prevalent genotype of BVDV (Fulton et al., 2003; Tajima, 2006). In contrast, genotyping of BVDV isolates from South America revealed considerable genetic heterogeneity (Vilcek et al., 2004). The sanitary barrier to prevent the introduction of Foot and Mouth Disease into Costa Rica prohibits the import of cattle from South America. However, an increase of import of livestock from Canada, USA and Mexico to Costa Rica occurred mainly in the 60s. This might be the reason why Costa Rican isolates were homogenous and all in genogroup BVDV-1b (OIE, 2009).
The predominance of BVDV-1b in Costa Rica may have an impact on vaccination programmes. All vaccines used in Costa Rica in 2006, until today, contained BVDV-1a and BVDV-2a. These vaccines may not properly protect bovines against BVDV-1b, since only low neutralizing antibody titers against BVDV-1b were induced by BVDV-1a in commercial vaccines (Fulton et al., 2002; Fulton et al., 2003); however, a commercially VDVB-1b vaccine is not yet available (Ishmael, 2009). It will be interesting in future studies to determine the predominant clinical signs caused by this genogroup in susceptible bovine herds from Costa Rica, since BVDV-1b is generally isolated from calves with respiratory symptoms (Fulton et al,. 2002).
This research was supported by Cooperativa de Productores de Leche R.L., FUNDAUNA Project No. 081602, the School of Veterinary Medicine at Universidad Nacional (Costa Rica), the Institute of Virology at Tierärztliche Hochschule Hannover (Germany), and Deutscher Akademischer Austauschdienst (DAAD) (Bonn, Germany). We would like to thank Rocío Cortés for her technical assistance and all veterinarians involved for their help with the sample collection, especially Luis Diego Rodríguez, Jaime Murillo and Julio Murillo. Farm owners’ collaboration is also gratefully acknowledged. Andrea Chaves received a fellowship from the Costa Rican Ministry of Science and Technology (MICIT) and the National Council of Science and Technology (CONICIT). This research was performed as a partial requirement for the M.Sc. degree of Andrea Chaves at Universidad Nacional.
This article was originally published in Ciencias Veterinarias, 26(1), 37-46, 2008. This is an Open Access article distributed under the terms of the Creative Commons Attribution License.