Bovine parainfluenza 3 virus is a nonsegmented, negative-strand RNA virus of the genus Respirovirus in the family Paramyxoviridae.1 Other viruses of the genus Respirovirus include Sendai virus of mice and HPV types 1 (HPV1) and 3 (HPV3), which are genetically and antigenically related to BPI3V.1 The single-stranded genome of BPI3V has approximately 15,500 nucleotides that comprise 6 genes, which encode 9 proteins. Host antibody responses are generated against the HN and fusion envelope proteins, which are poorly conserved among viral isolates and are involved in the attachment and penetration of the host cell. Antibody titers against BPI3V are more common and generally higher in older animals than in younger animals2,3; however, the antibody titer required for protection against disease is unknown, as is the duration of protection provided by those antibodies.
Bovine parainfluenza 3 virus was first identified in the United States in 1959 and is now endemic in cattle herds, although its adverse health effects are often underappreciated.4 In cattle, clinical signs associated with BPI3V infection range from none (subclinical) to acute respiratory disease characterized by pyrexia, nasal discharge, and coughing. Cattle infected with BPI3V that develop severe disease are generally coinfected with other viral and bacterial respiratory tract pathogens.5–7 Bovine parainfluenza 3 virus is frequently involved in the pathogenesis of bovine respiratory disease complex, where its primary role is to predispose infected animals to secondary infections.
Serologic evidence of BPI3V infection has been reported in several domestic and free-ranging ungulate species such as cattle, goats,8 sheep,9 bighorn sheep,10 camels,11 and New World camelids.12,13 Bovine parainfluenza 3 virus has also been isolated from swine with clinical disease, but it appears to be unable to establish endemicity in porcine populations.14 Bovine parainfluenza 3 virus is not considered a serious zoonotic pathogen1,2 and holds potential as a vaccine candidate for the closely related HPV3.15 Although Shibuta et al16 described the use of monoclonal antibodies to identify antigenic variations among BPI3V isolates in 1986, only recently has genetic sequencing of the matrix protein allowed for the differentiation of BPI3V isolates into 3 distinct genotypes (ie, A, B, and C).17,18 With advances and improvements in sequencing technology, full genome sequencing is now commonly used for identification of viral genotypes. For BPI3V, the identity percentage for whole genome sequencing generally ranges from 91% to 99% within genotypes and from 81% to 84% between genotypes.19 Differences in the virulence of BPI3V have not been well characterized on the basis of genotype,4 and aside from genomic analysis, genotypes cannot be differentiated clinically. Historically, large differences in the geographic distribution of various genotypes of BPI3V have been reported. Genotype A (BPI3V-A) has been identified in cattle populations worldwide, whereas genotype B (BPI3V-B) is found primarily in Australia,17 and genotype C (BPI3V-C) is found in China,18 Agrentina,20 and Korea.21 Recently, BPI3V-B and BPI3V-C were identified by the use of whole genome sequencing in US cattle with clinical disease19; that was the first time those 2 genotypes had been isolated in the United States. The geographic distribution of non-A genotypes of BPI3V in the United States is unknown. Consequently, the purpose of the study reported here was to conduct a serologic survey for antibodies against the 3 genotypes of BPI3V in unvaccinated ungulates in Alabama.
Materials and Methods
Animals
All study procedures were approved by and carried out in accordance with requirements established by the Auburn University Institutional Animal Care and Use Committee. Serum samples were evaluated for antibodies against BPI3V genotypes A, B, and C from 6 distinct animal groups that had not been vaccinated against BPI3V. Group 1 consisted of 12 randomly selected Angus and Angus-crossbred heifers from a population of approximately 60 similar heifers housed at the North Auburn Beef Unit in Auburn, Ala. A random-number generator was used to randomly select 20% of the population for evaluation. All heifers were approximately 18 months old and were born at either the North Auburn Beef Unit or Upper Coastal Plain Agricultural Research Center in Winfield, Ala. Both herds are managed by the Animal Health Research Service of the Auburn University College of Veterinary Medicine. Groups 2, 3, 4, and 5 were likewise managed by the Animal Health Research Service but were housed in separate pastures in Auburn, Ala. Group 2 consisted of 11 (7 mature cows and 4 yearling heifers) Miniature Zebu–crossbred cattle, which represented all mature cows and yearling heifers in that herd. All cattle evaluated in group 2 were born and raised at the university-owned facility in Auburn, Ala. Group 3 consisted of 4 Holstein steers that were approximately 2 years old. Each of the steers had a surgically created rumen fistula. The steers were born at the Auburn University Dairy in Shorter, Ala, or the Auburn University College of Veterinary Medicine Dairy in Auburn, Ala, and had been comingled together since they were 3 months old. Group 4 consisted of all 6 mature Boer and Boer-crossbred does housed at the university research facility. All 6 goats were born and raised at the facility. Group 5 consisted of the 26 alpacas and 3 llamas that comprised the Auburn University College of Veterinary Medicine camelid teaching herd. One alpaca was a yearling, and the remaining 25 alpacas and 3 llamas were mature females or geldings. The herd had been assembled from donated animals from throughout the southeastern United States. All animals evaluated in group 5 had been part of the herd since birth or for at least 3 years. Group 6 consisted of a convenience sample (n = 21) of 40 white-tailed deer that were captured between 2005 and 2009 as described.22
Virus neutralization assay
A blood sample (approx 8 mL) was collected by jugular venipuncture from each animal in groups 1 through 5. Serum was harvested from each sample and assayed within 4 hours after collection. Banked serum samples were assayed for the deer evaluated in group 6, and those samples were stored at −80°C until analysis. All serum samples for each group were assayed in 1 batch to minimize batch-to-batch and between-day variation. All assays were performed by the same investigators (PKG, KPR, and YZ) within a 3-week period. The samples for groups 1 and 3 were assayed on the same day as were the samples for groups 2 and 4. The samples for groups 5 and 6 were assayed on separate days. Thus, serum samples were assayed on 4 separate days during that 3-week period.
For each sample, a standard virus neutralization microtiter assay was used to determine the antibody titer against each of the 3 BPI3V genotypes (BPI3V-A, BPI3V-B, and BPI3V-C). Each sample was assayed in triplicate for each genotype. Briefly, serum samples were heat-inactivated for 30 minutes at 56°C. For each sample, serial 2-fold dilutions were created with 50 μL of culture medium and placed in wells of a microtiter plate. To each well, 50 μL of stock solution that contained the appropriate indicator virus (4 cell culture infectious doses/μL) was added. For each BPI3V genotype, the indicator virus was isolated and sequenced by whole genome sequencing as described.19 The indicator viruses selected for the study were recent clinical isolates from US cattle. For each indicator virus, genome sequencing revealed a lack of ambiguous base sequence results and the absence of other viruses, which suggested that it was a pure culture of that isolate. The indicator viruses used for BPI3V-A, BPI3V-B, and BPI3V-C were TVMD60, TVMD15, and TVMD19, respectively (Appendix). Phylogenetic analysis of the HN gene of each indicator virus confirmed the separation of the isolates by genotype and revealed that TVMD60 was closely related to SF4, a BPI3V-A strain previously isolated from US cattle and commonly used as the reference (indicator) strain in virus neutralization assays (Figure 1). Following inoculation, the plates were incubated at 37°C in a humidified atmosphere with 5% CO2 for 1 hour. Then, 2.5 × 103 Madin-Darby bovine kidney cells in 50 μL of culture medium was added to each well. The plates were incubated at 37°C in a humidified atmosphere with 5% CO2 for up to 5 days before they were examined for CPE. All plates were examined for CPE by the same investigator (PKG). For each sample, the antibody titer for each genotype was expressed as the reciprocal of the greatest serum dilution at which > 50% of the wells were free of CPE. Animals with an antibody titer ≤ 2 for a particular genotype were considered seronegative for that genotype. The indicator virus for each genotype was used as the positive control for the virus neutralization assay involving that genotype. Cells in wells that were not treated with the indicator virus served as the negative controls.
Phylogenetic tree for BPI3V strain SF4 and 3 other BPI3V isolates that were used as the indicator (reference) viruses in virus neutralization assays to measure titers of antibodies against each of the 3 viral genotypes (BPI3V-A [TVMD60], BPI3V-B [TVMD15], and BPI3V-C [TVMD19]). All 4 strains were clinical isolates from US cattle; SF4 is a BPI3V-A isolate commonly used as the reference strain in virus neutralization assays for BPI3V. The phylogenetic tree was created on the basis of genomic sequencing results for the HN gene. Notice that the 2 BPI3V-A strains (SF4 and TVMDL60) were most closely related. The scale at the bottom represents the difference in the nucleotide sequence of the HN gene among isolates (eg, 0.02 represents a 2% difference in nucleotide sequence).
Citation: American Journal of Veterinary Research 78, 2; 10.2460/ajvr.78.2.239
Statistical analysis
All antibody titer data underwent a log2 transformation for analysis. For each BPI3V genotype, a 1-way ANOVAa was used to compare the geometric mean antibody titer among the 6 groups. The Tukey-Kramer methodb was used for post hoc analyses when necessary. For all analyses, values of P ≤ 0.05 were considered significant.
Results
All animals in groups 2 (11 Miniature Zebu–crossbred cattle), 4 (6 Boer and Boer-crossbred goats), and 6 (21 captured white-tailed deer) were seronegative for antibodies against all 3 genotypes of BPI3V. All 12 heifers evaluated in group 1 were seropositive for all 3 genotypes. The highest antibody titer for individual heifers in group 1 was against BPI3V-B in 10 heifers; BPI3V-A and BPI3V-C titers were each highest in 1 heifer. All 4 steers in group 2 had an antibody titer of 64 against BPI3V-B and were seronegative for antibodies against BPI3V-A and BPI3V-C. Seven of 26 alpacas and 2 of 3 llamas in group 5 had an antibody titer ≥ 4 against BPI3V-B; the remaining 19 alpacas and 1 llama were seronegative for antibodies against BPI3V-B. One alpaca had an antibody titer of 4 against both BPI3V-A and BPI3V-C, but was seronegative for antibodies against BPI3V-B. The geometric mean antibody titer for all animals in group 5 was 0.1 for both BPI3V-A and BPI3V-C and 2.8 for BPI3V-B. The geometric mean antibody titers for the seropositive animals in groups 1, 3, and 5 were summarized (Table 1). For all 3 groups with seropositive animals, the geometric mean antibody titer against BPI3V-B was significantly greater than that against BPI3V-A or BPI3V-C, although it varied substantially among the groups.
Geometric mean antibody titers against BPI3V genotypes A, B, and C for seropositive animals in 3 of 6 groups of goats, camelids, and deer.
| Genotype | |||
|---|---|---|---|
| Group | A | B | C |
| 1 | 50.8 | 171.9* | 40.2 |
| 3 | 0 | 64* | 0 |
| 5† | 0.3 | 5.7* | 0.3 |
None of the animals were vaccinated against BPI3V prior to the study. Each group, except group 6, was individually housed in separate pastures in Auburn, Ala. Group 1 consisted of twelve 18-month-old Angus and Angus-crossbred heifers, and all 12 heifers were seropositive for antibodies against all 3 BPI3V genotypes (BPI3V-A, BPI3V-B, and BPI3V-C). Group 3 consisted of four 2-year-old Holstein steers, each of which had a surgically created rumen fistula. All 4 steers had an antibody titer of 64 against BPI3V-B, but were seronegative for antibodies against BPI3V-A and BPI3V-C. Group 5 consisted of 26 alpacas and 3 llamas, of which only 7 alpacas and 2 llamas had an antibody titer > 4 against BPI3V-B; the remaining 19 alpacas and 1 llama were seronegative for antibodies against BPI3V-B. One alpaca had an antibody titer of 4 against both BPI3V-A and BPI3V-C, but was seronegative for antibodies against BPI3V-B. All animals in groups 2 (11 Miniature Zebu-crossbred cattle), 4 (6 Boer and Boer-crossbred goats), and 6 (21 captured white-tailed deer) were seronegative for antibodies against all 3 genotypes of BPI3V.
Within a group, value differs significantly (P ≤ 0.05) from that for the other genotypes.
Values represent the mean for seropositive animals only.
Discussion
Results of the present study suggested that BPI3V-A may no longer be the predominate genotype of BPI3V circulating among unvaccinated ungulates in Alabama. This finding could have clinical significance because all currently available vaccines against BPI3V contain only BPI3V-A antigens. None of the animals surveyed in this study had been previously vaccinated against BPI3V, and all were well beyond the age when maternally derived antibodies would be expected to persist.4 Therefore, we believe the observed antibody titers in the seropositive animals were the result of natural exposure to the respective strains of BPI3V.
Bovine parainfluenza 3 virus is endemic in cattle populations throughout the world, and it is not un-common for unvaccinated herds to have a high prevalence of cattle seropositive for antibodies against BPI3V. To our knowledge, the present study was the first to evaluate the seroprevalence of antibodies against the 3 genotypes of BPI3V in unvaccinated ungulates. An unpublished review of whole genome sequencing results for 60 clinical isolates of BPI3V submitted to the Texas A&M Veterinary Medical Diagnostic Laboratory by 1 investigator (JDN) indicated that < 25% of the isolates were BPI3V-A and approximately 50% of the isolates were BPI3V-C. By contrast, results of the present study suggested that unvaccinated ungulates in Alabama were exposed to BPI3V-B more frequently than BPI3V-A or BPI3V-C. Regardless, those collective results appear to indicate that infections caused by BPI3V-A are decreasing relative to infections caused by non-A genotypes of the virus, at least in the southeastern and south-central United States. Further investigation of animals in a wider geographic area than that evaluated in the present study is necessary to determine whether a similar shift in the prevalence of BPI3V genotypes is occurring elsewhere.
The generation of antibodies against BPI3V has been reported in animals infected with the virus since it was first described in 1959.23 In the present study, the geometric mean antibody titer against BPI3V-B was significantly greater than that for BPI3V-A and BPI3V-C in all 3 groups (1, 3, and 5) with seropositive animals. The serum antibody concentration against BPI3V is important for limiting the clinical severity of disease once an infection has been established; however, mucosal antibody concentrations might be more important than serum antibody concentrations for prevention of BPI3V infections.24–26 Serum antibodies primarily moderate disease virulence, whereas mucosal antibodies negate viral challenge.24–26 We did not attempt to measure the mucosal antibody titers against the various genotypes of BPI3V because the scope of the present study was simply to conduct a serologic survey for antibodies against the 3 genotypes of BPI3V in unvaccinated ungulates in Alabama. Although low serum antibody titers against BPI3V may be clinically irrelevant from a disease-mitigation standpoint, the fact that 17 of the 18 seropositive animals in groups 3 and 5 had detectable antibodies against BPI3V-B but not BPI3V-A provided evidence that BPI3V-A is no longer the predominate field strain of BPI3V circulating within the surveyed area. Continued active sampling and surveillance of unvaccinated ungulates in that area will need to be performed to confirm that supposition.
The extent of cross-reaction among antibodies against the various BPI3V genotypes is unknown. Bovine parainfluenza 3 virus is closely related to both Sendai virus, the prototypical member of the genus Respirovirus, and HPV. Human parainfluenza virus has 4 distinct serotypes, and the host antibody response is fairly specific for the infecting strain.1 Results of another study19 indicate significant antigenic differences among BPI3V genotypes and subgenotypes. In that study,19 serum neutralization assays that used an APHIS BPI3V-A isolate as the reference (indicator) virus revealed a 2- to 5-fold decrease in serum antibody titers for BPI3V isolates that were genetically dissimilar to the reference strain. Interestingly, the magnitude of serum antibody titer decrease was most pronounced for BPI3V-B and BPI3V-A strains that were of a different subgenotype than the reference strain.19 In the present study, all 4 steers of group 3 had fairly high antibody titers against BPI3V-B but were seronegative for antibodies against the other 2 BPI3V genotypes, which suggested that there was not extensive cross-reaction among the genotypes. However, all heifers evaluated in group 1 were seropositive for antibodies against all 3 genotypes of BPI3V. The geometric mean antibody titer against BPI3V-B for that group was significantly greater than that for either BPI3V-A or BPI3V-C, and the antibody titer against BPI3V-B was highest among the 3 genotypes tested for all but 2 of the 12 heifers. It is unknown whether those heifers were naturally exposed to all 3 BPI3V genotypes or there was some degree of cross-reaction among the 3 genotypes. Further research is necessary to elucidate the extent of cross-reaction among antibodies against the 3 BPI3V genotypes and determine whether the virus has distinct serotypes.
It has been proposed that non-A genotypes of BPI3V arose as a result of geographic isolation.18 Initial reports of genotypes B and C were limited to Australia17 and China,18 respectively. Since then, BPI3V-B has been reported in the United States and Argentina,20 and BPI3V-C has been reported in the United States, Argentina, and Korea.21 Sequence homology is < 85% between US non-A genotypes and US BPI3V-A strains but ranges from 94% to 98% between US and foreign non-A genotypes.19 Although the origin of US non-A genotypes of BPI3V is unknown, it is unlikely that multiple independent mutations resulted in similarly divergent strains in multiple geographic regions. It is possible that the BPI3V population dynamics shifted because of vaccine pressure. In the United States, historic and current commercial BPI3V vaccines (both killed-virus and modified-live virus vaccines) contain BPI3V-A antigens only. Given that cross-protection appears to be minimal among antibodies against the various BPI3V genotypes, widespread use of vaccines will presumably decrease the number of animals susceptible to BPI3V-A and provide non-A genotypes a competitive advantage. Thus, non-A genotypes could eventually replace BPI3V-A as the predominate genotype circulating among US ungulates, which could have important implications for US livestock producers. Further research is necessary to investigate the extent of protection provided by current vaccines against the various genotypes of BPI3V.
In the present study, serum antibody titers against BPI3V-B were significantly greater than those against BPI3V-A in unvaccinated cattle and New World camelids in Alabama. That finding suggested that BPI3V-A is no longer the predominate genotype circulating among ungulates in that region. Although the extent of cross-protection among antibodies against the various BPI3V genotypes is unknown, this apparent shift in genotype among field strains of BPI3V could have substantial implications for livestock producers given that the virus is frequently involved in the pathogenesis of bovine respiratory disease complex and currently available vaccines contain antigens against BPI3V-A only.
Acknowledgments
Supported by an Animal Health Research and Disease Grant from the Office of Research and Graduate Studies, College of Veterinary Medicine, Auburn University.
ABBREVIATIONS
| BPI3V | Bovine parainfluenza 3 virus |
| CPE | Cytopathic effect |
| HN | Hemagglutinin-neuraminidase |
| HPV | Human parainfluenza virus |
Footnotes
Excel 2013, Microsoft, Redmond, Wash.
JMP Pro, version 11.1.1, SAS Institute Inc, Cary, NC.
References
1. Chanock RM, Murphy BR, Collins PL. Parainfluenza viruses. In: Knipe MK, Howley PM, eds. Fields virology. 4th ed. Philadelphia: Lippincott Williams & Wilkins, 2001;1341–1379.
2. Woods GT, Mansfield ME, Cmarik G, et al. Vaccination of beef calves before weaning with a live-virus bovine myxovirus parainfluenza-3 vaccine. Am J Vet Res 1968; 29: 1349–1353.
3. Frank GH, Marshall RG. Relationship of serum and nasal secretion-neutralizing antibodies in protection of calves against parainfluenza-3 virus. Am J Vet Res 1971; 32: 1707–1713.
4. Ellis JA. Bovine parainfluenza-3 virus. Vet Clin North Am Food Anim Pract 2010; 26: 575–593.
5. Bryson DG, McFerran JB, Ball HJ, et al. Observations on outbreaks of respiratory disease in calves associated with parainfluenza type 3 virus and respiratory syncytial virus infection. Vet Rec 1979; 104: 45–49.
6. Fulton RW, Purdy CW, Confer AW, et al. Bovine viral diarrhea viral infections in feeder calves with respiratory disease: interactions with Pasteurella spp, parainfluenza-3 virus, and bovine respiratory syncytial virus. Can J Vet Res 2000; 64: 151–159.
7. Gagea MI, Bateman KG, van Dreumel T, et al. Diseases and pathogens associated with mortality in Ontario beef feed-lots. J Vet Diagn Invest 2006; 18: 18–28.
8. Li W, Mao L, Cheng S, et al. A novel parainfluenza virus type 3 (PIV3) identified from goat herds with respiratory diseases in eastern China. Vet Microbiol 2014; 174: 100–106.
9. Giangaspero M, Savini G, Orusa R, et al. Prevalence of antibodies against parainfluenza virus type 3, respiratory syncytial virus and bovine herpesvirus type 1 in sheep from northern prefectures of Japan. Vet Ital 2013; 49: 285–289.
10. Rudolph KM, Hunter DL, Rimler RB, et al. Microorganisms associated with a pneumonic epizootic in Rocky Mountain bighorn sheep (Ovis canadensis canadensis). J Zoo Wildl Med 2007; 38: 548–558.
11. Intisar KS, Ali YH, Khalafalla AI, et al. Respiratory infection of camels associated with parainfluenza virus 3 in Sudan. J Virol Methods 2010; 163: 82–86.
12. Rivera H, Madewell BR, Ameghino E. Serologic survey of viral antibodies in the Peruvian alpaca (Lama pacos). Am J Vet Res 1987; 48: 189–191.
13. Marcoppido G, Parreno V, Vila B. Antibodies to pathogenic livestock viruses in a wild vicuña (Vicugna vicugna) population in the Argentinean Andean altiplano. J Wildl Dis 2010; 46: 608–614.
14. Qiao D, Janke BH, Elankumaran S. Complete genome sequence and pathogenicity of two swine parainfluenzavirus 3 isolates from pigs in the United States. J Virol 2010; 84: 686–694.
15. Sato M, Wright PF. Current status of vaccines for parainfluenza virus infections. Pediatr Infect Dis J 2008; 27(suppl 10): S123–S125.
16. Shibuta H, Suzu S, Shioda T. Differences in bovine parainfluenza 3 virus variants studied by monoclonal antibodies against viral glycoproteins. Virology 1986; 155: 688–696.
17. Horwood PF, Gravel JL, Mahony TJ. Identification of two distinct bovine parainfluenza virus type 3 genotypes. J Gen Virol 2008; 89: 1643–1648.
18. Zhu YM, Shi HF, Gao YR, et al. Isolation and genetic characterization of bovine parainfluenza virus type 3 from cattle in China. Vet Microbiol 2011; 149: 446–451.
19. Neill JD, Ridpath JF, Valayudhan BT. Identification and genome characterization of genotype B and genotype C bovine parainfluenza type 3 viruses isolated in the United States. BMC Vet Res 2015; 11: 112.
20. Maidana SS, Lomonaco PM, Combessies G, et al. Isolation and characterization of bovine parainfluenza virus type 3 from water buffaloes (Bubalus bubalis) in Argentina. BMC Vet Res 2012; 8: 83.
21. Oem JK, Lee EY, Lee KK, et al. Molecular characterization of a Korean bovine parainfluenza virus type 3 isolate. Vet Microbiol 2013; 162: 224–227.
22. Ditchkoff SS, Raglin JB, Smith JM, et al. From the field: capture of white-tailed deer fawns using thermal imaging technology. Wildl Soc Bull 2005; 33: 1164–1168.
23. Reisinger RC, Heddleston KL, Manthei CA. A myxovirus (SF-4) associated with shipping fever of cattle. J Am Vet Med Assoc 1959; 135: 147–152.
24. Omar AR, Jennings AR, Betts AO. The experimental disease produced in calves by the J-121 strain of parainfluenza virus type 3. Res Vet Sci 1966; 7: 379–388.
25. Bryson DG, McNulty MS, Ball HJ, et al. The experimental production of pneumonia in calves by intranasal inoculation of parainfluenza type III virus. Vet Rec 1979; 105: 566–573.
26. Marshall RG, Frank GH. Neutralizing antibody in serum and nasal secretions of calves exposed to parainfluenza-3 virus. Am J Vet Res 1971; 32: 1699–1706.
Appendix
Description and GenBank accession numbers of the 3 BPI3V isolates used as the indicator (reference) strains in virus neutralization assays to measure titers of serum antibodies against BPI3V in unvaccinated ungulates in Alabama.
| Virus | GenBank accession No. | Sequence description | Genotype | Country of origin |
|---|---|---|---|---|
| TVMD60 | KJ647289 | Full-length genome | A | United States |
| TVMDI5 | KJ647284 | Full-length genome | B | United States |
| TVMDI9 | KT935304 | Fusion gene | C | United States |
| KT935305 | HN gene |
