Respiratory illness in young and adult cattle caused by bovine viral diarrhea virus subgenotype 2b in singular and mixed bacterial infection in a BVDV-vaccinated dairy herd

Abstract

Bovine respiratory disease (BRD) is a common global health problem in dairy cattle. The definitive diagnosis of BRD is complex because its etiology involves several predisposing and determining factors. This report describes the etiology of a BRD outbreak in a dairy herd in the mesoregion of Central Eastern Paraná, which simultaneously affected young (calves and heifers) and adult (cows) Holstein-Friesian cattle. Nine biological samples, consisting of five lung samples from two cows and three suckling calves, and four nasal swab samples from heifers, were used for etiological diagnosis. The nucleic acids extracted from lung fragments and nasal swabs were subjected to PCR and RT-PCR assays for partial amplification of the genes of five viruses [bovine viral diarrhea virus (BVDV), bovine alphaherpesvirus 1 (BoAHV1), bovine respiratory syncytial virus (BRSV), bovine parainfluenza virus 3 (BPIV-3), and bovine coronavirus (BCoV)] and four bacteria (Mycoplasma bovis, Mannheimia haemolytica, Pasteurella multocida, and Histophilus somni) involved in the etiology of BRD. All nine biological samples from the animals with BRD tested negative for BoAHV1, BRSV, BPIV-3, BCoV, and H. somni. Therefore, the involvement of these microorganisms in the etiology of BRD outbreak can be ruled out. It was possible to identify the presence of BVDV and M. bovis in singular and mixed infections of the lower respiratory tract in cattle. BVDV was also identified in two nasal swabs: one as a single etiological agent and the other in association with two bacteria (P. multocida and M. haemolytica). The phylogenetic analysis conducted in the nucleotide sequence of the 5’UTR region and N^pro gene of the BVDV amplicons demonstrated that the BVDV field strains of this BRD outbreak belong to subgenotype 2b. To the best of our knowledge, this is the first report of BVDV-2b involvement in the etiology of BRD in Brazil. Finally, it is necessary to highlight that the cattle were obtained from an open dairy herd with biannual vaccinations for BVDV-1a and - 2a.

Introduction

Bovine respiratory disease (BRD) is a common global health problem in dairy cattle. BRD is a multifactorial disease that affects young (calves and heifers) and adult (cows) animals [1–3]. Aspects related to animals (age, nutrition, and immunity) and management (animal density and biosecurity practices) may predispose dairy cattle herds to BRD. However, infections of the upper and lower respiratory tracts caused by viruses and bacteria, which occur in single and mixed infections, are the major causes of clinical signs of BRD [2, 4, 5].

Early diagnosis of respiratory infections in cattle is fundamental for establishing the appropriate therapy and reducing the financial costs. Calves with BRD have impaired body development, and cows have reduced milk production; BRD can lead to death when severe [3, 6].

In addition to the predisposing factors, multiple etiologies of BRD complicate laboratory diagnosis. Viruses [bovine viral diarrhea virus (BVDV), bovine respiratory syncytial virus (BRSV), bovine alphaherpesvirus 1 (BoAHV1), bovine coronavirus (BCoV), and bovine parainfluenza virus 3 (BPIV-3)] and bacteria (Mannheimia haemolytica, Pasteurella multocida, Histophilus somni, and Mycoplasma bovis) are the most frequently identified microorganisms in infections of the upper and lower respiratory tract of cattle [2, 7–9].

BVDV belongs to the family Flaviviridae and genus Pestivirus, and the species involved in bovine diseases include Pestivirus bovis (BVDV-1), Pestivirus tauri (BVDV-2), and Pestivirus brazilense (BVDV-3) [10]. Currently, 24 subgenotypes of BVDV-1 (BVDV-1a to BVDV-1x) and 5 subgenotypes of BVDV-2 (BVDV-2a to BVDV-2e) have been described [11–13].

BVDV-1 and BVDV-2 were identified in respiratory infections in dairy cattle in several countries, including Brazil [11, 14–17]. Among the BVDV subgenotypes identified in cattle with BRD in Brazil, only BVDV-1a, -1b, and - 1d were described [2, 18–20].

BVDV-2b has been previously described in Brazilian cattle herds from cattle with gastrointestinal and reproductive clinical signs, mucosal-like disease, persistently infected (PI) animals, and serum samples [18, 19, 21–26]. However, this subgenotype is not associated with the occurrence of BRD in Brazil.

This report describes the identification of BVDV-2b in singular and mixed bacterial infections in calves, heifers, and cows from a BVDV-vaccinated dairy herd in the State of Paraná.

Materials and methods

Biological samples

The Laboratory of Animal Virology at the Universidade Estadual de Londrina received five lung fragments from two cows, three suckling calves, and four nasal swab samples from heifers for the etiological diagnosis of BRD. The samples were stored at − 80 °C until processing. Biological samples were collected from a dairy cattle herd in the Central Eastern mesoregion of Paraná, southern Brazil, by a consulting veterinarian. All evaluated animals showed clinical signs of BRD (nasal discharge, dyspnea, lethargy, and depression). Respiratory problems began in newly calved cows and spread to calves and heifers. The outbreak of BRD lasted for about 15 days, affecting other cows, heifers, and calves. Despite showing clinical signs, they survived with early therapy.

All the cows and heifers in the herd were biannually vaccinated with a commercial vaccine containing the prototype strains of BVDV-1a and BVDV-2a, according to the manufacturer’s recommendations. Concerning biosecurity standards, the dairy cattle herd was classified as an open herd because heifers and cows from neighboring herds were incorporated into the herd.

Nucleic acid extraction

Lung samples were mechanically disrupted using a TissueLyser LT (Qiagen, Hilden, Germany) and homogenized with 1 mL of sterile physiological solution (0.9% NaCl). After centrifugation, 500 µL of the lung and swab sample solutions were pretreated with 0.2 mg/mL proteinase K (Ambion, Grand Island, NY, USA) and sodium dodecyl sulfate at a final concentration of 1% (volume/volume). Nucleic acids were extracted from the lungs using a combination of the phenol/chloroform/isoamyl alcohol (25:24:1) and silica/guanidine isothiocyanate methods [27, 28]. Nucleic acids in the swab samples were extracted using the silica/guanidine isothiocyanate method.

Molecular diagnostics

Nucleic acids extracted from nasal swabs and lung fragments were subjected to PCR and RT-PCR assays to amplify the nucleic acids of nine etiologic agents involved in BRD. The following sequences were amplified: 288 bp of the 5’-untranslated (5’-UTR) region of pestivirus [29] and 428 bp of the partial BVDV N^pro gene [30]; 371 bp of the BRSV G gene [31]; 251 bp of the BCoV N gene [32]; 647 bp of the BPIV-3 HN gene [33]; 425 bp of the BoAHV1 D gene [34]; 460 bp of the P. multocida ORF clone KMT1 [35]; 408 bp of the H. somni 16 S gene [36]; 385 bp of the M. haemolytica lktA-artJ intergenic region [37]; and 488 bp of the M. bovis 16–23 S intergenic region [38]. The amplified products were evaluated by electrophoresis on 2% agarose gels, stained with ethidium bromide, and observed under ultraviolet light.

Sequencing and phylogenetic analysis

The amplified products were purified using a PCR Purification Combo Kit (Invitrogen^® Life Technologies, Carlsbad, CA, USA) and quantified using a Qubit^® fluorometer (Invitrogen^® Life Technologies, Eugene, OR, USA). The products were sequenced using a Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems^®, Foster City, CA, USA) on an ABI3500 Genetic Analyzer sequencer. Quality and contig analyses were performed using the PHRED and CAP3 software, respectively. Similarity searches were performed against sequences deposited in GenBank using the basic local alignment search tool, and phylogenetic analyses of the nucleotide (nt) sequences were performed using the MEGA 7.0 software. The nt sequence identity matrices were obtained using BioEdit software version 7.08.0.

Results and discussion

BVDV and M. bovis were identified in lung tissue from a cow and three calves. These pathogens have been observed in the bovine lower respiratory system as singular infections and associations [8, 39]. Zhou et al. [39] described the risk of developing BRD due to an association between BVDV and M. bovis in the lungs of Chinese cattle. A North American study reported that M. bovis was the third most common agent detected in cattle that died of BRD, followed by M. haemolytica and BVDV. The association of these three infectious agents was the most prevalent [40]. BVDV infection can cause immunosuppression and lesions in the bovine respiratory system, leading to secondary infections by commensal (M. haemolytica, P. multocida) or infectious (M. bovis) respiratory bacteria [41].

BVDV was also identified in the following two nasal swabs: (1) as a single etiological agent and (2) associated with P. multocida and M. haemolytica. These two bacteria are frequently identified in the upper respiratory tract of cattle with BRD [42–44] and in asymptomatic animals [45, 46]. Table 1 shows the PCR and RT-PCR results for the partial amplification of genes from nine potential pathogens involved in BRD etiology.

Table 1.

Results of PCR and RT-PCR techniques for partial amplification of nucleic acid from respiratory pathogens (viruses and bacteria) in biological samples from cattle with bovine respiratory disease in a dairy herd, state of Paraná, southern Brazil

Biological sample	Animal Category	Results
Lung	Cow (n = 2)	BVDV + M. bovis
		Negative(#)
	Calf (n = 3)	BVDV
		M. bovis
		BVDV + M. bovis
Nasal swab	Heifer (n = 4)	BVDV
		BVDV + P. multocida + M. haemolytica
		P. multocida + M. haemolytica
		Negative(#)

BVDV (Bovine viral diarrhea virus); M. bovis (Mycoplasma bovis); P. multocida (Pasteurella multocida); M. haemolytica (Mannheimia haemolytica)

^(#) Negative samples for the nine respiratory pathogens evaluated (BVDV, M. bovis, P. multocida, M. haemolytica, Histophilus somni, Bovine respiratory syncytial virus, Bovine alphaherpesvirus 1, Bovine coronavirus, and Bovine parainfluenza virus 3)

All nine biological samples (lung fragments, n = 5; nasal swabs, n = 4) from the animals with clinical respiratory signs, evaluated using PCR and RT-PCR for the etiological diagnosis of BRD, were negative for BRSV, BoAHV1, BCoV, BPIV-3, and H. somni. Therefore, the involvement of these microorganisms in the etiology of BRD outbreak can be ruled out. The lung fragment from a cow and the nasal swab from a heifer were negative for the nine microorganisms included in the analyses.

The BVDV amplicons of the 5’UTR region from one lung and the 5’UTR region and N^pro gene from another lung (UEL12-BR/17) were sequenced. Molecular characterization of the BVDV strain identified in one lung sample was not performedbecause of the low quality of the nt sequence. Two BVDV-positive nasal swab samples were also sequenced in the 5’UTR region. The four sequenced samples were classified as BVDV-2b (Fig. 1) and showed 100% nt identity in this genomic region. The sequences of the BVDV UEL12-BR/17 field strain are available in GenBank (accession numbers: MG004720 and MG004721) and showed 98.7% nt identity with the Japanese Hokudai-Lab/09 strain (accession number: AB567658) and with the Brazilian LV/N1156/13 strain (accession number: KP715136). Compared to other BVDV-2b strains, nt identity of our BVDV-2b field strains varied between 92.6% and 98.7%. According to the N^pro gene, the BVDV UEL12-BR/17 field strain presented the highest (97.2%) nt identity with the Japanese Hokudai-Lab/09 strain (accession number: AB567658) and 86.2–97.2% with other BVDV-2b strains available in the GenBank database. The nt identity of the BVDV-2b field strains from this study with the prototype strain New York93 of BVDV-2a was 89.4% and 81.6% for the 5’UTR region and the N^pro gene, respectively. Figure 1 shows the phylogenetic trees reconstructed with partial 5’UTR region and N^pro gene nucleotide sequences of the BVDV-2b respiratory field strains described in this report and available in the public database (GenBank).

Fig. 1.

Phylogenetic trees reconstructed with partial 5’UTR region (A) and N^pro gene (B) with nucleotide sequences of representative BVDV-2 strains. The BVDV-2b Brazilian strains were marked with hollow triangles (Δ) and the strains from this study are marked with a black circle (⬤). The analyses were based on the neighbor-joining method with the Kimura 2-parameter model and bootstrapping was statistically supported with 1,000 replicates using MEGA 7.0. Bootstrap values of less than 50% are not shown. All GenBank accession numbers are between parentheses

The first description of BVDV-2 was in gastroenteric disease in North America [47]. BVDV-2b was first identified in Brazil by Canal et al. [21] in a bovine with clinical signs of mucosal disease (prototype Soldan strain). Other descriptions of BVDV-2b in Brazil were in PI animals [18, 19, 22] and cattle with gastrointestinal [18, 19] and reproductive clinical signs [19]; in fetal bovine serum [23] and bovine serum samples [24, 26].

Descriptions of BVDV-2b in BRD are rare. In Slovakia, the BVDV-2b subgenotype was described in animals that died of catarrhal pneumonia, bronchopneumonia, and liver and spleen hemorrhage [48]. In Argentina, BVDV-2b was detected in the nasal swabs of cattle [49]. Additionally, in Uruguay, BVDV-2b and H. somni were detected in the lung fragments of dairy heifers with bronchopneumonia [15].

Several vaccines have been developed to control and prevent BVDV infection in cattle. The first vaccines were developed in the early 1960s. Currently, there are hundreds of vaccines registered worldwide. These include monovalent vaccines with only BVDV-1a and multivalent vaccines containing BVDV-1a and  -2a and other microorganisms associated with reproductive and respiratory diseases of cattle. Vaccine formulations with inactivated or attenuated BVDV strains are available in international and domestic markets. However, results from experimental and field studies describing the effectiveness of vaccines vary considerably [50].

A meta-analysis study describes the impact of BVDV vaccination on the rate of reproductive losses in cattle. It evaluated 46 experiments from 41 scientific articles published in various countries, including Brazil. The analysis yielded results from various attenuated, inactivated, monovalent, or polyvalent vaccines and challenges with homologous, heterologous viruses, and field strains. The study concluded that any vaccination provided significant protection against reproductive disease and fetal infection, reducing the risk by up to 85% [51].

In the same way, Patel et al. [52] reported that the cow vaccination followed by the challenge with the same BVDV subgenotype (1a) prevented the birth of PI calves.

However, several studies describe vaccine failures to protect against infection depending on the type of vaccine (inactivated, attenuated), viral species (BVDV-1, BVDV-2), and viral subgenotype. Failure of fetal protection is frequently reported with loss of embryo/fetus or with the generation of PI calves in cows vaccinated and subsequently field challenged with BVDV-1 or BVDV-2 strains with a subgenotype different from that present in the vaccine formulation [49, 50, 53–55].

In addition to differences in the nt sequence identified in BVDV strains classified into different subgenotypes, differences in the nt sequence are also described in strains of the same subgenotype. Bolin [56] reported differences in the nt sequence in 16 BVDV-2a field strains that ranged between 3 and 10% when compared with the reference strain (BVDV-890) and from 3 to 11% when compared between them.

The BVDV-1 or -2 strains classification into subgenotypes is determined by comparing the nt identity of two regions, 5’UTR and N^pro, of the viral genome. Studies based on molecular characteristics (viral species and subgenotype) of BVDV strains present in the vaccines available on the market and field strains are relatively common. However, these regions do not encode the two main immunogenic proteins, E2 and NS2-3, of BVDV [50].

Studies focusing on the antigenic characteristics of BVDV strains and evaluating the presence and titer of neutralizing antibodies are less frequent. Fulton et al. [57] described those animals vaccinated with a strain of BVDV-1a showed low titers of virus-neutralizing antibodies to the BVDV-1b strain. In another study, the absence of neutralizing virus antibodies to BVDV-1e was found in animals vaccinated with strains of BVDV-1a and  -1b [58]. Antigenic differences can be found even in strains classified under the same subgenotype, such as those described in eight Argentine field strains of BVDV-1b evaluated by cross-neutralization tests [49].

This case report is the first description of the BVDV-2b subgenotype in nasal swabs and lung fragments of cows, heifers, and calves with respiratory clinical signs in Brazil. The animals were from a dairy cattle herd with biannual BVDV-1a and  -2a vaccinations.

We conclude that in open dairy cattle herds that incorporate neighboring heifers and cows into the herd, regular vaccination against BVDV-1a and  -2a is insufficient to control the infection. Additional internal and external biosecurity measures are essential to prevent the introduction and spread of different subgenotypes of BVDV-1 or -2 across the several categories of animals (suckling and weaned calves, heifers, and cows) present in a dairy herd.

Author contributions

Conceptualization and design: JTTF, CYY, EL, and AAA; Methodology: IVS and JTTF; Writing - original draft preparation: JTTF, IVS, and CYY; Writing - review and editing: JTTF and EL; Supervision: AFA and AAA. All authors have read, critically analyzed, and approved the final draft of this manuscript.

Funding

This study was supported by the National Institute of Science and Technology of Dairy Production Chain (INCT-Leite / CNPq) [Grant number 465725/2014–7].

Data availability

The datasets generated during and/or analyzed during the study are available from the corresponding author upon reasonable request.

Declarations

Conflict of interest

The authors declared no potential conflicts of interest concerning the research, authorship, and/or publication of this article.