Introduction

The Flaviviridae family consists of positive-sense, single-stranded, and unsegmented enveloped RNA viruses. The family is divided into four genera: Flavivirus, Pestivirus, Hepacivirus, and Pegivirus1. Currently, the genus Pestivirus includes four species: Bovine viral diarrhea virus 1 (BVDV-1), Bovine viral diarrhea virus 2 (BVDV-2), Border disease virus (BDV), and Classical swine fever virus (CSFV)2.

The current taxonomic classification based on phylogenetic analysis of conserved amino acid sequences revealed 19 different Pestivirus species. Among species, Pestivirus bovis (formerly “BVDV 1”) is further classified into 23 subspecies. There are also Pestivirus tiauri (formerly “BVDV 2”) and Pestivirus brazilense (formerly “BVDV 3 or HoBi-like virus”)3. Furthermore, BVDV is classified as cytopathic (cp.) and noncytopathic (ncp) based on its lytic effect on cell culture4,5. Cytopathic viruses have been shown to originate from ncp viruses through either the incorporation of cellular RNA sequences into the NS2/NS3 coding region or duplication of the NS3 region, and point mutation can cause cleavage of NS2/NS36. Consequently, the increased accumulation of the NS3 protein induces a cytopathic effect in infected cells7,8.

The genome of BVDV is approximately 12.3 kb in size, and has a single long open reading frame (ORF), which lacks a cap structure at its 5’ and 3’ UTR, and is not polyadenylated at the 3’ end. The ORF encodes a polyprotein that is post-translationally cleaved into structural and non-structural proteins. Following the co- and post-translational processes, this polyprotein is processed by both cellular and viral proteases, and 12 mature proteins, including four structural proteins (capsid, Erns, E1, and E2) and eight nonstructural proteins (Npro, p7, NS2, NS3, NS4a, NS4b, NS5a, and NS5b) are synthesized9,10 In cattle-producing countries, bovine viral diarrhea (BVD) is a highly problematic infection caused by Bovine viral diarrhea virus-1 and Bovine viral diarrhea virus-211. This infection not only results in significant reproductive losses but also perpetuates the spread of the virus to naïve animals through persistently infected calves. As a result, imposes huge economic loss in cattle farming12. Additionally, BVD infections are responsible for a range of health issues worldwide, including abortion, stillbirths, and congenital tremors in newborn animals13.

The high mutation rate and extensive genetic variability of the BVDV species significantly affect BVD control programs14. Hence, it is essential to identify BVDV species and sub-genotypes circulating at the national or regional levels15. To improve the protection provided by existing vaccines and safeguard fetuses from infection, it is necessary to compare the existing wild-type viral strain with the vaccine strain at the antigenic and genetic levels16. Despite the high seroprevalence of BVD in Ethiopia, the circulating species and sub-genotypes of BVDV have not yet been understood. To the authors’ knowledge, this is the first report on the circulating BVDV species and sub-genotypes in the dairy cattle population in Ethiopia.

Materials and methods

Study areas

The study was conducted in central (Addis Ababa, Sebeta, Suluta, Sendafa, Bishoftu, and Holeta), southern (Hawassa), and northern (Bahir Dar and Gondar) parts of Ethiopia during August 2022 and January 2023 (Fig. 1). The high potential for dairy production, the existence of a large number of improved dairy animals, and the difference in agroecology favor the selection of the study areas. Although local or indigenous breeds were the majority of the livestock population, 9% were exotic or hybrid cattle17. The total dairy cattle for each study area was obtained from the agriculture office in each region. Accordingly, the dairy cattle (exotic and cross-breeds) population in the study areas was estimated at 244,045. Of which 20,855, 18,000, 4500, 20,000, 28460,55000, 35,000, and 81,000 were found in Addis Ababa, Suluta, Sendafa, Bishoftu, Holeta, Hawassa, Bahir Dar, and Gondar, respectively.

Fig. 1
figure 1

Map of Ethiopia showing the study areas from where samples were collected. Panel “a” represents the northern part (Bahr Dar and Gondar); “b” denotes the central (Addis Ababa, Holeta, Suluta, Sebeta, Bishoftu, and Sendafa), and “c” designate the southern part (Hawassa) of Ethiopia. Created with Quantum Geographic Information System (QGIS) version 3.30.3 (https://qgis.org/project/visual-changelogs/visualchangelog330/).

Full size image

Sample collection and Preparation

A total of 576 nasal swab samples (northern = 256; central = 254; southern = 57) were collected from non-vaccinated and BVD-suspected cattle exhibiting clinical signs, such as fever (40 °C), depression, decreased milk production, transient inappetence, rapid respiration, excessive nasal secretion, excessive lacrimation, and diarrhea during the acute phase of infection. Furthermore, during severe infection, animals may exhibit oral ulcerations, eruptive lesions of the coronary band and interdigital cleft, and petechiae hemorrhages. Additionally, high morbidity with a mortality above 25% is commonly seen in herds affected with a highly virulent strain of BVDV. Clinical history of fetal mummification, premature birth, stillbirth, congenital malformation, and birth of weak calves also characterizes BVDV infection. Swabs were immersed in viral transport medium (1x Hanks Balanced Salt Solution, 2%fetal bovine serum, 100ug/mL gentamicin, and 0.5ug/mL amphotericin B) in properly labeled tubes and transported to the Animal Health Institute (AHI) at Sebeta, Ethiopia, at 4 °C in a cooler with an ice pack. On arrival, samples were stored at −80 °C until laboratory analysis was conducted. Samples were vortexed and filtered with sterile 0.22 µL pore syringe filters before further analysis.

One-step RT-PCR assays

The total viral RNA was isolated using NucleoSpin RNA Virus Mini kit (Macherey-Nagel, Germany) according to the manufacturer’s instructions. Subsequently, SuperScript™ IV One-Step RT-PCR (Invitrogen, USA) was used to amplify the 5’ untranslated region of the virus. The forward and reverse primers (F1 5′-GCCATGCCCTTAGTAGGACT-3′, and R1 5′-CACCCTATCAGGCTGTRTYC-3′) were adapted from Gong et al.18 Briefly, a 50 µL reaction volume containing 2x Platinum SuperFi RT-PCR master mix (25 µL), 10 µmol of forward primer, 10 µmol of reverse primer, 0.5 µL of SuperScript IV RT mix, 2 µL of template RNA, and 17.5 µL of nuclease-free water were used during PCR amplification. The PCR cycling involved reverse transcription at 50 °C for 10 min, RT inactivation/initial denaturation at 98 °C for 2 min, followed by 40 cycles at 98 °C for 10 s, 57 °C for 10 s, and 72 °C for 1 min, with a final elongation step at 72 °C for 5 min. The amplicons were subjected to 1.5% agarose gel electrophoresis and visualized following ethidium bromide staining.   Virus isolation

Madin-Darby Bovine Kidney (MDBK) cells were grown in Dulbecco’s modified Eagle medium (DEME) containing 10% fetal calf serum (HyClone, Logan, UT) and maintained in DMEM containing 2% fetal calf serum). Those swab samples that tested positive during PCR were filter sterilized before infecting cells. A monolayer of 2.5 × 10^5 MDBK cells per well was used to isolate and characterize the virus. The infected cell was examined every 24 h for cytopathic effect (CPE). After three passages at five-days intervals, the 3rd passages of infected cells were used as viral stock for downstream analysis.

Immunofluorescence assay (IFA)

A 24-well microtiter plate (Costar, NY, USA) was seeded with MDBK cells in DMEM with 7% FBS and incubated overnight at 37 °C in 5% CO2. When cells reached 70–80% confluency, they were infected with 20-fold diluted field BVDV isolates from the third passage. Uninfected cultures were used as negative controls. After 72 h of incubation at 37 °C in 5% CO2, the plate was fixed with 80% cold acetone/methanol and washed with 1x PBS. Fluorescein Isothiocyanate (FITC)-conjugated Anti-Bovine Viral Diarrhea Virus polyclonal antiserum (VMRD, USA) was added to the plate and incubated for 30 min at 37 °C in a humid box. After three washes with PBS, 50% glycerol in PBS was added to each well. An inverted Fluorescence microscope (CETI, UK) was used to examine the fluorescence signal.

Sequence and phylogenetic analysis

The band corresponding to 5’ UTR regions (230 bp) were excised from the gel and purified using the Wizard SV Gel and PCR Clean-Up kit (Promega, USA). The purified amplicons were quantified using the SimpliNano spectrophotometer (BioChrom, USA) and subsequently cloned into pJET1.2/blunt vector using the cloneJET PCR cloning kit (Thermo Fisher Scientific Inc., Waltham, USA). Recombinant plasmids were sequenced using pJET1.2 forward and reverse primers at Eurofins Genomic Services (Germany, Konstanz). The raw sequence data were edited using BioEdit v7.2.519. The nucleotide sequences were blasted into the National Center for Biotechnology Information (NCBI), and reference strains were retrieved in FASTA format. The FASTA format was imported into MEGA version 1120 to construct rooted phylogenetic trees using the Neighbor-Joining method21 and the Kimura 2-parameter statistical model22. Bootstrap analysis was performed on each tree (1000 replicates). FigTree v. 1.4.4 was used to display and edit the phylogenetic tree generated by the MEGA software to produce publication-ready figures (http://tree.bio.ed.ac.uk/software/figtree/). Pairwise sequence identity percentage was performed using sequence Demarcation Tool (SDT) v1.2 software was used to calculate pairwise identities with MUSCLE alignments, and cutoff values of 94 and 78 were applied to categorize identity level as high (> 94%), medium (78–94%), and low (< 78%)23.

Results

Viral isolation and characterization

Eighteen (n = 18) PCR positive samples were cultured and passaged in MDBK cells, with no obvious CPE observed after three passages at five-day intervals. This indicates that the virus is a ncp biotype (Fig. 2).

Fig. 2
figure 2

IFAT for BVDV isolates on MDBK cell line. The confirmation of BVDV infection was made using FITC-labeled BVDV antibody. The fluorescent greenish color indicates BVDV-infected cells. The specific immunofluorescent signals were detected in the cytoplasm of cells inoculated with BVDV (a, b, c), and mock-infected cells showed no bright fluorescence (d).

Full size image

Phylogenetics and sequence analysis

Of the 18 viral isolates from samples collected in the central and northern parts of Ethiopia, eight (44.4%) isolates were used for phylogenetic analysis and sequence comparison (Fig. 3). However, we failed to detect and identify the genotype and sub-genotypes of the virus from the southern part of Ethiopia. Based on the 5’ UTR region of the virus and comparison to a set of reference strains, 87.5% (7/8) of the viral isolates were classified as BVDV 2a sub-genotypes, and 12.5% (1/8) were identified as BVDV 1b sub-genotype. Although the bootstrap and pairwise sequence identity estimates were low, a BLASTn search identified that the BVDV2a Ethiopian isolates were somewhat (71%−75%) related to isolates from, Spain (isolate GZ424). Furthermore, the BVDV 2a Ethiopian isolate exhibited a pairwise sequence identity ranging approximately from 83% to 94% (medium). In addition, the Chagni_ET_2022 isolate showed low similarity (72%) to the Chinese isolate CB7, although this comparison was supported by low bootstrap values and a relatively low pairwise sequence identity (Fig. 3; Table 1; Fig. 4). Representative sequences were deposited into NCBI GenBank under the following accession numbers: PQ096497 (BVDV_Chagni), PQ096499 (Addis_ET_2022), PQ096503 (Holeta_ET_2022), PQ096504 (Suluta2_ET_2022), PQ096500 (Suluta1_ET_2022), PQ096501 (Gondar_ET_2022), PQ096498 (Bishoftu_ET_2022), and PQ096502 (Bahrdar_ET_2022).

Fig. 3
figure 3

Rooted phylogenetic tree based on the 5′ UTR sequences. The phylogenetic tree was constructed by the neighbor-joining (NJ) (Kimura two-parameter method) with the sequences published in GenBank. The numbers at the phylogenetic branches indicate the bootstrap values (1000 replicates) in percentage supporting each group. The bar represents a genetic distance. The tree was viewed and designed using FigTree v. 1.4.4. The clades highlighted in light blue represent BVDV 2a and 2; green indicates BVDV 2b; silver corresponds to BVDV 1; pink denotes BVDV 1b and 1. Species marked with triangles and highlighted in purple color indicate Ethiopian isolates.

Full size image
Table 1 Percentage identity of Ethiopian isolates with selected reference strains.
Full size table

As shown in Fig. 4, the pairwise identity matrix for 5’UTR regions was calculated using SDT v1.2 software. Based on a color-coded pairwise identity matrix, 83% and 100% sequence identity among the Ethiopian BVDV2a isolates, while, the BVDV1b isolate was found very distinct compared to others. Compared to reference strains, Ethiopian BVDV2a isolates showed percentage pairwise sequence identity below 78% suggesting low genetic identity.

Fig. 4
figure 4

Color-coded matrix indicating the correspondence between pairwise sequence identities. SDT matrix of pairwise identity scores generated by alignment of a 230 bp fragment of the 5’UTR gene for 56 BVDV viruses. Each colored cell represents the percentage of identity between two nucleotide sequences, one horizontally and the other vertically, intersecting in the cell.

Full size image

Discussion

BVD is an economically important disease in the cattle industry worldwide24. It mainly affects cattle; however, the virus infects other even-toed ungulates, including sheep, goats, swine, yaks, deer, and camels. Moreover, wild ruminants and wild boars are also susceptible to the virus25. The disease is caused by BVDV, an RNA virus highly prone to mutation, like most other RNA viruses26. Studying viral species and sub-genotypes enables an understanding of the viral evolution and the source of infection. Furthermore, this information is an important input towards designing and implementing prevention and control programs against the disease15,27.

In Ethiopia, several pieces of serological evidence show that the virus is occurring in the country with a higher prevalence28,29,30,31,32. Until now, the genotype of the BVDV circulating in the country has not been studied. To the best of our knowledge, this study is the first to identify the species and subgenotypes of BVDV circulating in selected regions in Ethiopia. Our findings revealed that BVDV 2a is the predominant subtype circulating in dairy farms. Meanwhile, BVDV 1b was detected in samples collected from one breeding center, which reported a history of high calf mortality and morbidity.

Although isolates from Ethiopia had lower pairwise percentage identity with reference strains and low bootstrap value, 83–94% (medium) sequence identity was recorded within the Ethiopian BVDV 2a isolates. Suggesting that there is moderate genetic identity among Ethiopian isolates. This may be attributed to dairy cattle trading across the areas. Furthermore, BVDV 1b was very distinct because samples were collected from breeding farms that reported high calf mortality during this study period. The lower pairwise identity and weak bootstrap values may be linked to the low phylogenetic signal of the 5’ UTR marker, errors in sequences, limited detailed information on the molecular epidemiology of BVDV in Africa, as well as the greater sequence divergence from other reference strains. These limitations hindered us from constructing robust phylogenetic trees using all publicly available sequences and validating genotype assignment through a standard genetic clustering algorithm.

However, our results were overall consistent with the current literature, suggesting a continuing increase in the diversity of the BVDV 1 subgenotype5. Sub-genotypes, including BVDV 1b, BVDV 1a, 1c, and 2a have been reported predominantly in cattle-producing countries33. BVDV-1a is the predominant isolate in South Africa and is widely circulating in the United States, Korea, and Japan, while BVDV 1c has been reported as predominant genotype in Australia34. BVDV 1a and 1c are among the most commonly detected subgenotypes in different regions of China35. In Italy, BVDV1b and 1e are among the highly prevalent subtypes, exhibiting a wide range of temporo-spatial distribution across the country36.

In Africa, scattered studies on the genetic diversity of BVDV among cattle and other ruminants have been reported. Sub-genotypes, BVDV 1a, 1b, and 1f have been identified from cattle, buffaloes, and goats in Egypt37,38,39. BVDV 1 and BVDV 2 were reported in South Africa40,41,42. In Botswana, BVDV 1a was detected in serum samples collected from viremic cattle43. Although the subspecies for BVDV 2 from South Africa has not been studied, the current isolates from Ethiopia were consistent with the previous reports from those African countries.

A combination of vaccination and strict biosecurity measures is the best strategy to minimize the impact of BVD in cattle-producing countries. However, genetic heterogeneity of the viral isolates constrains the prevention and control of BVD. The efficacy of BVDV vaccines may vary according to the type of viral strain isolated from different regions44. Despite the availability of various types of BVDV vaccines in global markets, there is no vaccination program against BVDV in place in Ethiopia. As a result, the dairy industry has continued to be significantly affected by the virus. This study contributes to the basic knowledge of BVD in Ethiopia to design and implement prevention and control strategies against the disease.

Conclusions

In this study BVDV 2a and BVDV 1b were identified based on sequencing of the 5’-UTR region of the virus isolated from dairy farms and a breeding center, respectively. Although the bootstrap and pairwise sequence identity values were low, the BVDV 2a and 1b isolates showed some similarity to species previously detected in Spain and China, respectively. The species circulating in Southern Ethiopia have not been detected during this study. These findings offer valuable insights for developing strategies to prevent and control the disease, as well as for guiding the selection or development of an effective vaccine. To better understand the epidemiology of BVDV, further studies are recommended on the role of cattle kept under extensive farming systems in disease epidemiology, whole-genome sequencing, and the molecular epidemiology of the virus in other regions, including South Ethiopia, using 5’ UTR, Npro, and E2 markers.