Bovine parainfluenza virus type 3 (BPIV3), a member of the Respirovirus genus in the family Paramyxoviridae, is one of the most critical viruses causing bovine respiratory disease complex (BRDC), which results in significant economic losses to the global cattle industry (1, 2). BPIV3 primarily infects cattle, damaging the lower respiratory tract, triggering interstitial pneumonia, and leading to immunosuppression of the host organism (3). In addition, BPIV3 is associated with abortion in dams and has attracted increasing attention for its role in the reproductive health (4, 5). BPIV3 was first detected in BRDC cases in 1959 (6). BPIV3 is an enveloped, non-segmented, negative-strand RNA virus with a genome approximately 15 kb in size. It encodes six structural proteins (N, P, L, M, HN, and F) and three non-structural proteins (C, V, and D) (7–9). Currently, BPIV3 is classified into three genotypes (A, B, and C) based on phylogenetic analyses (10, 11), and it has been characterized as having a worldwide distribution (12–15).

BPIV3 was isolated from aborted fetal cows in Ohio as early as 1965 (16). Subsequent isolations of BPIV3 from aborted fetal cows have been reported in Portugal and Brazil, but the genome and genotype of these strains have not been reported (16, 17). Research data on BPIV3 related to abortion in pregnant cattle remain scarce. One study showed that natural infection with BPIV3 genotype A in a Holstein cattle resulted in abortion, with the lungs of the aborted fetus exhibiting diffuse reddening and a rubbery texture (5). Additionally, several experimental studies of in-utero infections have found that BPIV3 was able to cause abortions in two out of six dams and that surviving calves were weak, unable to stand after birth, and developed interstitial pneumonia (17, 18). However, the precise mechanism of transmission remains unclear, particularly whether BPIV3 can cross the placental barrier to directly infect the fetus following maternal infection. A study conducted in Argentina provided new insights by detecting BPIV3 genotype B in the vaginal tissues of infected dams, suggesting a potential transmission route (13).

Despite the long-standing recognition of BPIV3’s association with miscarriages, there remains a scarcity of genomic studies exploring the pathogenesis and transmission of its strains. In particular, there is a significant gap in global genomic sequence data concerning BPIV3 derived from aborted samples, which limits our comprehensive understanding of its molecular characteristics. Therefore, this study aims to fill this gap by isolating and characterizing BPIV3 from aborted fetus, and conducting genomic analysis. By elucidating the genome sequences of these strains, we aim to uncover the genetic evolutionary characterization of BPIV3, thereby enhancing our understanding of the molecular mechanisms underlying BPIV3-induced abortions.

Out of six samples, BPIV3 positivity was detected only in lung samples from aborted fetuses. The strain was successfully isolated using MDBK cells and was named LC1. After five passages, the isolate demonstrated stable cytopathic effects, such as nuclear condensation and cytoplasmic fragmentation when cultured for 48 h (Figure 1A). No CPE was observed in the negative control culture (Figure 1B). The virus titer was 10^7.5TCID₅₀/mL. Indirect immunofluorescence assays showed strong BPIV3 antigen specific fluorescence in the cytoplasm (Figure 1C), and electron transmission electron microscopy revealed that the isolate had a diameter of approximately 250 nm and vesicle membranes (Figure 1D). The HA titer of the LC1 strain was 1:8. After primer amplification, we obtained a complete genome sequence of 15,474 bp in length, containing all coding regions for six structural proteins (N, P, L, M, HN, and F) and three non-structural proteins (C, V, and D). Comparative genomic analysis revealed that the number, transcriptional orientation, and sequential arrangement of genes were highly conserved among BPIV3 strains, consistent with the typical genomic organization of BPIV3 genotype C (3′-N-P(C/V/D)-M-F-HN-L-5′). This sequence has been uploaded in the GenBank database under the accession number PQ074215.

Phylogenetic analysis based on the complete genome revealed that this isolate belonged to genotype C and clustered into a branch with Chinese BPIV3 genotype C strain SD2020 (Figure 2A). The complete genome of the LC1 isolate showed nucleotide similarity ranging from 80.8 to 99.6% and complete coding sequences of all viral proteins showed amino acid similarity ranging from 87.6 to 99.7% compared to BPIV3 strains representing all genotypes in the GenBank database; the nucleotide similarity of LC1 to all BPIV3 C strains in the GenBank database was 81.8 to 99.6%, and complete coding sequences of all viral proteins showed amino acid similarity was 88.2 to 99.7%, showing the highest amino acid homology to the BPIV3 genotype C strain SD2020, which was isolated in Shandong Province, China (Figure 2B). The recombination analyses revealed no recombination events for this genome.

Complete-genome sequence analysis is a powerful tool for studying pathogenicity, pathogenic mechanisms, and transmission routes of the pathogen. Further analysis of all BPIV3 genotype C complete genome sequences available in the GenBank database revealed a total of six unique nucleotide mutations in its P (T890C, T1191C), M (C13A), and L (C483T, G5046A, C6183T) genes. Two of these mutations were sense mutations, resulting in unique changes at position 297 for the P (I297T) protein, and at position 5 for the M (P5N) protein. Modeling of P and M proteins and subsequent analysis after introducing mutations showed that the amino acid mutations reduced the number of hydrogen bond (H-bond) in the P and M proteins by one at these sites (Figure 3).

Selection pressure analyses of the BPIV3 N, P, M, F, HN, and L genes revealed that the dN/dS values varied across genes: the N gene ranged from 0 to 0.1133, the P gene from 0.005 to 0.595, the M gene from 0.006 to 0.624, the F gene from 0 to 0.5906, the HN gene from 0 to 0.521, and the L gene from 0.031 to 0.227. Notably, the dN/dS values for all genes were significantly less than 1, suggesting that these genes are predominantly subjected to negative selection pressure. Further analysis of mutations at unique amino acid sites in the P (I297T) and M (P5N) proteins indicated that mutations in the P protein were predicted to experience positive selection pressure by three algorithms (MEM, FEL, and FuBAR). In contrast, the mutation in the M protein was predicted to be under positive selection pressure by only the MEM algorithm.

BPIV3 is one of the pathogens causing reproductive disorders in cattle (15), but the research data on BPIV3 related to abortion in dams remain scarce. Previous research has demonstrated that BPIV3 genotype A strains have been identified in aborted Holstein fetuses in the United States, causing tissue damage, including necrotizing bronchiolitis/alveolitis, intrabronchial/alveolar fibrin exudation, non-suppurative-peribronchitis, perivascular interstitial pneumonia, and necrosis of intestinal epithelial cells (13). BPIV3 genotype B strains were isolated from vaginal swabs of buffaloes with reproductive diseases in Argentina (10), implying that BPIV3 can be transmitted through the reproductive tract. Given the close anatomical connection between the vagina and the uterus, there is a potential for the virus to be transmitted to the uterus through retrograde infection, which could subsequently impact fetal health. However, to our knowledge, this study is the first to report the isolation of a genotype C strain from an abortion sample. Therefore, further research is needed on the pathogenicity, pathogenic mechanisms, and transmission routes of BPIV3 genotype C strains in the bovine reproductive system. In addition, we amplified the nearly complete genome of the abortion-derived BPIV3 and uploaded it to the GenBank database. The release of this genome sequence fills a gap in the existing data and lays a foundation for further molecular biology research and exploration of the pathogenic mechanism.

Phylogenetic analysis confirmed that the LC1 isolate belongs to BPIV3 genotype C, forming a closely related clade with SD2020 (GenBank accession number MZ359081.1), a respiratory-derived strain from China. This genetic relationship is supported by high nucleotide and amino acid sequence identities between LC1 and SD2020, despite their isolation from geographically distant provinces, Heilongjiang and Shandong, respectively, which are separated by over 1,000 km in a straight line. These findings underscore the necessity of comprehensive, regionally inclusive genomic surveillance of BPIV3 to effectively track viral evolution and transmission dynamics. However, further sequence analysis revealed that the abortion-derived BPIV3 genome had four synonymous mutations in the P and L genes, as well as one missense mutation each in the P (I297T) and M (P5N) proteins. These missense mutations resulted in the loss of an H-bond at the mutation sites in the P and M proteins, respectively. H-bonds play a critical role in biochemical reactions, such as protein folding, protein-ligand, protein-lectin, or protein–protein interactions, which may lead to an alteration in their conformation and function (22). In addition, synonymous codon substitutions are known to control the level of ribosomally expressed proteins affecting protein folding (23). Therefore, both synonymous and missense mutations may have important biological significance, and further analyses are needed to determine the functional characterization of these mutations. Notably, the genomic information for the strain we studied was obtained from an abortive sample, representing only genomic data from an abortive source, while all other known BPIV3 genomic comes from respiratory samples. This unique source emphasizes the possibility that this mutant strain may possess specific biological characteristics associated with abortion. More genomic sequences from aborted isolates will be needed in the future to identify this unique mutation.

Selection pressure plays a crucial role in viral evolution, influencing genetic diversity and adaptability (24). The selection pressure analysis of BPIV3 genes (N, P, M, F, HN, and L) provides valuable insights into the evolutionary dynamics of this virus. The results of this study indicate that all the examined BPIV3 genes are under negative selection pressure, which is essential for eliminating deleterious mutations and maintaining the functional integrity of key viral proteins. Additionally, predictive analyses of unique amino acid mutation sites in the P (I297T) and M (P5N) proteins revealed that localized sites may experience positive selective pressure. This suggests that these mutations could confer adaptive advantages, aiding the virus in adjusting to host environments or enhancing its transmission capabilities. Collectively, these findings underscore the intricate interplay of selection pressures in shaping BPIV3 evolution.

Currently, research on the P and M proteins of BPIV3 strains is limited. Existing studies indicate that the open reading frame of the P gene not only translates the phosphoprotein but also encodes three non-structural proteins (V/C/D), which primarily function to inhibit host cell interferon production (25). The M protein plays a critical role in virus assembly, budding, and release (26). A recent study revealed that the first 10 amino acids at the N-terminus and the last 4 amino acids at the C-terminus of the BPIV3 M protein are crucial for the release of viral particles (27). In this study, the isolate shows a mutation at the fifth amino acid at the N-terminus of the M protein. Therefore, we hypothesize that this mutation could potentially affect the release of viral particles. This potential impact on viral release merits further investigation to understand the biological consequences associated with this mutation. In conclusion, the unique amino acid mutations identified in the P and M proteins of the isolates in this study necessitate further investigation to elucidate their impact on the biological properties of these strains.

In this study, we successfully isolated an abortion-derived BPIV3 genotype C strain from the aborted fetus of a Holstein sample and obtained the nearly complete genome. This work confirms the existence of an abortion-derived BPIV3 genotype C strain and provides valuable data for its molecular characterization. Complete genome sequence analysis identified unique amino acid mutations in the P and M proteins of the isolate. These mutations may significantly impact the biological properties of the isolate; therefore, future research should focus on elucidating the specific effects of these mutations on the virus’s biological characteristics and their role in the BRDC and reproductive disorders.