Molecular characterization and genetic diversity of parvoviruses prevalent in cats in Central and Eastern China from 2018 to 2022

Cats are a potential source of genetic diversity for parvoviruses. Herein, 134 samples were collected from cats with clinical gastroenteritis and analyzed for the presence of viral DNA via polymerase chain reaction, which revealed 48 positive samples. Identity analysis of VP2 nucleotide sequences indicated that these 48 strains, belonging to feline panleukopenia virus (FPV) and canine parvovirus type-2 (CPV-2; including new CPV-2a and CPV-2c genotypes), shared 94.59–99.94% nucleotide identity with the reference strains. The FPV strain F8 (isolated from Vietnam) appeared to be a recombinant of strains HB2003 and JS1901, whereas the Chinese CPV-2b strain BM-(11) isolated in 2011 was believed to be a recombinant of strains AH2008 and JS1901. In phylogenetic tree analysis based on VP2 nucleotide sequences, all obtained FPV strains and most reference FPV strains were clustered together, except strain BJ-22, which originated from monkeys. Further, two new CPV-2a strains (AH2005 and AH2008) were close to the newly reported Chinese CPV-2a strains but were distant from the other CPV-2a strains, namely CPV-339 (from the United States) and K022 (from South Korea). Additionally, the FPV and CPV-2 strains had high mutation rates in the antigenic regions of the VP2 protein. According to model prediction of the CPV–VP2 protein, these mutations may cause changes in the tertiary structure of VP2. The findings of this study can be used to improve the pre-evaluation of vaccination efficacy against diseases caused by FPV and CPV-2 in domestic cats and understand their genotypic transmission and mutation trends.


Introduction
Parvoviruses are nonenveloped single-stranded DNA viruses that infect various animals (1). Feline panleukopenia virus (FPV) and canine parvovirus type-2 (CPV-2) are members of the genus Protoparvovirus belonging to the family Parvoviridae (2). FPV and CPV-2 can cause severe intestinal diseases in cats and dogs with high mortality and infectivity (3). They are mainly transmitted through feces, urine, and oral secretions from infected animals and can cause severe symptoms such as increased body temperature, vomiting, viral diarrhea, and a marked decrease in leukocyte count (4). Since the early 20th century, FPV has been recognized as one of the main pathogens that cause feline viral diarrhea (5). In general, cats infected with FPV have high morbidity and mortality, and kittens aged <3 months are highly susceptible to this virus (6,7). Further, CPV-2, which was discovered in the 1970s (8), causes a highly contagious and fatal disease in dogs (2). CPV-2 was previously believed to be evolved from FPV in domestic cats after cross-species transmission to wild carnivores as a viral intermediate (9), but the directionality of the mutations was not considered (10,11). The genome coding region of FPV/CPV-2 contains two major expression cassettes, encoding nonstructural (NS1 and NS2 for the left open reading frame) and structural (VP1 and VP2 for the right open reading frame) proteins (12). VP2 is a significant component of the FPV/CPV-2 capsid protein and represents approximately 90% of the virion (11). It is a critical protein that determines the antigenic properties, host range, and receptor binding of FPV/ CPV-2 (13). Multiple genotypes have been reported owing to the antigenic differences in VP2 across several CPV-2 strains. The genotypes of these strains are classified based on the key amino acid (aa) residues of VP2. The key aa residues of VP2 in different genotypes are as follows: classical CPV-2 containing 87Met, 101Ile, 300Ala, 305Asp., and 375Asn; CPV-2a containing 426Asn and 297Ser; CPV-2b containing 426Asp and 297Ser; the new CPV-2a strain containing 426Asn and 297Ala; the new CPV-2b strain containing 426Asp and 297Ala; CPV-2c containing 426Glu and 297Ala; and FPV containing 80Lys, 93Lys, 103Val, 323Asp., and 568Ala (14).
Initially, CPV-2 could not replicate in cats; however, currently, all CPV variants (CPV-2a to 2c) can infect cats and cause subclinical disease or feline panleukopenia (10). This study aimed to understand the evolution of FPV/CPV-2 circulating in domestic cats in Central and Eastern China. Samples collected from domestic cats were subjected to FPV/CPV-2 screening and molecular characterization of VP2.

Sample preparation and viral DNA/RNA extraction
A total of 134 rectal swabs or fecal samples were collected from cats with clinical gastroenteritis at pet hospitals in Henan, Anhui, Jiangsu, and Hubei Provinces. Each sample was suspended in phosphate-buffered saline, and viral DNA/RNA was extracted using EasyPure Viral DNA/RNA Kit (TransGen Biotechnology, Inc., Beijing, China) according to the manufacturer's instructions. The extracted viral DNA/RNA was stored at −80°C until analysis.

VP2 sequencing
For FPV/CPV-2, a pair of primers (targeting the nucleotides 2,761-4,536 according to the FPV strain HF1; accession number: MT614366) was used to amplify VP2 via polymerase chain reaction (PCR), as described previously (15). The amplicons from each positive sample were ligated into the pMD18-T easy vector, and the recombined plasmids were sequenced by Syn-Biotechnology (Syn-Biotechnology, Suzhou, China). Clinical information on the sources of these sequenced parvoviruses is provided in Supplementary Table S1.

Identity, recombination, and phylogenetic analyses
Sequences were assembled using the SeqMan module of Lasergene 7 (DNASTAR, Inc., Madison, WI, United States). To display the results more intuitively, BioAider Version 1.314 was used to analyze the 48 strains and 41 reference strains (Supplementary Table S2) for differences, and the results were annotated using Chiplot online software 1 (16).
Based on complete VP2 nucleotide sequences, a phylogenetic tree was constructed using MEGA-X Version 10.1 (bootstrap replicates = 1,000) via the maximum likelihood method (Tamura 3 parameter model) (24).

Mutation sites, antigen, and tertiary structure prediction
The aa variations in the obtained and reference strains were predicted as potentially dominant B cell antigen epitopes via DNAMAN Version 5.2.2 (Lynnon Biosoft, Canada). The 11 obtained CPV-2 strains were compared with the original CPV-2 strain (790,312 strain; accession number: M38245). The classical and variant CPV-2 strains were structurally modeled using homology via SWISS-MODEL 2 and were visualized using PyMOL Molecular Graphics System.

Detection and genotypes of parvoviruses
The PCR assay revealed that 48 samples (35.82%, 48/134) were positive for FPV or CPV-2, and FPV accounted for 77.08% (37/48) of the positive samples. The obtained FPV or CPV-2 strains were confirmed to contain full-length VP2 genes (1755 bp). The CPV-2 strains included two new 2a strains from Henan and Jiangsu Provinces and nine CPV-2c strains from Henan Province.
Frontiers in Veterinary Science 03 frontiersin.org

Nucleotide identity of the VP2 gene
The VP2 nucleotide sequences of the 48 obtained strains and 41 reference strains were aligned and cluster-analyzed ( Figure 1). The results revealed that all strains shared 94.59-99.94% nucleotide identity with each other; the FPV strains showed 97.72-99.94% identity, whereas the CPV-2 strains displayed 96.7-99.89% identity. No evident differences were noted in strain sequences across different provinces or years.

Recombination analysis
Based on RDP analysis, three major recombination events (1-3) were identified. Simplot software was used to analyze the recombination events. The results were consistent with those of RDP4 (

Phylogenetic analysis of FPV and CPV
The evolutionary tree of parvovirus was constructed based on the VP2 nucleotide sequences of 89 FPV and CPV-2 strains (Figure 3). The FPV strains obtained in this study were grouped into three phylogenetic clusters and were closely related to most reported FPV strains but were distant from the isolated BJ-22 strain, which originated from a monkey with diarrhea (accession number: FJ231389). Based on the evolutionary relationship, the nine obtained CPV-2c strains were distant from the reported reference CPV-2c strains from Germany (G367-97 strain, accession number: FJ005202 and G333-99 strain, accession number: FJ005204) and Ecuador (2c-ME28-ECU2012 strain, accession number: KF149984). Two new CPV-2a strains (AH2005, accession number: OQ868526 and AH2008, accession number: OQ868527) obtained in this study were close to certain new

Antigenic epitope predictions
To provide an immune-response base for FPV and CPV, the B cell epitope prediction of VP2 was performed in these 89 strains, revealing  Table S3). The epitope with the highest score (1.207) was located at the aa residue position of 247-256.

Discussion
In this study, domestic cats with suspected FPV/CPV-2 infections in Central and Eastern China were evaluated; 37 FPV and 11 CPV-2 strains were detected, indicating that CPV-2 is also widely distributed in domestic cats in China. Additionally, based on our previous study, trends in CPV-2 infections revealed that CPV-2c was the most widespread genotype (15). Considering the complex prevalence and code translation of CPV-2 in dogs in China (25), the genetic variation of CPV-2 in domestic cats requires continuous monitoring.
The obtained FPV strains showed 97.72-99.94% identity, whereas the CPV-2 strains displayed 96.7-99.89% identity. The 11 FIGURE 3 A phylogenetic tree based on the VP2 gene from the strains obtained in this study and reference strains. The 48 strains obtained from domestic cats in this study are indicated with a red star. Bootstrap values calculated from 1,000 replicates are indicated by the size of the red triangle on the respective branches. Different colors represent clade taxanomy of strains and the distance scale of the branch of the phylogenetic tree.
Frontiers in Veterinary Science 06 frontiersin.org CPV-2 strains obtained in this study were all from Henan and Jiangsu Provinces. The high similarity of the strains in different regions indicates the high infectivity of CPV-2 with the impact of human activity. To further explore the evolutionary trend in FPV and CPV-2 strains, we performed phylogenetic and recombination analyses. Evolutionary analysis revealed a relationship between FPV and CPV strains. The FPV strains were grouped into three branches in the evolutionary tree, suggesting the emergence of different subtypes. In general, CPV-2 harbors more mutations and evolves more rapidly than FPV (26). The new CPV-2a genotype is still the predominant epidemic genotype of CPV-2 in dogs from some provinces of China (27,28). The two new CPV-2a strains obtained in this study were significantly different from the other new CPV-2a strains reported abroad, and further research is warranted to expand and determine the evolutionary trend. The results of CPV-2 genotyping were not completely consistent with those of phylogenetic analysis, which showed that changes in a single base do not affect the overall evolutionary characteristics of CPV-2.
To understand the evolutionary pathways of new strains, we further analyzed and predicted the recombination possibilities of the obtained and reference strains. Recombination analysis revealed that complex recombination events might occur between FPV and CPV-2 in cats. The HN2103 strain (CPV-2c type) was a recombinant of the FPV strain HN1806 (collected in 2018) and the CPV-2c strain AH2008 (collected in 2020), with AH2008 also being a recombinant of the JS1901 (a CPV-2c type strain collected in 2019) and BM-(11) (a new CPV-2b type strain collected in 2011) strains. The presence of recombinants between the FPV and CPV-2 strains with different genotypes in different regions or years suggests the role of cats as reservoirs for CPV-2 transmission. Moreover, the Chinese CPV-2c strain JS1901 (collected in 2019) and the FPV strain HB2003 (collected in 2020) were the recombinant precursors of the Vietnamese FPV strain F8 (collected in 2018), demonstrating the complex evolution and recombination between the FPV and CPV-2 strains. Furthermore, the Ser297Ala mutation in the antigenic epitope of VP2 protein was observed in all 11 CPV-2 strains isolated from domestic cats in this study. This mutation has been proven to be related to antigenicity changes in new CPV-2a/2b and may have a pronounced effect on continuous host adaptation (29). However, it remains unclear whether the 11 feline CPV-2 strains originated from antibody pressure in vaccinated animals and underwent key point mutations to develop feline tropism. For the prevention and control of CPV-2 in the same residence, the cross-species transmission between cats and dogs may be a challenge (30).
Moreover, the tertiary structure of the VP2 protein of CPV-2 showed that mutations (Met87Leu, Ile101Thr, and Asn426Glu) in the antigenic region may cause structural changes in the VP2 protein, which may result in cross-host transmission (31). Taken together, mutations in the VP2 protein may be responsible for the pathogenicity and evolution of FPV and CPV-2 (32).
In conclusion, our study identified 48 parvovirus strains and confirmed the cocirculation of FPV, new CPV-2a, and CPV-2c strains in domestic cats from Central and Eastern China. This study contributes to the research on the evolution and monitoring of parvovirus in domestic cats, indicating that continuous surveillance may help update the currently available vaccines.

Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://www.ncbi.nlm.nih.gov/genbank/, OQ868522-OQ868569.

Ethics statement
The animal study was reviewed and approved by Animal Welfare and Ethics Committee of Nanyang Normal University (No. 14027). Written informed consent was obtained from the owners for the participation of their animals in this study. Frontiers in Veterinary Science 07 frontiersin.org