Rectal swabs from 230 cats (45 healthy cats and 185 cats with diarrhea) were obtained from pet hospitals in Guangdong, Henan, Anhui, and Zhejiang provinces and the Inner Mongolia Autonomous Region from June 2018 to March 2021. This study adhered to the ethical policies and was approved by the Committee on the Ethics of Animal Experiments of South China Agricultural University (SYXK 2019-0136).

For nucleic acid extraction, sodium phosphate buffered saline was added to swabs collected from cats. They were then vortexed and centrifuged at 4°C. Viral DNA/RNA was extracted from these samples using the EasyPure Viral DNA/RNA Kit (TransGen Biotechnology, Beijing, China). In this investigation, diagnostic primers for the nonstructural protein 1 (NS1) protein-coding gene (sited 2138–2448 according to the IDEXX-1 strain; accession no. MN396757) were used to determine the presence of FeChPV DNA through a previously reported hemi-nested polymerase chain reaction (PCR) method. The primers used for the analysis are listed in Supplementary Table S1 (14). First-round PCR primers (FeChPVF1 and FeChPVR1) amplified a 332 bp region, and second-round primers (FeChPVF2 and FeChPVR2) amplified a 311 bp region. The reaction conditions were as follows: 95°C for 5 min, 30 cycles at 95°C for 30 s, 58°C for 30 s, 72°C for 30 s, and a final extension at 72°C for 10 min.

The samples were examined for feline gastroenteritis–associated viruses other than FeChPV, such as cachavirus (CachaV) (12), feline astrovirus (FeAstV) (16), feline bocavirus (FBoV) (16), feline kobuvirus (FeKoV) (17), and feline parvovirus (FPV) (18), using reported primer pairs that exhibit extensive reactivity with various viruses. The primer pairs used are listed in Supplementary Table S1. Moreover, the UpSet diagram produced by TBtools and Wayne diagram¹ were jointly used to indicate the viral coinfection status (19, 20).

Based on the genome sequences of the FeChPV strain IDEXX-1, eight primer pairs amplifying overlapping fragments were designed and synthesized (Supplementary Table S1) to amplify the whole genome of FeChPV. For PCR, a 20-μL reaction mixture was prepared, which included a template DNA (>100 ng/L), 6 pmol of upstream and downstream primers, primer STAR HS DNA polymerase, and supporting reaction buffer (TaKaRa). For sequence amplification, the following cycle conditions were employed: predenaturation at 95°C for 5 min, followed by 35 cycles of 95°C for 30 s, 55°C for 30 s, and 72°C for 1 min, and the final extension step at 72°C for 10 min. The harvested amplicons were inserted into pMD18-T cloning vector (TaKaRa) and sequenced by Hongxun Company (Jiangsu, China).

The complete nucleotide sequence of the genome and the amino acid (aa) sequence of NS1 of the obtained and reference parvovirus strains were aligned via the Muscle algorithm in MEGA software (21). Phylogenetic trees were constructed using the aa sequences of NS1 and VP1 via the maximum likelihood method in MEGA software with the JTT + G + I model and 1,000 bootstrap replicates. Additionally, ChiPlot² was used to construct and display the sequence alignment heatmap (22).

The prediction of recombination events in the obtained strains compared with other ChPVs was performed using RDP4.36 with default parameters via RDP, GENECONV, MAXCHI, and BOOTSCAN. The final results were confirmed using SimPlot 3.5.1 (23, 24).

To comprehensively understand the impact of mutation sites on the tertiary structure, CHN190305 and CHN191011 harboring representative variant sites were selected to construct structural models of NS1 and VP1, respectively. Moreover, tertiary structural models of NS1 and VP1 were constructed for the reference strain IDEXX-1. WeMol³ and SWISS-MODEL⁴ were used to model the altered aa sequences of the obtained strains. Furthermore, PyMOL was used to gather and preserve the modeled Protein Data Bank files.

Supernatants from FeChPV-positive samples were cocultured with CRFK and MDCK cell lines for 4 days. The cultures were split every 4 days for five passages. PCR assay was performed using the method described above to detect FeChPVs.

The prevalence of FeChPV was compared between healthy and diarrheal cats using Fisher’s exact test. GraphPad Prism 8.0 (San Diego, California, United States) was used for all statistical analyses. Statistics were deemed significant at p < 0.05.

Of the 185 collected samples, the 6 samples from cats with diarrhea tested positive for FeChPV. Samples from healthy cats did not test positive for FeChPV. In addition, 147 (79.4%), 34 (18.5%), 18 (9.7%), 5 (2.7%), and 3 (1.6%) samples tested positive for FPV, FBoV, FeAstV, FeKoV, and CachaV in cats with diarrhea, respectively (Figure 1). All FeChPV-positive samples (6 of 185, 3.2%) were from cats with symptoms of diarrhea. However, statistical analysis did not suggest an association between FeChPV infection and clinical symptoms (p > 0.05).

Additionally, cultivated cells did not exhibit cytopathic effect until the fifth passage, and FeChPV DNA could not be identified using PCR.

The near full-length genome sequences (4,092 nucleotides) of six strains identified in this study were deposited in GenBank (accession nos. OQ694028, OQ694029, OQ694030, OQ694031, OQ694032, and OQ694033). The comparison of whole-genome sequences of the six strains and the reference FeChPV strains deposited in the National Center for Biotechnology Information database demonstrated that the FeChPV strains identified in this study contained two main open reading frames encoding a 658-aa NS1 and 508-aa VP1.

Sequence analyses revealed that the six strains possessed a complete genome nucleotide identity of 98.2%–99.1%. In addition, they exhibited 83.2%–98.9% and 61.2%–69.1% genome identity with other relevant FeChPV and CachaV strains, respectively. Moreover, the genome identities of the other viruses belonging to the genus Chaphamaparvovirus varied from 32.2% to 54.3% (Supplementary Figure S1). According to aa identity analysis of NS1 and VP1, the six FeChPVs shared overall aa identities of 96.6%–98.1% for NS1 and 97.8%–99.1% for VP1. The six FeChPVs had aa identities of 97.1%–99.1% for NS1 and 97.4%–99.2% for VP1 with FeChPV; 63.9%–65.3% for NS1 and 63.5%–65.2% for VP1 with CachaV strains, and 30.5%–47.5% for NS1 and 6.2%–48.6% for VP1 with other representative strains.

Complete genome phylogenetic trees of the six obtained strains and 38 reference strains (FeChPV, CachaV, and other chaphamaparvovirus) were constructed to assess the genetic association (Figure 2A). The trees of FeChPV indicated that the two obtained strains CHN200228 and CHN200523 were closely related to reference FeChPVs, which were primarily identified from China with identities of 82.96%–98.66%. The remaining four FeChPVs were clustered into another sub-branch with CHN20201025 and CHN20201226 (detected in dogs, shared identities of 98.12%–99.02%) and reference strains detected in other countries. For the aa phylogenetic tree of NS1, the six obtained strains and other FeChPV strains were grouped into one major branch and were distantly related to cachavirus strains and reference ChPV strains (Figure 2B). The phylogenetic tree constructed using the aa sequences of VP1 showed that all the six obtained strains were clustered with two FeChPV strains from dogs and other Chinese FeChPV strains (except C8-2) previously reported (Figure 2C) (15, 25).

RDP4 and SimPlot were used to predict the recombination events in FeChPV strains (Supplementary Table S2). The recombination analysis predicted four recombinant events, indicating the two obtained strains CHN200228 and CHN180917 as minor and major recombinant parents, respectively, for CHN20201025 detected in dogs (Figure 3).

The comparison of aa sequences of NS1 and VP1 in the six obtained FeChPVs and reference FeChPV strains revealed some unique variants. There were three unique sites in NS1, namely, Phe21Leu, Lys267Arg, and Lys267Glu/Arg, which were identified only in the obtained strains. Thr75Ala was found only in the NS1 protein of the obtained strains and few Chinese FeChPVs. The VP1 protein of the obtained FeChPVs demonstrated alterations at His419Thr and Asp479Gly. His45Tyr, Ala57Ser, and His419Asn were only harbored in the VP1 protein of the obtained strains and few Chinese FeChPVs. Supplementary Tables S3, S4 list the additional variant sites in NS1 and VP1, respectively. The variation in the tertiary structure of NS1 and VP1 of the obtained strains is illustrated in Figure 4.

Based on a previous study (26), it was inferred that the NS1 protein of the six obtained Chinese strains possessed two conserved endonuclease (replication initiator) motifs: 95-FHIHVIMAL-102 and 149-SLIAYMCK-156. Furthermore, motif related to the helicase domains 311-GCSNTGKS-318, 349-IGVWEE-354, 366-KQIFEGMECSIPVK-379, and 391-IIMTTN-396 was conserved in NS1 of all obtained strains except the CHN180917 strain, and it harbored the mutation of Met372Thr that did not lead to tertiary-structure change (27). The VP1 protein of other members of the subfamily Hamaparvovirinae did not possess the motifs of phospholipase A2 (PLA2) (28). Meanwhile, an 18 aa-elongation located at the N-terminus of the VP1 protein was present in all six strains obtained in this study and FeChPVs previously detected in Canada and Italy (14). A relatively long capsid protein has also been observed in other members of the genus Chaphamaparvovirus, which has been hypothesized to be related to low expression in host (29).