Introduction

Polyomaviruses (PyVs) are nonenveloped DNA viruses; a total of 86 PyV species have been classified [1]. PyVs are ubiquitous viruses that frequently infect cattle [2, 3], fish [4, 5], horses [6], pigs [7,8] and other vertebrates including humans [9, 10], and their effects on distinct hosts differ [11]. PyVs replicate in animal tonsils and adenoids [12] and may cause benign or malignant tumors in some animals [13].

Lyon IARC PyV (LIPyV) was first reported in 2017 in skin samples from humans. A genetic analysis indicated a 65% nucleotide identity with previously reported raccoon polyomavirus (RacPyV) [14]. In a study by Brostoff et al., LIPyV was found in 1/127 benign skin tumor samples from liver transplant recipients in Finland [14]. In another study by Gheit et al. 9 skin swabs (9/445; 2.0%) and 1 eyebrow hair sample (1/439; 0.2%) tested positive for LIPyV from 448 American skin cancer screening patients and 3/140 gargle samples from healthy French adults [15]. Recently, no evidence of LIPyV presence was found in 689 noncancerous human tonsil samples or 139 gargle samples from France [16]. The same result was reported in a serological survey of 156 people in the Netherlands [17]. LIPyV has also been found in the feces of diarrheic cats [18]; however, only two complete genomes of LIPyV strains have been sequenced to date. Furthermore, epizootiology and molecular data on LIPyV in China have yet to be reported. In this study, LIPyV was identified in fecal samples collected from cats with diarrhea from Guangxi Province, China between 2016 and 2018 by the Guangxi Center for Animal Disease Control and Prevention and further characterized by genomic similarity and phylogenetic analysis.

Stool supernatants were extracted according to the instructions of the DNA extraction kit. Polymerase chain reaction (PCR) amplification was performed by using the detection primers for LIPyV (F-R) (Table 1) (Fahsbender et al. 2019). The PCR products were purified from agarose gels and sequenced. A BLAST analysis of the sequences identified that 2% (2/100) of the samples contained a virus most closely related to LIPyV. The LIPyV-positive samples were also examined for feline panleukopenia virus (FPV) [19], feline bocavirus (FBoV) 1 to 3 [20], feline astrovirus (FeAstV) [21], feline kobuvirus (FeKoV) [22] and feline coronavirus (FCoV) [23] with the previously described PCR/RT-PCR assays. The LIPyV-positive samples were also evaluated according to specific primers (16S mam1/16S mam2) as previously described [24].

Table 1 List of primers
Full size table

The sequences obtained from the two PCR-positive samples were identical. Moreover, the two cats were related, therefore it was assumed that most likely identical viruses were present in both cases, so only one was selected for analysis. Five pairs of PCR primers that cover the complete genome sequences of previously reported LIPyV strains were designed (Table 1). The target fragments were amplified by using 2 × VazymeLAmp Master Mix (Vazyme, China). DNA fragments of 1072, 1296, 1305, 1297 and 695 bp were produced. The PCR-amplified products were ligated into pMD18T vectors (Takara, Japan), and the positive plasmids were sent to Sangon Biotech (Shanghai) Co., Ltd. for sequencing. After obtaining the raw sequencing data, DNASTAR SeqMan was used to assemble the whole genomic sequences of LIPyV-GXNN01 (accession number: MW054655). A sequence analysis of the entire genome and the genomes of representative strains of mammalian PyVs in GenBank was performed (Table S1).

The genome of LIPyV-GXNN01 is circular and 5263 bp in length, and it contains open reading frames (ORFs) for all the major PyV proteins (Fig. S1). An analysis of the complete nucleotide sequence showed that the LIPyV genome shares the features of other known PyVs, with an early region that consists of ST-Ag and LT-Ag and a late region coding for the VP1, VP2, and VP3 structural proteins with lengths of 1307, 743 and 515 aa, respectively, (Table S2).The G + C content of LIPyV-GXNN01 was 39.6%, and the VP1, VP2 and VP3 sequences of LIPyV-GXNN01 exhibited a 95.9–99.4%, 97.6–99.2%, and 97.1–99.2% identity, respectively, with their counterparts in other LIPyVs reported previously (Table 2). However, with the corresponding sequences of other PyVs, such as those from chimpanzees or raccoons, they shared less than a 65% identity.

Table 2 Genetic sequence similarity among different polyomavirus hosts (%)
Full size table

In addition, compared with the reference LIPyV sequences identified in the USA, several amino acid substitutions (T41A in VP2; S334N in VP1 Y435F, I436F, T679I, and K692R in LT-Ag; and A126S in ST-Ag) were observed in the genomic sequences of LIPyV-GXNN01.A noncoding control region (NCCR, nucleotide positions1–401) that shared the characteristics of the ori regions of most mammalian PyVs was found. This region contains four LT-Ag-binding sites [15], specifically, one GAGGC and three reverse complement GCCTC motifs. To visually illustrate the evolutionary relations among the LIPyVs and other PyVs, after estimation with the Model Finder program, a maximum likelihood (ML) phylogenetic tree was created with gamma distributed with invariant sites’ (G + I) substitution and Kimura 2-parameter models, respectively, with 1000 bootstrap replicates that used MEGA7 [25] (Fig. 1).The results based on full-length genomes demonstrated that LIPyV-GXNN01 formed a single clade with LIPyV reference strains LI polyomavirus and Lyon-IARC polyomavirus. Lyon-IARC polyomavirus strains clustered significantly with different mammalian PyVs including human polyomaviruses, bat polyomaviruses, and primate polyomaviruses.

Fig. 1
figure 1

Phylogenetic analyses of PyVs based on the full-length genomes using the maximum likelihood algorithm and 1000 bootstrap replications in a heuristic search with MEGA 7. Black circles represent sequenced strains from this study. Animal graphical indicate the virus host

Full size image

Here, we present the first evidence for the occurrence and possible circulation of LIPyV among cats in China. The proportion of the positive samples (2%) is comparable to that reported in other countries, including the USA (0.2%), France (2.14%) and Finland (0.79%). Although the presence of other viral agents (FPV, FeBoV 1 and 3, FeAstV, FeKoV, and FeCoV) was excluded in these two samples, the eventual etiological role of the detected LIPyV in feline diarrhea has yet to be clarified. In previous investigations, LIPyV was identified during a metagenomics analysis of feline diarrheic samples in which FPV, FBoV 3 and numerous bacteria were also detected [18]. The effect of coin fection with other enteric viruses on pathogenesis is still unclear. Despite the apparent distribution of LIPyV in skin biopsies, the debate regarding the possibility of the environmental contamination of human skin biopsy samples remains. To answer this question, LIPyV isolations necessary; however, no study has reported the successful isolation of the virus. We failed to use F81 cells for isolation.

Accordingly, this study is the first report of LIPyV in China and analyzes the presence of LIPyV, which provides a basis for research on PyV etiology and epidemiology. The threat of this virus to human health remains unknown. To investigate the pathogenicity of LIPyV, further studies including viral isolation are necessary.