Phylogeny and VP2 amino acid sequence substitutions
The ML phylogenetic tree constructed in the present study was inferred based on 144 reference VP2 sequences of Carnivore protoparvovirus 1 and those of the parvoviruses obtained from the two pangolins examined in this study. The original tree is shown in Fig. S1, and a summarized version of the tree is shown in Fig. 3. The VP2 sequences in samples obtained from pangolin P1 were classified into the CPV-2c subclade, with 99.8% identity to a CPV-2c strain (MN832850) isolated from a Taiwanese pangolin in China, whereas those from pangolin P2 were classified into the CPV-2a subclade, with 99.8% identity to a CPV-2a strain (KY386858). Notably, CPV-2a and CPV-2c are the dominant strains prevalent among domestic dogs in southern China (Chiang, Wu, Chiou, Chang, & Lin, 2016; Hao et al., 2020; Qi, Zhao, Guo, & Sun, 2020) and in neighboring tropical countries, respectively (Charoenkul et al., 2019; Hoang et al., 2019; Inthong, Kaewmongkol, Meekhanon, Sirinarumitr, & Sirinarumitr, 2020). In the present study, we established that the two Chinese pangolins were respectively infected with CPV-2a and CPV-2c. The VP2 sequences obtained from pangolin P2 were classified into the CPV-2a subtype group, closely related to CPV-2a strains, such as CPV-2a/Dog/GY-9/China/16 (KY386858), CPV-2a/Dog/CN/JS1704/China/17 (MK518017), and CPV-2a/Dog/CN/HN1702/China/17 (MK517985) isolated from dogs in Guiyang, Jiangsu, and Henan provinces in the central and southern regions of China (Fig. 3). Moreover, the VP2 sequences obtained from pangolin P1 were classified into the CPV-2c subtype group, closely related to CPV-2c strains, such as CPV-2c/Dog/Vietnam/HN07/17, CPV-2c/Pangolin/China/Taiwan/2018/18, and CPV-2c/Dog/South Korea/K01708-1/17, which were isolated in the neighboring countries of Vietnam and South Korea (Fig. 3). Accordingly, these findings provide convincing evidence to indicate that the CPV-2 strains infecting the two pangolins were of dog origin.
The amino acid sequence of the VP2 viral protein detected in pangolin P2 is characterized by an asparagine (N) residue at position 426 and refers to other residues (such as residues 80, 103 and 297) in the VP2 protein, on the basis of which, it should be identified as CPV-2a (new) subtype. In contrast, the sequence in pangolin P1 has a glutamic acid (E) residue at position 426, and accordingly should be recognized as the CPV-2c subtype. Consistently, the P2 sequence has glutamine (Q) and alanine (A) residues at positions 370 and 440, respectively, which are the same as those of the CPV-2a subtype, whereas the P1 sequence is characterized by arginine (R) and threonine (T) residues at positions370 and 440, as in CPV-2c.
Most CPV-2 variants are characterized by a glycine residue at position 300 of the VP2 sequence, which has been demonstrated to be necessary for the infection of canine and felid species (Allison et al., 2016; Voorhees et al., 2019). The findings of numerous studies have indicated that the CPV-2a variant can infect wild civet cats (Paradoxurus musangus ) (Mendenhall et al., 2016), red pandas (Ailurus fulgens )(Qin, Loeffler, Li, Tian, & Wei, 2007), masked palm civets (Paguma larvata ), and Chinese ferret badgers (Melogale moschata ) (Chang & Chen, 2021) in East Asian countries. The CPV-2c variant has also been confirmed to infect masked palm civets and Chinese ferret badgers (Chang & Chen, 2021). Moreover, CPV-2 has been established to be highly prevalent in farmed raccoon dogs and has evolved in only 10 years (Lu et al., 2020).