Predominance of canine parvovirus 2b in Japan: an epidemiological study during 2014-2019

Canine parvovirus 2 (CPV-2) is an important pathogen of domestic dogs and wild canids. In Japan, CPV-2 infection is one of the most common infectious diseases of dogs. We analyzed samples collected between 2014 and 2019 to identify antigenic variants of CPV-2 in dogs in Japan. Our results demonstrated that the CPV-2b variant was predominant. The CPV-2c variant was not found among our samples. Our findings demonstrate that the distribution of CPV-2 antigenic variants in Japan was more similar to that in Australia than to that in neighboring countries in Asia.

Canine parvovirus 2 (CPV-2) is an important pathogen of domestic dogs and wild canids. It is small non-enveloped virus with a single-stranded DNA genome belonging to the family Parvoviridae, subfamily Parvovirinae, genus Protoparvovirus, species Carnivore protoparvovirus 1 [1]. The CPV-2 genome encodes four proteins: NS1, NS2, VP1, and VP2. NS1 and NS2 are non-structural proteins that regulate replication of the viral genome, while VP1 and VP2 are structural proteins that constitute the capsid of the virion [2,3]. Amino acid residues in VP2 define the antigenic variants of CPV-2 [3]. CPV-2 infection is characterized by gastroenteritis-like symptoms such as fever, anorexia, diarrhea, and vomiting [4]. Unvaccinated adult dogs usually have subclinical infections [5].
CPV-2 was identified in the late 1970s in dogs [6]. CPV-2 is genetically similar to other carnivore parvoviruses and was presumably derived from feline panleukopenia virus (FPLV) [7]. Within a few years after the outbreak of CPV-2 infection across the United States and Australia in 1978, CPV-2 infection was identified in dogs across a number of countries. In the 1980s, two antigenic variants of CPV-2, namely CPV-2a and CPV-2b, emerged [8]. In 2000, a third variant, CVP-2c, was detected in dogs in Italy [9,10]. These three antigenic variants have a different amino acid at position 426 of the VP2 protein (CPV-2a, N; CPV-2b, D; CPV-2c, E) [8,10,11]. Epidemiological studies over the past 20 years have identified these three antigenic variants in the dog population. However, the first CPV-2 (original CPV-2) has virtually disappeared in the dog population. Studies have suggested that there is no difference among the antigenic variants in terms of pathogenicity [12]. However, the licensed CPV2 vaccine is thought to be less effective in protecting dogs against CPV-2c [13]. According to a report by Doan et al. [14], there are two major epitope regions recognized by B cells (B-cell epitope A and B-cell epitope B) in VP2 of CPV-2. B-cell epitope B includes amino acids 87, 101, 297, 300, 305, and 426, which are located at the tops of the threefold spikes of VP2 and are involved in the receptor binding, host tropism, and the neutralizing response.
An epidemiological study by Soma et al. [15] demonstrated that CPV-2b was the dominant antigenic variant among dogs in Japan between 2009 and 2011. However, a recent study demonstrated that other antigenic variants are more common in other Asian countries that are geographically close to Japan. For example, the dominant antigenic variant among dogs in Taiwan has shifted from CPV-2b to CPV-2c [16,17]. Similarly, an epidemiological study in Laos and Vietnam demonstrated that over 90% of infected dogs had the CPV-2c variant [18,19]. CPV-2c was identified in various countries in Asia except in Japan [16][17][18][19][20][21][22][23]. Collectively, these studies suggest that the composition of antigenic variants in Japan differs from that in neighboring countries. However, the data about CPV-2 antigenic variants in dogs in Japan has not been updated in the last 10 years. Therefore, the distribution of CPV-2 antigenic variants in Japan may have shifted in recent years. In the present study, we analyzed samples collected between 2014 and 2019 to identify antigenic variants of CPV-2 among dogs in Japan. A total of 206 fecal samples were collected from dogs with diarrhea symptoms between 2014 and 2019. These samples were sent to Marupi Lifetech Co., Ltd., (Osaka, Japan) by veterinary clinics in Japan. Fecal samples were stored at 4℃, and fecal homogenates in PBS were prepared within 2 days. These homogenates were stored at -45℃ until DNA extraction. Viral DNA was extracted from the samples using a QIAamp DNA Mini Kit (QIAGEN, Valencia, CA, USA) following the manufacturer's instructions. PCR to diagnose CPV-2 infection using the primer pair (Pabs-Pabas) was performed as described by Senda et al. [24]. The fulllength VP2 gene sequence was generated using four overlapping PCR amplicons obtained using four sets of primers, the sequences of which are shown in Supplementary Table S1. The reaction conditions for PCR were set according to the published protocols for each primer [9,24,25]. The reaction conditions for the primer pair 2304F-3148R were as follows: DNA was denatured at 95°C for 5 min, followed by 40 cycles of denaturation at 95°C for 1 min, primer annealing at 52°C for 1 min, and 72°C for 1 min, with a final extension at 72°C for 10 min. Nucleotide sequences were determined by a commercial service (Fasmac Co., Ltd., Kanagawa, Japan). The following samples were excluded from this study: 1) 43 samples in which the CPV2 gene was not detected using the F4-R4 and 555F-555R primer pairs (the antigenic variant strain and vaccine strain were indistinguishable) and 2) five samples that contained the vaccine strain of CPV2 based on its amino acid sequence (only the original CPV-2 is used as a vaccine in Japan). Thus, 158 samples were used to identify the antigenic variants of CPV-2 based on residue 426 of VP2. Our results demonstrated that 12 (7.6%) of the samples contained variant CPV-2a, 146 (92.4%) contained CPV-2b, and 0 (0.0%) contained CPV-2c. Supplementary Table S2 summarizes the information for all 158 samples with Gen-Bank accession numbers.
Phylogenetic analysis was performed on the amino acid sequence of CPV-2 VP2. Multiple sequence alignment of the translated amino acid sequences was performed using MAFFT (version 7.475). The aligned VP2 sequences were used for construction of a phylogenetic tree using the maximum-likelihood method based on the Jones-Taylor-Thornton (JTT) model in MEGA X [26]. Phylogenetic analysis of the full VP2 gene revealed that CPV-2a can be categorized as either East Asian CPV-2a isolates or European/American CPV-2a isolates, except for three samples (Figs. 1 and  2). CPV2-7-VP2 could not be categorized into any of the groups. All of the CPV-2b isolates from Japan belonged to the same major group.
Additionally, Japan was divided into two regions (Western and Eastern Japan) for the purpose of the analysis to determine the proportion of the antigenic variants in the two regions (Supplementary Table S3). In both regions, over 90% of all samples contained CPV-2b. Age, sex, and vaccination status were not associated with the antigenic variant.
CPV-2 infection is very common in dogs throughout the world. CPV-2 is categorized into three antigenic variants: CPV-2a, CPV-2b, and CPV-2c. These variants have different mutation frequencies and are found across the world [8][9][10]. The genomic substitution rate of CPV-2 is approximately 1 × 10 -4 substitutions per site per year [27]. Thus, although CPV-2 is a DNA virus, the rate at which variants are produced is similar to that of an RNA virus. Continued mutation of the CPV-2 genome may result in the emergence of a CPV-2 antigenic variant for which the current vaccine is no longer effective. Currently, none of the known CPV-2 antigenic variants has a negative impact on the vaccine efficacy. However, given the genetic diversity of CPV-2, it is important to continue tracking the presence of CPV-2 antigenic variants in dogs.
Researchers in a number of countries have performed phylogenetic analysis of CPV-2 based on the complete sequence of VP2. VP2 is the major structural protein that determines the antigenicity of CPV-2 [9,[28][29][30]. In Japan, phylogenetic analysis of CPV-2 based on the complete sequence of VP2 has not been performed for over 10 years [31]. We analyzed the complete sequence of VP2 to provide information about CPV-2 antigenic variants found in Japan.
Our sequence analysis demonstrated that CPV-2b was the dominant variant in Japan. Among 146 samples containing CPV-2b, 116 had the same complete sequence for VP2. CPV-2c strains are currently circulating in East and South East Asian countries [17][18][19][20][21][22][23]. However, we demonstrated that none of the dogs in Japan were infected with CPV-2c. To our knowledge, among the Asian countries that conduct regular epidemiological studies of CPV-2, Japan remains the only country in which CPV-2c has not been detected. This could be attributed to various possible factors: 1) Import of CPV-2c-infected dogs to Japan is geographically challenging, as Japan is isolated from countries where CPV-2c is prevalent. 2) It is unlikely that dogs that enter Japan are infected with CPV-2c, as it is strongly recommended that they be vaccinated. Interestingly, CPV-2c has not been detected in New Zealand, which is an island country like Japan [32]. CPV-2c has been identified in wild carnivores such as leopard cats and Asian palm civets [33,34]. Thus, it is possible that dogs in countries where CPV-2c is prevalent may frequently come in contact with these wild carnivores. The CPV-2 vaccine efficacy tends to be lower against CPV-2c than against other antigenic variants. Thus, epidemiological studies should be conducted regularly to ensure that CPV-2c outbreaks do not occur in Japan.
Our findings, as well as those of a previous study by Soma et al. [15], indicate that, in contrast to other countries, CPV-2 has been the dominant antigenic variant of in Japan CPV-2b for approximately 20 years, since 2000. Furthermore, the VP2 of the dominant CPV-2b strains identified in the present study had the same amino acid sequence as the CPV-2b strain identified in 2003 in Japan (accession no. AB128923; Fig. 2). This suggests that the dominant CPV-2b strain in Japan has undergone little to no antigenic drift for over 17 years and that the vaccine used in Japan, which is based on the original CPV-2 sequence, may not be effective against this strain. Moreover, most dogs infected with CPV-2a and CPV-2b have been vaccinated with the original CPV-2. Thus, it is important to determine whether dogs that have been vaccinated for CPV-2 have sufficient neutralizing antibodies against CPV-2a and CPV-2b and to consider the presence of maternally derived antibodies at the time of vaccination.
Seventeen CPV-2b sequences had several mutations compared with the amino acid sequence of the dominant CPV-2b. None of these viruses were found to be mutated in the amino acids located at the tops of the threefold spike domain of VP2. On the other hand, the amino acid mutations found in nine CPV-2b sequences were located in a region (loop 1, aa 84-89; loop 2, aa 216-235; loop 4, aa 409-444) that affects the shape of the threefold spike domain of VP2. It is necessary to investigate whether the amino acid mutations found in these viruses affect the efficacy of CPV2 vaccines in Japan.
Recent studies in Australia have indicated that the dominant antigenic variant is shifting from CPV-2a to CPV-2b [35]. While it remains unclear whether CPV-2b is becoming dominant, it is important to continue monitoring the data from Australia.
In the present study, we analyzed samples collected from 2014 to 2019 in Japan and demonstrated that CPV-2b was the dominant antigenic variant. CPV-2 infections remain one of the most common infections among dogs in Japan. In order to eliminate CPV-2 infections, a number of steps need to be taken, including improved vaccination, development of an appropriate vaccination protocol, and appropriate management of dogs. Moreover, in order to prevent CPV-2c outbreaks in Japan, it is important to ensure that dogs and cats that are imported to Japan have received vaccination against CPV-2 and to identify possible CPV-2 infections among wild carnivores. Epidemiological studies are important and should be performed regularly to prevent novel CPV-2 epidemics.

Conflict of interest
The authors declare that there is no conflict of interest.
Ethical approval This research did not use human participants or other animals.