Dynamic evolution of canine parvovirus in Thailand.

Background and Aim: According to the previous study, the circulating canine parvovirus (CPV) in Thailand is 2a and 2b. Nowadays, CPV mutants, including CPV-2c, have been identified in many parts of the world. This study aimed to investigate the genetic diversity of the circulating CPV in Thailand. Materials and Methods: Eighty-five CPV-positive fecal samples were obtained from dogs with either acute hemorrhagic diarrhea or diarrhea. The complete VP2 gene of these samples was amplified using VP2 specific primers and polymerase chain reaction (PCR). The obtained full-length VP2 sequences were analyzed and a phylogenetic tree was constructed. Results: Sixty and 25 CPV-positive fecal samples were collected in 2010 and 2018, respectively. Thirty-four samples were new CPV-2a and 31 samples were new CPV-2b due to amino acids substitution at position 297 (Ser-Ala). In 2018, 5 new CPV-2a, 19 CPV-2c, and 1 feline panleukopenia virus (FPV) were found, but no new CPV-2b was detected. Moreover, most of the CPV in this study had amino acids mutations at positions 324 and 440. The phylogenetic construction demonstrated the close relationship between the current new CPV-2a with the previous CPV-2a reported from Thailand, China, Uruguay, Vietnam, Singapore, and India. Interestingly, the current new CPV-2b in this study was not closely related to the previous CPV-2b reported in Thailand. The CPV-2c in this study was closer to Asian CPV-2c and further from either European or South America CPV-2c. Interestingly, FPV was identified in a diarrhea dog. Conclusion: The evolution of CPV in Thailand is very dynamic. Thus, it is important to monitor for CPV mutants and especially the clinical signs relating to these mutants to conduct surveillance for the emergence of new highly pathogenic CPV in the future.


Introduction
Canine parvovirus (CPV) is one of the most common viruses in domestic dogs. It causes acute hemorrhagic gastroenteritis, leukopenia, nausea, diarrhea, and sometimes fatal myocarditis in young puppies [1]. CPV belongs to the family Parvoviridae, subfamily Parvovirinae, and genus Parvovirus. It is a non-enveloped, icosahedral, linearized, and single-stranded DNA virus. The genome of CPV is approximately 5.2 kb in length. The virus encodes two nonstructural proteins (NS1 and NS2) and three structural proteins (VP1, VP2, and VP3). The VP2 capsid protein is the main capsid protein and plays an important role in the determination of the antigenicity and host range of CPV [2]. CPV type 2 (CPV-2) was first identified in the USA in 1978 and was found to have spread worldwide in domestic and wild canid populations [3]. CPV is genetically close to feline panleukopenia virus (FPV); however, CPV has at least seven amino acid differences from FPV which determine the canine or feline host range, such as amino acid positions 80 (Lys-Arg), 93 (Lys-Asn), 103 (Val-Ala), 232 (Val-Ile), 323 (Asp-Asn), 564 (Asn-Ser), and 568 (Ala-Gly) [4].
Molecular surveillance may be used as a tool for the detection of the new mutant, prediction of disease severity and providing the important data for the development of the better vaccine or the better diagnostic test in the future. Moreover, the knowledge of the current genotype of the CPV in Thailand is limited. This study aimed to investigate the current genotype of CPV circulating in Thailand and to determine the existence of CPV-2c and other CPV strains.

Ethical approval
This study was approved by the Animal Ethics Committee of the Faculty of Veterinary Medicine, Kasetsart University, Thailand (ACKU62-VET-007).

Samples
Eighty-five fecal samples were used in this study based on a positive result for CPV according to routine polymerase chain reaction (PCR) testing. In 2010 and 2018, 60 and 25 positive samples, respectively, were collected (Table-1). These fecal samples were collected from dogs that displayed either acute hemorrhagic diarrhea or diarrhea and nausea at the Veterinary Teaching Hospital, Kasetsart University, Rattanatibeth Referral Animal Hospital, Bangkok, Thailand, and the Amphawa Pet Hospital, Samut Songkhram, Thailand. These dogs were either vaccinated or unvaccinated and were aged from 1 month to 5 years. The fecal samples were stored at −80°C until used for DNA extraction.

DNA extraction and PCR
DNA was extracted from the fecal samples using the acid guanidinium thiocyanate-phenol-chloroform extraction method. A set of primers was designed to amplify the whole VP2 gene: F (5'-ATG AGT GAT GGA GCA GTT CA) and R (5 '-TTA ATA TAA  TTT TCT AGG TGC TAG TTG). The PCR mixture (25 µl) was composed of 1× buffer (20 mM Tris-HCl [pH 8.4], 50 mM KCl 2 ), 0.2 mM dNTPs, 2.5 mM MgCl 2 , 100 pmol of each of the forward and reverse primers, 1 unit Taq DNA polymerase (Invitrogen, Carlsbad, CA, USA), and 2.5 µl of DNA template to give a total volume of 25 µl. After an initial denaturing at 94°C for 5 min, the amplification was performed using 35 cycles at 94°C for 40 s, annealing at 50°C for 40 s, and extension at 72°C for 90 s, and a final extension at 72°C for 10 min. The expected PCR products were 1755 bps in size. The PCR products were analyzed using 1% agarose gel electrophoresis at 100 V for 30 min and visualized under ultraviolet illumination. The PCR products were purified using an UltraClean ® 15 DNA Purification Kit (MO BIO Laboratories, Inc., Carlsbad, CA, USA) and cloned into plasmid pGEM-T Easy (Promega Corporation, Madison, WI, USA). The sequences of the cloned full-length VP2 were determined at First BASE Laboratories Sdn Bhd, Selangor, Malaysia.

Analysis and phylogenetic construction of full-length VP2 gene of CPVs
The nucleotide sequences were translated into amino acid and multiple alignments of the amino acid sequences using the Bioedit biological sequence alignment editor computer package (version 7.1.3; Ibis Biosciences, Carlsbad, CA, USA). The phylogenetic analysis was constructed from the amino acid sequences of all 85 samples in this study and other full-length VP2 sequences obtained from the GenBank database (Table-2) using the MEGA program (version 7.0, The Biodesign Institute, Tempe, AZ, USA) with the neighbor-joining method and running 1000 replicates in the bootstrap. Bayesian phylogenetic analysis was also performed for more extensive amino acids analysis to analyze the selective pressures on certain amino acids using mixed model analysis. The phylogenetic tree was created by MrBayes version 3.2.6 (https://nbisweden.github.io/ MrBayes/download.html) [17]. The tree was viewed using FigTree software version 1.4.3 (http://tree.bio. ed.ac.uk/software/figtree/).

Results
Thirty-nine and nine samples out of 85 positive samples were from non-vaccinated and vaccinated dogs, respectively. The other 37 positive samples did not have a history of vaccination. The youngest CPV-infected dog was aged 21 days, and the oldest was 3 years. The youngest age of positive vaccinated dogs was 2 months; however, the individual histories of booster vaccination for these positive, vaccinated dogs were not available (Table-1).
For the 2010 data, the amino acid sequences analysis revealed that 29 samples were new CPV-2a and 31 samples were new CPV-2b due to amino acids substitution at position 297 (Ser-Ala) ( Table-1). The number of positive samples for new CPV-2a and new CPV-2b was nearly equal in 2010. For 2018, 5 new CPV-2a, 19 CPV-2c, and 1 FPV were found, but CPV-2b was not detected (Table-3   440 (Thr-Ala) substitution due to an A-to-G transition at nucleotide 1318 (Table-1 Figures-2 and 3), but distanced from the CPV-2c in Europe and South America. Interestingly, one dog with clinical signs of diarrhea was positive for FPV. The six amino acids (at 80, 103, 232, 323, 564, and 568) out of the seven amino acids which determine the canine or feline host range were similar to the reference FPV; however, there was an amino acid substitution at position 93 (Lys-Asn). According to this amino acid substitution, FPV in this study was in the different clades compared to the reference FPV (Figures-2 and 3).

Discussion
The number of samples positive for new CPV-2a and 2b circulating in central Thailand in 2010 was approximately equal. This result contrasted with a study in Thailand (data collected between 2003 and 2008) in which the predominant genotype in central Thailand was CPV-2a [11]. Interestingly, the majority CPV genotype in 2018 was CPV-2c, but new CPV-2b was not found. This study showed that CPV-2c had become the predominant genotype during the year 2018 in Thailand. Besides of t he diminishing of CPV-2b, there was only one amino acid difference at position 426

Year of collection
Moreover, most of the current new CPV-2a had alanine at amino acid position 440; however, most of the new CPV-2b still had threonine. Amino acid substitution at position 440 (Thr-Ala) was also reported in the previous studies in China [19], India [21,22], Nigeria [24,31], Pakistan [32], South Africa [33], South Korea [25], and Uruguay [26]. Amino acid position 440 is also important because it is located at the top of the three-fold spike, the main antigenic site of the virus [34,35]. Interestingly, this mutation was not detected in a previous study in Thailand [11]. In the current study, two new CPV-2a ( (phenylalanine, tyrosine, aspartic acid, and threonine, respectively), as reported in the previous study in Thailand [11]. However, new CPV-2b in this study had isoleucine at amino acid 324. Thus, CPV-VT56 might also represent the transition evolution of the original new CPV-2b in the current new CPV-2b.
The CPV-2c in this study was similar to CPV-2c in Asia, such as in China [36], Laos [14], and Taiwan [15,37] due to amino acid substitution at amino acid positions 267 (Tyr), 324 (Ile), and 370 (Arg). CPV-2c in this study differs from CPV-2c in Europe and South America at three amino acid positions (Phe267Tyr, Tyr324Ile, and Gln370Arg). The amino acid substitution at position 440 (Thr-Ala) has been found in Argentina, but this change was not found in this study [12]. Interestingly, FPV was found in a dog that had clinical signs of diarrhea. This FPV sample had an amino acid substitution at position 93 (Lys-Asn). The amino acid at position 93 is important because it is one of the amino acids that determine the canine host range [38]. FPV infection in dogs has also been reported in Pakistan [39]. These findings demonstrated that parvoviruses in Thailand have been dynamically evolving as those in the other part of the world [40,41]. This mutation rate is as high as seen in RNA viruses [41]. The rapid mutation of CPV has resulted in growing concern about the effectiveness of vaccines regarding the new mutant or genotype of CPV. Molecular surveillance of CPV is crucial for the prediction of disease severity and may be important for the development of more effective vaccines or diagnostic tests in the future.

Conclusion
Two genotypes of CPV (new CPV-2a and CPV-2c) are circulating in Central Thailand and the predominant circulating genotype of CPV has been changed from CPV-2a in the past to CPV-2c at present. Currently, CPV-2b has not been found in Central Thailand. The current new CPV-2a circulating in Thailand has amino acid substitutions at positions 324 (Tyr-Ile) and 440 (Thr-Ala). FPV was found in a dog that had acute diarrhea; however, the importance of this finding remains to be determined. Our results provided additional information on the dynamic evolution of CPV in Thailand, which is following the same evolutionary trend observed in the others part of the world.

Authors' Contributions
NI and TS designed the experiment and made DNA extraction, PCR, multiple alignment, and phylogenetic study. NM, SK, KS, and TS were involved in scientific discussion and provided suggestions for the overall work. All authors read and approved the final manuscript.