Genetic Diversity of Porcine Circovirus Isolated from Korean Wild Boars

In Korea, three genotypes of porcine circovirus type 2 (PCV2a, PCV2b, and PCV2d) have been identified on domestic pig farms, while two genotypes (PCV2a and PCV2b) have been identified in wild boar populations. Here, we investigated genotype diversity and genotypic shift in 91 PCV2 isolates from 1340 wild boars captured in South Korea between 2013 and 2017. Phylogenetic analyses based on the complete ORF2 showed that the 91 PCV2 strains were detected as four genotypes by qPCR screening assay: PCV2a (2.2%, 2/91), PCV2b (16.5%, 15/91), PCV2d (80.2%, 73/91), and PCV2h (1.1%, 1/91). Only one intergenotype recombinant event was detected between PCV2 ORF2 in wild boars (PCV2b) and domestic pigs (PCV2a). Amino acid positions 86–89 within ORF2, which distinguishes the different genotypes, were conserved in all PCV2 genotypes isolated from South Korean wild boars, including TNKI in PCV2a/PCV2h, SNPR in PCV2b, and SNPL in PCV2d. The estimated nucleotide substitution rates in the ORF2 region of viruses from South Korean wild boars and domestic pigs were 5.8145 × 10−4 and 4.5838 × 10−4 substitutions per site per year (s/s/y), respectively. The times to the most recent common ancestor (tMRCA) for South Korean domestic pig PCV2 were 1937 (PCV2a), 1972 (PCV2b), 1999 (PCV2d-1), and 2000 (PCV2d-2). By contrast, the tMRCA for South Korean wild boar PCV2b and PCV2d were 1989 and 2001, respectively. Thus, the PCV2d genotype is prevalent among South Korean wild boars and domestic pigs.


Rates of Nucleotide Substitution per Site per Year and the Time to the Most Recent Common Ancestor
We used a Bayesian MCMC approach to estimate the rate of nucleotide substitution per site per year (s/s/y) and the time to the most recent common ancestor (tMRCA) for 264 PCV2 strains. Of the 264 PCV2 strains examined, 91 were isolated from South Korean wild boars in this study, 19 from South Korean wild boars captured between 2010 and 2012, and 109 from South Korean domestic pigs (1999-2017; data for 2010-2012 (wild boars) and 1999-2017 (domestic pigs) were obtained from GenBank). The estimated nucleotide substitution rates in the ORF2 region of PCV2 from South Korean wild boars and domestic pigs were 5.8145 × 10 −4 (s/s/y) and 4.5838 × 10 −4 (s/s/y), respectively (  Figure 3).

Rates of Nucleotide Substitution per Site per Year and the Time to the Most Recent Common Ancestor
We used a Bayesian MCMC approach to estimate the rate of nucleotide substitution per site per year (s/s/y) and the time to the most recent common ancestor (tMRCA) for 264 PCV2 strains. Of the 264 PCV2 strains examined, 91 were isolated from South Korean wild boars in this study, 19 from South Korean wild boars captured between 2010 and 2012, and 109 from South Korean domestic pigs (1999-2017; data for 2010-2012 (wild boars) and 1999-2017 (domestic pigs) were obtained from GenBank). The estimated nucleotide substitution rates in the ORF2 region of PCV2 from South Korean wild boars and domestic pigs were 5.8145 × 10 −4 (s/s/y) and 4.5838 × 10 −4 (s/s/y), respectively ( Table 2).      , with a 10% burn-in. CN5130 strain (blue color) was contained in PCV2h group with BGO-1, GXYL1208, and GXLZ1208a strains. The recombination of G2284 strain (red color) was found between the major parent (03R955 strain) and the minor parent (CB5006) strain (black color) by the RDP 4 program. PCV2 strains detected from South Korean wild boars and South Korean domestic pigs are marked with gray and pink background colors, respectively. Node labels indicate branch times. with a 10% burn-in. CN5130 strain (blue color) was contained in PCV2h group with BGO-1, GXYL1208, and GXLZ1208a strains. The recombination of G2284 strain (red color) was found between the major parent (03R955 strain) and the minor parent (CB5006) strain (black color) by the RDP 4 program. PCV2 strains detected from South Korean wild boars and South Korean domestic pigs are marked with gray and pink background colors, respectively. Node labels indicate branch times.

Bayesian Skyline Plot Analysis
The phylodynamics of PCV2 isolated from South Korean wild boars and domestic pigs were estimated using a Bayesian skyline plot (BSP) based on the ORF2 (capsid protein) gene. The effective population size (median) of South Korean wild boar PCV2 strains was stable (between 40

Bayesian Skyline Plot Analysis
The phylodynamics of PCV2 isolated from South Korean wild boars and domestic pigs were estimated using a Bayesian skyline plot (BSP) based on the ORF2 (capsid protein) gene. The effective population size (median) of South Korean wild boar PCV2 strains was stable (between 40

Amino Acid Sequence Analysis of the PCV2 ORF2 Protein from South Korean Wild Boars and Domestic Pigs
The nucleotide and amino acid sequences of PCV2 ORF2 strains aligned using the CLUSTAL X alignment program [39]. The nucleotide (nt) and amino acid (aa) sequence identity between each genotype isolated from South Korean wild boars was 89.6-99.7% and 89.4-100%, respectively, whereas that between strains from South Korean domestic pigs was 88.8-99.4% and 86.8-99.6%, respectively. A commercial PCV2 vaccine based on the PCV2a strain was manufactured. Among PCV2 strains isolated from South Korean wild boars, the amino acid sequences at position 86-89, which is thought to distinguish between PCV2a and PCV2b, were 86-TNKI-89 (n = 3) for PCV2a and PCV2h; 86-SNPR-89 (n = 31) for PCV2b and 86-SNPL-89 (n = 76) for PCV2d (Table 3). AA residues important for antibody recognition (residues 173-175 and 179) were conserved (173-YFQ-175 and 179-K) in all genotypes isolated from South Korean wild boars and domestic pigs (Table 3)

Amino Acid Sequence Analysis of the PCV2 ORF2 Protein from South Korean Wild Boars and Domestic Pigs
The nucleotide and amino acid sequences of PCV2 ORF2 strains aligned using the CLUSTAL X alignment program [39]. The nucleotide (nt) and amino acid (aa) sequence identity between each genotype isolated from South Korean wild boars was 89.6-99.7% and 89.4-100%, respectively, whereas that between strains from South Korean domestic pigs was 88.8-99.4% and 86.8-99.6%, respectively. A commercial PCV2 vaccine based on the PCV2a strain was manufactured. Among PCV2 strains isolated from South Korean wild boars, the amino acid sequences at position 86-89, which is thought to distinguish between PCV2a and PCV2b, were 86-TNKI-89 (n = 3) for PCV2a and PCV2h; 86-SNPR-89 (n = 31) for PCV2b and 86-SNPL-89 (n = 76) for PCV2d (Table 3). AA residues important for antibody recognition (residues 173-175 and 179) were conserved (173-YFQ-175 and 179-K) in all genotypes isolated from South Korean wild boars and domestic pigs (Table 3) (Table 3).

Recombination Analysis Using Recombination Detection Program
One intergenotype recombinant was detected in ORF2 gene between the PCV2a of domestic pigs and PCV2b of wild boars. G2284 strain detected from South Korean domestic pigs in 2010 was selected as recombinant and the breakpoint of recombination was determined by MaxiChi, Chimaera, SIScan, and 3Seq methods (p < 0.01). The beginning and ending breakpoints probability of recombination were determined for positions 4 and 339 in alignment. The major parent was the 03R955 strain detected from South Korean domestic pigs in 2003 and the minor parent was the CB5006 strain detected from South Korean wild boars in Chungbuk (CB) region in 2014. No intragenotype recombinant event was detected in wild boars and in domestic pigs.

Discussion
Previous studies suggest that PCV2 circulates at a high rate among wild boar populations in Europe; indeed, it is speculated that these animals might serve as important reservoirs for PCV2 [31,33,34]. In particular, genotype PCV2d-2, which infects both domestic pigs and wild boars, was reported in Italy [35]. PCV2a and PCV2b genotypes were circulating within wild boars in Romania [40] and PCV2b was also detected in wild boars in Uruguay [38]. Our results showed that PCV2d has been detected in wild boars in mountains nationwide in South Korea; since 2015, the main genotypes of PCV2 that have been detected are PCV2d and PCV2b (minor levels), but no PCV2a has been detected. PCV2a, PCV2b, and PCV2d were first isolated from South Korean wild boars in 2013, 2010, and 2011, respectively. Interestingly, the CN5130 strain detected in Korean wild boars in 2014 was contained within the PCV2h group in Vietnam (BGO-1 strain) and China (GXYL1208, and GXLZ1208a strains).
A BSP showed that the effective population size (median) of PCV2 in South Korean wild boars was stable (between 40.1 and 70.3) from 2001 to 2017. Over the past 10 years, the number of wild boars in South Korea has been estimated to be about 350,000 (250,000-450,000); the lack of rapid population increase suggests that the PCV2 levels have remained constant. However, the increase in the effective population size (201.7-305.4) of South Korean domestic pigs from 1999 to 2008 led to sharp increases in the incidence of PMWS and PCVAD [29].
ORF2-based intra-and intergenotype recombination in wild boar populations and the possible recombination between PCV2 strains of wild boars and domestic pigs were detected 5 cases [33]. In Uruguay, PCV2 recombinant strain circulation revealed a predominance, suggesting that PCV2 recombination can lead to the emergence of strains able to complete and potentially displace parental ones [34]. The different patterns of natural intergenotype recombination found between PCV2 parental strains (PCV2a and PCV2b) in China [13,43,44], USA [45], and South Korea [46]. We first reported the recombination of intergenotypes (PCV2a and PCV2b) between wild boars and domestic pigs in South Korea. However, there seems to be a low probability of current intragenotype and intergenotype recombination within wild boars.
AA residues 86-89 within the ORF2 region of PCV2 are thought to distinguish between PCV2a and PCV2b [47]. In South Korean wild boars, aa positions 86-89 were conserved for TNKI (PCV2a and PCV2h), SNPR (PCV2b), and SNPL (PCV2d). However, this does not hold true in domestic pigs. This may be because the PCV2 capsid protein has been subjected to different selective pressures that act on the different genotypes [14]. Residues 190,191,206, and 210 within the capsid protein are crucial for PCV2 replication in vitro [47]. These residues in South Korean PCV2a and PCV2h wild boars were SRID (n = 3), while those in PCV2d were TGID (n = 75) and SAID (n = 1). The residues in most PCV2b isolates from South Korean wild boars were AGIE (n = 24); however, changes in residues of PCV2b were noted in seven strains (TGIE (n = 5) and AGKE (n = 2)), which may have an effect on replication. Residues 173-175 and 179 are crucial for antibody recognition [48]; we found that the aa values in these positions in PCV2a, PCV2b, and PCV2d were conserved (YFQ and K) in South Korean wild boars and domestic pigs. This is consistent with the finding that aa 169-180 of the capsid protein are highly conserved among all PCV2 genotypes [49].
In conclusion, PCV2 strains isolated from South Korean wild boars have four genotypes; of these, PCV2d strains appeared mainly to be from 2015. Each genotype (PCV2a, PCV2b, and PCV2d) in South Korean wild boars seems to have evolved in a stable and independent manner, whereas some strains might show unusual changes, such as PCV2h. The data suggest that the PCV2d genotype will continue to circulate in wild boars.

Sample Collection and qPCR
A total of 1340 whole blood samples were collected from captive wild boars in 9 provinces of South Korea from 2013 to 2017. Total DNA was extracted directly from 250 µL of each blood sample using a DNeasy Blood and Tissue kit (Qiagen Inc., CA, USA) following the manufacturer's instruction and eluted with 200 µL of elution buffer. For PCV2 detection in extracted viral DNA samples, quantitative real-time PCR (qPCR) was performed on an C1000 thermocycler (BIO-RAD, CA, USA) using the TaqMan-based VDx PCV2 qPCR kit (MEDIAN Diagnostics Inc., South Korea) according to the manufacturer's instructions. Briefly, amplification was conducted in a 20 µL reaction mixture containing 5 µL extracted DNA and 15 µL TaqMan reaction mixture. The thermal profile for the amplification was 95 for 5 min, followed by 40 cycles of 95 for 10 s and 60 for 40 s. Analysis of each data sample was performed with Bio-Rad CFX Manager software version 3.1 (CA, USA). The cut-off CT value was 35, with values less than 35 considered PCV2 positive. PCV2 positive control DNA was supplied by the manufacturer. The entire ORF2 gene of PCV2 was performed the PCR amplification as describe previously study [50]. All PCR reaction was performed using the AccuPower ® ProFi Taq PCR PreMix kit (Bioneer Inc., Daejeon city, South Korea). All purified PCR products were sequenced using the sequencing service of the professional molecular analysis company (CosmoGENTECH Inc., Seoul, South Korea).

Sequence Analysis of the ORF2 Protein
The complete ORF2 nucleotide sequences of 91 PCV2 strains detected from wild boars were submitted to the GenBank under accession numbers MT376295-MT376380 and MT501728-MT501732.

Phylodynamic and Phylogeographic Analysis
A BEAST input file was generated using BEAUti within the BEAST package v.1.8.1 [51]. The rate of nucleotide substitutions per site per year and the time from the most recent common ancestor (tMRCA) were estimated using a Bayesian MCMC approach. Each dataset was simulated using the following options: generation = 100,000,000; 10% burn-in; and ESSs > 200. The exponential clock and expansion growth population model in the BEAST program was used to obtain the best-fit evolutionary model, while the maximum clade credibility (MCC) tree was visualized using Figtree 1.4 [52].

Recombination Analysis
The recombination between PCV2 from South Korean wild boars and PCV2 from South Korean domestic pigs was analyzed by the recombination detection program (RDP4 version 4.22) [53]. Briefly, this program identifies recombinants from a group of aligned DNA sequences using a number of recombination detection and analysis algorithms. The aligned sequences were evaluated by the RDP, GeneConv, SiScan, MaxChi, BootScan, Chimera, and 3Seq methods in RDP version 4.22, while the highest acceptable result was obtained when the p-value was set to 0.01. We have used 264 PCV2 strains for RDP4 analysis. Specifically, we used 91 strains identified from wild boars in this study (2 PCV2a,15 PCV2b, 73 PCV2d, and 1 PCV2h) and 172 other PCV2 strains obtained from NCBI Genbank database (27 PCV2a, 72 PCV2b, 1 PCV2c, 11 PCV2d-1, 53 PCV2d-2, 1 PCV2e, 3 PCV2f, 2 PCV2g, and 3 PCV2h). Detailed information of all PCV2 strains were indicated in Figure 2.

Ethics Approval
The authors confirm that the work complies with the ethical policies of the journal.