Detection of PCV3 in various sample types from wild boars in Poland

Abstract

Background

In 2016, the new circovirus, PCV3, was discovered in pigs displaying various clinical symptoms. To date, this virus is believed to be pathogenic for pigs and considered equal or greater economic significance for the porcine sector compared to PCV2. Besides pigs, other species, including wild boars, have been identified as susceptible to PCV3 infection. Infection in wild boars is not linked with any clinical outcome. However, the data regarding the detailed pathogenesis of PCV3 infection in wild boars contain some knowledge gaps. Available data indicate that PCV3 is highly prevalent in wild boars, and this species can represent a potential reservoir of the virus for pigs. Therefore, this research aimed to assess the prevalence of PCV3 in different samples collected from wild boars.

Results

154 of 285 (54%) sampled wild boars tested positive for PCV3 DNA presence. There were no statistically significant differences between the proportion of positive individuals culled in different regions of the country. PCV3 genetic material was detected in every tested sample type, and the sample type significantly influenced the detection rate (p < 0.05), with the highest and lowest detection rates in liver (47.02%) and feces (2.5%). No statistically significant differences (p > 0.05) were found in the detection rate of different organs between animals of various age categories. The highest viral load expressed as Ct value was determined in the liver, however, it was not significantly different from that noted for serum (p > 0.05). The obtained genome sequences were similar to those previously determined for animals infected in Poland.

Conclusions

PCV3 is highly prevalent in the Polish wild boar population and circulates among animals of different ages. Due to the viral prevalence and the quantity in the liver, this organ can represent the target site for replication and one of the most suitable samples for PCV3 detection in wild boars.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12917-025-05256-z.

Background

Porcine circovirus type 3 (PCV3) was discovered in 2016 in commercial pig farms in the USA by the next-generation sequencing (NGS). It was identified as the third species within the Circovirus genus of the Circoviridae family [1, 2]. Subsequently, its presence has been confirmed in Europe [2, 3], Asia [4], South America [5], and Africa [6]. Several attempts to define genotypes and subgroups within PCV3 have been undertaken, but the proposed nomenclature has often been inconsistent and difficult to reproduce using alternative methods [7–9]. Fux et al. [9] analyzed full genome sequences from German and international strains, suggesting a clear division of PCV3 strains into two main subgroups (a and b) which they proposed might be considered PCV3 genotypes [9]. These groups were defined based on specific amino acid marker patterns in open reading frame 1 (ORF) 1 and ORF2; described as the motif A V K S I for group a, and motifs S A R S I and S A R T L in subgroup a and subgroup b respectively. On the other hand, Li et al. [7] analyzed concatenated coding sequences (ORF1 + ORF2) and identified two stable clades, which they formally named PCV3a and PCV3b, and other authors proposed alternative grouping [7]. To standardize PCV3 classification, rigorous criteria were proposed by Franzo et al. [10] for formally accepting new genotypes: requiring a maximum within-genotype genetic distance of 3% at the complete genome level and 6% at the ORF2 level, alongside strong bootstrap support (higher than 90%) in phylogenetic trees. Based on these stringent criteria, only a single genotype, PCV3a, had been formally proposed at the time of publication of the criteria.

The emergence of PCV3 is considered an outcome of intensive farming development [11]. So far, PCV3 has been detected in pigs displaying numerous clinical disorders and presentations, including porcine dermatitis and nephropathy syndrome (PDNS) as well as a range of reproductive, respiratory, neurological, gastrointestinal or systemic disorders [1, 2, 12]. The genetic material of PCV3 has also been detected in clinically healthy pigs of different ages and countries [3, 12]. Therefore, although the PCV3 is believed to be pathogenic, the precise outcomes of the PCV3 infection remain uncertain. In 2021, Saporiti et al. [13] proposed that the two most common clinical outcomes of PCV3 infections are reproductive disease and systemic disease.

Not only pigs are susceptible to PCV3 infections. Its genetic material has also been confirmed in wild boars (Sus scrofa) without clinical/pathological signs of infection [14]. PCV3 DNA has also been identified in some non-Suidae species (wild ruminants, ticks, dogs, laboratory mice, cattle); however, with low infection rates [15–18]. The detection of PCV3 in wild boars is not surprising, as wild boars and domestic pigs belong to the same species (Sus scrofa) and many infectious agents found in domestic pigs also circulate in the wild boar population [19, 20]. Moreover, wild boars serve as a reservoir for various important pathogens with potential for transmission to other animals and humans [21]. Although infections with PCV3 in wild boars are not currently linked with specific clinical outcomes [14], their role should not be diminished. Previously published data indicate that PCV3 prevalence in wild boars is similar or even higher compared to domestic pigs, especially those kept on commercial farms [22, 23]. The recent findings of Franzo et al. [22] demonstrated the role of wild boars in the maintenance and spread of infection, and suggested that wildlife reservoirs may be responsible for the introduction of new strains into intensive farms, either directly or via rural pigs. In recent years, an increase in the wild boar population has been observed in forested and urban areas in many European countries, including Poland [24]. The increase in the wild boar population enhances the contact opportunities with domestic animals, which, collectively with high viral prevalence, makes wild boars a potential reservoir of PCV3 for pigs [14]. The evidence of the apparent long‐lasting PCV3 infection among wild boars favors this hypothesis [25]. The recent study by Frant et al. [26] confirmed the presence of PCV3 DNA in the wild boar population in Poland. The aforementioned study aimed to determine the prevalence of PCV3 in wild boars and assess the probability of African swine fever virus (ASFV) infection based on the PCV carrier status in free-ranging suids in Poland. The objectives of the current study were to determine, through a molecular survey, the occurrence and tissue predilection of PCV3 by analyzing the presence of PCV3 DNA in sets of four types of samples collected from wild boars of different ages. Further investigation into the epidemiology and genetic diversity of PCV3 strains detected in wild boars in Poland was also the focus of the study.

Results

In total, 154 out of 285 (54%) wild boar were positive for PCV3 DNA in at least one tested sample type (Table S1). Analyzing the results in terms of the frequency of PCV3 DNA-positive individuals in different Regional Directorates of State Forests (RDSFs), positive samples were detected in all examined regions. The percentage of positive individuals was highest in RDSF Szczecin and lowest in RDSF Radom; however, the statistical difference was not significant (Table 1).

Table 1.

The distribution of the analyzed samples in Poland (territory of Poland divided into RDSF)

Regional Directorates of State Forest (RDSF)	Province	Number of tested animals	% of Positive animals*	95% confidence interval (%)
Katowice	Opole Voivodeship, Silesia	30	53.3	34.33–71.66
Olsztyn	Warmia-Masuria	31	48.4	30.15–66.94
Szczecin	Lubusz Voivodeship, West Pomerania	27	77.8	57.74–91.38
Szczecinek	Pomerania, West Pomerania	23	69.6	47.08–86.79
Białystok	Podlasie Voivodeship, Warmia-Masuria	8	62.5	24.49–91.48
Gdańsk	Pomerania, Warmia-Masuria	7	57.1	18.41–90.10
Krosno	Subcarpathia	21	61.9	38.44–81.89
Lublin	Lublin Voivodeship, Subcarpathia	6	50.0	11.81–88.19
Łódź	Lodzkie Voivodeship, Mazovia	19	36.8	16.29–61.64
Piła	Greater Poland	26	34.6	17.21–55.67
Poznań	Greater Poland	15	40.0	16.34–67.71
Radom	Mazovia, Świętokrzyskie Voivodeship	10	30.0	6.67–65.25
Toruń	Kuyavia-Pomerania	21	57.1	34.02–78.18
Kraków	Lesser Poland	10	60.0	26.24–87.84
Warszawa	Mazovia	15	46.7	21.27–73.41
Zielona Góra	Lubusz Voivodeship	16	68.8	41.34–88.98
Total		285	54.0	48.06–59.93

^*The animal was considered positive if genetic material of PCV3 was detected in at least one sample type

Genetic material of PCV3 was detected in all types of tested specimens, however, sample type had a significant effect on the detection rate of positivity (Table S1). The highest frequency of PCV3 DNA was detected in liver samples (47.02%), followed by serum (34.03%), and lungs (20.35%). The lowest viral DNA frequency rate was observed in feces (2.5%). In addition, statistically significant differences in the PCV3 occurrence were observed between all of tested matrices (p < 0.05).

Regarding the age and frequency of PCV3 DNA detection, no statistically significant differences were observed between particular age groups, regardless of specimen type. The aforementioned findings have been summarized in Table S1.

The viral load expressed as Ct values showed a similar trend as was observed for the frequency detection in different types of specimens (p ≤ 0.01) (Fig. 1). The median of Ct value of positive liver samples was 33.64, with a maximum value of 36.93 and a minimum value of 24.97. In the case of positive serum samples, the median of Ct value was 33.98, the maximum value was 36.87, and the minimum value was 27.89. In turn, for positive lung and feces samples, the median of Ct value was 35.1 and 36.72, the maximum value was 37.00 and 36.92, and the minimum value was 28.82 and 36.00, respectively.

Fig. 1.

Box Plot Graph of cycle threshold (Ct) values in different types of specimens of positive samples for PCV3. Vertical lines indicate the range of values, and the median Ct value is represented as a thicker horizontal line within the box plot. The box indicates the 25th and 75th percentiles. An asterisk denotes a statistically significant difference between the test types of specimens. **p ≤ 0.01

The highest viral loads detected in liver samples differed significantly from those observed for lungs and feces (p ≤ 0.01) (Fig. 1). No significant discrepancy was detected between liver samples and serum samples (p > 0.05). The lowest viral loads were found in fecal samples and differed significantly from those determined for serum and liver. A statistically significant difference was also observed between PCV3 loads in serum and lungs (Fig. 1).

Eight complete coding sequences (CDS) of ORF2 and 13 CDS of ORF1 gene obtained in the present study were deposited in GenBank under accession numbers PV541639-PV541659 (Table 2).

Table 2.

GenBank accession numbers for porcine circovirus 3 (PCV3) isolates obtained in the present study (PV541639–PV541659)

Name of isolate	ORF1 sequence (Replication Associate Protein)	ORF2 sequence (Capsid Protein)	Province
22–27	PV541639	Not sequenced	Silesia
22–28	PV541640	Not sequenced	West Pomerania
22–29	PV541641	Not sequenced	West Pomerania
22–30	PV541643	PV541642	West Pomerania
22–31	PV541644	Not sequenced	West Pomerania
22–32	PV541646	PV541645	Greater Poland
22–33	PV541647	Not sequenced	Subcarpathia
22–34	PV541649	PV541648	Subcarpathia
22–35	PV541651	PV541650	Silesia
22–36	PV541653	PV541652	Silesia
22–37	PV541655	PV541654	West Pomerania
22–38	PV541657	PV541656	Silesia
22–39	PV541659	PV541658	West Pomerania

Both the representative subset (61 sequences) and comprehensive validation dataset (2,340 sequences) independently selected the Tamura-Nei model with gamma-distributed rate heterogeneity (TN + F + G4) as the best-fit substitution model (BIC), with similar gamma shape parameters (α = 0.576 and 0.624, respectively). This concordance in model selection validates that both datasets share identical underlying evolutionary patterns. Phylogenetic analyses of both datasets revealed extensive polytomies and limited phylogenetic resolution, characteristic of recently emerged viral populations. The representative subset contained 34 near-zero internal branches (< 0.0015 substitutions/site), representing 59% of all nodes, while the comprehensive dataset exhibited 1,041 near-zero branches. Bootstrap support analysis showed that 31.0% of nodes in the representative subset and 48.0% of nodes in the comprehensive dataset achieved UFBoot support ≥ 70%, with mean UFBoot values of 48.0 ± 31.8% and 54.9 ± 39.3%, respectively. The majority of nodes in both analyses fell below conventional support thresholds, confirming that limited phylogenetic structure reflects genuine biological signal rather than sampling artifacts. The star-like topology with extensive polytomies was consistent across both dataset scales, indicating rapid evolutionary radiation following the virus's emergence. Despite large difference in sequence number, both analyses recovered identical phylogenetic patterns: minimal geographic structuring, high sequence conservation, and low number of well-supported clades.

Phylogenetic analysis revealed four informal genetic groupings (Fig. 2). Group 1 comprised all eight Polish wild boar sequences, clustering with other European wild boar strains from Spain, Russia, and Italy, suggesting regional wildlife circulation patterns. Group 2 consisted predominantly of Asian domestic pig sequences. Groups 3 and 4 showed mixed geographic composition, including early emergence sequences from 2015–2016. However, moderate bootstrap support (46–88%) across main defining nodes preclude formal phylogenetic classification of these groups.

Fig. 2.

Maximum likelihood phylogenetic tree of PCV3 ORF2 sequences constructed using IQ-TREE with TN + F + G4 substitution model. Polish wild boar sequences from this study are indicated with blue circles. Node support values are shown as SH-aLRT/UFBoot percentages. Support values were only displayed for nodes where SH-aLRT ≥ 70% or UFBoot ≥ 60%. All sequences represent PCV3a, the only genotype meeting formal classification criteria [10], reflecting PCV3's highly uniform genetic structure. Informal genetic groupings based on sequence clustering were marked. Observed groupings are provisional and descriptive in nature and should not be interpreted as statistically supported phylogenetic clades or formal genotype designations. Scale bar represents nucleotide substitutions per site.

All the sequences obtained in the present study were highly similar and localized within the same cluster in the phylogenetic tree (Fig. 2). The analysis of ORF1 and ORF2 sequences showed a high level of genetic identity to each other, 98.88–100% and 98.76–100%, respectively. Two sequences collected in the same geographic area (Silesia RDSF), 22–36 and 22–38, exhibited 100% of genetic identity both within ORF1 and ORF2. These two strains shared the same two point mutation change from AGA to AAG mutation within the second ORF2 codon, only detected in one other strain from Korea (MN698820), changing amino acid R2K in the predicted Cap sequence. Other variable positions in ORF2 included nucleotide (nt) position 44 within codon 15 (change R15K in strain 22–35), position 225 within codon 75 (change A75T in strain 22–32), and nt position 292 within codon 98 (change Q98K in strain 22–35).

In general, the obtained sequences exhibited high similarity to previously published sequences derived from infected wild boars (OQ533864-OQ533873) and swine from Poland (ON376246-ON376262). In case of ORF1 region, the percentage of genetic identity to previously reported sequences from wild boars and swine reached 98.22–99.85% and 97.98–99.85%, and for ORF2-99.1–100% and 98.77–100%, respectively. Analysis of previously described marker positions in predicted capsid and replicase protein sequences revealed an amino acid pattern consistent with strains from PCV3 subgroup b1: S in position ORF1-122, A/ORF2-24, R/ORF2-27, S/ORF2-77, and I/ORF2-150 [26]. Although no amino acid changes were detected within conserved regions of ORF1, two synonymous mutations were identified within DDFYGWVP (T663C) and DRYP (G678A) motifs, in strains 22–33 and 22–27, respectively.

Discussion

The available research results indicate that PCV3 is highly prevalent in wild boar populations, and wild Suidae can represent an important reservoir of this virus [14, 22, 23, 25]. The first detection of PCV3 DNA in the samples collected from wild boars was reported by Franzo et al. [14]. A later study by Klaumann et al. [25], however, demonstrated that PCV3 had been circulating in the populations of wild boars since at least 2004. In Germany, PCV3 was detected in two distinct wild boar groups, indicating its presence before the division of Berlin [27]. Until now, PCV3 genetic material has been confirmed in wild boar samples from several countries, including Italy (2018), Spain (2018), Germany (2019), Brazil (2021), Korea (2021), Austria (2022), Japan (2023), and very recently in Poland (2024) and Portugal (2025) [14, 25–32]. Herein, we further explored its circulation in the Polish wild boar population, in the country where PCV3 has been known to be circulating among pigs since at least 2016 [3]. Similar to our results, a few other studies demonstrated the presence of PCV3 in the serum samples of wild boars [14, 23, 25, 29, 33, 34]. In the study of Franzo et al. [14], 62 sera of wild boars tested positive for PCV3, revealing a high viral occurrence (33.15%) that was similar to the one observed in our study (34.03%). However, serum accounted for three-quarters of all the samples tested [14]. On the other hand, in the remaining studies, the prevalence of PCV3 in the sera was higher (42.66% and 66.7%) or lower (10% and 0.09%) when compared to our results [23, 25, 33, 34]. Moreover, the observed prevalence was higher compared to that detected in domestic pigs in Poland (25%), as well as in other European countries, which is in line with some previous studies [3, 12, 14, 23, 32]. A range of biosecurity measures implemented in commercial pig farms in different regions can be responsible for this phenomenon [24].

A comparison with recent Polish data revealed a similar frequency of PCV3 detection (34.03% vs. 37.8%) [26]. In the study by Frant et al. [26], the reported prevalence referred to both serum and organ samples, without distinguishing serum alone. The authors also analyzed whether sampling location affected PCV3 detection rates. They found no significant influence of voivodeships overall, although in one model the Lubusz voivodeship showed a 2.5-fold higher chance of detecting positive animals compared with Podlasie voivodeship. This effect, however, became insignificant when additional regions were included [26].

In our study, the frequency of PCV3-positive samples also varied between RDSFs, ranging from 30% in Radom to 77.8% in Szczecin, with no statistically significant differences (p > 0.05). The Szczecin RDSF covers areas of northwestern Poland, overlapping regions with higher wild boar densities, which may partly explain the higher detection rate. This observation is consistent with the results of Frant et al. [26], who also reported the highest PCV3 frequency in the Lubusz voivodeship [26].

The knowledge regarding the pathogenesis mechanisms of PCV3 infection, target organs, and routes of virus dissemination throughout the body is still limited. It was observed that during both natural and experimental infections, replication of PCV3 and inflammation simultaneously occur in different tissues [35]. The presence of PCV3 DNA in the blood indicates systemic infection. However, the main target organs and cell tropism remain to be elucidated [25, 35]. Apart from serum samples, PCV3 DNA was also detected in other tissues obtained from wild boars, including kidney, liver, lungs, lymph nodes, spleen, tonsils, and feces samples [25]. In pigs, the heart and kidneys were considered important replication sites of PCV3 [35]. It is speculated that in some scenarios, tissue tropism and virus titer can be associated with pathogen virulence [35]. In the present study, PCV3 genetic material was demonstrated in all tested types of specimens, including serum, liver, lungs, and feces, however, with different frequencies of positivity. The highest PCV3 frequency was revealed in the liver (47.02%), followed by serum (34.03%) and lung (20.35%), and the lowest in feces (2.50%). A similar study was previously conducted by Klaumann et al. [25]. Authors analysed the presence of PCV3 genetic material in 9 different sample types (sera, faeces, brain, kidney, liver, lung, submandibular lymph node, spleen, tonsil) [25]. According to the quoted study, the types of samples in which PCV3 DNA was most commonly detected were as follows: lungs (57.14%), liver and spleen (54.29%), and tonsils (42.86%), and suggested that these samples are most useful in the PCV3 detection. Moreover, PCV3 prevalence was higher in all types of tested tissue, compared to serum (17.86%), except feces (9.09%) [25]. In the present study, the occurrence of PCV3 DNA in sera was relatively high and second only to the liver (34.03%). On the other hand, we documented the relatively low overall viral prevalence in the lungs (20.4%), contrary to the previously published results in wild boars and domestic pigs [25, 36]. However, a study conducted in Sardinian wild boars revealed the highest frequency of viral genome in the spleen (75%), followed by blood (66.67%), and lowest in the lung (50%), which partially supports our results [23]. Similarly, the results of a recent study conducted on Brazilian wild boars were partially consistent with our findings, showing a higher viral prevalence in serum compared to lung tissue [34]. Additionally, Tan et al. [36] examined archived lung samples from wild boars for the presence of PCV3 genetic material, alongside samples from pigs. Their results were surprising, as all 28 lung samples from the wild boar population in Peninsular Malaysia gave a negative result for PCV3 DNA. The authors speculated that the low number of tested samples may have influenced the obtained prevalence [36]. The results of our study, along with most of the aforementioned studies, support Klaumann et al.'s [25] theory that testing only serum samples may result in underestimating the PCV3 genetic material occurrence in wild boars. Given that samples from wild boars are predominantly collected post-mortem, it seems reasonable to include internal organs, not just serum, for assessing the occurrence of PCV3. This approach would be a more comprehensive evaluation.

According to the present study, the liver is the most suitable sample type for PCV3 detection in wild boars. Also, viral load expressed as a Ct value reaches the highest level in liver samples. This may potentially imply that the liver is among the most useful tissues for PCV3 detection, and may represent the target organ for its replication. Although the frequency of PCV3 DNA detection in serum was moderate, the Ct value was relatively high without statistically significant differences compared to the liver Ct value. Interestingly, in the Klaumann et al. [25] study, the frequency of viral genome detection rate was highest in lung samples, meanwhile, the highest median viral load level was detected in the submandibular lymph nodes, followed by liver, spleen, kidney, and lung. In the abovementioned study, the lowest median viral load was detected in tonsil samples. The discrepancies between our study and the previously mentioned may result from a different number of samples tested. Also, the age of the sampled wild boars may have affected the obtained results.

To date, previous studies have observed significant differences in the frequency of PCV3 occurrence across age categories. Franzo et al. [14] observed a lower virus prevalence in the sera obtained from the group of animals younger than 12 months of age. Similarly, Klaumann et al. [25] demonstrated that the detection frequency of PCV3 in serum was significantly higher in samples collected from adult individuals compared to subadults or juvenile ones. This statement was further evidenced by Souza et al. [34], as PCV3 was detected in samples collected only from adult free-living wild boars. Auer et al. [30] documented that PCV3 infection was significantly associated with adult age. Another study displayed that the frequency of PCV3-positive wild boars tested via lymph nodes was also higher among adult animals compared to juveniles [31]. Such findings indicate less intensive virus circulation among juvenile wild boars [34]. It can be a result of a low viremia during sampling, possibly due to the protective impact of maternal-derived immunity or a higher infection exposure likelihood among adult wild boars [30, 34]. These results seem in contrast with our observations, in which the higher viral prevalence in the sera was noted among the group of wild boars below 12 months. Such differences may result from, among others, variances in the ages of sampled animals. Precise age data for the juvenile wild boars examined in this study were not available. However, if majority of individuals classified as juveniles were approximately 12 months, they were likely beyond the period of maternal immunity protection, which would increase their susceptibility to infection and potentially influence the obtained results. It is worth pointing out that in our study, the group of juvenile wild boars was underrepresented compared to the two remaining groups. On the other hand, our findings seem consistent with results reported regarding commercial, domestic pigs [3, 12, 37]. Importantly, in the present study, PCV3 prevalence in the liver of juvenile animals was higher compared to adults and wild boars aged 12–24 months. Similarly, PCV3 prevalence in the lungs of juveniles was higher compared only with adults.

Several attempts to provide a classification system for PCV3 were undertaken [7–9]. However, proposed nomenclature and genotypes/subgroups were often non-consistent and difficult to reproduce with alternative methods. Although ORF2 sequence is commonly used for phylogenetic inference for PCV2 sequences, some studies have shown differences between classification of PCV3 strains based on this region compared to the whole genome clustering [7]. However, it was also proved that the divergence within PCV3 ORF2 was higher than that of complete genomes, which makes it a viable candidate for phylogenetic inference [38]. Phylogenetic analysis revealed extensive polytomies and limited bootstrap support observed in PCV3 phylogenies, most probably reflecting the virus's recent emergence (first detected 2015) and rapid global dissemination. The validation across vastly different dataset scales (61 vs 2,340 sequences) with independent model selection and consistent topology confirms that the observed shallow phylogenetic structure represents genuine evolutionary history rather than methodological limitations. These results are in agreement with Franzo et al. [10], who revised multiple, highly variable classification systems previously presented by different authors and proposed criteria for distinguishing genotypes within PCV3 species.

All the sequences obtained in this study were highly similar to each other and localized in the same cluster on the phylogenetic tree, together with other Polish wild boar as well as swine-obtained sequences available in GenBank. The detection of well-supported local transmission clusters, such as the identical sequences 22–36 and 22–38 from Polish wild boars, demonstrates that phylogenetic signal is present where genuine evolutionary divergence has occurred, validating the sensitivity of our analytical approach. Analysis of previously described marker positions in the sequences obtained in the present study revealed a pattern S A R S I (aa positions ORF1-122, ORF2-24, ORF2-27, ORF2-77, and ORF2-150) consistent with sequences previously allocated in PCV3 group b1 cluster [37]. Although this classification does not fulfill the criteria to constitute formal genotypes and could not be attributed to any geographical clustering, we denote marker positions detected in our strains to provide additional characteristic comparable to previous studies.

Apart from two identical sequences obtained from the same location and suggesting the local virus spread, no other correlations with geographical distribution were observed.

Despite the high level of genetic similarity, some polymorphism in predicted Cap amino acid positions 2, 15, 75, and 98 was observed between sequences obtained in this study. Substitutions in positions 2 and 15 localized within nuclear localization signal (NLS) included R/K change, which may be considered conservative due to similar characteristics of both amino acids. However, substitutions in positions 75 and 98 (A/T and Q/K) are much less conservative, as they involve change between amino acids of different polarity and charge, respectively. Such substitutions may impact protein folding, function and interactions. These results are in agreement with observations of Pranoto et al. [39], who suggested that polymorphism in Cap positions 56, 75, 77, 98, 101, 150 and 155 may alter antigenicity and biological properties of Cap protein due to localization of these positions within immunogenic B-cell epitope. It cannot be excluded that also two mutations identified within conserved regions in rep sequence may impact virus biological properties. These conserved regions include the motifs DDFYGWVP (where a T663C mutation was detected in strain 22–33) and DRYP (where a G678A mutation was detected in strain 22–27). The ORF1 (Rep) sequence encodes the replication associate protein responsible for viral replication. Conserved motifs in the replicase protein are critical for viral function. Although the mutations were synonymous (silent) and did not change the amino acid sequence of DRYP and DDFYGWVP motifs, even synonymous mutations in highly conserved regions might impact codon usage, mRNA stability or evolution rate [40]. More studies are necessary to elucidate the actual impact of observed polymorphism on biological function of PCV3 protein and evolutionary processes.

Conclusion

Summarising, the current study approved the circulation of PCV3 with high prevalence among different age groups of the Polish wild boar population. Moreover, PCV3 was detected in all tested types of specimens within all age groups, except feces samples collected from juvenile boars. To the best of our knowledge, it is also the first study that pictures the viral prevalence in various types of tissues in different age categories. Detection of the highest viral prevalence of all tested age groups in the liver indicates that the liver may serve as the most suitable specimen for PCV3 detection in wild boars, and it may represent the PCV3 target organ for replication. Additionally, considering that the quantity of the virus in liver samples was higher compared to other matrices, this indicates that liver serves as a good sample type for virus isolation in cell culture. However, there is still a paucity of data about PCV3 pathogenesis in wild boars, and further research is demanded in this area.

Methods

Sample collection

The present study was carried out based on a total of 285 wild boar culled during the hunting seasons 2017–2018 and 2019–2020, inhabiting 46 forestry management units belonging to 16 out of 17 RDSFs in Poland (Table 1). The hunting of animals was in compliance with the Polish hunting law on animal protection. Ethical authority approval was not required as the samples were collected post-mortem by the certified hunters.

Blood, liver, lungs, and feces samples from 285 wild boars were collected. Blood samples, mainly as blood clots, were taken from the main vein trunks or the animal’s chest cavity, whereas livers, lungs, and feces were collected during the carcass evisceration. The material was sent in refrigerated conditions to the laboratory. Upon arrival, the livers, lungs, and feces were stored frozen at − 20 °C until use. The blood clot was centrifuged at 3000 rpm for 10 min in order to obtain serum, which was kept at − 20 °C until further usage. According to body weight, wild boars were categorised as juveniles (up to 35 kg, less than 1 year); adolescents (36 to 70 kg, between 1 and 2 years); or adults (more than 70 kg, more than 2 years).

DNA extraction and qPCR

Viral DNA was extracted from 200 μL serum samples and tissue homogenates (liver, lungs) using a QIAmp DNA Mini Kit (QIAGEN®, USA) according to the manufacturer’s instructions. Briefly, 20% suspension of each tissue in PBS (w/v) was submitted to mechanical tissue homogenization with the use of TissueLyser LT (QIAGEN®, USA). Next, 200 μl of clarified homogenate was used in the extraction protocol. Regarding fecal samples, DNA was extracted from 200 mg of feces with QIAmp DNA Stool Mini Kit (QIAGEN®USA) following instructions provided by the supplier (QIAGEN®, USA).

Amplification of viral DNA was performed by real-time PCR assay targeting rep gene region (ORF1) of PCV3 using primers and probe described previously [11] and the QuantiTect probe PCR kit (QIAGEN®, USA). The reaction mix consisted of 2 μl of extracted DNA, 12.5 µL of QuantiTect Master Mix, 0.5 μl of each primer (20 µM), 0.5 μl of probe (10 µM), and sterile water added to reach the final volume of 25 μl.

The thermal profile of the PCR reaction included 95 °C for 15 min and 40 cycles of 95 °C for 15 s and 60 °C for 60 s. The reactions were conducted using Mx3005P qPCR System (Agilent Technologies, Santa Clara, USA). Samples with a cycle threshold (Ct) > 37.0 were considered as negative.

Sequencing and phylogenetic analysis

To enhance the likelihood of effective sequencing PCR-positive samples with the Ct value lower than 30, corresponding to high viral load, were first subjected to rolling-circle amplification (RCA). Total DNA extracted from selected samples was amplified using TempliPhi 100 Amplification Kit (GE Healthcare) according to manufacturer’s recommendation. Secondly, attempts were made to obtain the complete PCV3 genome by performing PCR assays with three specific primer pairs (PCV-374–1144, PCV-31137–1561, and PCV-31427–433) reported by Fux et al. [9].The obtained amplicons were submitted for nucleotide sequencing. The PCR reaction contained 2 µL of DNA template, 2.5 µL of each primer (10 µM), 25 µL of Q5 High-Fidelity Master Mix (2X) (BioLabs), and sterile water to bring the final volume up to 50 µL. The thermal protocol was as follows: activation of DNA polymerase at 98 °C for 30 s, 35 cycles of denaturation at 98 °C for 10 s, annealing at 60 °C for 30 s and extension at 72 °C for 30 s, followed by a final extension at 72 °C for 2 min.

Phylogenetic analysis was performed on 2,340 PCV3 ORF2 sequences (661 nucleotides) retrieved from GenBank (accessed 2024.07.09) and aligned using MAFFT v7.490 with the FFT-NS-2 algorithm in Geneiuous Prime v.2024 0,7. Maximum likelihood phylogenetic reconstruction was conducted in IQ-TREE v1.6.12 web server (http://iqtree.cibiv.univie.ac.at/) using the Tamura-Nei substitution model with empirical base frequencies and gamma-distributed rate heterogeneity (TN + F + G4, α = 0.6238). Tree robustness was assessed using 1,000 ultrafast bootstrap replicates. The best-fit nucleotide substitution model was selected independently for each dataset using ModelFinder based on the Bayesian Information Criterion (BIC). Branch support was assessed using 1,000 ultrafast bootstrap (UFBoot) replicates and SH-aLRT (Shimodaira-Hasegawa approximate Likelihood Ratio Test) with 1,000 replicates. Nodes were considered well-supported when UFBoot values ≥ 70% or SH-aLRT values ≥ 80%. For visualization purposes, an initial dataset was reduced by selecting a subset of PCV3 sequences (n = 61) representing all main genetic clusters distinguished within genotype 1. To validate that the representative subset accurately captured the phylogenetic signal present in the comprehensive dataset, we compared: (1) substitution model selection, (2) bootstrap support value distributions, (3) tree topology patterns including polytomy prevalence, and (4) phylogenetic placement of Polish wild boar sequences between selected subset and the full dataset. Statistical comparisons of bootstrap support distributions were performed using Python 3.x with NumPy and Matplotlib libraries. ORF1 sequences obtained in this study were not included into the phylogenetic analysis, however, the nucleotide identity to each other was evaluate to estimate the presence of local genetic clusters.

Statistical analysis

The method described by Clopper and Pearson (Clopper and Pearson, 2006) was used to determine the exact confidence intervals (CI) at a 95% level of confidence for the obtained PCV3 PCR positive frequencies in the analysed sample type and age category. Differences in the frequency of PCV3 DNA detection among the different age categories (juveniles, adolescents, and adults) within each sample type, between sample types, and between RDSFs were compared with the Chi-square test with Bonferroni correction for pairwise comparisons or Fisher´s exact test. The difference in the mean (± SD) of Ct value was calculated using the Kruskal–Wallis test, with Dunn’s post hoc analysis. Microsoft Excel 2019 (version 2409) and the Real Statistics Resource Pack for Excel (Release 9.1.1) and Graph Pad vs. 10.0 were used for data analysis. The values of p < 0.05 were considered statistically significant.

Abbreviations

ASFV

African swine fever virus

CDS

Complete coding sequences

CI

Confidence intervals

Ct

Cycle threshold

NLS

Nuclear localization signal

NGS

Next-generation sequencing

ORF

Open reading frame

PCV2

Porcine circovirus type 2

PCV3

Porcine circovirus type 3

RDSF

Regional Directorates of State Forests

Authors’ contributions

E.C-D.– designing of the study, research on available knowledge about the topic, laboratory analyses, interpretation of data, funding acquisition, writing– original draft, writing—review & editing. A.D.– research on available knowledge about the topic, statistical analyses, interpretation of data, writing– review & editing, A.A.—research on available knowledge about the topic, writing– review & editing, A.J. – coordination of samples collection, writing– review & editing, E.S.—samples preparation, writing—review & editing. K.P– results analyses, interpretation of data, writing original draft, writing– review & editing. All authors reviewed the manuscript. All authors read and approved the final manuscript.

Funding

The study was financed by statutory funding 506.514.05.00 of the Department of Preclinical Sciences and Infectious Diseases, statutory funding 506.513.05.00 of the Department of Internal Diseases and Diagnostics, Faculty of Veterinary Medicine and Animal Science, Poznan University of Life Sciences, Poland, financed by the Polish Ministry of Science and Higher Education. The publication was financed by the Polish Minister of Science and Higher Education as part of the Strategy of the Poznan University of Life Sciences for 2024–2026 in the field of improving scientific research and development work in priority research areas.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval was not required for this study as blood and tissue samples were obtained from culled wild boars legally hunted as part of the hunting economy program and within hunting limits (Dz. U. z 1995 r. Nr 147 poz. 713, Ustawa z dnia 13 października 1995, “Prawo Łowieckie” z póź zm. (Journal of Laws 1995, No. 147, item 713; Bill of 13 October 1995, “The Hunting Law” as amended)).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.