Molecular survey of duck circovirus infection in poultry in southern and southwestern China during 2018 and 2019

Background Duck circovirus (DuCV) is a potential immunosuppressive virus that causes feather disorders in young ducks. In this study, DuCV obtained from various species of ducks was investigated by polymerase chain reaction (PCR) in southern and southwestern China (Guangdong, Guangxi and Yunnan provinces) from 2018 to 2019. Results A total of 848 bursa samples were collected from dead Mulard, Cherry Valley Pekin, Muscovy and Mallard ducks from duck farms. The positivity rate of DuCV in the total sample was approximately 36.91%. We found that the prevalence of DuCV in Yunnan (43.09%) was higher than those in Guangxi (34.38%) and Guangdong (34.4%). However, the positivity rates of DuCV in the four duck species were not significantly different (P > 0.05). Nineteen randomly selected complete viral genomes were sequenced. The complete genomes of the DuCV were 1987 to 1995 nt in length, and were 81.7–99.3% homologous to the other 57 sequences in GenBank. Phylogenetic analyses based on the complete genomes of 76 DuCVs showed that the 19 novel DuCV sequences from Guangdong and Guangxi provinces mainly belonged to the DuCV-1 and DuCV-2 genetic groups, respectively. However, the two genotype groups coexisted in Yunnan Province. In addition, recombination analysis showed putative recombination sites in 3 strains in Yunnan that originated from strains Guangdong and Guangxi. Interestingly, the epidemiological investigation showed that Mulard ducks, Cherry Valley Pekin ducks and Muscovy ducks more than 4 weeks old were more susceptible to infection with the novel DuCV than ducks less than 4 weeks old. Conclusions These data provide insight into the molecular epidemiology and genetic diversity of DuCVs circulating in southern and southwestern China for the first time.

low body weight (BW), lymphocyte depletion, necrosis, and histiocytosis in the bursa of Fabricius (BF) [10]. DuCV can be divided into two genotypes, DuCV-1 and DuCV-2, based on the complete genome and cap gene [17,19]. Worldwide, there are more reports of DuCV-1 in Germany, Hungary, the US, China, South Korea, and Poland than DuCV-2 reports in Taiwan and China [17]. Recent research has suggested that DuCVs-1 and 2 are mainly concentrated in eastern coastal cities and in migrating wild ducks in China [20]. Epidemics of DuCVs-1 and 2 have been associated with increasingly serious harm to the poultry industry in northern and southern China, including Hebei, Shandong, Fujian and Zhejiang provinces [15,17,21]. However, there have been few reports of DuCV epidemic outbreaks in Guangdong, Guangxi and Yunnan provinces. Therefore, this research aimed to better understand the epidemiological and genetic characteristics of DuCV in southern and southwestern China in 2018 and 2019.

Molecular assays of clinical samples
In this study, a total of 313 (36.91%) ducks were positive for DuCV. The positivity rates of DuCV were 34.4% (86/250), 34.38% (121/352) and 43.09% (106/246) in Guangdong, Guangxi and Yunnan provinces, respectively (Table 1) Table 1). The positivity rate was not significant among the various species of ducks (P > 0.05). Among the positive samples, 64 (21.92%) of 292, 162 (41.65%) of 389 and 87 (52.1%) of 167 ducks were in the age ranges of 0 to 4, 4 to 8 and more than 8 weeks, respectively ( Table 2). The DuCV prevalence in ducks more than 8 weeks old was 2.4-and 1.25-times greater than that in ducks 0 to 4 and 4 to 8 weeks old, respectively. The positivity rates were 23.23, 40.12 and 53.03% in Mulard ducks; 17.74, 42.11 and 51.16% in Cherry Valley Pekin ducks; and 22.58, 45.12 and 53.33% in Muscovy ducks aged 0 to 4 weeks, 4 to 8 weeks and more than 8 weeks, respectively. The positivity rates were 23.19 and 41.12% in Mallard ducks aged 0 to 4 weeks and 4 to 8 weeks; positivity in ducks aged more than 8 week was not recorded in this study ( Table 2). The results prompted us to determine that ducks more than 4 weeks old were more susceptible to DuCV than those less than 4 weeks old in Mulard ducks, Cherry Valley Pekin ducks and Muscovy ducks (P < 0.05). We analyzed 19 full-length sequences, and a putative recombination sites (at nt 989 and 1122) was detected in the genome of recombinant 3 strains (YN190411-China-Yunnan, YN190412-China-Yunnan, and YN190415-China-Yunnan) originating from the strains GD150501-China-Guangdong and GX190512-China-Guangxi.

Phylogenetic analysis of DuCVs
Phylogenetic analyses of the full-length genomes of the 19 novel sequences and 57 reference sequences obtained from GenBank indicated that the DuCVs could be divided into two distinct genetic groups: DuCV-1 (Group 1: Germany lineage) and DuCV-2 (Group 2: Taiwan lineage) (Fig. 1). The DuCVs from Guangdong (MK814571-MK814577) formed a distinct clade within DuCV-2, while the DuCVs from Yunnan (MK814584 and MK814585) clustered in one branch. The DuCVs from Guangxi (MK814586-MK814589) and Yunnan (MK814578-MK814583) clustered in another branch within DuCV-1. The results of the phylogenetic analysis indicated that two genetic groups of DuCVs had caused epidemics in ducks in three provinces.

Discussion
As stated in previous reports, DuCV may be an immunosuppressive virus that may increase the pathogenicity of coinfecting agents. The classic symptoms are generally considered to be feather disorders, poor body condition and low weight for age. The virus can persist in the thymus, liver, spleen, kidney and BF, and among these organs, the bursa is the most predominant location for circovirus replication [5,11,13,15,16,22,23]. There are no existing vaccines or drugs to cure or prevent DuCV, resulting in serious economic and production losses in the poultry industry. The prevalence of DuCV has been extensively reported in multiple species of ducks and from many provinces and cities in China [20,24].
In this study, we aimed to elucidate the epidemiology of DuCVs circulating in southern and southwestern China. PCR methods were used in a molecular epidemiological investigation of DuCV in the BF of dead ducks, and the results confirmed a high overall DuCV infection rate in four species of ducks from Guangdong, Guangxi and Yunnan Province. However, the positivity rates of DuCV by species and province were not significantly different. These results were similar to previous reports in which DuCV infection was ubiquitous, showing no species dependency [15,18]. In addition, the positivity rates showed that Mulard ducks, Cherry Valley Pekin ducks and Muscovy ducks more than 4 weeks old were more susceptible than ducks less than 4 weeks old, consistent with a previous report [15].
Similar to those of other members of the DuCV family, the genomic sequence of the 19 DuCVs contained three major ORFs: ORF 1, ORF 2 and ORF 3 [22,25]. According to the phylogenetic analyses of the complete genomes, the DuCV in this study belonged to two genotypes [6]. Compared to previous reports, the 19 DuCVs detected in this study belonged to the Group 1 Germany lineage or Group 2 Taiwan lineage [15]. Based on these results, we confirmed that among these 19 DuCVs, Group 1 was prevalent in Guangdong, Group 2 was prevalent in Guangxi and both groups were prevalent in Yunnan province. Interestingly, there were differences between the DuCVs identified in this study and other DuCVs from previous reports, and the genome lengths of the novel sequences GD150501, GD150502 GD180503, GD180504, GD190401, GD190402 and GD190403 were all 1987 nt, which is the shortest observed DuCV genome on record [15,20,24,26]. When the sequences were compared with the other genomes in GenBank, it became clear that there was a deletion at the 966 site in the intragenic region in 7 DuCVs from Guangdong, and other DuCVs also showed multiple Fig. 1 Phylogenetic analysis of novel duck circovirus (DuCV) wholegenome sequences using the neighbor-joining (NJ) method with 1000 bootstrap replications (Mega 7.0). DuCV-1-duck circovirus genotype 1, DuCV-2-duck circovirus genotype 2. The sequence of the circovirus detected in this study is marked with a symbol. DuCV strains from Guangdong are indicated with circles, those from Guangxi with a triangle, and those from Yunnan with a square. The other sequences were obtained from GenBank, and the accession numbers of these sequences are included in the phylogenetic tree mutations in the complete sequences. This suggests that there are mutations in the novel DuCV gene in the poultry of southern and southwestern China. In addition, we found that potential recombination events occurred in 3 strains from Yunnan, originating in strains from Guangdong and Guangxi. Geographically, Guangxi, Guangdong, and Yunnan are adjacent to each other, and the DuCV sequences were similar, leading to recombination events occurring in Yunnan. Our hypothesis is that virus transmission may have occurred due to trade within the province. A limitation of this study was that we collected only dead ducks; and novel DuCV should be collected and detected in Mallard ducks in the future. Additionally, other virus genotypes may infect ducks but produce no clinical symptoms. Therefore, we cannot rule out the existence of other genotypes in this region, but whether they cause clinical symptoms requires further investigation.

Conclusion
This study will help elucidate the epidemiological and molecular characteristics of DuCV in southern and southwestern China. The acquired data suggest that DuCVs are highly prevalent and that there were two distinct genetic groups of novel DuCV in 4 species of ducks in 2018 and 2019. The novel DuCVs can infect different breeds of ducks, and ducks over 4 weeks old were the most susceptible. We were the first to report variations in the complete genome lengths of and recombination in a novel DuCV in southern and southwestern China [15,20,[24][25][26]. DuCVs in additional species of ducks remain to be detected, and no wild bird samples were collected. In the future, we need to investigate additional species in poultry and wild-birds to determine the exact origin and analyze the transmission and pathogenesis of DuCVs to aid in the development of more effective vaccines against various DuCVs.

Sample collection
A total of 848 BF samples were collected from four species of ducks ranging from 2 to 10 weeks in age that presented clinical symptoms (feathering disorders, growth retardation and general sickness) and later died. Upon death, BF extraction was conducted according to standard care guidelines. The BFs from 327 Mulard ducks, 186 Cherry Valley Pekin ducks, 159 Muscovy ducks, and 176 Mallard ducks from 53 duck farms in Guangdong (20), Yunnan (17) and Guangxi (16) provinces in China from 2018 to 2019 were removed and sectioned (Table 1).

Statistical analysis
All one-sample t tests were performed using Graph-Pad Prism 6 (GraphPad Software, Inc.), and two-tailed P < 0.05 was defined as statistically significant.

Sequence analysis
Genetic variation and phylogenetic relationships with the complete genomes from 76 DuCV sequences available in GenBank were analyzed (online Technical Appendix Table S1). Nucleotide sequence alignments and homology comparisons were made using the Clustal W method in the MegAlign program (DNASTAR, Madison, USA). The neighbor-joining (NJ) method was used to generate phylogenetic trees based on the aligned nucleotide sequences in MEGA v7.0 [4,27]. To detect putative recombination breakpoints in the DuCV genome and to identify sequences that possibly originated from a recombination event, the RDP 4.24 program was used [28].
Additional file 1: Figure S1. Deduced amino acid sequence comparisons of the 19 DuCV strains in Cap (A) and Rep (B) protein. Table S1. Details of DuCV isolates used in this study and other isolates available in GenBank.  (KLPREAD201801-07). The funding body was solely involved in infusion and had no role in the design of the study; in the collection, analysis, or interpretation of the data; or in writing the manuscript.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Ethics approval and consent to participate
The duck samples used in this study were obtained from the Military Veterinary Institute, Academy of Military Medical Sciences, and the experiment was approved by the Institutional Animal Care and Use Committee (IACUC) of the Chinese Academy of Military Medical Science, Changchun, China (10ZDGG007). Written consent for the use of duck samples before participation in the study was obtained from the duck's owners.

Consent for publication
Not applicable.