Introduction

Porcine epidemic diarrhea virus (PEDV), a member of the genus Alphacoronavirus of the family Coronaviridae, is an enveloped, single-stranded RNA virus [2]. Porcine epidemic diarrhea (PED) caused by this virus is a highly infectious and destructive enteric disease, characterized by acute watery diarrhea, vomiting, and dehydration, which can rapidly lead to death in piglets, especially in neonatal piglets, with high morbidity and mortality (up to 100%) [7, 19]. PED was first observed in the early 1970s in England, and it subsequently spread widely to many swine-breeding areas in Europe and Asia [35, 46]. The PEDV genome is ~28 kb in length with seven open reading frames (ORFs) encoding four structural proteins (membrane [M], nucleocapsid [N], envelope [E], and spike [S]) and three non-structural proteins (ORF1a, ORF1b, and ORF3). The S glycoprotein, which is 1383 amino acids (aa) in length, is the most important viral antigen with respect to immunity [6, 25]. It is of central importance for viral attachment and entry, virulence, and induction of neutralizing antibodies [3, 35]. The S gene is highly variable due to frequent nucleotide substitutions, insertions, and deletions [11, 37, 38], and it is therefore used to study the genetic evolution of PEDV.

Porcine circoviruses (PCVs), which belong to the genus Circovirus of the family Circoviridae, are the smallest known pathogenic autonomously replicating viruses [1]. To date, four porcine circoviruses (PCV1, PCV2, PCV3, and PCV4) have been identified. PCV1 is non-pathogenic to pigs [34], whereas PCV2 is one of the most common pathogens causing immunosuppressive diseases in pigs and poses a major threat to the swine industry [9, 21, 29]. PCV3 was detected on swine farms in the United States in 2016 [28], but it is not clear whether it is pathogenic to pigs. PCV4 was first identified in 2019 in pigs with severe diarrhea, respiratory symptoms, and dermatitis nephrotic syndrome in Hunan province of China [42]. Since then, PCV4 has been reported in other provinces of China and in South Korea [12, 31, 32]. Like PCV1, PCV2, and PCV3, PCV4 is a single-stranded circular DNA virus without a viral envelope. Its genome size is 1,770 nucleotides (nt), and it has two main open reading frames: ORF1 and ORF2 [42]. ORF1 encodes a protein of 296 aa, namely Rep, which is believed to be responsible for replication of the viral genome, and ORF2 encodes a 228-aa Cap protein, which is the main viral antigen and is highly immunogenic [5, 17, 36]. Recently, PCV4 was successfully rescued from an infectious clone and shown to be pathogenic to piglets [23].

A previous study demonstrated that coinfection with PEDV and PCV4 occurs on Chinese pig farms [14], and this might result in more-severe clinical signs and higher mortality in pigs. It is therefore important to survey the prevalence and genetic characteristics of these two pathogens. Although there have been reports on the prevalence and genetic diversity of PEDV in Henan province of China before 2019 [6, 18, 43], there has been little information about new variant PEDV strains in recent years. PCV4 is a recently discovered circovirus, and more information about its epidemiology and pathogenicity is needed. Therefore, a duplex SYBR Green I–based quantitative real-time PCR (qPCR) assay was established to detect PEDV and PCV4 simultaneously, and 65 samples collected from diseased piglets in Henan province from 2019 to 2021 were screened for the presence of PEDV and PCV4. In addition, the full-length S gene of PEDV and a portion of the PCV4 genome were analyzed to investigate the genetic diversity of these two viruses in Henan province from 2019 to 2021.

Materials and methods

Viruses and clinical samples

PEDV- and PCV4-positive samples that were confirmed by PCR were used for the development of a duplex qPCR assay. Porcine reproductive and respiratory syndrome virus (PRRSV), PCV2, pseudorabies virus (PRV), porcine parvovirus (PPV), classical swine fever virus (CSFV), porcine transmissible gastroenteritis virus (TGEV), porcine bocavirus (PBoV), and porcine rotavirus (PoRV) were provided by the Key Laboratory for Animal-Derived Food Safety of Henan province, China.

From 2019 to 2021, a total of 65 clinical samples (including feces and intestinal tissues) were collected from piglets suffering from vomiting, diarrhea, and porcine dermatitis and nephrotic syndrome on 19 large-scale pig farms in 15 cities in Henan province of China. All clinical samples were homogenized and then diluted with phosphate-buffered saline (PBS). After three freeze-thaw cycles, the samples were centrifuged at 12,000 × g for 5 min, and the supernatants were collected. Viral RNA and DNA were extracted from 200 μL of the supernatant using a MiniBEST Viral RNA/DNA Extraction Kit Ver.5.0 (Takara, Dalian, China) according to the manufacturer’s instructions. Subsequently, viral RNA was reverse transcribed to cDNA using a TIANScriptII RT Kit (TIANGEN, Beijing, China), following the manufacturer’s instructions. The DNA and cDNA were stored at −20°C until testing.

Preparation of standard plasmids

The genome nucleotide sequences of PEDV strains (AF353511, KT591944, LC022792, and KT199103) were retrieved from the GenBank database and aligned using the MegAlign program in DNASTAR 7.01 software (DNASTAR Inc., Madison, WI, USA). A pair of specific primers (forward, 5′-CTCGGCTTGCATCACTCT-3′; reverse, 5′-GACCCA GTAGCAACCTTAT-3′) for amplification of a fragment of 225 bp was designed based on the conserved sequence of the PEDV M gene using the Primer Select program of DNASTAR 7.01 software. Primers for PCV4 (forward, 5′-CCACATAGTCTCCATCCAGTTG-3′; reverse, 5′-TACAGCCTCCCATTTGCATATTA-3′) were synthesized as described previously [40] and used to amplify a portion of the Cap gene of PCV4, with a product size of 124 bp.

PCR amplification of the PEDV M and PCV4 Cap genes was performed using a Veriti™ 96-Well Thermal Cycler (Thermo Fisher Scientific, Waltham, MA, USA). The PCR was carried out in 25-μL reaction volume, which consisted of 12.5 μL of 2× Taq Master Mix (Novizan, Nanjing, China), 3 μL of DNA template, 0.5 μL of forward primer (25 μM), 0.5 μL of reverse primer (25 μM), and 8.5 μL of ddH2O. The PCR amplification began with a pre-denaturation step at 95°C for 5 min, followed by 35 cycles of 95°C for 30 s, 60°C for 30 s, and 72°C for 20 s and a final extension step at 72°C for 10 min. The positive PCR products corresponding to the PEDV M gene and the PCV4 Cap gene were purified separately using a SanPrep Column DNA Gel Extraction Kit (Sangon, Shanghai, China) in accordance with the manufacturer’s instructions and introduced into the vector pMD18-T (Takara) for sequencing. The resulting plasmids (pMD-PEDV and pMD-PCV4) were quantified by measuring the optical density at 260 and 280 nm wavelength using a Nano-100 microspectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), and the concentrations of the two standard plasmids pMD-PEDV and pMD-PCV4 were 126.9 ng/µL (5.52 × 1010 copies/μL) and 126.8 ng/µL (4.11 × 1010 copies/μL), respectively.

Optimization of the duplex qPCR

First, separate singleplex qPCR assays for detecting PEDV and PCV4 were developed on a CFX96™ system (Bio-Rad Inc., USA), and the duplex qPCR was then optimized by adjusting the annealing temperature, time, and primer concentrations. PEDV and PCV4 were easily distinguishable based on their different melting peaks. The melting curves were acquired by monitoring fluorescence signals from 65°C to 95°C with increments of 0.5°C.

Standard curves were prepared by making tenfold serial dilutions of standard plasmids, and these were used to measure the sensitivity of the duplex qPCR assay. Nucleic acids obtained from PEDV, PCV4, TGEV, PPV, PRV, CSFV, PRRSV, PCV2, PBoV, and PoRV were applied for evaluating the specificity of this assay, and ddH2O was used as a negative control. Three dilutions (10-3-10-6) of the standard PEDV and PCV4 plasmids were used to evaluate the reproducibility of the duplex qPCR, and three parallel tests were performed for each dilution at three different times. Coefficients of variation of the Ct values were calculated to assess the intra- and inter-batch variability.

Detection of PEDV and PCV4 in clinical samples

Sixty-five clinical samples were screened for the presence of PEDV and PCV4 using the duplex qPCR. When the PEDV and/or PCV4 copy number in the sample was higher than the limit of detection (LOD), the sample was considered positive for PEDV and/or PCV4.

Cloning of partial genome sequences and phylogenetic analysis

The full-length S gene of PEDV and a portion of the PCV4 genome were amplified from positive samples as described previously [6, 14]. The purified products were ligated into the vector pMD18-T (Takara) to obtain recombinant plasmids, which were sequenced by Sangon Biotech Shanghai Co., Ltd. All sequencing reactions were performed in duplicate.

Sequences of 40 PEDV reference strains (Table 1) and 42 PCV4 reference strains (Table 2) were downloaded from the GenBank database, and the S gene sequences of the PEDV strains and partial genome sequences of the PCV4 strains were aligned with those of reference strains using the MegAlign program of the LaserGene software package (DNASTAR, Inc., Madison, WI). A phylogenetic tree based on the S gene sequences of PEDV was constructed using DNASTAR. A phylogenetic tree based on partial genome sequences of PCV4 was constructed using the neighbor-joining method (NJ) in Molecular Evolutionary Genetics Analysis (MEGA) software (version 7.0) with the p-distance model and 1,000 bootstrap replicates. Possible recombination events were analyzed using RDP4.0 software, and the result was corroborated using the Simplot program.

Table 1 Information about PEDV strains used for sequence alignment and phylogenetic analysis
Full size table
Table 2 Information about PCV4 strains used for sequence alignment and phylogenetic analysis
Full size table

Results

Optimization of the duplex qPCR assay

To develop a duplex qPCR assay for detection of PEDV and PCV4 simultaneously, various annealing temperatures (50°C, 55°C, and 60°C) and concentrations of each primer (15 µM, 25 µM, and 50 µM) were tested. The results showed that the optimal annealing temperature and primer concentration were 60°C and 25 µM, respectively. The optimized amplification conditions of the duplex qPCR included pre-denaturation for 5 min at 95°C, followed by 35 cycles of 95°C for 30 s, 60°C for 30 s and 72°C for 20 s and a final extension at 72°C for 10 min. A linear standard curve was prepared from 5.52 × 101 to 5.52 × 107 copies/μL for PEDV and 4.11 × 101 to 4.11 × 107 copies/μL for PCV4, and the equations for the standard curves were y = -3.3927X + 39.817 and y = -3.2747X + 36.243 for PEDV and PCV4, respectively, with strong linear correlations (R2) of 0.9999 (PEDV) and 0.9998 (PCV4) (Fig. 1). To determine the LODs of PEDV and PCV4, standard plasmids were serially diluted tenfold in elution buffer to 10-10. Amplification was observed when the plasmids were diluted to 10-9, but no amplification was observed at 10-10. Identical results were obtained when the experiments were repeated 10 times. Therefore, at the limiting dilution of 10-9, the LOD was 55.2 copies/μL and 44.1 copies/μL for PEDV and PCV4, respectively. Specific melting peaks were observed at 84°C and 79.5°C for PEDV and PCV4 respectively, while no amplification or fluorescence signal was detected for PCV2, TGEV, PBoV, PoRV, PRV, PPV, CSFV, or PRRSV. The intra-batch coefficient of variation of standard products with different concentrations was 0.132-0.313%, and the inter-batch coefficient of variation was 1.149-1.758%, i.e., both less than 2%, with a high rate of reproducibility.

Fig. 1
figure 1

SYBR Green I–based qPCR standard curves for PEDV and PCV4. The standard curve for PEDV was determined to be y = -3.3927X + 39.817, with an R2 value of 0.9999. The PCV4 standard curve was determined to be y = -3.2747X + 36.243, with an R2 value of 0.9998

Full size image

Testing of clinical samples

Sixty-five clinical samples collected from swine in Henan province were tested using the duplex qPCR assay. The results showed that 26 samples (40%, 26/65) were positive for PEDV, and 25 samples were positive for PCV4, with a positive rate of 38% (25/65). The rate of coinfection with PEDV and PCV4 was 34% (22/65). The Ct values for the positive samples are shown in Table 3.

Table 3 Ct values for samples that were positive for PEDV and PCV4
Full size table

Genetic analysis of PEDV

The complete S genes of eight PEDV strains from positive pig samples were amplified and sequenced as described previously [6], and their full-length sequences were assembled using the EditSeq program of the LaserGene software package based on overlapping sequences. Seven of these isolates (HNZMD-25, HNZMD-26, HNKF-23, HNKF-21, HNJZ-29, HNJZ-27, and HEXX-24) had an S gene that was 4,149 nt in length, encoding a 1,383-aa protein, but the S gene of the remaining strain (HNXX-24XIA) was only 3,555 nt in length (1,185 aa) and contained a wide range of nucleotide deletions. Notably, the isolates HEXX-24 and HNXX-24XIA were present in the same sample. A multiple sequence alignment showed that the nucleotide sequence identity of the eight PEDV strains ranged from 98.2% to 99.8%, and these strains shared 93.6% to 99.3% nucleotide sequence identity with 40 PEDV reference strains (Table 1). Phylogenetic analysis revealed that the 48 PEDV strains were divided into two groups: G1 and G2 (Fig. 2). G1 was comprised of classical strains, including a vaccine strain (CV777), a South Korean strain (virulent DR1), and two Chinese strains (SD-M and LZC). G2 was composed of emerging non-S INDEL strains (G2a and G2b) and emerging S INDEL strains (G2c). All strains identified in the present study belonged to the subgroup G2a, together with most of the PEDV reference strains from China from 2011 to 2021.

Fig. 2
figure 2

Genotyping analysis of 48 PEDV strains based on S gene sequences. A phylogenetic tree was constructed using DNASTAR software. The positions of the eight PEDV strains from the present study are indicated in red. The recombinant strain HEXX-24 is indicated in bold

Full size image

Interestingly, among the seven strains with longer S genes, only one common aa deletion was observed at residue 135 in the N-terminal domain (NTD, aa 19-233) of the S protein when compared with CV777. However, isolate HNXX-24XIA displayed a number of aa deletions from positions 31 to 229. Notably, when compared with the S proteins of BJ-2011-1 (G2a) and AJ1102 (G2b), the other seven strains displayed common aa deletions (59CGVN62 and 140ND141) or insertions (154DG155), frequently occurring in the N-terminal domain (NTD, aa 19-233) of the S protein, but a distinct change (G155R) in the insertion site was observed in HNKF-21 (Fig. 3). A previous study demonstrated the presence of four principal neutralization antigen determinants on the surface of the PEDV S protein: COE (aa 499–638), SS2 (aa 748–755), SS6 (aa 764–771), and 2C10 (aa 1368–1374) [30]. It should be noted that aa mutations were frequently found in the neutralizing epitopes of PEDV (Table 4), especially in the neutralizing epitope COE (aa 499–638). Recombination analysis revealed that a recombination event had possibly occurred in isolate HEXX-24 and might have involved the strains PC22A and HUBY. The putative beginning and ending breakpoints were located at nucleotide positions 58 and 665, respectively (Supplementary Figs. S1 and S2).

Fig. 3
figure 3

Amino acid sequence analysis of the PEDV S protein. The red open boxes represent deletions or insertions of amino acids in the N-terminal domain of the S protein in comparison with CV777, BJ-2011-1, and AJ1102

Full size image
Table 4 Amino acid mutations in neutralizing epitopes of the putative S protein of eight PEDV strains from this study compared with strain CV777
Full size table

Genetic analysis of PCV4

The partial genomes (1,274 nt) of three PCV4 strains were sequenced, and the resulting sequences contained the Cap gene. A nucleotide sequence alignment showed that the nucleotide sequence identity was 99.8-99.9% among the three PCV4 strains and 98.6-99.7% between the three PCV4 strains and 42 PCV4 reference strains (Table 2). Phylogenetic analysis of the partial genome sequence showed that the 45 PCV4 strains were divided into two major genotypes, PCV4a and PCV4b (Fig. 4), confirming that the genotyping system (PCV4a and PCV4b) proposed by Xu et al. [40] was still applicable despite the increase in the number of PCV4 sequences. All three PCV4 strains from the present study belonged to genotype PCV4a, together with 34 Chinese strains and one South Korean strain (E115). The remaining seven PCV4 strains belonged to genotype PCV4b. Phylogenetically, PCV4a was further classified into PCV4a1 and PCV4a2, and the three PCV4 strains from this study and 22 reference strains were distributed within the PCV4a1 cluster. In addition, as reported by Wang et al. [36], five antibody recognition regions were located at aa 77F-88F, 104N-112Y, 122D-177N, 199N-205V, and 219F-225P, respectively. Notably, an aa mutation (K128R) in isolate ZZ-2019 from this study was found in the antibody recognition region (122D-177N) (Fig. 5).

Fig. 4
figure 4

Genotyping analysis of three PCV4 strains from the present study and 42 PCV4 reference strains, based on the Cap gene. A phylogenetic tree was generated by the neighbor-joining method. Bootstrap values are indicated at each node, based on 1000 replicates. The positions of the three PCV4 strains in the present study are indicated by solid black circles

Full size image
Fig. 5
figure 5

Amino acid mutations in the Cap proteins of 45 PCV4 strains. The PCV4 strains were divided into three genotypes, PCV4a1 (gold), PCV4a2 (blue), and PCV4b (orange). Potential genetic markers are indicated by red rectangles. Amino acid sites in potential epitope regions are highlighted in yellow. Black solid circles (●) represent the PCV4 isolates from the present study

Full size image

Discussion

Recently, PED was listed as one of the most threatening infectious diseases in the domestic swine industry in China [4]. PCV4 was first discovered in Hunan province in 2019 and is now prevalent in most areas of China [42]. Coinfection with multiple pathogens has become the main epidemic form on large pig farms in China, and this often leads to high morbidity and mortality of pigs. Coinfection with PEDV and PCVs is common and often causes more-severe disease than a single infection. Tian et al. [33] reported that the infection rates of PEDV, PCV2, PCV3, and PCV4 in 63 samples collected from 2018 to 2019 were 22.22% (14/63), 88.89% (56/63), 41.27% (26/63), and 25.40% (16/63), respectively. Among the samples that were positive for PCV4, the detection rate of PEDV was 18.75% (3/16), suggesting that PCV4 might promote PEDV infection. Hou et al. [14] reported that the detection rates of PEDV and PCV4 in 152 samples collected from diseased pigs in Henan province during 2011-2021 were 10.53% (16/152) and 45.39% (69/152), respectively. It is noteworthy that, in PEDV-positive tissue samples, the positive rate for PCV4 was 56.25% (9/16). Previous studies have shown that PCV4 is frequently detected in pigs with diarrhea and is often present in coinfections with PEDV. These reports suggest that PCV4 may be a potential cause of diarrhea in pigs, so it is necessary to monitor the prevalence of these two viruses.

In this study, a duplex SYBR Green I–based qPCR method was successfully established for detection of PEDV and PCV4 simultaneously. These two viruses are easily distinguished by their distinct melting temperatures, which are 84.0°C for PEDV and 79.0°C for PCV4. Furthermore, the limit of detection for PEDV was 55.2 copies/μL, which is lower than that of the fluorescent quantitative PCR assay developed by Zheng et al. [48]. The limit of detection for PCV4 was 44.1 copies/μL, which is comparable to those of previously reported qPCR assays [13, 40]. When 65 clinical samples were tested using the duplex qPCR assay, the detection rate of PEDV was 40% (26/65), which is similar to that reported by Zhang et al. [45]. The detection rate of PCV4 was 38% (25/65), which is much higher than that reported by Zhang et al. [41], suggesting that PCV4 might be widespread in Henan province.

Interestingly, the Ct values for most of the PEDV and PCV4 strains tested exceeded 25. We therefore determined the full-length S gene nucleotide sequences of eight PEDV strains and partial genome sequences including the capsid (Cap) gene of three PCV4 strains. Phylogenetic analysis demonstrated that PEDV strains mainly fell into two groups, G1 (classical strains) and G2 (variant strains), and the latter was further subdivided into G2a (non-S INDEL strains), G2b (non-S INDEL strains), and G2c(S INDEL strains) subgroups, as reported previously by Lin et al. [20]. Eight PEDV strains obtained in the present study belonged to subgroup G2a, which includes a number of Chinese strains identified in recent years. Notably, the isolate HNXX-24XIA exhibited deletions of 199 aa in the N-terminal domain of S protein when compared with strain CV777. A similarly large number of aa deletion events have also been reported in the United States and Japan [8, 15, 24, 44], implying that large-scale deletion of the N-terminal domain of the S protein might not be a rare occurrence. Moreover, the other seven strains displayed only one aa deletion and no aa insertions in the N-terminal domain of the S protein in comparison with CV777, and they were distinct from other G2a strains. Notably, when compared with BJ-2011-1 (G2a), six aa deletions and two aa insertions were observed in the N-terminal domain of the S protein. These results suggest that a novel strain of G2a is circulating in Henan province of China. Interestingly, the isolates HEXX-24 and HNXX-24XIA were present in the same sample, collected in 2021, and an aa sequence alignment of these two isolates showed that, except for the successive aa deletions, only one aa mutation was found in the S protein, indicating that the HNXX-24XIA strain with a large gene deletion may be evolved from the HEXX-24 strain. These results showed the eight strains differed genetically from the reference strains.

The eight PEDV strains displayed common aa mutations in neutralizing epitopes when compared with the CV777 strain, and two extra aa mutations were observed in the HNKF-21 strain. These aa mutations might increase the likelihood of immune escape, and further investigations should be carried out to determine whether these aa mutations influence the antigenicity and pathogenicity of PEDV. A recombination event was identified in the HEXX-24 sequence. Interestingly, the putative parental strains of HEXX-24 belonged to two different subtypes (G2a and G1). The possible circulation of recombinant strains of PEDV in Henan province might pose a challenge for the prevention and control of PEDV.

Phylogenetic analysis based on partial genome sequences of PCV4 isolates showed that they were divided into two major genotypes: PCV4a and PCV4b. PCV4a was further divided into PCV4a1 and PCV4a2, which could provide some reference for the genotyping of PCV4. Previous studies have shown that aa substitutions can serve as markers to distinguish different subtypes of viruses, such as CPV and PCV3 [10, 16, 26, 27, 47]. In this study, PCV4a1, PCV4a2, and PCV4b could be differentiated by a specific combination of amino acids in the Cap protein (27S/G, 28R, and 212L for PCV4a1; 27S, 28G, and 212L for PCV4a2; and 27N, 28R, and 212M for PCV4b). The three PCV4 isolates from the present study belonged to PCV4a1. Intriguingly, among the PCV4 reference strains, six PCV4 strains (Hebei-Fox1 and Hebei-Rac1-5) were from fur animals (foxes and raccoon dogs), demonstrating that pigs are not the only hosts of PCV4 and suggesting that cross-species transmission of PCV4 can occur. Nucleotide sequence alignment showed that all of the PCV4 strains had similar sequences (>98% identity), and this high degree of sequence conservation may have an impact on genotype classification. To formulate accurate and useful classification schemes, greater effort must be made to carry out representative and structured sampling and to increase the sharing of appropriately annotated sequences in freely accessible databases.

It has been reported that the N-terminus (aa 1-37) of the Cap protein of PCV4 contains a putative nuclear localization signal (NLS), which could mediate nuclear targeting of the viral genome [49]. Notably, the same aa substitution (N27G) was observed in the putative NLS of three of the isolates (XC-2021, LY-2020, and ZZ-2019) from the present study, which might affect its function. Nguyen et al. [22] previously identified two motifs (P-x-x-P and Y-x-x-φ) in the Cap protein of PCV4 that are critical for host cell entry by members of the family Circoviridae [39]. When compared with HNU-AHG1-2019, the isolate ZZ-2019 contains an amino acid substitution in a previously identified antibody recognition region, which might affect its antigenicity. More experimental data are needed to shed light on the biological functions and infection mechanism of PCV4.

In summary, the duplex qPCR assay established in this study can detect PEDV and PCV4 simultaneously and has good specificity, sensitivity, and reproducibility, which provides technical support for rapid diagnosis and monitoring of PEDV and PCV4. In addition, characteristics of PEDV and PCV4 were analyzed to provide a reference for the prevention and control of PEDV and PCV4.