Development of a PRRSV Detection Assay Using Multienzyme Isothermal Rapid Amplification and Genetic Analysis of PRRSV in Southwest China

Abstract

Porcine reproductive and respiratory syndrome (PRRS) is a highly contagious viral disease that causes substantial economic losses in the swine industry. This study aimed to develop a PRRS virus (PRRSV) detection assay using multienzyme isothermal rapid amplification (MIRA) and to analyze the genetic variation of PRRSV in Southwest China. A total of 13,863 samples, including blood and lung tissues from pigs suspected of PRRSV infection, were collected. The MIRA assay was designed with primers and probes targeting conserved regions of the PRRSV-M gene, demonstrating high specificity with no cross-reactivity to other swine pathogens and an estimated detection threshold sensitivity of 1.0 copy/μL. Prevalence analysis revealed that, although vaccinated pigs showed relatively high antibody levels, the virus continued to circulate, particularly in unvaccinated herds. Genetic analysis of the predominant PRRSV strains indicated an increasing prevalence of NADC30-like strains and notable genetic variation in genes such as ORF5 and nonstructural protein 2 (NSP2), including amino acid deletions and alterations of glycosylation sites. Recombination events were also observed in some isolates. These findings provide essential insights into the epidemiology and genetic diversity of PRRSV in Southwest China, contributing critical data for the development of more effective control and prevention strategies.

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

1. Introduction

Porcine reproductive and respiratory syndrome (PRRS) is a highly contagious viral disease that poses a significant threat to the global swine industry, causing substantial economic losses and impacting animal welfare [1–3]. The disease can lead to severe reproductive failures in sows, such as abortions, stillbirths, and the birth of weak or premature piglets with high mortality rates. Additionally, it causes respiratory distress and pneumonia in pigs of all ages, which can reduce feed intake, slow growth, and increase susceptibility to other infections [4, 5]. These factors can lead to a significant decline in herd productivity and impose considerable financial burdens on pig farmers. The causative agent of PRRS is the PRRS virus (PRRSV), an enveloped, positive-sense, and single-stranded RNA virus classified under the family Arteriviridae within the order Nidoviralase [6]. Recently, PRRSV strains are reclassified into two genetically distinct species: betaarterivirus suid 1 (European type or PRRSV-1) and betaarterivirus suid 2 (North American type or PRRSV-2), each with further subdivisions [7, 8].

In China, PRRSV has been a persistent challenge for the swine industry since its emergence. After the initial detection of the classic PRRSV strain (CH-1a-like) in 1990s [9], the virus has continued to evolve, resulting in new genotypes and variants such as highly pathogenic PRRSV (HP-PRRSV) and NADC30-like and NADC34-like PRRSV [10–13]. The genetic variation of PRRSV is a key factor in its pathogenicity and ability to evade immunity. Particularly, mutations in the nonstructural protein 2 (NSP2) and the glycoprotein 5 (GP5, encoded by ORF5) genes are significant [14]. The NSP2, involved in viral replication, has been associated with modulation of viral virulence and persistence [15, 16]. GP5, on the other hand, is crucial for virus entry into host cells and is a major target for neutralizing antibodies [17]. Despite efforts in vaccine development and implementation of control measures [18–21], PRRSV remains a formidable adversary for pig farmers and the veterinary community in China. The virus’s mutability means that existing vaccines may not always provide protection against new strains, emphasizing the need for continuous monitoring and surveillance to track the virus’s evolution and to devise effective control strategies.

Current diagnostic methods for PRRSV, including RT-PCR, ELISA, quantitative PCR (qPCR), and sequencing, are essential for managing the disease [22]. However, these methods can be time-consuming and labor-intensive. There is a pressing need for the development of novel rapid diagnostic tools that can provide accurate results in a shorter time frame. Multienzyme isothermal rapid amplification (MIRA) is an innovative nucleic acid amplification technique that operates at a constant temperature, relying on multiple enzymes for rapid and efficient amplification [23]. MIRA simplifies the diagnostic process, as it can amplify target DNA/RNA directly from samples within 5–20 min, offering high sensitivity and specificity [24]. This method is versatile and can be applied in various fields, including research, food safety testing, and animal disease detection.

In this study, we aimed to determine the prevalence of PRRSV in Southwest China, contributing key data for its control and prevention efforts. Moreover, we strived to develop an innovative rapid detection method for PRRSV using MIRA technology, with the goal of improving the efficiency of PRRSV surveillance.

2. Materials and Methods

2.1. Virus Strains and Clinical Samples

All viral strains utilized in this study, including PRRSV strains (CH-1a, VR2332, BJ-4, HP-PRRSV, NADC30-like PRRSV, and NADC34-like PRRSV), CSFV, PEDV, PCV2, and PRV, were stored in our laboratory. A total of 13,863 samples, comprising 11,479 blood and 2433 lung tissues, were collected from pigs suspected of PRRSV infection across various farms from January of 2021 to June of 2024. Nucleic acid extraction from both viral strains and clinical samples was performed using the DNA/RNA extraction kit purchased from Bioflux (Hangzhou Bioer Technology Co., Ltd, China).

2.2. Primer and Probe Design

By aligning membrane (M) gene sequences of various PRRSV isolates collected from GenBank, we utilized the MegAlign software to identify highly conserved regions. With these conserved areas as a guide, we employed Oligo 6.0 to design one pair of primers and one probe. Simultaneously, another pair of primers were designed for amplifying the full-length PRRSV-M gene to construct a standard plasmid. Furthermore, we designed primers for ORF5 and NSP2 to facilitate sequence analysis of randomly selected prevalent PRRSV strains. To obtain a full-length genome, we designed eight pairs of primers. All primers are listed in Supporting Information 1: Table S1. These primers were synthesized by Sangon Biotech Ltd (Shanghai, China).

2.3. Construction of the Standard Plasmid

The PRRSV-M gene was amplified from the PRRSV strain HuN4 (GenBank Accession number: EF635006) and subsequently cloned into the pMD-18T Vector. The recombinant plasmid pMD-18T-M was verified through DNA sequencing. The concentration of the correct plasmid was determined by measuring ultraviolet absorbance at 260 nm using a Nano Drop spectrophotometer. The copy number of the plasmid was then calculated with the formula: copies/μL = (6.02 × 10²³ × (value of A₂₆₀ ng/μL) × 10⁻⁹)/(plasmid length × 650).

2.4. Establishment of MIRA Assay for PRRSV Detection

The MIRA reaction was performed using the RNA isothermal rapid amplification kit (catalog number WLRE8208KIT, Amp-Future Biotech Co. Ltd, Weifang, China) following the manufacturer’s protocol. Each reaction mixture comprised 5 µL of template, 29.5 µL of buffer A, 8.5 µL of nuclease-free water, 2 µL M–F (10 µM), 2 µL M–R (10 µM), 0.6 µL probe (10 µM), and 2.5 µL buffer B. The amplification was carried out for 20 min at 42°C with one cycle every 30 s. In each experiment, a negative control containing only nuclease-free water was included to ensure specificity.

2.5. Assessment of Specificity, Sensitivity, and Comparability of the MIRA Assay

To assess the specificity of the MIRA assay, we conducted cross-reactivity tests with CSFV, PEDV, PCV2, and PRV to ensure there was no interference with the assay’s performance. Additionally, various PRRSV strains were used for confirmation testing. The limit of detection (LOD) of the MIRA assay for PRRSV was determined using a tenfold serial dilution of a standard plasmid, with concentrations ranging from 10⁻¹ to 10⁶ copies/μL as templates.

We collected a total of 20 clinical samples from pig farms in the Southwest China region and compared the detection results from the MIRA method developed in this study with those obtained using the gold standard qPCR methods in China, as per national standards, to verify the accuracy of our established detection method.

2.6. Prevalence Analysis of PRRSV Infection in Southwest China

Serum samples were evaluated for PRRSV antibodies using a commercial ELISA kit (Jinnuo, Beijing, China). Additionally, a subset of tissue samples from pigs in the southwestern region of China was tested using our established MIRA assay to confirm the presence of PRRSV and to assess its prevalence in the region.

2.7. Genetic Diversity Analysis of Dominant PRRSV Strains in Southwest China

In order to explore the genetic diversity among the circulating PRRSV strains in Southwest China, we randomly selected positive samples and carried out full-length sequencing of the ORF5 and the hypervariable region (HVR) of the NSP2 gene, which are recognized for their high genetic variability. After obtaining the sequences, we aligned them with reference strains from the GenBank database (Supporting Information 2: Table S2) using the Clustal W alignment method of the MegAlign program of the DNASTAR Lasergene.v7.1. Subsequently, we employed MEGA software (version 11.0) to construct phylogenetic trees using the neighbor-joining (NJ) method with a p-distance model, incorporating 1000 bootstrap replicates, in order to comprehensively analysis the genetic variation among the dominant PRRSV strains in this region.

2.8. N-Glycosylation Analysis

To explore the temporal N-glycosylation patterns of GP5, GP5 sequences were submitted to the NetNGlyc web server for analysis (https://services.healthtech.dtu.dk/services/NetNGlyc-1.0/) [25].

2.9. PRRSV Genomic Full-Length Sequencing

In order to obtain the full-length genome sequence of PRRSV, viral RNA was extracted from porcine lung tissues infected with PRRSV. Subsequently, the viral genome was amplified in segments using RT-PCR. Multiple pairs of primers were designed to cover the entire genome of PRRSV. The segments were amplified using a high-fidelity enzyme. After amplification, the PCR products were sent to a sequencing company for sequencing. The sequencing results were then assembled and corrected using bioinformatics software, ultimately yielding the full-length genome sequence of PRRSV.

2.10. Recombination Analysis

The complete genome sequences were aligned using the ClustalW program of MEGA 11 software for recombination screening. RDP v4.8 containing seven algorithms (RDP, BootScan, GENECONV, MaxChi, Chimera, SiScan, and 3Seq) was used to predict the putative recombination events and precise recombination breakpoints. Recombination events were only considered as significant ($p<0.05$) when supported by at least four out of the seven detection methods as previously described [26–28]. SIMPLOT v3.5.1 with a window size of 200 bp and a step size of 20 bp was used to check the recombination signals further.

3. Results

3.1. Development of the Standard Curve for PRRSV Detection

A recombinant plasmid containing the M gene fragment of PRRSV was successfully constructed. DNA sequencing was performed to verify the plasmid’s integrity and sequence accuracy. To generate a standard curve, this recombinant plasmid was diluted in a tenfold serial dilution series, creating standard samples with copy numbers logarithmically spaced from 1 × 10⁷ to 1 × 10¹ copies/μL. These samples were employed as templates to create a standard curve specific to PRRSV in the MIRA assay. As shown in Figure 1, the efficiency of amplicon and the correlation coefficients met the reliability standards, with a coefficient of determination (R²) value of 0.9946, indicating a high degree of correlation. This standard curve was then applied for the precise detection of PRRSV using the MIRA assay.

[figure(s) omitted; refer to PDF]

3.2. Evaluation of Sensitivity, Specificity, and Concordance of the MIRA Assay

The sensitivity of the MIRA assay was determined by analyzing its ability to detect minimal PRRSV quantities, with results indicating an estimated detection threshold of ~ approximately 1.0 copy/μL based on a standard dilution series, as shown in Figure 2A. However, this threshold was not validated through repeated independent experiments and should be further confirmed in future studies. Specificity evaluations involved testing the assay with nucleic acids from non-PRRSV pathogens, including CSFV, PEDV, PCV2, and PRV. Positive controls consisted of viral RNA from various PRRSV strains, such as CH-1a, VR2332, BJ-4, HP-PRRSV, NADC30-like PRRSV, and NADC34-like PRRSV, along with five clinical samples confirmed to be PRRSV-positive. Nuclease-free water served as the negative control. All PRRSV isolates and positive samples produced distinct amplification curves, while non-PRRSV pathogens and the negative control showed no amplification, as depicted in Figure 2B.

[figure(s) omitted; refer to PDF]

In a comparative analysis with the gold standard qPCR methods across 20 clinical samples, the MIRA assay correctly identified 15 samples that were positive by qPCR, demonstrating a high level of agreement between the two detection methods (Table 1). This consistency suggests that the MIRA assay is a reliable method for PRRSV detection, matching the sensitivity and specificity of the qPCR methods. The coefficient of determination (R²) value of 0.9946 further validates the MIRA assay’s reliability, reflecting a robust correlation with qPCR results. These results confirm the high specificity and reliability of the MIRA assay for detecting PRRSV.

Table 1

Comparative analysis of the MIRA and qPCR assay.

Sample number	MIRA (×30 s)	qPCR (Ct)	Sample number	MIRA (×30 s)	qPCR (Ct)
130	—	—	279	5.5	35.06
285	—	—	291	2.16	30.18
315	3.21	32.07	330	—	—
489	—	—	1	—	—
823	2.43	30.50	2	2.11	30.52
116	4.09	33.24	3	3.59	32.85
197	3.42	32.45	4	2.43	30.22
200	6.12	35.76	5	2.39	30.87
2268	7.37	36.84	6	9.28	38.17
2326	5.12	35.12	Negative control	—	—
2532	9.13	37.95	—	—	—

3.3. Prevalence of PRRSV Infection in Southwest China

To evaluate the prevalence of PRRSV among pig farms in Southwest China with varying vaccination statuses, we collected a set of 13,863 samples. The serum samples were subjected to testing for PRRSV antibodies using a commercial ELISA kit, while the lung tissues were examined using the MIRA assay developed in this study. As detailed in Table 2, during the period from 2021 to 2024, it was observed that pigs that had received vaccinations exhibited significantly higher antibody levels, ranging from 87.18% to 91.76%, suggesting the relative effectiveness of the currently employed PRRSV vaccines in eliciting an immune response. However, the detection of PRRSV antibodies in 16.12%–23.60% of nonvaccinated pigs indicates a substantial exposure to the virus even in the absence of vaccination. The MIRA assay for viral RNA detection revealed a presence rate of PRRSV ranging from 13.99% to 21.10%. This result not only confirms the ongoing circulation of PRRSV within the region but also highlights its persistent significance as a potential health risk to the swine population.

Table 2

Detection of PRRSV by ELISA and the MIRA assay.

Year of detection	Vaccination status	PRRSV antibodies using ELISA			PRRSV RNA using MIRA
		Test no.	Positive no.	Percentage (%)	Test no.	Positive no.	Percentage (%)
2021	Yes	2002	1796	89.71	693	97	13.99
	No	1723	298	17.30	—	—	—
2022	Yes	2307	2117	91.76	732	127	17.35
	No	1903	385	20.23	—	—	—
2023	Yes	1338	1171	87.52	839	177	21.10
	No	1364	322	23.60	—	—	—
2024	Yes	476	415	87.18	120	19	15.83
	No	366	59	16.12	—	—	—

3.4. Genetic Variation of Prevalent PRRSV Strains

To understand the genetic variation of prevalent PRRSV strains, we carried out a molecular analysis. A randomly selected subset of samples was subjected to RT-PCR amplification targeting the PRRSV-ORF5 and NSP2 gene fragments. Upon successful amplification, the obtained DNA fragments were cloned and sequenced. This process yielded a total of 52 sequences for the ORF5 gene and 28 sequences for the NSP2 gene, all of which were deposited in GenBank (details summarized in Supporting Information 3: Table S3). A thorough analysis of these sequences provided insights into the circulation patterns of PRRSV strains in Southwest China. The results indicated that VR-2332-like and HP-PRRSV-like strains still existed in the pig population, but the proportion of NADC30-like strains continued to increase. Specifically in 2023, the proportion of NADC30-like strains reached 61.9% (13/21), establishing it as the predominant strain of PRRSV epidemics. Although the proportion of NADC34-like strains was relatively lower, it reached 42.1% (8/19) in 2022, highlighting that its epidemiologic potential should not be ignored.

3.4.1. Phylogenetic Analysis and Genetic Diversity of ORF5

To investigate the phylogenetic relationships and genetic diversity of the ORF5 gene among PRRSV strains, a phylogenetic tree was constructed. The results showed that among the 52 PRRSV isolates in this study, one strain was classified as PRRSV-1 type, while the remaining 51 strains clustered into PRRSV-2 (Figure 3). These PRRSV-2 strains were further divided into different lineages, specifically lineage 1 (including sublineages 1.5 and 1.8), lineage 5, and lineage 8. The nucleotide identity among these strains ranged from 82.5% to 100%. Among the 51 PRRSV-2 isolates, eight strains exhibited a close genetic relationship with lineage 5, sharing a nucleotide identity of 98.5% to 99.7% with the reference strain VR2332; seven strains were closely related to the lineage 8, sharing 99.2%–99.7% nucleotide identity with the reference HP-PRRSV strain HuN4. Twelve strains were grouped into sublineage 1.5 and 24 strains into sublineage 1.8. The strains in sublineage 1.5 shared 87.2%–96.8% nucleotide identity with the representative strain NADC34, while those in sublineage 1.8 shared a nucleotide identity of 89.6%–94.7% with the representative strain NADC30 (details are summarized in Supporting Information 4: Table S4).

[figure(s) omitted; refer to PDF]

Detailed multiple amino acid sequence alignments were conducted to explore the characteristics of the GP5 in the isolated PRRSV strains. The results revealed that the isolated strains consisted of 199–200 amino acids, among which seven NADC30-like strains exhibited a single amino acid deletion at the decoy epitope or neutralizing epitope positions at amino acid 32, 33, or 42. The 51 PRRSV-2 strains shared 77.1%–99.5% amino acid sequence identity (Supporting Information 5: Table S5). Specifically, 12 strains belonging to sublineage 1.5 shared a relatively high amino acid identity of 93.5%–95% with the representative strain NADC34. In contrast, 24 strains of sublineage 1.8 had 86%–95.5% amino acid identity with the reference strain NADC30. Notably, the VR2332-like and HP-PRRSV-like strains exhibited extremely high similarity to their corresponding reference strains. For instance, eight strains belonging to lineage 5 shared an amino acid similarity of 96.5%–99% with VR2332 and seven strains of sublineage 8.7 demonstrated an amino acid similarity of 98%–100% with the HP-PRRSV strain HuN4.

Analysis of the amino acid sequence mutations in the GP5 protein across different lineages revealed that they occurred at various sites, with mainly in the signal peptide region, decoy epitope, the HVR, the transmembrane (TM) domain, T cell epitope region, and B cell epitope region [28–30] (Figure 4). The amino acid sites of the VR-2332-like and HP-PRRSV-like strains are relatively conserved compared to their corresponding reference strains, with only 1–4 amino acid mutations observed in the B cell epitope region. In contrast, the newly emerging NADC34-like and NADC30-like strains exhibited a relatively higher variation in each functional region compared to their reference strains.

[figure(s) omitted; refer to PDF]

The glycosylation of GP5 plays a pivotal role in the biology of PRRSV, so we analyzed the potential N-glycosylation sites of GP5. Among the 51 PRRSV-2 isolates analyzed, each strain was found to contain three–five N-linked glycosylation sites (NGSs) and the potential NGSs at positions N44 and N51 had a high degree of conservation across all isolates. In comparison to the reference strain VR2332, all isolates belonging to lineage 5 harbored additional NGSs at N30 and N33, and CHJC22102 showed a novel NGS at N34. However, CHJC230203C012 and CHJC230203C011 lacked NGS at N33, but had one at N35. When contrasted with the HP-PRRSV strain HuN4, the isolates of lineage 8 consistently possessed five NGSs, with no new mutations observed in NGSs at N30, N34, or N35. It is noteworthy that the lineage 1 strains contained three–five NGSs, with a significantly higher degree of variability in the NGSs at N30 and N32/33/34. Moreover, the NADC30-like strain CHJC22083 (within sublineage 1.8) and certain NADC34-like strains (within sublineage 1.5) were found to have a distinct glycosylation at N57 (Supporting Information 6: Table S6).

3.4.2. Phylogenetic Analysis and Genetic Diversity of NSP2

To gain insights into the phylogenetic relationships and genetic diversity of the NSP2 gene among the PRRSV isolates, a phylogenetic tree was constructed based on the high-variable region of NSP2. The results revealed that all 28 isolates were classified as PRRSV-2 and could be further divided into three lineages: lineage 5, lineage 8, and lineage 1 (sublineage 1.8; Figure 5A). Sequence alignments indicated that the nucleotide and amino acid sequence homologies between the four strains of lineage 5 and the representative strains of the same lineage VR-2332 were in the range of 99.4%−99.7% and 97.7%–99%, respectively. For the 11 strains in sublineage 1.8, their nucleotide and amino acid sequence homologies with the representative strain NADC30 were 89.2%−91.4% and 84.2%−86.5%, respectively. In the case of the 13 strains of lineage 8, their nucleotide and amino acid sequence homologies with the HuN4 representative strain were 98.4%−99.1% and 96.1%−98.1%, respectively (details are summarized in Supporting Information 7: Table S7 and Supporting Information 8: Table S8).

[figure(s) omitted; refer to PDF]

To further draw the amino acid sequence patterns of NSP2, the 28 isolates were compared with the PRRSV-2 representative strain VR2332. It was found that all four strains of lineage 5 were highly similar to VR2332, and thus, classified as VR2332-like strains. All 13 isolates of lineage 8 had 30 (1 + 29) discontinuous amino acid deletions, typical of HP-PRRSV-like strains. Additionally, the remaining 11 isolates of lineage 1.8 exhibited 131 (111 + 1 + 19) discontinuous amino acid deletions, a characteristic feature of NADC30-like strains (Figure 5B). Interestingly, five isolates of the lineage 1.8, CHQQHR112022, CHLJCD212022, CHLJCD262022, CHLJCD232022, and CHQQHR3002022, initially classified as NADC34-like strains based on GP5 typing, were found to belong to the NADC30-like strains according to the NSP2 sequence alignment results. Moreover, the strain CHJC220523C02 of the same lineage exhibited a 5-amino-acid deletion at positions aa 464–468, similar to the deletion observed in the NADC31 strain, suggesting the occurrence of a novel recombination pattern involving NADC30, NADC31, and NADC34 during the transmission and evolution of PRRSVs.

3.4.3. Homology and Recombination Analysis of the Full-Genome of Four Isolates

Based on the analysis of GP5 and NSP2 sequences, we selected several strains for full-length sequencing and successfully obtained four full-length sequences, namely, CHLJCD212022, CHCQWL12022, CHCQWL22022, and CHCQWL32022. Sequence alignment of these four strains revealed that their nucleotide sequence similarity ranged from 87.7% to 98.8%, with the predicted amino acid sequence similarity between 83.1% and 97.9%. Subsequently, a phylogenetic tree was constructed using the full-length sequences. The results showed that all four strains belonged to subgroup 1.8 (NADC30-like; Figure 6).

[figure(s) omitted; refer to PDF]

To further investigate the potential recombination events, we conducted recombination analyses using RDP4 and SimPlot software. The RDP4 analysis of CHLJCD212022 revealed that this strain was a recombinant virus with three different parental sources. Notably, NADC30 was the major parental strain, while the HP-PRRSV strain JX and IA/2014/NADC34 strain were identified as the minor parental strains (Supporting Information 9: Table S9). SimPlot results corroborated those of RDP4, identifying two recombination events within CHLJCD2122022 (Figure 7A). Recombination event 1 was located at nucleic acid positions 10524–11426 (Figure 7A), and event 2 occurred at positions 13568–14286 (Figure 7A). Phylogenetic analysis based on breakpoints provided consistent results. The nucleotide sequences before position 10524 and after position 14286 were found to belong to sublineage 1.8 (NADC 30-like; Figure 7B), whereas the sequences from 10524 to 11426 were affiliated with sublineage 8.7 (HP-PRRSV-like; Figure 7C), and those from 13568 to 14286 were attributed to sublineage 1.5 (NADC34-like; Figure 7D).

[figure(s) omitted; refer to PDF]

In the case of the three strains CHCQWL12022, CHCQWL22022, and CHCQWL32022, their nucleotide similarity was above 95% and the predicted amino acid similarity exceeded 94%. Therefore, CHCQWL12022 was selected for further recombination analysis. The RDP4 results indicated that NADC30 was the major parental strain of CHCQWL12022, with JX as the minor parental strain, and their recombination occurred at positions 1–501 nt (Supporting Information 9: Table S9). Simplot was used to validate the RDP4 findings (Figure 8A). Through phylogenetic tree analysis of the 1–501 nt breakpoint and the remaining part, it was determined that the strain and NADC30 were in the same branch outside the breakpoint, while the strain of breakpoints 1–501 nt JX and the strain were not in the same branch (Figure 8B,C). According to these observations, we speculate that the minor parents may have evolved from intermediate strains and share homologous with JX, although their exact identities remain to be determined.

[figure(s) omitted; refer to PDF]

4. Discussion

Since its initial discovery in China during the 1990s, the PRRSV has continuously threatened the country’s pig industry due to its high variability and complex epidemiological characteristics. Currently, the prevalent lineages of PRRSV are gradually shifting from HP-PRRSV to strains similar to NADC30 and NADC34 [11, 31]. Particularly in Southern China, NADC30-like strains have become the predominant circulating strains and frequently exhibit recombination [27, 32]. These changes not only increase the complexity of PRRSV but also significantly weaken the protective efficacy of existing vaccines. This study aims to develop a new detection method for PRRSV and integrate it with existing detection technologies to investigate the prevalence of PRRSV in Southwest China. It also aims to deeply analyze the molecular evolutionary patterns of the strains, with the goal of providing a scientific basis for disease prevention and control.

4.1. Advantages and Limitations of the MIRA Method

In contemporary clinical practice, detection techniques for PRRSV primarily encompass virus isolation and identification, RT-PCR, fluorescent qPCR, ELISA, immunoperoxidase monolayer assay (IPMA), and indirect immunofluorescence antibody (IFA) test, among others [22]. These detection methods all necessitate stringent operational procedures; if samples are not managed correctly, they can readily result in false positive or false negative outcomes. The MIRA method developed in this study has shown significant technical advantages. Compared to fluorescent quantitative RT-PCR, MIRA requires simpler equipment, is easier to operate, suitable for on-site detection, and has relatively low costs. These characteristics make it highly applicable in resource-limited environments. Additionally, this technology provides strong support for rapidly and accurately assessing the prevalence of PRRSV. Despite the reduced costs of MIRA technology, the costs are still relatively high, which may be one of the reasons why the technology has not been widely promoted.

4.2. Implications of PRRSV Prevalence and Genetic Variation

In China, PRRSV exhibits a high overall seropositivity rate. For example, studies have reported an average seropositivity rate of 50.89% across four regions of China—northeast, central, east, and north—between 2016 and 2018 [33]. Similarly, the seropositivity rate in southwestern provinces reached 62.56% during 2017–2021 [34]. The vaccinated group maintained a high antibody positive rate across all years, peaking in 2022 at 91.76%, with slight decreases observed in 2023 (87.52%) and 2024 (87.18%). Despite potential influencing factors such as variations in vaccine batches, fluctuations in immune responses, and differences in rearing management, the overall positive rate remained stable, indicating that the existing PRRSV vaccines have reliable immune-inducing effects. In contrast, the unvaccinated group demonstrated significantly lower seropositivity compared to vaccinated counterparts, with antibody positive rates of 17.3% (2021), 23.6% (2023), and 16.1% (2024). This suggests persistent viral circulation, potentially attributable to suboptimal biosecurity protocols or persistent subclinical viral transmission. Notably, the 2024 seroprevalence decline (16.1%) may reflect improved containment measures implemented following the 2023 surge, though sampling size variability across surveillance periods warrants consideration in interpreting these longitudinal fluctuations.

The prevalence of the NADC30-like strain has increased annually, reaching 61.9% in 2023. This trend suggests that the NADC30-like strain has a selective advantage in immune evasion and transmissibility, warranting close attention. Although the NADC34 strain remains less prevalent, its increase to 42.1% in 2022 indicates significant epidemic potential. Commercial vaccines currently used in China are primarily based on the classic PRRSV VR2332 strain (lineage 5) modified live vaccines (MLVs) and attenuated vaccines from lineage 8 [19, 35]. In our study, VR2332-like and highly pathogenic strains exhibited high nucleotide similarity to widely used respPRRS MLV and highly pathogenic vaccine strains. This suggests that vaccine strain reversion may be one of the main reasons for the prevalence and evolution of these strains, thereby exacerbating the epidemic situation of PRRSV. Furthermore, the current vaccines have limited immunogenic efficacy against certain strains such as NADC30-like and NADC34-like strains, revealing issues of immune escape and inadequate vaccine stability.

GP5, as the principal structural protein of PRRSV, occupies a pivotal position in the virus’s antigenicity and immune evasion capabilities [30, 36, 37]. Mutations in the amino acids of GP5 are predominantly clustered in the signal peptide region, decoy epitope, HVR, TM domain, and T-cell epitope regions [28, 38]. Here, our data reveal that NADC30-like and NADC34-like strains exhibit notable mutations and variations in these functional domains, whereas VR2332-like and HP-PRRSV-like strains remain relatively conserved. These variations not only mirror PRRSV’s remarkable adaptive prowess under immune selection pressure but may also be intricately linked to alterations in transmission efficiency and virulence among hosts [39, 40]. These lineage-specific variation patterns imply that distinct strains may possess unique evolutionary trajectories, particularly as changes in the antigenic structure of NADC30-like and NADC34-like strains could significantly bolster their immune evasion capabilities, thereby diminishing the efficacy of existing vaccines.

The glycosylation of the PRRSV GP5 is crucial for viral infection and immune evasion. The presence and diversity of glycosylation sites not only influence the virus’s immunogenicity but may also affect vaccine effectiveness [41–43]. Our study further examined the variations in potential NGSs of the GP5 protein and found that the N44 and N51 sites are highly conserved across all isolates, while glycosylation sites like N30, N32/33, and N34 demonstrate significant strain-specific variability. This finding aligns with previous research [38, 44], indicating that NGSs in the GP5 exhibit distinct variation patterns among different strains, with the conservation of N44 and N51 sites potentially tightly linked to the virus’s biological functions [30, 37, 45]. This dynamic variation in glycosylation sites may not only modulate viral antigenicity but also play a pivotal role in the interaction between the virus and host cells.

NSP2, the most variable nonstructural protein in the PRRSV genome, exhibits unique amino acid deletion patterns that serve as vital markers for lineage classification [46]. In this study, a phylogenetic tree was constructed based on the HVR sequences of NSP2, revealing that the 28 isolates could be categorized into lineage 5, lineage 8, and sublineage 1.8. Each lineage displayed characteristic amino acid deletion patterns consistent with those identified in previous studies; for instance, lineage 8 HP-PRRSV features a 30-amino acid discontinuous deletion and sublineage 1.8 NADC30-like harbors a 131-amino acid discontinuous deletion [46, 47]. Notably, some strains initially classified as NADC34-like based on GP5 (such as CHQQHR112022 and CHLJCD232022) were reclassified as NADC30-like strains based on NSP2 sequences, highlighting potential discrepancies in typing when relying on GP5 versus NSP2. This typing divergence may stem from the differing evolutionary rates and functional constraints imposed on these two proteins within the viral genome. Additionally, our study suggested potential recombination among NADC30, NADC31, and NADC34 strains of PRRSV; regrettably, however, we were unable to confirm the full-length sequence of this strain.

Recombination serves as a significant driving force in PRRSV evolution. Through comprehensive whole-genome analysis, our study identified two recombination events in the CHLJCD212022 strain and one recombination event in the CHCQWL12022 strain, with recombination breakpoints located at nucleotide positions 10524–11426 nt, 13568–14286 nt, and 1–501 nt, respectively. These recombination events encompass ORF1b (nsp12), ORF5 (GP5), 5^′UTR, and ORF1a (nsp1), indicating that these regions may represent recombination hotspots within the PRRSV genome. These hotspots are distributed across 5^′UTR ~ 1500 nt (Nsp1), 5374–8177 nt (Nsp4~Nsp9), and GP2-GP6 [46, 48, 49], which are critical for viral replication and infection. In the recombination patterns of PRRSV, interlineage recombination is a prevalent phenomenon. For example, studies conducted in China and the United States have demonstrated frequent recombination among lineage 1, lineage 2, and lineage 8 [50]. Intralineage recombination also occurs in PRRSV; for instance, recombination within the lineage 1 primarily involves the NSP2 and NSP9 regions [48]. In lineage 1.8, NADC30 serves as the primary parental strain providing the genetic framework, while sublineage 1.5 NADC34 and lineage 8 JX strains contribute essential exogenous fragments. This multiparental recombination pattern may endow the virus with novel antigenic properties or adaptive advantages. These adaptive changes may confer selective benefits in host-to-host transmission and may significantly contribute to NADC30 strains becoming the predominant epidemic strains [15]. Notably, the identification of recombination hotspots offers valuable insights into the evolutionary mechanisms of PRRSV. Especially in high-density farming environments where multiple strains coinfect, the coexistence of different strains may accelerate the occurrence of recombination events. Therefore, in-depth exploration of hotspot regions can not only elucidate the adaptive evolution patterns of the virus but may also provide potential targets for future vaccine development.

Currently, PRRSV typing primarily relies on phylogenetic analysis of GP5 or NSP2 sequences. However, our findings indicate that relying solely on the sequence characteristics of a single protein may lead to inconsistent typing results. For example, some strains in our study exhibited conflicting typing outcomes between GP5 and NSP2, particularly as some samples initially classified as NADC34-like based on GP5 were reidentified as NADC30-like based on NSP2 analysis. This typing discrepancy may reflect differences in functional selection pressure and evolutionary rates between the two proteins; as a viral surface protein, GP5^′s variations are largely driven by host immune selection [30, 51], while NSP2, as a replication-related protein, may undergo variations associated with viral adaptive mutations or recombination events [46, 52]. To enhance the accuracy and consistency of virus typing, future studies should consider combining analyses of GP5, NSP2, and other structural proteins (e.g., GP3 [53]) or nonstructural proteins to establish a more comprehensive typing framework. Furthermore, given the accelerating pace of viral evolution and the frequent emergence of recombinant strains, traditional lineage classification methods may struggle to fully capture the intricate relationships among strains. Consequently, a typing approach based on the full genome sequence may represent a more reliable option, not only improving typing accuracy but also providing valuable information for studies on viral evolution and transmission.

4.3. Future Directions for PRRSV Research and Control

In recent years, significant progress has been made in the study of PRRSV, but its persistent global prevalence and high degree of variability still pose a serious threat to the pig industry. In this study, we found that PRRSV continues to spread in Southwest China, with NADC30-like strains becoming the predominant prevalent strains and the presence of frequent recombination events, which drives the wide diversity of the virus. For the control of PRRSV, biosecurity management is the primary measure, including strengthening the hygiene management of pig farms and strictly controlling the entry and exit of personnel and vehicles. Meanwhile, vaccination is an important means of preventing PRRSV infection, but the protective effect of the vaccine may be affected due to the high variability of the virus. Therefore, future research should strengthen the monitoring of PRRSV epidemiology and track the virus mutation in time to provide a scientific basis for vaccine development and the formulation of prevention and control strategies. In the future, with deeper research and technological advances, we have reason to believe that greater breakthroughs will be made in the research and control of PRRSV, which will provide a strong guarantee for the healthy development of the global pig industry.

5. Conclusion

In summary, this study established a novel method for detecting PRRSV and investigated the genetic characteristics and epidemiology of PRRSV in the southwest region of China. The NADC30-like strain was identified as the predominant circulating lineage in this area, with an increasing number of NADC34-like strains also being detected. Key mutations potentially enhancing immune evasion were observed in GP5 and NSP2, along with recombination events between lineage 1 and lineage 8 and within the lineage 1. These findings provide valuable insights into the evolution of PRRSV and offer support for future disease control strategies and vaccine development.

Author Contributions

Juan Zhang, Xinrong Wang, and Jun Zhou contributed equally to this work.

References

[1] J. J. M. Meulenberg, "PRRSV, the Virus," Veterinary Research, vol. 31 no. 1, pp. 11-21, DOI: 10.1051/vetres:2000103, 2000.

[2] G. Wensvoort, "Lelystad Virus and the Porcine Epidemic Abortion and Respiratory Syndrome," Veterinary Research, vol. 24 no. 2, pp. 117-124, 1993.

[3] E. Albina, "Epidemiology of Porcine Reproductive and Respiratory Syndrome (PRRS): An Overview," Veterinary Microbiology, vol. 55 no. 1–4, pp. 309-316, DOI: 10.1016/S0378-1135(96)01322-3, 1997.

[4] S. Done, D. Paton, "Porcine Reproductive and Respiratory Syndrome: Clinical Disease, Pathology and Immunosuppression," Veterinary Record, vol. 136 no. 2, pp. 32-35, DOI: 10.1136/vr.136.2.32, 1995.

[5] K. D. Rossow, "Porcine Reproductive and Respiratory Syndrome," Veterinary Pathology, vol. 35 no. 1,DOI: 10.1177/030098589803500101, 1998.

[6] J. J. Zimmerman, K.-J. Yoon, R. W. Wills, S. L. Swenson, "General Overview of PRRSV: A Perspective From the United States," Veterinary Microbiology, vol. 55 no. 1–4, pp. 187-196, DOI: 10.1016/S0378-1135(96)01330-2, 1997.

[7] J. H. Kuhn, M. Lauck, A. L. Bailey, "Reorganization and Expansion of the Nidoviral Family Arteriviridae," Archives of Virology, vol. 161 no. 3, pp. 755-768, DOI: 10.1007/s00705-015-2672-z, 2016.

[8] W. H. Lin, K. Kaewprom, S. Y. Wang, "Outbreak of Porcine Reproductive and Respiratory Syndrome Virus 1 in Taiwan," Viruses-Basel, vol. 12 no. 3,DOI: 10.3390/v12030316, 2020.

[9] P. Jiang, P. Y. Chen, Y. Y. Dong, J. L. Cai, B. X. Cai, Z. H. Jiang, "Isolation and Genome Characterization of Porcine Reproductive and Respiratory Syndrome Virus in P. R. China," Journal of Veterinary Diagnostic Investigation, vol. 12 no. 2, pp. 156-158, DOI: 10.1177/104063870001200211, 2000.

[10] L. Zhou, J. Han, H. Yang, "The Evolution and Diversity of Porcine Reproductive and Respiratory Syndrome Virus in China," Veterinary Microbiology, vol. 298,DOI: 10.1016/j.vetmic.2024.110252, 2024.

[11] H. Z. Zhao, F. X. Wang, X. Y. Han, "Recent Advances in the Study of NADC34-Like Porcine Reproductive and Respiratory Syndrome Virus in China," Frontiers in Microbiology, vol. 13,DOI: 10.3389/fmicb.2022.950402, 2022.

[12] Z. Zhou, J. Ni, Z. Cao, "The Epidemic Status and Genetic Diversity of 14 Highly Pathogenic Porcine Reproductive and Respiratory Syndrome Virus (HP-PRRSV) Isolates From China in 2009," Veterinary Microbiology, vol. 150 no. 3-4, pp. 257-269, DOI: 10.1016/j.vetmic.2011.02.013, 2011.

[13] Y. Yu, Q. Zhang, Z. Cao, "Recent Advances in Porcine Reproductive and Respiratory Syndrome Virus NADC30-Like Research in China: Molecular Characterization, Pathogenicity, and Control," Frontiers in Microbiology, vol. 12,DOI: 10.3389/fmicb.2021.791313, 2021.

[14] C. Li, J. Zhao, W. Li, "Prevalence and Genetic Evolution of Porcine Reproductive and Respiratory Syndrome Virus in Commercial Fattening Pig Farms in China," Porcine Health Management, vol. 10 no. 1,DOI: 10.1186/s40813-024-00356-y, 2024.

[15] C. Kong, D. Li, Y. Hu, "The Genetic Variation of Porcine Reproductive and Respiratory Syndrome Virus Replicase Protein NSP2 Modulates Viral Virulence and Persistence," Journal of Virology, vol. 97 no. 3,DOI: 10.1128/jvi.01689-22, 2023.

[16] W. Zhang, X. Wang, H. Zhang, "Comparison of Pathogenicity and Host Responses of Emerging Porcine Reproductive and Respiratory Syndrome Virus Variants in Piglets," Journal of Virology, vol. 98 no. 11,DOI: 10.1128/jvi.01542-23, 2024.

[17] J. Li, S. Tao, R. Orlando, M. P. Murtaugh, "N-Glycosylation Profiling of Porcine Reproductive and Respiratory Syndrome Virus Envelope Glycoprotein 5," Virology, vol. 478, pp. 86-98, DOI: 10.1016/j.virol.2015.02.013, 2015.

[18] J. Li, L. C. Miller, Y. Sang, "Current Status of Vaccines for Porcine Reproductive and Respiratory Syndrome: Interferon Response, Immunological Overview, and Future Prospects," Vaccines, vol. 12 no. 6,DOI: 10.3390/vaccines12060606, 2024.

[19] L. Zhou, X. Ge, H. Yang, "Porcine Reproductive and Respiratory Syndrome Modified Live Virus Vaccine: A “Leaky” Vaccine With Debatable Efficacy and Safety," Vaccines, vol. 9 no. 4,DOI: 10.3390/vaccines9040362, 2021.

[20] C. Chae, "Commercial PRRS Modified-Live Virus Vaccines," Vaccines, vol. 9 no. 2,DOI: 10.3390/vaccines9020185, 2021.

[21] J. Jeong, S. Kim, C. Park, "Commercial Porcine Reproductive and Respiratory Syndrome Virus (PRRSV)-2 Modified Live Virus Vaccine Against Heterologous Single and Dual Korean PRRSV-1 and PRRSV-2 Challenge," Veterinary Record, vol. 182 no. 17,DOI: 10.1136/vr.104397, 2018.

[22] J. Pan, M. Zeng, M. Zhao, L. Huang, "Research Progress on the Detection Methods of Porcine Reproductive and Respiratory Syndrome Virus," Frontiers in Microbiology, vol. 14,DOI: 10.3389/fmicb.2023.1097905, 2023.

[23] H. Chen, C. Sun, Y. Wang, "Rapid Detection of SARS-CoV-2 Using Duplex Reverse Transcription-Multienzyme Isothermal Rapid Amplification in a Point-of-Care Testing," Frontiers in Cellular and Infection Microbiology, vol. 11,DOI: 10.3389/fcimb.2021.678703, 2021.

[24] Y. Wang, J. Niu, M. Sun, "Rapid and Sensitive Detection of Streptococcus iniae in Trachinotus ovatus Based on Multienzyme Isothermal Rapid Amplification," International Journal of Molecular Sciences, vol. 24 no. 9,DOI: 10.3390/ijms24097733, 2023.

[25] R. Gupta, S. Brunak, "Prediction of Glycosylation Across the Human Proteome and the Correlation to Protein Function," Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing, vol. 7, pp. 310-322, 2002.

[26] D. P. Martin, B. Murrell, M. Golden, A. Khoosal, B. Muhire, "RDP4: Detection and Analysis of Recombination Patterns in Virus Genomes," Virus Evolution, vol. 1 no. 1,DOI: 10.1093/ve/vev003, 2015.

[27] T. Tu, Y. Li, G. Zhang, "Isolation, Identification, Recombination Analysis and Pathogenicity Experiment of a PRRSV Recombinant Strain in Sichuan Province, China," Frontiers in Microbiology, vol. 15,DOI: 10.3389/fmicb.2024.1362471, 2024.

[28] Q. Luo, Y. Zheng, Y. He, "Genetic Variation and Recombination Analysis of the GP5 (GP5a) Gene of PRRSV-2 Strains in China From 1996 to 2022," Frontiers in Microbiology, vol. 14,DOI: 10.3389/fmicb.2023.1238766, 2023.

[29] M. Veit, A. K. Matczuk, B. C. Sinhadri, E. Krause, B. Thaa, "Membrane Proteins of Arterivirus Particles: Structure, Topology, Processing and Function," Virus Research, vol. 194, pp. 16-36, DOI: 10.1016/j.virusres.2014.09.010, 2014.

[30] Q. Luo, Y. Zheng, H. Zhang, "Research Progress on Glycoprotein 5 of Porcine Reproductive and Respiratory Syndrome Virus," Animals, vol. 13 no. 5,DOI: 10.3390/ani13050813, 2023.

[31] Y. Ouyang, Y. Du, H. Zhang, "Genetic Characterization and Pathogenicity of a Recombinant Porcine Reproductive and Respiratory Syndrome Virus Strain in China," Viruses, vol. 16 no. 6,DOI: 10.3390/v16060993, 2024.

[32] Z. Guo, X.-X. Chen, R. Li, S. Qiao, G. Zhang, "The Prevalent Status and Genetic Diversity of Porcine Reproductive and Respiratory Syndrome Virus in China: A Molecular Epidemiological Perspective," Virology Journal, vol. 15 no. 1,DOI: 10.1186/s12985-017-0910-6, 2018.

[33] B. Yin, S. Qi, W. Sha, "Molecular Characterization of the NSP2 and ORF5 (ORF5a) Genes of PRRSV Strains in Nine Provinces of China During 2016–2018," Frontiers in Veterinary Science, vol. 8,DOI: 10.3389/fvets.2021.605832, 2021.

[34] P. Zhao, C. Wang, W. Cao, R. Fang, J. Zhao, "Risk Factors and Spatial-Temporal Analysis of Porcine Reproductive and Respiratory Syndrome Seroprevalence in China Before and After African Swine Fever Outbreak," Frontiers in Veterinary Science, vol. 9,DOI: 10.3389/fvets.2022.929596, 2022.

[35] H. Wang, W. Feng, "Current Status of Porcine Reproductive and Respiratory Syndrome Vaccines," Vaccines, vol. 12 no. 12,DOI: 10.3390/vaccines12121387, 2024.

[36] H. Tian, Y. Cheng, J.-Y. Wu, J.-H. He, Y.-J. Shang, X.-T. Liu, "Application of GP5 Protein to Develop Monoclonal Antibody Against Porcine Reproductive and Respiratory Syndrome Virus," Virologica Sinica, vol. 26 no. 4, pp. 267-272, DOI: 10.1007/s12250-011-3198-5, 2011.

[37] W. V. Breedam, H. V. Gorp, J. Q. Zhang, P. R. Crocker, P. L. Delputte, H. J. Nauwynck, "The M/GP5 Glycoprotein Complex of Porcine Reproductive and Respiratory Syndrome Virus Binds the Sialoadhesin Receptor in a Sialic Acid-Dependent Manner," PLOS Pathogens, vol. 6 no. 1, 2010.

[38] K. Fang, S. Liu, X. Li, H. Chen, P. Qian, "Epidemiological and Genetic Characteristics of Porcine Reproductive and Respiratory Syndrome Virus in South China Between 2017 and 2021," Frontiers in Veterinary Science, vol. 9,DOI: 10.3389/fvets.2022.853044, 2022.

[39] B. Fan, X. Liu, J. Bai, T. Zhang, Q. Zhang, P. Jiang, "Influence of the Amino Acid Residues at 70 in M Protein of Porcine Reproductive and Respiratory Syndrome Virus on Viral Neutralization Susceptibility to the Serum Antibody," Virology Journal, vol. 13, 2016.

[40] D. He, X. Niu, A. Jiang, "SwIFN- β Promotes Genetic Mutation of Porcine Reproductive and Respiratory Syndrome Virus in Marc-145," Veterinary Microbiology, vol. 166 no. 1-2, pp. 174-178, DOI: 10.1016/j.vetmic.2013.06.002, 2013.

[41] P. Liu, C. Yue, B. Meng, "Spike N354 Glycosylation Augments SARS-CoV-2 Fitness for Human Adaptation Through Structural Plasticity," National Science Review, vol. 11 no. 7,DOI: 10.1093/nsr/nwae206, 2024.

[42] J. Chaudhari, H. L. X. Vu, "Porcine Reproductive and Respiratory Syndrome Virus Reverse Genetics and the Major Applications," Viruses, vol. 12 no. 11,DOI: 10.3390/v12111245, 2020.

[43] J.-A. Lee, B. Kwon, F. A. Osorio, "Protective Humoral Immune Response Induced by an Inactivated Porcine Reproductive and Respiratory Syndrome Virus Expressing the Hypo-Glycosylated Glycoprotein 5," Vaccine, vol. 32 no. 29, pp. 3617-3622, 2014.

[44] L. Zhou, Y. Yang, Q. Xia, "Genetic Characterization of Porcine Reproductive and Respiratory Syndrome Virus From Eastern China During 2017–2022," Frontiers in Microbiology, vol. 13,DOI: 10.3389/fmicb.2022.971817, 2022.

[45] J.-C. Choi, M.-S. Kim, H.-Y. Choi, "Porcine Reproductive and Respiratory Syndrome Virus Engineered by Serine Substitution on the 44th Amino Acid of GP5 Resulted in a Potential Vaccine Candidate With the Ability to Produce High Levels of Neutralizing Antibody," Veterinary Sciences, vol. 10 no. 3, 2023.

[46] Y. Liu, J. Li, J. Yang, "Emergence of Different Recombinant Porcine Reproductive and Respiratory Syndrome Viruses, China," Scientific Reports, vol. 8 no. 1,DOI: 10.1038/s41598-018-22494-4, 2018.

[47] H. Chang, J. Zheng, Y. Qiu, "Isolation, Identification, and Pathogenicity of a NADC30-Like Porcine Reproductive and Respiratory Disorder Syndrome Virus Strain Affecting Sow Production," Frontiers in Veterinary Science, vol. 10,DOI: 10.3389/fvets.2023.1207189, 2023.

[48] X.-Y. Cui, D.-S. Xia, L.-Z. Luo, T.-Q. An, "Recombination of Porcine Reproductive and Respiratory Syndrome Virus: Features, Possible Mechanisms, and Future Directions," Viruses, vol. 16 no. 6,DOI: 10.3390/v16060929, 2024.

[49] J. Guo, Z. Liu, X. Tong, "Evolutionary Dynamics of Type 2 Porcine Reproductive and Respiratory Syndrome Virus by Whole-Genome Analysis," Viruses, vol. 13 no. 12,DOI: 10.3390/v13122469, 2021.

[50] F. Yu, Y. Yan, M. Shi, "Phylogenetics, Genomic Recombination, and NSP2 Polymorphic Patterns of Porcine Reproductive and Respiratory Syndrome Virus in China and the United States in 2014–2018," Journal of Virology, vol. 94 no. 6, pp. e01813-01819, 2020.

[51] H. Xu, L. Xiang, Y. D. Tang, "Genome-Wide Characterization of QYYZ-Like PRRSV During 2018–2021," Frontiers in Veterinary Science, vol. 9,DOI: 10.3389/fvets.2022.945381, 2022.

[52] B. Li, L. Fang, Z. Xu, "Recombination in Vaccine and Circulating Strains of Porcine Reproductive and Respiratory Syndrome Viruses," Emerging Infectious Diseases, vol. 15 no. 12, pp. 2032-2035, DOI: 10.3201/eid1512.090390, 2009.

[53] S. Jiao, J. Zhang, J. Wang, "Whole-Genome Analysis of the Recombination and Evolution of Newly Identified NADC30-Like Porcine Reproductive and Respiratory Syndrome Virus Strains Circulated in Gansu Province of China in 2023," Frontiers in Veterinary Science, vol. 11,DOI: 10.3389/fvets.2024.1372032, 2024.