Establishment of a multiplex RT-PCR assay to detect different lineages of swine H1 and H3 influenza A viruses

Abstract

Classical swine H1N1, emerging European avian-like H1N1 and human-like H3N2 lineages are co-circulating in the swine population in China. The reverse transcriptase polymerase chain reaction (RT-PCR) assay is an effective method for use in influenza surveillance. In this study, a multiplex RT-PCR method was developed for simultaneous identification of hemagglutinin (HA) genes derived from the three lineages of swine influenza viruses. Three primer sets were designed and aimed specifically at HA genes of these viral lineages. The specificity of the assay showed that the established methods could efficiently differentiate the HA genes of classical swine H1N1, European avian-like H1N1, and human-like H3N2 viruses while other viruses such as classical swine fever virus, porcine reproductive and respiratory syndrome virus, pseudorabies virus, and porcine circovirus type 2, could not be detected. The assay showed a sensitivity of 1 × 10^2.5 50% egg infectious dose for each virus lineage. The comparison of the results with those obtained from the analysis of 300 swine tracheal swab samples by means of virus isolation showed a high level of agreement. This multiplex RT-PCR method provides a rapid and specific swine influenza diagnostic tool that also has the potential for investigating the epidemiology of different lineages of swine influenza virus prevalent currently in China.

Introduction

Pigs are proposed to be vessels for the reassortment of human and avian influenza viruses and are the best candidates for the generation of pandemic influenza viruses [1]. The transmission of 2009 swine origin influenza H1N1 virus to humans, further increases the threats to public health [2–4]. Currently, classical H1N1 and human-like H3N2 swine viruses are prevalent in pigs throughout large parts of China [5]. A recent report revealed the emergence of avian-origin European H1N1 swine influenza virus (SIV) in China [6]. The fact that many subtypes of influenza viruses infect pigs in China emphasizes the importance of influenza surveillance in the pig population.

To date, various diagnostic methods for the detection of SIVs have been established. Conventional laboratory diagnosis involves virus isolation in embryonated chicken eggs or on Madin-Darby canine kidney (MDCK) cells from clinical specimens followed with subtype determination by hemagglutination inhibition (HI) and neuraminidase inhibition (NI) tests using monospecific antiserum for each subtype [7]. Immunofluorescent staining and enzyme-linked immunosorbent assays have also been applied to detect virus nucleoprotein antigen [8]. These methods are time consuming with poor sensitivity. Reverse transcriptase polymerase chain reaction (RT-PCR) and multiplex RT-PCR assays have been used to detect influenza A viruses in infected humans and pigs [9–12]. Applying these methods, SIVs from clinical samples could be detected and subtyped into H1, H3, N1, and N2 in a single tube reaction. No reports have demonstrated the use of multiplex RT-PCR assays to detect and subtype different lineages of classical H1N1, European avian-1ike H1N1 and human-like H3N2 SIVs. The purpose of this study was to establish a one tube assay that could identify different subtypes and lineages of SIVs circulating in China thereby providing a useful tool to investigate the surveillance of SIVs.

Materials and methods

Viruses

The viruses used in this study included eight classical SIVs, four European avian-like SIVs, seven H3N2 SIVs, eight H1N2 SIVs, one H5N1 SIV, and one H9N2 virus stored in our lab (Table 1). The plasmid containing HA gene from the pandemic H1N1 virus was kindly provided by Dr. Yuelong Shu, the Chinese National Influenza Center. The influenza A viruses were all propagated in the allantoic cavities of 9–11-day-old embryonated chicken eggs. The EID₅₀ of each of the viruses was checked according to the Reed and Muench formula [13]. Other viruses, including classical swine fever virus (CSFV), porcine reproductive and respiratory syndrome virus (PRRSV), pseudorabies virus (PRV), and porcine circovirus type 2 (PCV-2), were used for the specificity tests of the RT-PCR.

Table 1 Influenza viruses used in this study

Samples

Three hundred tracheal swabs stored at −70°C were taken at random from a sample pool which was collected from apparently healthy domestic pigs at abattoirs in Fujian, Guangdong, and Shandong Provinces from 2006–2009. In addition, nine 5-week-old BALB/c mice were divided randomly into 3 equal groups. Three SIVs strains A/Swine/Guangdong/1/05 (Sw/GD/1/05, classical H1N1); A/Swine/Fujian/204/07 (Sw/FJ/204/07, European avian-like H1N1); and A/Swine/Guangdong/7/06 (Sw/GD/7/06, human-like H3N2) were inoculated intranasally in groups of mice, respectively. Lung tissues from the mice on 3 days post-inoculation (d.p.i.) were collected to evaluate the established multiplex RT-PCR assay.

Primer sequences

The HA sequence data of different lineage SIVs were obtained from the influenza database at http://www.flu.lanl.gov and aligned using the MegAlign program (DNAStar Inc., Madison, WI, USA) for the design of primers specific for HA genes of classical H1N1, European avian-like H1N1, and human-like H3N2 virus (Table 2). The three primer sets were analyzed with OLIGO 6.0 primer design software to ensure they could be used together in a multiplex format.

Table 2 Primers used in the multiplex RT-PCR assays

Viral RNA extraction and multiplex RT-PCR assays

Viral RNA was isolated from infectious allantoic fluid or MDCK cell cultures using Trizol LS according to a previously described procedure [14]. Briefly, 250 μl of the allantoic fluid was mixed with 750 μl of Trizol LS agent (Invitrogen, Carlsbad, CA, USA). Chloroform (200 μl) was added and the suspension was centrifuged for 10 min at 12,000×g. The RNA-containing aqueous phase was precipitated with an equal volume of isopropanol, maintained at 25°C for 10 min and centrifuged at 13,000×g for 10 min at 4°C. The RNA pellet was washed with 1 ml 75% ethanol, centrifuged at 7,500×g for 5 min at 4°C. The supernatant was removed, and the RNA dried, followed by resuspension in 25 μl of diethyl-pyrocarbonate (DEPC)-treated deionized water.

The multiplex RT-PCR was performed in a reaction mixture (25 μl) containing 5 μl 5× Reaction Buffer, 0.5 μl dNTP Mix (10 mM each dNTP), 1 μl 25 mM MgSO4 (Promega, Madison, WI, USA), 2.5 U AMV Reverse Transcriptase (Promega), 30 U RNase inhibitor (Promega), 2.5 U Taq polymerase (Promega), 3 μl primer mix (0.4 μmol each primer) (Table 2), and 3 μl RNA template. RNase-free water was added to make the volume up to 25 μl. Reverse transcription was carried out at 45°C for 45 min with AMV RT inactivation at 94°C for 2 min, followed by 40 cycles of PCR amplification involving denaturing at 94°C for 30 s, annealing at 58°C for 40 s, and extension at 72°C for 1 min. The PCR ended with a final extension step at 72°C for 10 min.

Agarose gel electrophoresis

Four microlitres of the amplified products were loaded onto a 1% (w/v) agarose gel containing 0.5 μg/ml ethidium bromide and electrophoresed in 1× TAE buffer. After the electrophoresis, the DNA bands were visualized by UV trans-illumination.

Sequencing

The PCR products were sequenced directly. In brief, the PCR products were purified from agarose gels with an AxyPrep DNA Gel Extraction Kit (Axygen Scientific Inc., Union City, CA, USA) according to the manufacturer’s instructions. Size-specific PCR products obtained from the multiplex RT-PCR were sequenced in both directions by the Beijing Genomics Institute (China) and analyzed with the ClustalW multiple sequence alignment programs to evaluate the specificity of the assay.

Results

Establishment of multiplex RT-PCR

Three SIVs strains Sw/GD/1/05 (classical H1N1), Sw/FJ/204/07 (European avian-like H1N1), and Sw/GD/7/06 (human-like H3N2), previously identified and subtyped [6] were selected as reference strains. Three primer sets were applied in the multiplex RT-PCR assays (Table 2). The primer sets for HA genes permitted amplification of a 536 bp amplicon for classical swine H1N1, 756 bp from avian-like H1N1, and 991 bp from human-like H3N2. The multiplex RT-PCR methods were performed for the aforementioned SIV reference strains, and the products visualized on 1.0% (w/v) agarose gels (Fig. 1 lane 1). The specificity of the three primer sets were also tested separately by using a single RT-PCR assay (Fig. 1 lane 2–4). The specificity of RT-PCR products was confirmed by sequencing (data not shown).

Fig. 1

Detection of different viruses by multiplex RT-PCR assay. M molecular marker, Lane 1 a mixture of classical swine H1N1 virus (Sw/GD/1/05), European avian-like H1N1 virus (Sw/FJ/204/07), and human-like H3N2 virus (Sw/GD/7/06), lane 2 classical swine H1N1 virus (Sw/GD/1/05), lane 3 human-like H3N2 virus (Sw/GD/7/06), lane 4 European avian-like H1N1 virus (Sw/FJ/204/07), lanes 5–7 classical swine H1N1 viruses (Sw/GD/33/06, Sw/GD/628/06, Sw/GD/446/06), lane 8 European avian-like H1N1 virus (Sw/SD/101/08), lanes 9–10 human-like H3N2 virus (Sw/SD/133/07, Sw/FJ/43/07), lane 11 H1N2 SIV (Sw/GD/1222/06), lane 12 H5N1 SIV (Sw/GD/271/03), lane 13 H9N2 SIV (Sw/SD/3/03), lane 14 PRRSV, lane 15 PCV-2, lane 16 CSFV, lane 17 negative control

Specificity test of the multiplex RT-PCR

The specificity of the three primer sets was confirmed by RT-PCR using templates extracted from different subtype SIVs and other swine infectious viruses. All viruses belonging to the three lineages, including eight classical SIVs, four European avian-like SIVs, and seven H3N2 SIVs, were detected by multiplex RT-PCR, and each amplicon was of the correct size. Representative amplicons are shown in Fig. 1. H1N2 viruses and a plasmid containing the HA gene of the pandemic H1N1 virus were also detected. None of the other viruses, including H5N1 SIV, H9N2 SIV, PRRSV, PCV-2, PRV, and CSFV yielded the expected RT-PCR products following amplification (Fig. 1 lane 12–16).

Sensitivity test of the multiplex RT-PCR

The sensitivity of the multiplex RT-PCR was determined by testing RNA extracted from 10-fold serially diluted reference virus. The detection limits of classical swine H1N1, European avian-like H1N1, and human-like H3N2 were 1 × 10^1.5, 1 × 10^2.5 and 1 × 10^2.5 EID₅₀/100 μl, respectively (Fig. 2a). In order to determine the sensitivity of the multiplex RT-PCR with a mixture of viruses, a viral mixture containing 1 × 10^4.5 EID₅₀/100 μl of each virus was serially diluted 10-fold in MEM media. All viruses were detected simultaneously at a concentration of 1 × 10^2.5 EID₅₀/100 μl (Fig. 2b).

Fig. 2

Sensitivity of multiplex RT-PCR. Sensitivity for classical swine H1N1 virus (SW/GD/1/05), European avian-1ike H1N1 virus (SW/FJ/204/07), and human-like H3N2 virus (Sw/GD/7/06) (a) and a mixture of SW/GD/1/05, SW/FJ/204/07, Sw/GD/7/06 (b). a Lanes 1–5 reactions performed with 10-fold serially diluted samples (1 × 10^4.5–1 × 10^0.5 EID₅₀/sample). Lane N.C. negative control. b Lane M: molecular marker. Lanes 1–4 reactions performed in 10-fold serial dilution of three viruses (from 1 × 10^4.5 –1 × 10^1.5 EID₅₀ of each virus/sample). Lane N.C. negative control

Assay of specimens by multiplex RT-PCR

In order to further evaluate the multiplex RT-PCR method, we analyzed nine experimental tissue specimens and 300 tracheal swabs, and compared the results with those obtained by virus isolation as described previously (6). The test results are shown in Table 3. Each of the nine mouse lung samples tested positive by PCR with the size of the amplicon matching the predicted size of the bands. Of the 300 tracheal swabs, four samples were found to contain virus, two of the samples were positive for classical swine H1N1 virus, one sample was positive for European avian-like virus and the last sample was found to contain H3N2 SIV. The comparisons of these results with those obtained by virus isolation were agreement. In order to further confirm whether the PCR products amplified by multiplex RT-PCR were the actual expected HA subtype of influenza virus, we sequenced the amplicons. All amplified DNAs had the correct sequence for the corresponding subtype (data not shown).

Table 3 Diagnosis of experimental specimens by multiplex RT-PCR

Discussion

In China, sustained surveillance has shown that classical swine H1N1, European avian-like H1N1, and human-like H3N2 subtypes co-circulate in the porcine population [5, 6]. We report here the development of a multiplex RT-PCR assay to rapidly detect and subtype SIVs. This assay is capable of simultaneously differentiating the three HA genes derived from different viral lineages including classical swine H1N1, European avian-like H1N1, and human-like H3N2 SIVs.

The specificity of the assay established here was confirmed using different subtype SIVs and other infectious swine viruses. The PCR products from the three pairs of primers were the actual HA genes of each SIV lineage, including the HA gene of H1N2 and pandemic H1N1 virus SIVs which belongs to the classical swine H1N1 lineage. The results of the sensitivity test for the multiplex RT-PCR showed that European avian-like H1N1 and human-like H3N2 were both detected at a level of 1 × 10^2.5 EID₅₀/100 μl, while the classical swine H1N1 could be detected at a concentration of 1 × 10^1.5 EID₅₀/100 μl. Except for the human-like H3N2, the sensitivity of the multiplex RT-PCR for the other two viruses was similar under single and mixed virus conditions.

Various detection methods aimed at the genome and antisera associated with SIVs have been used to investigate their prevalence in the pig population [7], among which the multiplex RT-PCR was found to be one of the most rapid, specific, and sensitive approaches for detection of SIVs in clinical samples [11]. Although the European avian-like swine H1N1 virus has prevailed in the European pig population since the late 1970s, it was not detected in Chinese swine until recently [6, 15]. The emergence of these SIVs and their associated diseases provides a strong reason for their surveillance. The assay established in this study facilitated the detection of European avian-like swine H1N1 virus and could be a suitable assay for an investigation into the prevalence of swine influenza A viruses in China. The specificity and sensitivity of the assay established here should be further validated and evaluated on a larger quantity of clinical specimens and compared with the conventional virus isolation method. The preliminary results show the potential application of this system for the rapid detection of HA derived from different lineages of the influenza A virus commonly co-infecting in the swine population. Furthermore, the established multiplex RT-PCR could be modified easily if a novel influenza virus emerged in swine populations in future.