Findings

HPAI viruses continue to evolve and pose a threat to animal and human health. Recent HPAI H5 virus variants of different NA subtypes have been implicated in disease outbreaks in poultry in China (H5N1, H5N2, H5N5, H5N6, H5N8) , South Korea (H5N8), Japan (H5N8), Laos (H5N8) and Vietnam (H5N1, H5N6) and an H5N6 subtype virus caused a human fatality in China [1-3]. These variants have previously been referred to as “clade 2.3.4.6” viruses and originate from H5N1 clade 2.3.4 viruses [4,5]. However, the WHO/OIE/FAO H5N1 Evolution Working Group recently designated that these viruses fall into a new clade 2.3.4.4 and not 2.3.4.6 [6]. Of particular significance, these emergent H5 viruses have a propensity for reassortment of the HA gene with different NA subtypes, including NA1, NA2, NA5, NA6, NA8 [7]. Accordingly there is concern that this clade may be spreading more widely [8], and although these variant strains are being detected by diagnostic RT-qPCR assays in current use, they cannot be readily distinguished from other H5N1 HPAI viruses.

Molecular identification of this new clade currently requires conventional RT-PCR amplification of the HA gene followed by nucleotide sequence determination. Likewise, identification of the N subtype requires nucleotide sequence determination following conventional RT-PCR of the NA gene or multiple N6 subtype-specific RT-qPCR assays. Here we report the development of RT-qPCR assays to facilitate the identification of A/duck/Laos/XBY004/2014(H5N6) [1] and other closely related H5N6 HPAI viruses that may have zoonotic potential [3].

Primers and probes for HA and NA specific RT-qPCR assays were designed to complement sequences of the A/duck/Laos/XBY004/2014(H5N6) and related viruses. Other HA sequences that belong to the new proposed clade 2.3.4.4 (Table 1) were identified using Basic Local Alignment Search Tool (BLAST) [9] analysis of A/duck/Laos/XBY004/2014(H5N6) against the GenBank nucleotide sequence database (nr/nt) [10]. Subsequently, sequence alignments were performed using Geneious software (Version 7.1.6, Biomatters). Potential primer and probe regions distinctive of proposed clade 2.3.4.4 viruses were identified from these alignments and candidate primers and probes assessed using the Primer Express 3.0.1 program (Applied Biosystems). The clade specific HA probes contained the codon for amino acid substitution 273H that has been identified as one of the distinguishing amino acids for clade 2.3.4.4 viruses [4]. Two candidate probes were designed using black hole quencher 1 (BHQ1) and minor groove binder (MGB) chemistry, respectively. Primer and probe sequences and optimized concentrations of the new and reference assays are listed in Table 2. These assays were developed using thermocycling conditions compatible with the reference avian influenza (AI) type A and the H5 duplex RT-qPCR reference assays [11], (diagnostic manual, CSIRO Australian Animal Health Laboratory, unpublished). Reactions were performed using AgPath-ID One-Step RT-PCR Kit (Ambion - Applied Biosystems) on an AB7500 Fast instrument using the thermal cycle: 1 cycle of 45°C 10 min, 95°C 10 min followed by 45 cycles of 95°C 15 sec, 60°C 45 sec. For interpretation of results samples producing a cycle threshold (CT) less than 40 were considered positive, CT values 40.1 to 45 were considered indeterminate.

Table 1 HPAI H5 proposed clade 2.3.4.4 strains containing different N types
Full size table
Table 2 Primer and probe oligonucleotides for qRT-PCR assays used in this study
Full size table

The assays used in this study were:

  • AI Type A (generic). This is a generic assay targeting the matrix gene to detect influenza A viruses from different animal hosts, including avian, equine and/or other species. This assay is in routine use for AI diagnosis at the CSIRO Australian Animal Health Laboratory and Australian state veterinary diagnostic laboratories.

  • H5 duplex (generic H5). This is a generic assay to detect H5 subtypes, consisting of two sets of primers and probes targeting two different regions (C-terminus and N-terminus) of the H5 HA gene.

  • H5 2.3.4.4-BHQ1 (candidate clade specific). This assay was designed specifically for detection of new proposed HA clade 2.3.4.4 strains and utilizes a BHQ1 probe.

  • H5 2.3.4.4-MGB (candidate clade specific). This assay was designed specifically for detection of new proposed HA clade 2.3.4.4 strains and utilizes a MGB probe. The same primers were used for both H5 2.3.4.4 assays.

  • AI N6. This assay design was based on A/duck/Laos/XBY004/2014(H5N6) to detect Asian lineage N6-NA of novel A(H5N6) viruses, however it is not specific for all N6 viruses.

Reaction efficiencies and limit of detection (LOD) of the RT-qPCR assays were determined for the A/duck/Laos/XBY004/2014(H5N6) isolate (5 × 109/ml median embryo infective dose) using a serial 10 fold dilution of template and assays were performed in triplicate. The reaction efficiencies were 80.5, 92.0, 87.0, 91.2 and 86.3% for the AI type A, H5 duplex, H5 2.3.4.6-BHQ1, H5 2.3.4.6-MGB and AI N6, respectively. The LOD for all assays was observed at the 10−7 dilution, equivalent to approximately five median embryo infective doses for the given template and displaying the last positive CT values between 37.9 and 38.5 (data not shown).

The capability of the candidate RT-qPCR assays to differentiate proposed clade 2.3.4.4 viruses from other H5 HPAI viruses was evaluated using A/duck/Laos/XBY004/2014(H5N6) as a representative of clade 2.3.4.4 viruses in comparison with other relevant AI reference virus strains (Table 3).

Table 3 Comparison of RT-qPCR assays
Full size table

Laos A(H5N6) isolates were detected by the current RT-qPCR AI type A and H5 duplex assays however, only the H5 2.3.4.4-MGB RT-qPCR was able to distinguish these novel A(H5N6) viruses belonging to the proposed clade 2.3.4.4 strain from other AI viruses including A(H5N1). In contrast, the H5 2.3.4.4-BHQ1 RT-qPCR detected A/duck/Laos/XBY004/2014(H5N6) but was not able to differentiate from some other A(H5N1) clades as detailed in Table 3. The AI N6 qRT-PCT assay differentiated the Laos H5N6 subtype virus from H5N1 strains, thus providing useful rapid diagnostic differentiation from A(H5N1) that co-circulate in domestic poultry in the region (Table 3).

This evaluation was based on A/duck/Laos/XBY004/2014(H5N6) that was available for testing in our laboratory. However, in silico sequence analysis indicated that other N6 and non-N6 viruses in proposed clade 2.3.4.4 were identical in the sequences of the HA primer and probe target region (Table 1), but distinguished from other A(H5N1) by at least one nucleotide change in the probe region. As experimentally proven in this study, the mismatches in the MGB probe target region were sufficient to provide the required analytical specificity for clade 2.3.4.4 virus differentiation.

Accordingly, this sensitive and clade-specific assay will be of particular value for surveillance programs and in diagnostic laboratories where routine DNA sequencing is not readily performed or available.

Availability of supporting data

The data sets supporting the results of this article are included within the article.