Detection and discrimination of influenza B Victoria lineage deletion variant viruses by real-time RT-PCR

Background During the 2016/17 influenza season, influenza B/VIC lineage variant viruses emerged with two (K162N163) or three (K162N163D164) amino acid (aa) deletions in the haemagglutinin (HA) protein. There are currently five antigenically distinct HA proteins expressed by co-circulating influenza B viruses: B/YAM, B/VIC V1A (no deletion), B/VIC V1A-2DEL (2 aa deletion) and two antigenically distinguishable groups of B/VIC V1A-3DEL (3 aa deletion). The prevalence of these viruses differs across geographical regions, making it critical to have a sensitive, rapid diagnostic assay that detects and distinguishes these influenza B variant viruses during surveillance. Aim Our objective was to develop a real-time RT-PCR (rRT-PCR) assay for detection and discrimination of influenza B/VIC lineage variant viruses. Methods We designed a diagnostic assay with one pair of conserved primers and three probes specific to each genetic group. We used propagated influenza B/VIC variant viruses and clinical specimens to assess assay performance. Results This rRT-PCR assay detects and distinguishes the influenza B/VIC V1A, B/VIC V1A-2DEL, and B/VIC V1A-3DEL variant viruses, with no cross-reactivity. This assay can be run as a multiplex reaction, allowing for increased testing efficiency and reduced cost. Conclusion Coupling this assay with the Centers for Disease Control and Prevention’s Human Influenza Virus Real-Time RT-PCR Diagnostic Panel Influenza B Lineage Genotyping Kit results in rapid detection and characterisation of circulating influenza B viruses. Detailed surveillance information on these distinct influenza B variant viruses will provide insight into their prevalence and geographical distribution and could aid in vaccine recommendations.


INTRODUCTION
Influenza B viruses co-circulate with influenza A strains during the annual influenza season, contributing to overall mortality and morbidity of influenza epidemics. In the 1980's, influenza B viruses evolved into two genetically and antigenically distinct lineages represented by B/Victoria/2/1987 (VIC) and B/Yamagata /16/1988 (YAM), that co-circulate in human beings during influenza seasons worldwide [1][2][3][4][5]. During the 2016-2017 influenza season, the Centers for Disease Control and Prevention (CDC) detected influenza B/VIC viruses in the United States (U.S) that were antigenically distinct from the World Health Organization (WHO) recommended vaccine virus, B/Brisbane/60/2008 (VIC) [6]. Genetic analysis confirmed these viruses have a deletion of six nucleotides in the hemagglutinin (HA) gene resulting in a two amino acid (AA) deletion at positions 162 and 163 (corresponding to nucleotide positions 529-534) [6]. This B Victoria lineage two AA deletion virus (V1A.1; K162N163 deletion) has since spread and been detected worldwide. Two B/Vic lineage variants with three AA deletions (V1A. 2 and V1A.3) at AA positions 162-164 (K162N163D164, and nucleotide positions 529-537) emerged shortly after and subsequently via parallel evolution. The 3DEL viruses have since been detected in Asia, Africa, Europe and America [7]. Hemagglutination inhibition (HI) assays demonstrated that both the V1A.1 and V1A 3DEL variant influenza B viruses are antigenically distinct from each other and to the no-deletion B/VIC virus (V1A) (Table 1) The most effective method for prevention and control of influenza infection is vaccination [3,4,8]. Licensed seasonal vaccines are updated annually and the WHO makes recommendations on the composition of influenza virus vaccines on the basis of surveillance, laboratory, and clinical observations [3]. This process occurs twice a year, in February for the northern hemisphere and in September for the southern hemisphere. It is important to identify the most prevalent influenza A subtype and B lineage viruses through influenza surveillance for influenza vaccine selection, especially for people who may not have been exposed to a subtype/lineage of influenza virus [9]. Identifying the optimal viruses to include in the next season's influenza vaccines is a significant challenge for the influenza surveillance network known as the Global Influenza Surveillance and Response System (GISRS). A diagnostic assay that allows for rapid identification of these genetically and antigenically distinct Influenza B viruses is beneficial to understanding how the prevalence of these viruses varies within different geographic regions of the world and will provide data that could be leveraged to guide relevant vaccine strain selection.
A real time RT-PCR (rRT-PCR)-based B/Victoria lineage deletion detection assay (Vic deletion assay) for detection and discrimination of these antigenically drifted B/Vic deletion variant viruses is detailed herein. The Vic deletion assay consists of a set of conserved primers and deletion-specific dual-labeled hydrolysis probes and has high sensitivity and specificity to distinguish between V1A, V1A.1 V1A.2 and V1A.3 groups without cross-reactivity, allowing for multiplex reactions for maximum testing efficiency. This assay is currently the only available method, to our knowledge, that can distinguish these four B/VIC genetic groups, outside of a pyrosequencing technique and a conventional RT-PCR method, both recently published [10,11]. Real-time RT-PCR equipment is readily available in the majority of diagnostic and surveillance laboratories, it is highly sensitive, specific, and extremely rapid, making it the optimal assay for diagnostic and surveillance purposes. When used in conjunction with the

Influenza Viruses, Antigenic analysis and RNA Extraction
Influenza viruses tested in this study were grown to high titer in either Madin-Darby Canine Kidney (MDCK) cells or embryonated chicken eggs (ECE) [12]. Infectious virus in culture supernatants or allantoic fluids was measured by tissue culture infectious dose-50% (TCID50/ml) or egg-infectious dose-50% (EID50 /ml), respectively [13]. Influenza B virus isolates Influenza B virus lineages were confirmed using antigenic characterization by HI assay on MDCK-propagated virus isolates and genetic sequence analysis (Table1, Figure 1).
The antigenic characteristics of virus isolates were determined by HI tests using post-infection antisera. The HI test was performed as described previously (Kendal and Cate, 1983) [14].
Viral RNA was extracted from 100µl of supernatant or allantoic fluid and eluted into 100µl of RNA elution buffer using MagNA Pure Compact RNA isolation kit on a MagNA Pure Compact instrument (Roche Applied Science) [15].

The B/Victoria lineage deletion detection assay primers/probes
Oligonucleotide primers and probes of Vic deletion assay were designed based on available nucleotide sequence data from GenBank database of National Centers for Biological Information (NCBI) and the Global Initiative on Sharing Avian Influenza Data (GISAID).
The Vic deletion assay includes a single set of conserved amplification primers and three deletion-specific dual-labeled hydrolysis probes, including VIC 2_Del, Vic 3_Del and Vic No_Del probes. Three probes, targeted on deletion region of the HA gene of influenza B viruses, were designed to specifically detect and differentiate V1A.1, V1A 3DEL and the V1A genetic group (No deletion) viruses (B/Vic No_Del), respectively (Figure S1); Vic 2_Del and Vic No_Del probes were designed using BHQplus TM dual-labeled hydrolysis probes (BHQplus) that were labeled at the 5'-end with the reporter molecule 6-carboxyfluorescein (FAM) or Hex and CAL Fluor Red 610 and with Blackhole Quencher™ 1 (BHQ™1) (FAM or Hex) or BHQ™2 (CAL Fluor Red 610) at the 3'-end. The Vic 3_Del probe was labeled at the 5'-end with the reporter molecule 6-carboxyfluorescein (FAM) and (BHQ™) 1 at the 3'-end, and included a triplet of Locked Nucleic Acids (LNA) [16,17] that centered on the mismatch bases from V1A.1 and V1A viruses.
Primer probe sequence specificity was also evaluated by sequence analysis of 51,759 gene segments available in NCBI or GISAID databases. Primers were designed to have annealing temperatures of approximately 60°C and probes were designed to have annealing temperatures of approximately 68°C using PrimerExpress 3.0 software (Applied Biosystems, Foster City, USA). Primers were synthesized by the Biotechnology Core Facility at the CDC (Table 2).

rRT-PCR Reaction Conditions
Reaction conditions for rRT-PCR were based upon the FDA-approved CDC Flu rRT-PCR Dx Panel [15,18]. PCR reaction parameters of the Vic deletion assay were optimized using Invitrogen SuperScript™III Platinum® One-Step quantitative RT-PCR (qRT-PCR) Kits (Singleplex) and TaqPath™ qPCR Multiplex Master Mix (Triplex) kits (Life Technologies) on the Applied Biosystems TM (AB) 7500 Fast Dx Real-Time PCR instrument. All rRT-PCR reactions performed had a total reaction volume of 25µl with primer and probe reaction concentrations at 0.8µM and 0.2µM, respectively. Thermocycling rRT-PCR conditions were as follows: 50°C for 30 min, Taq activation at 95°C for 2 min and 45 cycles of 95°C for 15 sec and 55°C for 30 sec.
All analytical performance data were collected on an ABI 7500 Fast Dx Real-time PCR instrument. Increases in fluorescent signal were registered during the annealing step of the reaction. All data were analyzed with the sequence detector software (SDS) v1.4.1 (Life Technologies).

Analytical Sensitivity and Specificity
In order to demonstrate rRT-PCR performance of the VIC deletion assay, five B/Vic lineage The analytical specificity was further assessed using these five B/Vic lineage deletion viruses and two B/Yam lineage viruses, 24 seasonal influenza A and two avian influenza A viruses.

Assay performance on clinical specimens
To demonstrate the performance of the Vic deletion assay against clinical specimens, 167 clinical specimens including 67 B/Vic, 30 B/Yam and 30 influenza A and B negative, and 30 seasonal influenza A (H1N1pdm-15 and H3N2-15) as determined by the CDC rRT-PCR Flu Panel, were tested with the Vic deletion assay (Table 5).

Real time RT-PCR assay establishment
The Vic deletion assay is an rRT-PCR assay developed using the ABI 7500 Fast Dx Real-time PCR system, and is used for the qualitative detection and characterization of B/Vic virus RNA in respiratory specimens from patients presenting with influenza-like illness (ILI).
The Vic deletion assay employs one pair of conserved primers and three probes specific to each genetic group ( Figure 2). The probes target the deletion region of AA position 162-164 (corresponding to nucleotide position 529-537) of the HA gene in influenza B viruses.
The AA 162-164 deletion region contains multiple adenines and repeating sequences, thus creating a challenge for designing probes for this region. We found that unmodified RT-PCR probes were unsuccessful in targeting the deletion region due to challenges in the surrounding sequence patterns (data not shown). BHQplus probes contain stabilizing chemistry that allows probe oligonucleotides to be a shorter length which more easily targets regions with challenging sequence patterns [19]. We used BHQplus chemistry to successfully design probes to the V1A and V1A.1 influenza B viruses (Table 2). When designing a probe for detection of the V1A 3DEL viruses, we attempted both a BHQplus design and a Minor Groove Binding (MGB) probe for increased sequence specificity, but neither of these probe modifications were successful (data not shown). Locked Nucleic Acids (LNAs) have been shown to increase stability and sequence mismatch detection [16,17], and we found that using a triplet of LNAs centered on the mismatch allowed for successful detection of the V1A.2 and V1A.3 viruses.

Analytical Sensitivity and Specificity
The performance of the Vic deletion assay was evaluated by comparisons to the performance of the universal influenza B assay (InfB) from CDC Flu rRT-PCR Dx Panel. The InfB assay is designed for detection of the nonstructural (NS) gene in all influenza B viruses by targeting highly conserved regions of the NS protein of the influenza B virus.

Analytical Sensitivity
Analytical performance studies, evaluated by testing ten-fold serial dilutions of RNAs extracted  (Table 3).
Analytical sensitivity of the Vic deletion assay is shown in Table 4 (Table S1).

Analytical Specificity (Exclusivity)
Analytical exclusivity was evaluated with high titer influenza B VIC and YAM lineage viruses (Table 4). No cross-reactivity was observed when V1A viruses were tested with the VIC 2_Del and VIC 3_Del assays. Specificity was also demonstrated in V1A.2 and V1A.3 viruses tested with the VIC 2_Del assay and vice versa. Cross reactivity from the VIC No_Del assay was not observed when used to test V1A.1, V1A.2 and V1A.3 viruses. Likewise, the three assays did not react with the B/YAM lineage viruses tested (Table 4).
In order to demonstrate the absence of cross reactivity with influenza A viruses subtypes,

Clinical performance of B/Victoria lineage deletion detection assay
To evaluate the clinical performance of the Vic deletion assay, we tested the assay on clinical specimens received at the CDC during influenza surveillance. The Vic deletion assay was successful in classifying the 77 B/Vic clinical specimens into their respective genetic groups (30 V1A-like, 27 V1A.1-like and 20 of V1A.2 and V1A.3-like), and as expected, the rRT-PCR results were confirmed by genetic analysis. All 30 B/Yam, 15 each of A/H1N1pdm09 and A/H3N2, as well as 30 influenza A and B negative samples tested negative for all B/Vic deletion assay targets (Table 5).

DISCUSSION
The Vic deletion assay presented here is intended for the qualitative detection of the influenza B/Victoria lineage HA gene deletion variant viruses using rRT-PCR technology. The analytical and clinical performance of the Vic deletion assay in either the single-or tri-plex configuration demonstrates that the assay is highly efficient, sensitive, and comparable to the gold standard CDC rRT-PCR Flu Panel InfB assay [18]. We have further shown that the primer and probes of the VIC deletion assay are specific to V1A, V1A.1 and V1A.2 and V1A.3 viruses, and do not cross react with B/Yam viruses, seasonal influenza A, or avian influenza A viruses including HPAI A/H5N1 and Asian lineage A/H7N9 influenza viruses.
The key genetic distinction between V1A, V1A.1, V1A.2 and V1A.3 viruses is within the same nucleic acid region, presenting a challenge for rRT-PCR probe design. We used two chemical modifications in probe design, LNA and BHQplus chemistry, which allowed for greater stability and specificity. Including LNAs in real-time PCR probes improves detection of mismatches significantly [15,16]. The Vic 3_Del probe was designed and optimized using a triplet LNA approach, as probes labelled with a triplet of LNA residues centered on the mismatch provide greater discriminatory power than probes with a single LNA modification [16,17]. The V1A and V1A.1 probes were designed using BHQplus chemistry, allowing for stabilization and enhanced mismatch detection with an overall shorter probe length [19]. We found that unmodified probes were not sufficient for this assay.
During development of the Vic deletion assay, we also evaluated a ZEN_MGB probe that includes an internal ZEN TM quencher located nine nucleotides away from the 5' FAM reporter dye in addition to a MGB residue quencher at the 3' end of the probe (Integrated DNA Technologies. Inc. Coralville, IA, U.S). The ZEN_MGB probes performed comparably to the probes presented here in discriminating these influenza B/Vic viral variants [20]. Thus, both BHQplus and ZEN_MGB fluorescent hydrolysis probe quencher chemistries can be used to synthesize probes for the Vic deletion assay.
Sequence alignments using HA gene sequences from 2010-2019 demonstrated the conserved primers to be stable with no conserved genetic changes observed (Data not shown). Although the areas chosen for the conserved amplification primers and the deletion type-specific probes are currently stable, genetic changes due to rapidly virus evolution and the variable nature of RNA viruses may require periodic updates of the B Vic deletion assay primer and probe sequences.