Outbreak Severity of Highly Pathogenic Avian Influenza A(H5N8) Viruses Is Inversely Correlated to Polymerase Complex Activity and Interferon Induction

Phylogenetic analysis influenza A(H5Nx) viruses.The sequences of all influenza A(H5Nx) subtype viruses collected between 1996 and 2018 were downloaded from GISAID (44). The sequence data set was curated to remove sequences with more than 1% ambiguous nucleotides, less than 80% sequence length, or sequences that had any of the eight segments missing. The sequence set was annotated with the WHO/OIE/FAO H5 nomenclature using LABELv0.5.2 with module H5v2015 (45). The influenza A(H5N1) A/goose/Guangdong/1/1996 (H5N1) virus-like lineage was systematically downsampled by a WHO/OIE/FAO-designated clade to obtain a representative influenza A/Goose/Guangdong/1996 (H5N1) virus-like background data set (n = 500 viruses), with the influenza A(H5N8) viruses from the Netherlands isolated in the 2014-15 and 2016-17 outbreaks automatically included in the final data set for each segment. To account for the possibility of reassortment with the internal genes of LPAI viruses, each of the influenza A(H5N8) viruses from the Netherlands was compared to the complete global database of avian influenza viruses for each segment with BLAST+ (46). The 50 most closely related LPAI sequences were included in the final data set for each segment (n = 597), with the exception of HA and NA. The final data sets were aligned with MAFFT v7.397 and trimmed to the start of the mature protein (47). Maximum likelihood phylogenetic trees were constructed for each segment with RAxML 8.2.12 under the GTR+GAMMA substitution model with 100 bootstrap replicates and rooted to the influenza A/Goose/Guangdong/1996 (H5N1) virus (48). Phylogenetic trees were annotated and visualized using ggtree (49). Pairwise mean interclade patristic distance distributions were calculated using the ape package in R (50). The files are available at https://github.com/AMC-LAEB/avflu_h5n8_polymerase.

Cell culture.Human HEK293T (ATCC CRL-3216) cells and HEK293T RIG-I-deficient cells (kindly provided Ben Berkhout, Amsterdam UMC) were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco) supplemented with 10% fetal calf serum (FCS; Sigma), 1% nonessential amino acids (NEAA; Gibco), 100 IU/ml penicillin, and 100 IU/ml streptomycin (Pen/Strep; Gibco) incubated at 37°C and 5% CO₂. The HEK293T RIG-I knockout cells were produced using CRISPR-Cas gene knockout. HEK293T cells were transfected with three px458 plasmids expressing the CRISPR-enhanced green fluorescent protein (CRISPR-eGFP) protein and three independent guide RNA sequences targeting genomic RIG-I at a ratio of 1:1:1. After 48 h, cells were harvested, and single eGFP fluorescent cells were sorted into a 96-well plate. Expanding cell colonies were cultured for 6 weeks under the same conditions as for regular HEK293T cells. Cells were harvested and frozen in aliquots. Gene knockout was checked by gene-specific PCR that confirmed the absence of the genomic region containing the RIG-I gene, using a positive control for DNA extraction efficiency and PCR. Neither luciferase background levels nor absolute polymerase complex activity was significantly different between wild-type HEK293T and RIG-I knockout cells (data not shown). DF-1 cells from Gallus gallus (ATCC CRL-12203) were cultured in DMEM supplemented with 10% FCS, 1% NEAA, 100 IU/ml penicillin, and 100 IU/ml streptomycin and incubated at either 37°C and 5% CO₂ or 39°C and 5% CO₂. LMH cells from Gallus gallus (ATCC CRL-2218) were cultured in DMEM supplemented with 10% FCS, 1% NEAA, 100 IU/ml penicillin, and 100 IU/ml streptomycin and incubated at 37°C and 5% CO₂. QT6 cells from Coturnix coturnix japonica (ATCC CRL-1708) were cultured in medium 199 with Earle salts and 2 mM l-glutamine (M199; Gibco) supplemented with 5% FCS, 1% chicken serum (Sigma), 1% of 100× GlutaMAX-I (Gibco), 5% tryptose phosphate broth (TPB; Gibco), 100 IU/ml penicillin, and 100 IU/ml streptomycin and incubated at 37°C and 5% CO₂. CGBQ cells from Anser anser (ATCC CRL-169) were cultured in Ham’s F-12 nut mix with 2 mM l-glutamine (Ham’s F-12K; Gibco) supplemented with 15% FCS, 1% NEAA, 0.15% sodium bicarbonate (Gibco), 100 IU/ml penicillin, and 100 IU/ml streptomycin and incubated at 37°C and 5% CO₂. Duck embryo cells from Anas platyrhynchos domesticus (ATCC CRL-2118) were cultured in minimum essential medium with Hanks salts and 2 mM l-glutamine (MEM; Gibco) supplemented with 10% FCS, 1 mM sodium pyruvate (100 mM; Gibco), 100 IU/ml penicillin, and 100 IU/ml streptomycin and incubated at 37°C and 5% CO₂. Madin-Darby canine kidney (MDCK) cells from Canis familiaris (ATCC CRL-2935) were cultured in Eagle’s minimal essential medium (EMEM; Lonza) supplemented with 10% FCS, 100 IU/ml penicillin, 100 IU/ml streptomycin, 2 mM l-glutamine (Gibco), 0.15% sodium bicarbonate (Gibco), 10 mM HEPES (Lonza), and 1% NEAA (Gibco) and incubated at 37°C and 5% CO₂.

Primary duck embryo fibroblast were isolated from 13-day-old embryonated eggs, cultured in medium 199 (M199; Lonza), 10% tryptose phosphate broth (TPB; MP Biomedicals), 10% FCS, 10% 100 IU/ml penicillin, and 100 IU/ml streptomycin, and incubated at 37°C and 5% CO₂. Duck lung homogenate (DLH) cells were isolated from the lungs of 21-day-old duck embryonated eggs, grown in DMEM (Gibco) supplemented with 10% FCS (Sigma), 1% NEAA (Gibco), 100 IU/ml penicillin, and 100 IU/ml streptomycin (Pen/Strep; Gibco), and incubated at 37°C and 5% CO₂.

Polymerase activity/minigenome assay.The open reading frames (coding region) of the PB2, PB1, PA, and NP genes were cloned into the pPPI4 expression vector (51) using Gibson Assembly (New England BioLabs). A plasmid coding for a model viral RNA (vRNA), consisting of the firefly luciferase (FF) open reading frame flanked by the noncoding regions of segment 8 of influenza A virus under the control of a human pPolI was used for minigenome assays (52). The production of mRNA of this FF vRNA is solely possible by transcription of the produced vRNA by the influenza virus polymerase complex. Therefore, the FF activity is a measure of the amount of mRNA produced and thereby of the polymerase complex activity. Transfection of pRL (Promega), a plasmid containing the Renilla luciferase gene under a cytomegalovirus promoter which is stably expressed in avian and mammalian cells independent from the influenza polymerase complex, served as an internal control to normalize variation in transfection efficiency and sample processing. Human (HEK293T) or avian (DF-1, LMH, QT6, CGBA, or duck embryo) cells were seeded 1 day prior to the experiment into 96-well plates. Twenty-five nanograms of the FF reporter plasmid, 50 ng of each of the plasmids encoding PB2, PB1, and PA, and 100 ng of NP, and 2 ng of the pRL expression plasmid in 50 μl Opti-MEM (Gibco, Thermo Fisher) were mixed with 50 μl Opti-MEM containing either Lipofectamine 2000 (Invitrogen, Thermo Fisher) for HEK293T cells or TransIT-X2 (Mirus) for avian cells in a 1:3 ratio and incubated for 20 to 30 min at room temperature. Twenty microliters of the transfection mixture was added to each well. Each transfection was performed in quadruplicate in at least three independent experiments. Twenty-four hours after transfection, or 48 h for duck embryo cells, luminescence was measured using the dual-luciferase reporter assay system (Promega) using a GloMax luminometer, according to the manufacturer’s instructions (Turner BioSystems).

Quantitative PCR luciferase vRNA, cRNA, and mRNA.Transfection of cells was performed as described above for the minigenome experiments. Briefly, 25 ng of the FF reporter plasmid, 50 ng of each of the plasmids encoding PB2, PB1, and PA, 100 ng of NP, and 2 ng of the pRL expression plasmid in 50 μl Opti-MEM (Gibco, Thermo Fisher) were mixed with 50 μl Opti-MEM containing Lipofectamine 2000 (Invitrogen, Thermo Fisher). HEK293T cells cultured in a 24-well format were transfected with 100 μl transfection mixture and incubated for 48 h; after this, medium was aspirated, and cells were lysed in 500 μl TRIzol (Thermo Fisher) and stored at –80°C until RNA isolation. Transfections were performed in duplicate, and for each transfection, separate cDNA synthesis and qPCR were performed twice. RNA isolation from TRIzol was performed according to the manufacturer’s protocol. To ensure complete removal of any residual DNA (genomic/plasmid), the isolated RNA was treated with 100 units of Turbo DNase (Thermo Fisher) for 2 h, and RNA was isolated a second time using TRIzol.

For cDNA synthesis, 1 μg of purified total RNA was used, to which 1 μl of 50 μM oligo(dT)₂₀ (Thermo Fisher) or 1 μl of 2 μM gene-specific primer and 1 μl of 10 mM dinucleoside triphosphate (dNTP) mix (Thermo Fisher) were added in a total volume of 13 μl. Oligo(dT)₂₀ was used to generate mRNA, gene-specific primers for luciferase vRNA (5′-AGTAGAAACAAGGGTGTTTTTTATCA-3′), and cRNA (5′-TATGGGCATTTCGCAGCCTACCGTGGTGTT-3′) (adapted from references 26 and 27). This mixture was heated to 65°C for 5 min and then incubated on ice for at least 1 min. Next, 4 μl of 5× first-strand buffer (FSB; Thermo Fisher), 1 μl of 0.1 M dithiothreitol (DTT; Thermo Fisher), 1 μl RNaseOUT (Thermo Fisher), and 1 μl of 200 U/μl SuperScript III reverse transcriptase (Thermo Fisher) were added to each reaction mixture, incubated at 55°C for 60 min and 75°C for 15 min, and cooled to 4°C. For the real-time quantitative PCR, 2 μl of cDNA was used, along with 10 μl of 2× SYBR FAST qPCR master mix for LightCycler480 (LC480) (catalog no. KK4610; Kapa Biosystems), 4 μl PCR-grade water; 2 μl of 2 μM forward primer and 2 μl of 2 μM reverse primer were used for each reaction. For the detection of luciferase vRNA, cRNA, and mRNA luciferase, qPCR primers were used (forward, 5′-TATGAACATTTCGCAGCCTACCGTAGTGTT-3′; reverse, 5′-CCGGAATGATTTGATTGCCA-3′) (26, 27). For quantification of the household gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA, qPCR primers were used (forward, 5′-AAAATCAAGTGGGGCGATGCT-3′; reverse, 5′-GGGCAGAGATGATGACCCTTT-3′) (53). Real-time quantitative PCR using Kapa SYBR FAST qPCR mix for LC480 was performed on a LC480 (Roche), according to the manufacturer’s protocol. Gene expression relative to the negative control was calculated using the ΔΔC_T method (54), using GAPDH as a household gene.

IFN-β reporter assay.The influenza A virus ribonucleoprotein complex (RNP) was reconstituted in HEK293T cells using vRNA encoding NA of the A/Puerto Rico/8/34 (H1N1) virus instead of the FF reporter. Similar to the minigenome assays, 50 ng of each of the plasmids encoding PB2, PB1, and PA, 100 ng of NP, 25 ng of NA of A/Puerto Rico/8/34 (H1N1) virus, and 2 ng of the Renilla luciferase expression plasmid in 50 μl Opti-MEM were mixed with 50 μl Opti-MEM containing Lipofectamine 2000 in a 1:3 ratio and incubated for 20 to 30 min at room temperature. HEK293T cells grown in a 24-well format were transfected at 80% to 90% confluence and incubated for 48 h, after which cells were aspirated and resuspended in 500 μl TRIzol. Total RNA was extracted using the Direct-Zol RNA miniprep kit (Zymo Research), according to the manufacturer’s instructions. Four hundred nanograms of total RNA with 25 ng of IFN-β reporter plasmid (kindly provided A. J. W. te Velthuis [20, 28]) and 2 ng of the pRL expression plasmid in 50 μl Opti-MEM were mixed with 50 μl Opti-MEM containing Lipofectamine 2000 in a 1:3 ratio and incubated for 30 min. As a positive control, we used poly(I·C) (Thermo Fisher) in 10-fold dilutions at final concentrations of 1 μg to 10 ng in 20 μl of transfection mixture. HEK293T and HEK293T RIG-I knockout cells grown in a 96-well format were transfected at 80% to 90% confluence with 20 μl of the transfection mixture for each well. Each transfection was performed in quadruplicate in at least three independent experiments. At 48 h after transfection, luminescence was measured using the dual-luciferase reporter assay system (Promega) using a GloMax luminometer, according to the manufacturer’s instructions (Turner BioSystems).

Viruses.Rescue plasmids for an isolate from the 2014-15 outbreak [A/Chicken/Netherlands/emc-3/2014 (H5N8)] were described previously (29, 55). The eight gene segments of an isolate from the 2016-17 outbreak [A/Eurasian Wigeon/Netherlands/4/2016 (H5N8)] were amplified from RNA by one-step reverse transcription-PCR (RT-PCR; Thermo Fisher) using specific primers and were cloned in the bidirectional reverse-genetics plasmid pHW2000, as described before (56, 57). Recombinant viruses were rescued by reverse genetics upon transfection of HEK293T cells, as previously described (57). For the infection experiments, 4 recombinant viruses were generated, as follows: full recombinant A/Chicken/Netherlands/emc-3/2014 virus; full recombinant A/Eurasian Wigeon/Netherlands/4/2016 virus; recombinant A/Chicken/Netherlands/emc-3/2014 virus with the PB2, PB1, PA, and NP gene segments derived from A/Eurasian Wigeon/Netherlands/4/2016 virus; and a recombinant A/Eurasian Wigeon/Netherlands/4/2016 virus with PB2, PB1, PA, and NP gene segments derived from A/Chicken/Netherlands/emc-3/2014 virus. Virus stocks were propagated in and titrated on MDCK cells, as described below.

Titrations.MDCK cells were inoculated with 10-fold serial dilutions of virus stocks or supernatant of DLH- or DEF-inoculated primary cells. The cells were washed with phosphate-buffered saline (PBS) 1 h after inoculation, cultured in infection medium consisting of EMEM supplemented with 100 IU/ml penicillin, 100 IU/ml streptomycin, 2 mM l-glutamine (Gibco), 0.15% sodium bicarbonate (Gibco), 10 mM HEPES (Gibco), and 1% NEAA (Gibco), and incubated at 37°C and 5% CO₂. Three days after inoculation, the supernatants of the cell cultures were tested for agglutinating activity using turkey red blood cells (tRBCs) as an indicator of virus replication. Infectious virus titers were calculated by the method of Reed and Muench (58).

Viral replication kinetics.DEF and DLH cells were inoculated with a multiplicity of infection (MOI) of 0.001. The supernatant was harvested at 12, 24, 36, 48, and 72 h postinfection. Viral loads were determined by titration in MDCK cells, as described above.

IFN-β qPCR on HPAI A(H5N8) virus-inoculated primary duck cells.DEF and DLH primary cells were plated in 6-well plates at 1 to 2 days prior to inoculation. Cells were washed with PBS and inoculated in duplicate for DEF cells and triplicate for DLH cells with an MOI of 0.001 for each of the 4 recombinant viruses in infection medium consisting of M199 and DMEM, respectively, containing 100 IU/ml penicillin and 100 IU/ml streptomycin. Uninfected DEF and DLH cells were used as a negative control. Cells were aspirated at 24 h postinfection and were lysed in 500 μl lysis buffer using the Total RNA isolation kit (Roche). RNA was isolated according to the manufacturer’s protocol. cDNA synthesis was performed as described above, except that 250 ng total RNA was used for cDNA synthesis using 1 μl of 50 μM oligo(dT)₂₀ (Thermo Fisher). Quantitative PCR was performed as described above, using duck IFN-β qPCR primers (forward, 5′-ACCTCCTCAACCAGCTCAA-3′; reverse, 5′-GAAGTGTTGGATGCTCCTGA-3′) and duck GAPDH qPCR primers (forward, 5′-ATGTTCGTGATGGGTGTGAA-3′; reverse, 5′-CTGTCTTCGTGTGTGGCTGT-3′) (53). IFN-β gene expression relative to the negative control was calculated with the ΔΔC_T method (54) using GAPDH as a household gene.

Statistical analysis.The polymerase complex activity of viruses from the 2014-15 and 2016-17 outbreaks, the polymerase complex activity of reassortants thereof, the IFN-β reporter activity, and the expression levels of vRNA, cRNA, and mRNA were compared by using a Mann-Whitney t test (Prism 8.0.2; GraphPad). IFN-β expression levels in inoculated primary duck cells and viral titers were compared by using a two-tailed unpaired t test (Prism 8.0.2; GraphPad). A P value of <0.05 was considered statistically significant. *, P < 0.05; **, P < 0.005; ***, P < 0.0005, ns not significant.