Cells and viruses

All cell lines in this study were procured from ATCC: MDCK (CCL34), A549 (CCL-185), and 293T (CRL-3216). Human embryonic kidney cells (293T) and lung adenocarcinoma cells (A549) were grown in Dulbecco’s minimal essential medium (DMEM), enriched with 10% fetal bovine serum, 100 units/ml penicillin, and 100 μg/ml streptomycin sulfate (Life Technologies). Canine MDCK cells were cultivated in Eagle’s minimal essential medium (MEM), supplemented with identical quantities of serum and antibiotics. Three strains of influenza viruses, H1N1 virus A/California/04/2009, H5N1 virus A/dairy cattle/Texas/24-008749-003-original/2024, and A/mink/Spain/22VIR12774-14_3869-3/2022, were rescued using reverse genetics^48,49. Additionally, two DelNS1-H5N1 LAIVs were created, employing HA and NA from the H5N1 viruses. Three amino acids (RKR) were deleted from the polybasic cleavage motif (KRRKR↓GLF) in the hemagglutinin (HA) protein to attenuate the DelNS1-H5N1 LAIVs. All LAIVs were constructed and rescued according to the protocols described here and in our previous report³⁸. The viral gene segments underwent amplification and were subsequently cloned into pHW2000 plasmids, resulting in eight pHW2000 plasmids. These plasmids were transfected into mixtures of 293T/MDCK cells, and the rescued virus was amplified in either MDCK cells or embryonated chicken eggs. For the pre-existing immunity experiment, the H3N2 and H1N1 DelNS1 LAIVs were derived from A/Guangdong-Maonan/SWL1536/2019 (H1N1) and A/HongKong/2671/2019 (H3N2).

Generation and passage of DelNS1 viruses

293T/MDCK cells were transfected with eight pHW2000 plasmids, containing the DelNS1 segment and the other seven influenza virus genomic segments. In addition, an NS1 expression plasmid was added. After incubating overnight, the DNA mixture was removed, and MEM with 1 μg/ml N-tosyl-L-phenylalanine chloromethyl ketone (TPCK)-treated trypsin (Sigma) was added. The virus supernatant was collected 72 h later and labeled as passage 0 (P0) virus. It was then passaged in MDCK cells or embryonated chicken eggs. For the H5N1 reassortant DelNS1 viruses DelNS1-cH5N1 and DelNS1-mH5N1, six plasmids containing CA4-DelNS1 internal genes and H5N1 HA and NA plasmids derived from either A/dairy cattle/Texas/24-008749-003-original/2024 virus or A/mink/Spain/22VIR12774-14_3869-3/2022 virus were utilized for virus rescue. To confirm the deletion of the NS1 gene in all DelNS1 viruses, reverse transcription-PCR (RT-PCR) and sequencing were performed.

Growth kinetics and stability test

Confluent MDCK or A549 cells in 24-well plates were infected with viruses at specific multiplicities of infection (MOI). Following a one-hour adsorption period, the viral supernatant was removed. The cells were then washed twice with phosphate-buffered saline (PBS) and overlaid with MEM containing 1 μg/ml TPCK-treated trypsin. Cells were subsequently incubated at the temperatures indicated. Supernatants were collected at various time points, and titers were determined using plaque assays in MDCK cells. For the growth study in eggs, 8-day-old embryonated chicken eggs were infected with viruses at 7 × 10⁴ pfu or 7 × 10⁵ pfu, viruses were collected at 48 h post infection, and titers were determined using plaque assays in MDCK cells.

For the genetic stability test of DelNS1-cH5N1 and DelNS1-mH5N1 viruses, the rescued virus passage 0 (P0) DelNS1 viruses, DelNS1-cH5N1 and DelNS1-mH5N1were serially passaged 10 times at 33 °C in 9-day-old chicken embryonated eggs to evaluate their genetic stability. After each passage, viruses were diluted 100-fold in PBS and the passage was repeated. Following multiple passages (P5 and P10), viral RNA was extracted by viral RNA extraction kit (Qiagen). cDNA was synthesized using the High-Capacity cDNA reverse transcription kit (Invitrogen) and uni-12 primer. Viral genes were amplified by PCR using gene-specific primers, as detailed in the Supplementary Data 1 file. The complete genomes of the DelNS1 viruses were subsequently sequenced using the Sanger method to identify any potential mutations. Whole genome sequences were summited to GenBank (GenBank accession numbers: DelNS1-mH5N1-p1: PV124973-PV124980, DelNS1-mH5N1-p10: PV124993-PV125000).

Plaque assay and TCID₅₀ assay

For plaque assay, viruses were serially diluted by a factor of 10 and then added to confluent MDCK cells in 6-well plates. These plates were then incubated at 37 °C for one hour. After this, the supernatant was discarded, and the cells were washed twice with PBS. Next, the cells were overlaid with 1% MEM agarose that contained 1 μg/ml TPCK-treated trypsin. The plates were then incubated at a temperature of 33 °C for a period of 48 h. Following this incubation period, the plates were fixed with a 10% PBS-buffered formaldehyde solution for a minimum of 2 h. Finally, plaques were made visible by staining with a 1% crystal violet solution.

For TCID₅₀ assay, confluent MDCK cells in 96 well plates were washed and overlaid with MEM containing 1 μg/ml TPCK-treated trypsin, and then incubated with serially diluted virus. After 3 days, the presence of virus was detected by HA assay. TCID₅₀ values were calculated by the Reed–Muench Method.

Animal immunization and challenge

For mouse immunization, 6–8-week-old female BALB/c mice were anesthetized with ketamine and xylazine and then immunized intranasally with 2 × 10⁵ pfu (low titer dose, Fig. S7) or 1 × 10⁶ pfu (standard dose, all other experiments) of DelNS1-H5N1 LAIV vaccine. Three weeks after the immunization, blood serum was collected for ELISA and microneutralization assays. At week 4, mice were euthanized and bronchoalveolar lavage fluid (BAL), nasal washes, lungs and spleens were collected for ELISA and T cell response assays.

To determine the mouse lethal dose 50 (LD₅₀), 6–8-week-old female BALB/c mice were anesthetized with ketamine and xylazine and then intranasally inoculated with 20 μl of different titers of cattle H5N1 (n = 4 for each group; 5000 pfu, 500 pfu, 50 pfu, 5 pfu, <1 pfu), or intranasally inoculated with 20 μl of different titers of mink H5N1 (n = 4 for each group; 10⁴ pfu, 10³ pfu, 10² pfu, 10 pfu). Body weight and survival were monitored daily for 15 days. The mice were euthanized if they lost more than 20% of their initial body weight. LD₅₀ values were calculated based on the Reed-Muench method⁵⁰.

For virus challenge, after 4 weeks immunized mice were challenged intranasally with H5N1 virus (cattle H5N1: 5 × 10³ pfu, mink H5N1: 1 × 10⁴ pfu). Body weight and disease signs were monitored daily. At day 4 post infection, lungs and nasal turbinates (NT) or other organs were collected for virus titration and histopathological and immunohistochemical study and the remaining challenged mice were monitored for body weight and disease signs. For the virus titer detection, the collected organs (whole organ) were homogenized in 1 ml PBS, centrifuged and supernatants were then used for virus titration using TCID₅₀ assay (virus titters were calculated as TCID₅₀/ml). For the pre-existing immunity and immunity longevity studies immunization, sampling, challenge and sacrifice timings were adjusted, as detailed in the relevant sections.

For hamster immunization, 6–8-week-old male hamsters (Golden Syrian Hamster) were anesthetized with ketamine and xylazine and then immunized intranasally with 5 × 10⁶ pfu of DelNS1-cH5N1 LAIV vaccine. Three weeks after the immunization, blood serum was collected for ELISA and microneutralization assays. For virus challenge, after 4 weeks immunized hamsters were challenged intranasally with H5N1 virus (cattle H5N1, 1 × 10⁴ pfu). Body weight and disease signs were monitored daily. At 4 days post infection, lungs and nasal turbinates (NT) or other organs were collected for virus titration and the remaining challenged hamsters were monitored for body weight and disease symptoms. For the virus titer detection, the collected organs were then homogenized in 2 ml PBS, then centrifuged and supernatants were used for virus titration using TCID₅₀ assay.

All animal experiments were approved by the Committee on the Use of Live Animals in Teaching and Research of the University of Hong Kong (HKU) (CULATR) (# 22-092). All animal experiments related to H5N1 were performed in a biosafety level 3 laboratory at HKU. Animals were housed and bred under an AAALAC International accredited program at the Center for Comparative Medicine Research (CCMR), The University of Hong Kong. Animals were housed in open cages or individually ventilated cages under a 12:12 dark/light cycle within environmentally controlled rooms. Standard pellet feed and water were provided ad libitum.

Microneutralization (MN) assays

Serum from immunized BALB/c mice were treated with receptor-destroying enzyme (RDE; Denka Seiken). RDE-treated serum was then twofold serially diluted in 96-well plates and mixed with 100 50% tissue culture infective doses (TCID₅₀) of virus, then incubated for 1 h at room temperature. The mixture was added to confluent MDCK cells in MEM supplemented with 1 μg/ml TPCK-treated trypsin in a 96-well plate and incubated at 33 °C for 3 days. The presence of virus was confirmed by hemagglutination (HA) assay. The neutralizing titer was defined as the reciprocal of the highest dilution of serum that completely neutralizes infectivity of 100 TCID₅₀ of virus in MDCK cells.

Enzyme-linked immunosorbent assay (ELISA)

DelNS1-cH5N1 (4 × 10⁵ pfu/well), DelNS1-mH5N1 (4 × 10⁵ pfu/well), H5N1 HA1 (100 ng/well) (A/chicken/Ghana/AVL-763_21VIR7050-39/2021, Sino Biological 40933-V08B), the H3N2 DelNS1 LAIVs (4 × 10⁵ pfu/well), or the H1N1 DelNS1 LAIVs (4 × 10⁵ pfu/well) was coated onto MaxiSorp 96-well plates (Thermo Scientific) at 4 °C overnight. Plates were washed with PBS and blocked with 3% BSA at room temperature for 1 h. For serum IgG detection, heat-inactivated serum was tenfold serially diluted and added to the plate, then incubated at 37 °C for 30 min. For the detection of IgA in bronchoalveolar lavage fluid (BAL) or nasal washes, BAL and nasal washes were diluted fivefold and tenfold, respectively, in PBST, then undiluted or diluted BAL and nasal washes were added to the plate and incubated at 37 °C for 30 min. The plate was washed 5 times and then incubated with secondary antibody reagent at 37 °C for 30 min. After washing, color development solution was added and incubated at 37 °C for 15 min. Stop solution was added and absorbance at 450 nm was measured using a Varioskan LUX Multimode Microplate Reader (ThemoFisher). The endpoint for serum IgG detection and the area under the curve (AUC) for IgA detection were calculated using GraphPad Prism v10.

Intracellular cytokine staining (ICS) and NP₁₄₇ tetramer staining

To study tissue-resident memory T cells, 4 weeks after immunization, and following intravascular labeling by using injection of anti-CD45-PE-Cy5 (Biolegend-103110, Clone-30-F11, Lot-B392814) at the dose of 2 μg per mouse, 5 min prior to sacrifice, lungs and spleens of mice were collected. Splenocytes were isolated and homogenized through cell strainers (BD) and resuspended in RPMI medium (10% FBS and P/S). Lung tissue was chopped and digested in RPMI solution with Collagenase D (1 mg/ml) (Roche) for 1 h at 37 °C. Red blood cells were lysed by addition of Lysing solution (BD). After washing with RPMI, cells were counted and resuspended in RPMI. Cells were stimulated with RPMI solution containing influenza A virus-NP147 or -NP55 peptide at a final working concentration of 1 μg/ml per peptide (GenScript, NP147-TYQRTRALV, NP55-RLIQNSITIERMVLS). After 1 h, brefeldin A (BFA) was added to cells and incubated overnight. Cells were then washed with FACS buffer (2% FBS in PBS) and stained with Zombie (BioLegend-423102, Lot-B432360), anti-CD8-AF700 (BioLegend-100730, Clone-53-6.7, Lot-B433204), anti-CD4-FITC (BioLegend-100406, Clone-GK1.5, Lot-B374032), anti-CD69-BV711 (BioLegend-104537, Clone-H1.2F3, Lot-B406939) and CD103-BV421 (BioLegend-121422,Clone-2E7, Lot-B425334) for 30 min at 4 °C. Cells were washed and fixed with Perm/wash buffer (BD), and then stained with anti-IFNγ-APC (BioLegend-505810, Clone-XMG1.2, Lot-B335090) overnight at 4 °C. Cells were then washed twice with FACS buffer and resuspended in FACS buffer. For NP₁₄₇ tetramer staining, after IV labeling and isolation of cells from tissues, lung cells and splenocytes were stained with Zombie (BioLegend-423102, Lot-B432360), anti-CD8-BV785 (BioLegend-100750, Clone-53-6.7, Lot-B420449) and NP₁₄₇ tetramer-APC (MBL International-TB-M534-2, Lot-T2405016) for 30 min at 4 °C. Cells were washed and finally resuspended in FACS buffer. The samples were acquired with an Agilent NovoCyte Quanteon analyzer, and the generated data analyzed with FlowJo V10. Cytokine production was calculated by subtracting the no peptide control background value. The gating strategy is shown in Fig. S8.

RNA isolation and RT-qPCR

Total RNA was isolated from BALB/c mouse lungs using RNAzol RT reagent (MRC) according to the manual, and first-strand cDNA was generated from total RNA using HiScript III All-in-one RT SuperMix Perfect for qPCR (Vazyme) following the protocol provided. Real-time PCR was conducted with the TB Green Premix Ex Taq II (Takara) and gene-specific primers on a LightCycler® 480 System (Roche). PCR conditions were as follows: initial denaturation: 95 °C for 30 s, 40 cycles of amplification: 95 °C for 10 s, 60 °C for 10 s, 72 °C for 10 s, and melting curve analysis: 65 °C to 97 °C at 0.1 °C/s. Expression of target genes was normalized to internal reference genes (mouse β-Actin) and the comparative Ct method utilized to calculate the cytokine expression profile. All the primers have been described previously and showed in the Supplementary Data 1 file⁵¹.

Histopathology

Following fixation in 10% PBS buffered formalin, organs of infected BALB/c mice and Golden Syrian Hamsters were processed into paraffin blocks. Haematoxylin and eosin (H&E) staining was performed on tissue sections and subsequently examined under light microscopy. Pathological scores were given based on the pathological changes observed in the tissues according to a previous report⁵².

Immunohistochemistry

Formalin-fixed paraffin-embedded lung sections, nasal turbinate sections and brain sagittal sections of infected BALB/c mice and Golden Syrian Hamsters were labelled for immunohistochemical staining using a rabbit polyclonal anti-influenza A virus nucleoprotein antibody (GeneTex-GTX125989, Lot-44902). Briefly, sections were deparaffinized and rehydrated. Heat-induced epitope retrieval was performed using Antigen Unmasking Solution, Tris-Based (Vector Laboratories, H-3301-250) in a pressure cooker for about 1 min once the cooker had pressurized. Sections were then incubated with hydrogen peroxide to inactivate endogenous peroxidase activities prior to applying the primary antibody (1:2000) at 4 °C overnight, followed by a secondary antibody (Dako EnVision+ System- HRP Labelled Polymer Anti-Rabbit, Agilent-K4003) for 30 min at room temperature. Diaminobenzidine chromogen detection was completed using Dako DAB for 2 min (Agilent-K3468). Sections were then rinsed with deionized water and counterstained using hematoxylin. Finally, slides were cleared through a gradient of alcohol and xylene prior to mounting and coverslipping. Sections were examined under a light microscope.

Statistical analysis

Statistical analysis was carried out using GraphPad Prism. Data were presented as the mean values ± SD of at least 3 replicates. Statistical significance was analyzed by either Student’s t-test or one-way analysis of variance (ANOVA) followed by Dunn’s multiple comparisons test. For all tests: ****p <  0.0001, ***p <  0.001, **p <  0.01, *p < 0.05, ns-not significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.