{"paper_id":"28eb164d-3291-4e2f-bfee-5b9d6648b30d","body_text":"This is a preprint and has not been peer reviewed. Data may be preliminary.\nBroad mucosal and systemic immunity in mice induced by intranasal booster with a novel recombinant adenoviral based vaccine protects against divergent influenza A virus\nAbstract\nThe development of broad-spectrum universal influenza vaccines and optimisation of vaccination strategies to address the threats posed by pandemics and emerging influenza viruses are critical for public health. In this study, an adenovirus type 5 vector-based influenza vaccine carrying the hemagglutinin (HA) stem of H1, HA stem of H3, and neuraminidase of N1 from the influenza virus was constructed. Immune responses were evaluated in mice using various vaccination strategies: prime-only (intramuscular [IM] or intranasal [IN]) and prime-boost (IM+IN). Compared to the prime-only strategy, the prime-boost strategy significantly enhanced the systemic immune response, inducing higher levels of antigen-specific IgG, mucosal IgA, and T cell immunity in the spleen and lungs. Furthermore, the IN boosting strategy provided complete protection in mice challenged with the H1N1-PR8, rgH3N2-X31, and rgH5N1-Vietnam viruses, significantly reducing viral loads in the lungs and alleviating lung tissue pathologies. In conclusion, this study elucidates potential avenues for the development and application of universal influenza vaccines using customised mucosal boosting strategies.\n1. Introduction\nInfluenza virus infections remain a major global public health concern. Seasonal epidemics lead to 3–5 million cases of severe illness and an estimated 300,000–600,000 deaths annually. 1,2 Influenza viruses are classified into two groups (groups 1 and 2) based on the combination of surface glycoproteins, hemagglutinin (HA) and neuraminidase (NA), with only H1N1 and H3N2 subtypes causing seasonal epidemics. Vaccination is one of the most effective preventative measures; however, current commercialised flu vaccines offer varying protection against seasonal strains, ranging from 10 to 60%. 3 Therefore, there is an urgent need for a universal influenza vaccine that is resistant to antigenic drift and can alleviate the burden of seasonal influenza infections.\nCurrently approved influenza vaccines function primarily by inducing antibodies against the surface proteins of the virus, mainly the HA head and NA, offering protection against the vaccine strain with minimal cross-protection against other influenza strains or subtypes. 4 The HA head is immunodominant but prone to antigenic drift, whereas the HA stem is highly conserved across influenza subtypes, making it a promising target for a universal influenza vaccine. 5-7 The HA stem mediates viral fusion with the host cell membrane, facilitates viral entry, and contains broadly neutralising epitopes. Antibodies targeting the stem may cross-react with various influenza virus strains and subtypes, thereby offering protection against multiple influenza viruses. 8,9 The NA head contains an enzymatically active site that hydrolyses sialic acid receptors to prevent self-aggregation and promote the release of progeny viral particles. NA, a secondary antigen in influenza vaccines, provides cross-protection against strains carrying different HA antigens. 10-14\nInfluenza primarily infects the respiratory tract and spreads through respiratory droplets and aerosols. The viral load in oropharyngeal mucosal sites is a key determinant of disease transmission. Currently, most approved flu vaccines are administered via intramuscular (IM) injection, inducing a strong systemic immune response, but poor mucosal immunity in the respiratory tract. To address the challenges posed by emerging respiratory pathogens, vaccines that induce both mucosal and innate immune responses are urgently required. 15,16 Compared with IM vaccines, intranasal (IN) vaccines can induce local mucosal immune responses, which can block infection and transmission of respiratory pathogens. 17,18 Recombinant human adenovirus type 5 (rHAdV5) is a replication-defective respiratory virus that induces robust cellular, humoral, and mucosal immune responses and is a promising antigen delivery system for combating mucosal viruses, including influenza. Extensively tested as a gene therapy and vaccine vector, rHAdV5 has a high safety profile in humans and has been widely used as a gene delivery vector for the development of vaccines against infectious diseases, such as Ebola, 19,20 AIDS, 21 and COVID-19. 22\nIn this study, an adenovirus type 5 based vector vaccine (HAdV5-HNH) carrying the conserved influenza antigens HA stem and NA was constructed. Mucosal immunity, systemic immunity, and protection induced by the HAdV5-HNH vaccine were evaluated in mice using different immunisation routes. A combined IM and IN immunisation strategy effectively induced robust mucosal immune responses in the respiratory tract and significantly enhanced the protective efficacy of the vaccine compared to a single IM or IN vaccination (prime-only). This study provides valuable insights into the development and application of novel universal influenza vaccines and vaccination strategies.\n2. Materials and Methods\n2.1. Cells, Viruses, and Animals\nHEK-293, 293T, Hep-2, and MDCK cells were cultured in Dulbecco’s minimal essential medium supplemented with 10% foetal bovine serum (FBS) and antibiotics at 37°C with 5% CO 2 . The influenza viruses H1N1-PR8, rgH3N2-X31, rgH5N1-Vietnam, and rgH7N9-Anhui were stored in our laboratory. The rgH3N2-X31 and rgH5N1-Vietnam influenza viruses were generated by reverse genetics, containing six gene segments from the A/PR/8/34 (H1N1) virus and the HA and NA gene segments from A/X31/H3N2 and A/Vietnam/1203/2004, respectively. Six-to-eight-week-old female BALB/c mice were purchased from Beijing Sibeifu Biotechnology Co., Ltd. (Beijing, China). All vaccinations, animal breeding, and viral challenge experiments were conducted at Beijing Sinovac Biotech Co., Ltd. (Beijing, China), according to the guidelines of the Institutional Animal Care and Use Committee (Ethics Review Number 20231128092).\n2.2. Vaccine Construction\nThe multi-gene co-expression plasmid pcDNA3.1-HNH was constructed based on previous studies, in which three proteins, Mini-HA, 23 Tet-NA, 24,25 and H3Stem, 26 were connected in sequence using the self-cleaving peptides T2A and P2A. A glycine-serine-glycine sequence was added before the 2A peptide sequence to enhance cleavage efficiency. After codon optimisation for mammalian preferences, the construct was inserted into the eukaryotic expression vector pcDNA3.1. A Kozak sequence was inserted at the N-terminus to facilitate eukaryotic expression. The plasmid was synthesised by Nanjing GenScript Biotechnology Co. Ltd. Recombinant adenoviruses HAdV5-HNH and HAdV5-Mock (with foreign genes inserted) were constructed in collaboration with the Virus Morphology Laboratory (Figure 1), and amplified, purified, and titred as previously described. 27\n2.3. Western Blot\nWestern blotting was performed to assess the recombinant adenovirus protein expression. Hep-2 cell monolayers were infected with HAdV5-HNH or HAdV5-Mock. After 48 h of infection, cell supernatants were collected, analysed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis, and transferred to a polyvinylidene difluoride membrane. The membrane was blocked with 5% skim milk for 2 h and incubated overnight at 4℃ with anti-HA, anti-H3, or anti-NA antibodies (Sino Biological, Beijing, China) in 5% skim milk. Subsequently, the membrane was incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG, goat anti-human IgG, or goat anti-rabbit IgG at room temperature for 1 h and developed using chemiluminescent HRP substrate (Merck).\n2.4. Immunisation and Viral Challenge\nSix-to-eight-week-old female BALB/c mice were divided into seven groups: mock (IN), low-dose IN(L), high-dose IN(H), mock (IM), low-dose IM(L), mock (IM+IN), and IM+IN. Mice were immunised IN and via IM injection, followed by booster immunisation 3 weeks later. Serum and cellular immune responses were evaluated 14 days after the initial and booster immunisations. Mice were euthanised by cervical dislocation, and serum, bronchoalveolar lavage fluid (BALF), spleen, and/or lung tissues were collected. Three weeks after immunisation, mice were challenged IN with 3×LD 50 (50% mouse lethal dose) of H1N1-PR8, rgH3N2-X31, rgH5N1-Vietnam, or rgH7N9-Anhui influenza viruses. Clinical symptoms and weight loss were monitored daily, with the humane endpoint defined as 75% of the initial body weight. On day 5 post-infection, lung tissues from four mice were collected and stored at -80°C for viral load analysis. Lung tissues were fixed with 4% formaldehyde and stained with haematoxylin and eosin (H&E) (Beijing Zhongke Wanbang Biotechnology Co., Ltd., Beijing, China).\n2.5. Enzyme-Linked Immunosorbent Assay\nPurified H1Stem, H3Stem, and NA (TargelMol) proteins were coated onto 96-well ELISA plates using ELISA coating buffer (Solarbio), and incubated overnight at 4°C. IgG or IgA antibody levels in mouse serum and BALF were measured. For IgG detection, serum samples were initially diluted 1:50 and serially diluted three-fold, whereas BALF samples were initially diluted 1:10 and then serially diluted three-fold. After blocking with 10% goat serum at 37°C for 2 h, mouse serum was serially diluted using ELISA dilution buffer (phosphate-buffered saline [PBS] with 2% goat serum) and incubated at 37°C for 1 h. HRP-conjugated goat anti-mouse IgG (1:10000) or HRP-conjugated anti-mouse IgA (1:1500) was diluted in ELISA dilution buffer and incubated at 37°C for 1 h. For IgA detection, streptavidin-HRP-conjugated antibody (1:1500) was used, and incubated at 37°C for 1 h. TMB substrate (Solarbio) was added for colour development, and the reaction was terminated with 2 M H₂SO₄. Absorbance was measured at 450 nm. A sample with an absorbance value ≥2.1 times the blank well was considered positive. The highest dilution ratio determined as positive was considered the endpoint, and the reciprocal of the endpoint dilution represented the antibody titre of the sample.\n2.6. ELISpot Assay\nMouse IFN-γ was measured using the BD TM IFN-γ ELISpot kit. Single cells were isolated from the spleen and lungs, according to the manufacturer’s instructions. IFN-γ (1:200) was diluted with 1×PBS and plated at 100 μL/well in a 96-well ELISpot plate, followed by overnight incubation at 4°C. After preparing the splenocyte single-cell suspension, pools of H1Stem, H3Stem, or N1NA peptides were used as stimulants for IFN-γ ELISpot. Spot formation was quantified using an ImmunoSpot CTL reader, and the number of spot-forming units per million cells was calculated by subtracting the number of negative control spots.\n2.7. Intracellular Staining\nMouse spleen single-cell suspensions were prepared two weeks after immunisation at 2×10 6 cells/tube. H1Stem, H3Stem, and NAN1 peptide libraries (5 μg) were added to each tube and incubated for 30 min at 37℃ with 5% CO 2 . GolgiPlug (555029; BD) was added, mixed thoroughly, incubated for 12 h, and the cells centrifuged for 5 min at 350 × g . Cells were washed twice with 1× Dulbecco’s PBS (DPBS), and stained with 1 μL FVS780 dye (565388; BD) for 15 min at room temperature, protected from light. Cells were washed twice with DPBS containing 1% FBS, 2 mL 1% FBS added, centrifuged at 350 × g for 5 min, and the supernatant discarded. Next, 2 μL of Fc Block antibody (553141; BD) was added for 10 min at room temperature, protected from light. Antibodies against CD3 (551163; BD), CD4 (563106; BD), and CD8 (553030; BD) were added, and the cell surface stained by incubation at 4℃ for 15 min, protected from light. Subsequently, 1 mL of cell-staining buffer (554656; BD) and 250 μL of fixation/breaking solution (565388; BD) were added and incubated for 20 min at 4℃, protected from light. Intracellular staining was performed with antibodies to IL-2 (554428; BD), IL-4 (554436; BD), TNF (557644; BD), and IFN-γ (563376; BD) by incubation for 30 min at 4℃, protected from light. Data were collected using a FACS II flow cytometer and analysed using FlowJo software (version 10.8.1).\n2.8. Viral Load Measurement\n2.8.1. Real-Time RT-PCR\nLung tissues were homogenised in a tissue grinder to obtain uniform suspensions. After centrifugation at 1000 × g for 5 min, the supernatant was collected. Viral RNA was extracted from 100 μL of the supernatant according to the manufacturer’s protocol (Vazyme Biotech Co., Ltd., RM501). For influenza A virus (IAV) quantification, primers targeting the conserved region of IAV M were designed as follows:\nIAV-F: 5′-CAAGACCAATCCTGTCACCTYTR-3′,\nIAV-R: 5′-TCTACGCTGCAGTCCTCTCTC-3′,\nIAV-P: 5′-FAM-ACGCTCACCGTGCCCAGTGAGC-BHQ1-3′.\nAmplification was performed using the One-Step PrimeScript TM RT-PCR Kit (TaKaRa), according to the manufacturer’s protocol.\n2.8.2. Focus Forming Assay\nMDCK cells (2×10⁴ cells/well) were seeded in a 96-well plate and cultured at 37°C with 5% CO₂ until a monolayer was formed. The next day, the culture medium was discarded, and the cells were washed twice with 100 μL PBS. The viral samples to be tested were initially diluted 100 times, followed by 3-fold serial dilutions in subsequent wells. A total of 30 μL of each viral dilution was added to MDCK cells and incubated for 1 h at room temperature. The viral solution was discarded, and 100 μL of Avicel overlay was added. The plate was incubated at 37°C with 5% CO₂ for 18–24 h. Subsequently, virus titres were determined by FFA: each well was fixed with 200 μL 4% paraformaldehyde (Solarbio) for 40 min, followed by washing once with PBS. Next, 100 μL 0.1% Triton X-100 was added to permeabilise the cells at room temperature for 30 min. The wells were washed once with PBS, and 30 μL of HRP-conjugated monoclonal antibody against IAV nucleoprotein (1:2000; Sino Biological) was added to each well and incubated at room temperature for 1 h, followed by three washes with PBS. Finally, 30 μL of TRUEBLUE (5510-0030; KPL) HRP substrate was added, and the plate incubated in the dark at room temperature for 10–30 min. Excess liquid was removed by tapping the plate onto absorbent paper. Plaques were counted using a CTL ImmunoSpot S6 Universal Analyser.\n2.9. Histopathology\nLung tissues were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned into 4 μm-thick slices. The sections were stained with H&E and scored according to the International Nomenclature and Diagnosis Standard Coordination of Tissue criteria. Pathological scoring was based on the extent of alveolar oedema, intra-alveolar haemorrhage, and neutrophil infiltration, with scores ranging from 1 to 3.\n2.10. Statistical Analysis\nAll data are presented as means standard error of the mean. Groups were compared using Student’s t-test, a one-way ANOVA with Tukey’s multiple comparison test according to the distribution of data. All tests were performed using GraphPad Prism 9.5.1 software.\n3. Results\n3.1 Characterisation of HAdV5-HNH Vaccine\nA schematic diagram of the recombinant adenovirus HAdV5-HNH expressing H1Stem, Tet-NA, and H3Stem proteins is shown in Figure 1A. Purified recombinant adenovirus HAdV5-HNH was used to infect Hep-2 cells at an MOI of 0.001, and the supernatant was collected 24 h post-infection. Western blotting confirmed that all three proteins (H1Stem, H3Stem, and Tet-NA) were expressed, as expected (Figure 1B).\n3.2. IN Booster with HAdV5-HNH Induces Robust Humoral and Mucosal Immune Responses\nImmunisation was performed according to the schedule shown (Figure 1C). The immunised groups effectively induced specific IgG antibodies against H1Stem, H3Stem, and NA, with statistically significant differences compared to the control groups (Figure 2A). After a single IN immunisation, the IN(H) group showed significantly higher IgG levels against H1Stem and H3Stem than the IN(L) group, in a dose-dependent manner. The IM(L) group showed significantly higher IgG levels against H3Stem than the IN(L) group; however, no significant difference was observed between the two groups regarding H1Stemand NA-specific IgG antibodies. Compared to the prime-only groups, the intramuscular prime-IN boost regimen (IM+IN) significantly increased the levels of specific IgG antibodies in mice.\nAdditionally, specific IgG subclass profiles against H1Stem, H3Stem, and NA were analysed in the immunised groups (Figure 2B). All groups had high levels of IgG1 and IgG2a antibodies. Both the single IM injection and IN immunisation groups showed similar antibody subclass profiles, primarily inducing a Th2-type antibody immune response, characterised by IgG1 > IgG2a. In contrast, the IM+IN group induced a balanced Th1/2 immune response in the IgG subclass analysis of H1Stem and H3Stem, but a Th1-type response in the IgG subclass analysis of the NA protein.\nRespiratory mucosal immunity plays a critical role in defending against viral entry into the respiratory tract. To further investigate the mucosal immune response induced by different immunisation regimens, the production of mucosal IgA antibodies in BALF was measured (Figure 2C). Mucosal antibodies were not detected in the IN(L) or IM(L) groups. However, significantly higher levels of IgA were detected in the BALF of the IN(H) and IM+IN groups. Notably, the IM+IN group exhibited significantly enhanced mucosal IgA immune responses to IN adjuvants.\nIn summary, compared to the prime-only groups, IN boosting with HAdV5-HNH following IM priming induced higher levels of both IgG and IgA, resulting in a stronger mucosal and systemic immune response.\n3.3 IN Booster with HAdV5-HNH Triggers Enhanced Memory T Cell Immune Responses\nGiven the importance of a strong cellular immune response in controlling and clearing influenza virus infection, cellular immune responses in mouse spleens and lungs were assessed. The number of IFN-γ-secreting T cells was measured using ELISPOT after stimulation with the three peptide pools (Figure 3A–C).\nThe ELISpot assay for spleen cells (Figure 3A) revealed that the IM(L) and IN(L) groups did not exhibit significant differences in the induction of H1- and H3-specific cellular immune responses. However, the IM(L) group significantly outperformed the IN(H) group at inducing NA-specific cellular immune responses. This indicated that a single IN immunisation failed to effectively activate a splenic cell immune response comparable to that induced by IM immunisation. In contrast, the combined IM+IN immunisation regimen significantly induced specific IFN-γ secretion in spleen cells.\nRegarding mucosal cellular immune responses in mouse lungs (Figure 3B), both the IN(H) and IM+IN groups induced specific T cell immune responses. However, the IN(L) and IM(L) groups did not effectively activate pulmonary immune responses. The low-dose IM+IN regimen, however, effectively compensated for the weak mucosal immune responses triggered by IN(L) and IM(L) immunisations, successfully activating IFN-γ production.\nAdditionally, lymphocytes from mouse spleen tissue were isolated and stimulated with the three peptide pools. Intracellular cytokine staining was performed to analyse the levels of IFN-γ, TNF-α, IL-2, and IL-4, as well as the T cell subsets secreting these cytokines (Figure 3D). The prime-only groups showed mild splenic memory T cell responses. CD4 + T cells from immunised mice did not significantly secrete cytokines, whereas the IM+IN combined immunisation group showed activation of H1-specific IFN-γ-secreting CD8 + T cells. T cell activation was biased toward a Th1-type response, with IFN-γ secretion, but failed to effectively induce IL-4 secretion. Taken together, HAdV5-HNH IN boosting effectively induced stronger memory T cell immune responses in both mouse spleens and lungs.\n3.4. HAdV5-HNH IN Booster Provides Protection Against H1N1, rgH3N2, and rgH5N1 Challenge\nTo evaluate the protective effects of HAdV5-HNH against influenza virus infection, immunised mice were challenged with 3×LD 50 of H1N1-PR8, rgH3N2-X31, or rgH5N1-Vietnam and monitored for 14 days. In the prime-only immunisation strategy, mice experienced continuous weight loss following influenza virus challenge, with 100% mortality (Figure 4A–F). In contrast, the prime-boost immunisation strategy provided 100% protection to mice following challenge with H1N1-PR8 and rgH3N2-X31 (Figure 4A, B, D, E). For the rgH5N1-Vietnam challenge, the IM+IN group also showed protective effects compared to the mock group. The IM+IN group showed a gradual recovery in body weight, which dropped to 90% on day 3 and then rebounded, whereas the mock group experienced a weight loss of 75% by day 6 before gradually recovering (Figure 4C, F). Unfortunately, none of the immunization groups provided cross-protection against rgH7N9-Anhui (Figure. S1).\nTo further characterise the protective effects observed in the IM+IN vaccine group, lung viral load was measured and histopathological analysis performed on day 5 post-infection. Compared to the mock group, the IM+IN vaccine group showed a significant reduction in viral RNA copies of H1N1-PR8, rgH3N2-X31, and rgH5N1-Vietnam in lung tissue (Figure 4G–I). The FFA for lung tissues showed similar results (Figure 4J, K). In mice infected with rgH5N1-Vietnam, no live virus was detected in the lung tissue collected on day 5, which may be due to poor adaptation of this virus strain in the mouse lungs or because live viruses should have been assessed earlier.\nHistopathological examinations were conducted of lung tissues from mice infected with H1N1-PR8, rgH3N2-X31, or rgH5N1-Vietnam (Figure 5). Lung sections from the mock- and IM immunised groups showed typical clinical symptoms, with >50% of the lung parenchyma affected by multifocal, moderate-to-severe necrotising bronchitis and bronchiolitis, moderate-to-severe alveolitis with neutrophil-dominated mixed inflammatory cell infiltration, and pulmonary oedema. Interestingly, the IN groups showed significant differences in lung pathology scores compared to the Mock group in mice infected with rgH3N2-X31 and rgH5N1-Vietnam, with a dose-dependent effect (Fig.5E-F). These data suggest that, compared to intramuscular injection, mucosal delivery of the same vaccine provides better protection, consistent with a recent study. 28 In contrast, the IM+IN vaccine group exhibited minimal histopathological changes in the lungs after H1N1-PR8 infection, and the overall lung pathology score in this group was significantly lower than that of the Mock group (Fig.5D). Similarly, improvements in lung tissue pathology were observed in the IM+IN group in mice infected with rgH3N2-X31 and rgH5N1-Vietnam compared to the Mock group (Fig.5E-F).\nThese results indicate that HAdV5-HNH IN boosting effectively suppressed infection by H1N1-PR8, rgH3N2-X31, and rgH5N1-Vietnam in vivo.\nDiscussion\nThe global diversity and unpredictability of influenza strains with epidemic or pandemic potential necessitate the urgent development of novel therapeutic and preventative strategies. The development of a universal influenza vaccine offers promising prospects to address this challenge. Additionally, optimisation of vaccination strategies is crucial for the development and application of effective and broadly protective influenza vaccines. One effective approach to achieve a universal vaccine is to incorporate multiple conserved antigens from different viral proteins into the vaccine design. In response, we have developed a universal vaccine candidate, HAdV5-HNH, which leverages three conserved epitopes from the surface proteins of the influenza virus, delivered via an adenovirus vector-based platform. Co-immunization with HAdV5-HNH through both intramuscular (IM) and intranasal (IN) routes elicited strong, antigen-specific mucosal humoral immune responses and cellular immunity, providing 100% protection against challenges with divergent strains of influenza A virus.\nEnhancing the humoral immune response is crucial for the development of an effective and broadly protective influenza vaccine. Vaccination with HAdV5-HNH induced specific IgG responses against the H1 HA stalk, H3 HA stalk, and NA in mice, suggesting that the vaccine is systematically effective in boosting antibody responses. The IN(H) group showed significantly higher levels of specific antibodies than the IN(L) group, indicating a dose-dependent effect. Furthermore, the IM(L) group showed significantly higher levels of H3 HA stalk-specific antibodies than the IN(L) group, with a further increase in specific antibody levels after IN boosting. The HA stalk contains broad neutralising epitopes which induce neutralising antibodies that inhibit influenza virus entry into host cells. 29 However, neutralising antibodies were not detected in the present study using focus reduction neutralisation tests, likely because of the relatively low immunogenicity of the HA stalk. BALB/c mice typically exhibit a Th2-type immune response characterised by the production of IgG1 antibodies. 30 Th1-type immune responses, which produce IgG2a antibodies and recruit cytotoxic T lymphocytes, macrophages, and NK cells, play critical roles in the effective clearance of viral infections. 31 Serotyping results showed that HAdV5-HNH induced high levels of both HA stalk- and NA-specific IgG1 and IgG2a antibodies, indicating that the vaccine elicited both Th2-biased and Th1-biased immune responses. By activating these two types of immune responses, HAdV5-HNH holds promise for providing broad protection against influenza virus.\nIn addition to serum IgG, secretory IgA plays a crucial role in the early stages of viral infection, serving as the first line of defence to prevent pathogens from entering the body through mucosal surfaces. 32-34 Currently, most approved seasonal influenza vaccines are administered via IM injections, which induce strong humoral and splenic cellular immune responses. However, they have limited effectiveness in inducing mucosal immunity and protecting against mucosal infections. After IM administration of the COVID-19 mRNA vaccine, immune responses in the respiratory mucosa are suboptimal. In contrast, IN booster immunisation induces stronger mucosal humoral and cellular immune responses. 35-38 The present study confirmed this finding. Although single-dose IM(L) induced strong systemic humoral and splenic cellular immunity, it failed to effectively trigger mucosal IgA immune responses. In comparison, IN boosting with IM+IN not only significantly enhanced the IgG antibody levels induced by IM injection, but also resulted in higher levels of IgA in BALF compared to IN(H) immunisation alone. Therefore, in a context in which the majority of the population has already been vaccinated with IM injections, mucosal vaccines offer clear advantages as an immune-boosting strategy and represent an important direction for future respiratory virus vaccine development.\nIn addition to humoral and mucosal immune responses, immunisation of BALB/c mice with HAdV5-HNH significantly induced IFN-γ secretion by specific Th1-type CD4 + and CD8 + T cells. IFN-γ plays an important role in protective T cell immunity and is associated with enhanced immune memory, viral clearance, and breadth of protection. 39 Effector CD4 + T cells expressing IFN-γ and perforin possess cytolytic activity and can alleviate influenza infection severity, thereby promoting rapid recovery. 40,41 Furthermore, CD8 + T cells limit early viral replication and directly kill target cells by producing cytokines such as IFN-γ, thereby facilitating effective viral clearance. 42,43 ELISpot results for splenic T cells showed that single-dose IN immunisation was comparable to single-dose IM immunisation in terms of inducing cellular immune responses, whereas IM injection alone did not induce pulmonary T cell immune responses. Knudson et al. 44 further confirmed that, following respiratory viral infections, T cells primarily localise to the lung tissue, forming lung-resident memory T cells (TRMs). TRMs in the lungs enhance immune responses to reinfection with different influenza strains, and are key to developing influenza vaccines with long-lasting immunity. 45 The IM+IN group, through IN booster immunisation, exhibited higher levels of H1HA stem, H3HA stem, and NA-specific pulmonary immune responses, leading to IFN-γ induction in lung tissue. Intracellular cytokine staining showed that the IM+IN group induced a stronger H1HA stem-specific CD8 + T cell response, with a higher level of cellular immunity than the prime-only group. The high level of IFN-γ-mediated cellular immunity may explain the enhanced cross-protection observed in the IM+IN group. These data are consistent with the protection results from the H1N1-PR8, rgH3N2-X31, and rgH5N1-Anhui challenge studies.\nWidespread tissue tropism and potent expression of target antigens by adenoviral vectors make them one of the most immunogenic viral vector platforms. 46-48 However, early findings have shown that a significant portion of the global population has pre-existing immunity to adenovirus type 5, which can reduce the ability of the vector to deliver genes and induce an immune response to the target antigen. 49 In this study, the antibody response induced by the IM+IN combined immunisation strategy using HAdV5-HNH was not affected by pre-existing immunity to adenovirus. This is consistent with previous studies showing that mucosal vaccination can bypass pre-existing immunity to adenoviral vectors and induce effective immune responses against the target antigen. 50,51 An easily administered IN booster may be a better option for individuals who have already received an IM vaccine. Recent studies have shown that relying solely on IM vaccination does not provide local mucosal immunity. Booster doses using vaccines such as IN sprays, which trigger mucosal immunity, are necessary to form an infection barrier in the lungs and establish a strong immune response against new variants of SARS-CoV-2. 52-54 This may explain why the IM+IN group, with mucosal boosting, not only enhanced systemic immune responses generated by the initial vaccination but also induced mucosal immunity and lung-specific cellular immune responses, protecting mice from challenges with H1N1-PR8, rgH3N2-X31, and rgH5N1-Anhui influenza viruses.\nThis study has some limitations, including the failure to consider the IN+IM regimen in the immune design, and the comparison of immune response levels between the IM+IN and IN+IM regimens. However, previous studies have found that an IM+IN regimen more effectively stimulates immune responses in secondary germinal centres (GCs) than an IN+IM regimen, aiding the development of broad-spectrum COVID-19 vaccines and optimised vaccination strategies. Secondary GC responses are more crucial than primary responses for generating high-quality antibodies. In the present study, both the IM+IN and IN+IM regimens were homologous immunisation strategies. However, these results suggest that heterologous immunisation is superior in promoting antibody responses and significantly enhancing the switching and activation frequencies of antigen-specific memory B cells. 55-57 Furthermore, heterologous prime-boost immunisation is advantageous in overcoming adenoviral vector immunity. 58,59 Therefore, further research is required to explore more effective universal influenza vaccine immunisation strategies.\nIn the current landscape, mucosal vaccines represent a promising development direction for future respiratory virus vaccines. This study provides compelling evidence that a combined vaccine strategy can simultaneously induce robust systemic humoral and mucosal immunity, thereby providing direct protection against influenza virus infection. This strategy represents a promising approach for maximising the protective efficacy of respiratory viral vaccines. This study offers valuable insights into the customisation of mucosal boosting strategies to enhance vaccine efficacy and expand the scope of protection.\nFigure Legends\nFIGURE 1. HAdV5-HNH vaccine construction, expression verification, and immunisation scheme. (A) Schematic diagram of HAdV5-HNH vaccine construction. (B) Western blot analysis of the expression of H1Stem, H3Stem, and NA proteins in the supernatant of Hep-2 cells infected with purified HAdV5-HNH. HAdV5-Mock was used as a negative control. (C) Vaccination and challenge schemes in mice. IM, intramuscular; IN, intranasal; L, low dose; H, high dose.\nFIGURE 2. IN booster with HAdV5-HNH induces robust humoral and mucosal immune responses (A) Detection of serum hemagglutinin (HA) stem- and neuraminidase (NA)-specific IgG antibodies in immunised mice (n = 5). (B) HA stem and NA-specific IgG1 and IgG2a antibody titres 14 days post-boost (n = 5). (C) Detection of HA stem- and NA-specific IgG antibodies in bronchoalveolar lavage fluid (BALF) from immunised mice (n = 5). Each data point represents a single mouse. The black dashed line indicates the detection limit, and the geometric mean titre values are shown in (A) and (C). Statistical significance was assessed using one-way analysis of variance with Tukey’s multiple comparison test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.\nFIGURE 3. IN booster with HAdV5-HNH triggers enhanced memory T cell immune responses. (A, B) ELISpot analysis assessing the number of IFN-γ-secreting spots per million splenocytes (A) and lung cells (B) from BALB/c mice stimulated with H1, H3, and NA peptide pools (n = 5). (C) Representative ELISpot showing IFN-γ secretion by specific cytokine-producing cells. (D) Intracellular cytokine staining analysis of splenocytes showing the frequency of cytokine-producing CD4 + T cells (top) and CD8 + T cells (bottom) following stimulation with H1, H3, and NA peptide pools (n = 5). Each data point represents a single mouse. Statistical significance was assessed using one-way analysis of variance followed by Tukey’s multiple comparison test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.\nFIGURE 4. HAdV5-HNH IN booster provides protection against H1N1, rgH3N2, and rgH5N1 Challenge. (A–F) Body weight monitoring and survival rates of mice infected with H1N1-PR8, rgH3N2-X31, rgH5N1-Vietnam, and rgH7N9-Anhui (n = 6). (G–I) Lung viral load measured using real-time RT-PCR on day 5 post-infection in H1N1-PR8, rgH3N2-X31, and rgH5N1-Vietnam infected mice (n = 4). (J–K) Lung viral titres measured by focus forming assay on day 5 post-infection in H1N1-PR8, rgH3N2-X31, and rgH5N1-Vietnam infected mice (n = 4). Each data point represents a single mouse. Statistical significance was assessed using one-way analysis of variance followed by Tukey’s multiple comparison test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.\nFIGURE 5. Lung pathology on day 5 post-infection. (A) Representative lung pathology sections from four mice per group. (B) Histopathological scores of lungs on day 5 post-infection (n = 4). Each data point represents a single mouse. Statistical significance was assessed using one-way analysis of variance followed by Tukey’s multiple comparison test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.\nSupplementary Figure 1. Protective Efficacy of HAdV5-HNH Vaccine in rgH7N9-Anhui Infected Mice\n(A-B) Monitoring of body weight and survival rate in mice infected with rgH7N9-Anhui (n=6). (C) Lung viral load measured by real-time RT-PCR on day 5 post-infection with rgH7N9-Anhui (n=4). (D) Lung viral titers assessed by Focus Forming Assay (FFA) on day 5 post-infection with rgH7N9-Anhui (n=4). (E) Representative histopathological images of lung tissues from four mice per group.\n(F) Lung pathology scores on day 5 post-infection (n=4). Each point represents data from an individual mouse. Statistical significance was assessed by one-way ANOVA, followed by Tukey’s multiple comparisons test. * p < 0.05, ** p < 0.01, *** p < 0.001. Scale bar = 200μm.\nREFERENCES\n1. Palese P. Influenza: old and new threats. Nature Medicine . 2004/12/01 2004;10(12):S82-S87. doi:10.1038/nm11412. Iuliano AD, Roguski KM, Chang HH, et al. Estimates of global seasonal influenza-associated respiratory mortality: a modelling study. The Lancet . 3 2018;391(10127):1285-1300. doi:10.1016/S0140-6736(17)33293-23. Paules CI, Marston HD, Eisinger RW, Baltimore D, Fauci AS. The Pathway to a Universal Influenza Vaccine. Immunity . 10/17 2017;47(4):599-603. doi:10.1016/j.immuni.2017.09.0074. Krammer F, Palese P. Advances in the development of influenza virus vaccines. Nat Rev Drug Discov . 3 2015;14(3):167-82. doi:10.1038/nrd45295. Bullard BL, Weaver EA. Strategies Targeting Hemagglutinin as a Universal Influenza Vaccine. Vaccines . 3/13 2021;9(3):257. doi:10.3390/vaccines90302576. Kirkpatrick E, Qiu X, Wilson PC, Bahl J, Krammer F. The influenza virus hemagglutinin head evolves faster than the stalk domain. Sci Rep . 7/11 2018;8(1)doi:10.1038/s41598-018-28706-17. Casazza JP, Hofstetter AR, Costner PJM, et al. Phase 1 dose-escalation trial evaluating a group 2 influenza hemagglutinin stabilized stem nanoparticle vaccine. npj Vaccines . 9/17 2024;9(1)doi:10.1038/s41541-024-00959-08. Corti D, Voss J, Gamblin SJ, et al. A Neutralizing Antibody Selected from Plasma Cells That Binds to Group 1 and Group 2 Influenza A Hemagglutinins. Science . 8/12 2011;333(6044):850-856. doi:10.1126/science.12056699. Ekiert DC, Bhabha G, Elsliger M-A, et al. Antibody Recognition of a Highly Conserved Influenza Virus Epitope. Science . 4/10 2009;324(5924):246-251. doi:10.1126/science.117149110. Gao J, Li X, Klenow L, et al. Antigenic comparison of the neuraminidases from recent influenza A vaccine viruses and 2019–2020 circulating strains. npj Vaccines . 7/14 2022;7(1)doi:10.1038/s41541-022-00500-111. Johansson B. Supplementation of conventional influenza A vaccine with purified viral neuraminidase results in a balanced and broadened immune response. Vaccine . 5 1998;16(9-10):1009-1015. doi:10.1016/S0264-410X(97)00279-X12. Chen Z, Kadowaki S-e, Hagiwara Y, et al. Cross-protection against a lethal influenza virus infection by DNA vaccine to neuraminidase. Vaccine . 8 2000;18(28):3214-3222. doi:10.1016/S0264-410X(00)00149-313. Easterbrook JD, Schwartzman LM, Gao J, et al. Protection against a lethal H5N1 influenza challenge by intranasal immunization with virus-like particles containing 2009 pandemic H1N1 neuraminidase in mice. Virology . 10 2012;432(1):39-44. doi:10.1016/j.virol.2012.06.00314. Wohlbold TJ, Nachbagauer R, Xu H, et al. Vaccination with Adjuvanted Recombinant Neuraminidase Induces Broad Heterologous, but Not Heterosubtypic, Cross-Protection against Influenza Virus Infection in Mice. mBio . 03/10 2015;6(2)doi:10.1128/mBio.02556-1415. Neutra MR, Kozlowski PA. Mucosal vaccines: the promise and the challenge. Nature Reviews Immunology . 2 2006;6(2):148-58. doi:10.1038/nri177716. Lycke N. Recent progress in mucosal vaccine development: potential and limitations. Nature Reviews Immunology . 7/25 2012;12(8):592-605. doi:10.1038/nri325117. Thwaites RS, Uruchurtu ASS, Negri VA, et al. Early mucosal events promote distinct mucosal and systemic antibody responses to live attenuated influenza vaccine. Nature Communications . 12/5 2023;14(1)doi:10.1038/s41467-023-43842-718. Hartwell BL, Melo MB, Xiao P, et al. Intranasal vaccination with lipid-conjugated immunogens promotes antigen transmucosal uptake to drive mucosal and systemic immunity. Sci Transl Med . 7/20 2022;14(654):eabn1413-eabn1413. doi:10.1126/scitranslmed.abn141319. Li J-X, Hou L-H, Meng F-Y, et al. Immunity duration of a recombinant adenovirus type-5 vector-based Ebola vaccine and a homologous prime-boost immunisation in healthy adults in China: final report of a randomised, double-blind, placebo-controlled, phase 1 trial. The Lancet Global Health . 2017;5(3):e324-e334. doi:10.1016/S2214-109X(16)30367-920. Dolzhikova IV, Zubkova OV, Tukhvatulin AI, et al. Safety and immunogenicity of GamEvac-Combi, a heterologous VSV- and Ad5-vectored Ebola vaccine: An open phase I/II trial in healthy adults in Russia. Human Vaccines & Immunotherapeutics . 2017;13(3):613-620. doi:10.1080/21645515.2016.123853521. Behmardi A, Farazmandfar T. A recombinant adenoviral vector with a specific tropism to CD4-positive cells: a new tool for HIV-1 inhibition. Drug Deliv and Transl Res . 1/31 2022;12(10):2561-2568. doi:10.1007/s13346-021-01109-y22. Chew CK, Ruijie W, Bavanandan S, et al. Safety, efficacy and immunogenicity of aerosolized Ad5-nCoV COVID-19 vaccine in a non-inferiority randomized controlled trial. npj Vaccines . 10/31 2024;9(1)doi:10.1038/s41541-024-01003-x23. Impagliazzo A, Milder F, Kuipers H, et al. A stable trimeric influenza hemagglutinin stem as a broadly protective immunogen. Science . 09/18 2015;349(6254):1301-1306. doi:10.1126/science.aac726324. Ellis D, Lederhofer J, Acton OJ, et al. Structure-based design of stabilized recombinant influenza neuraminidase tetramers. Nature Communications . 2022;13(1)doi:10.1038/s41467-022-29416-z25. Streltsov VA, Schmidt PM, McKimm-Breschkin JL. Structure of an Influenza A virus N9 neuraminidase with a tetrabrachion-domain stalk. Acta Crystallographica Section F Structural Biology Communications . 2019;75(2):89-97. doi:10.1107/s2053230x1801789226. Sutton T, Chakraborty S, Mallajosyula V, et al. Protective efficacy of influenza group 2 hemagglutinin stem-fragment immunogen vaccines. npj Vaccines . 12/01 2017;2(1)doi:10.1038/s41541-017-0036-227. Guo X, Deng Y, Chen H, et al. Systemic and mucosal immunity in mice elicited by a single immunization with human adenovirus type 5 or 41 vector-based vaccines carrying the spike protein of Middle East respiratory syndrome coronavirus. Immunology . 8 2015;145(4):476-84. doi:10.1111/imm.1246228. Najwa DA, Levi T, Muhammad Yasir K, et al. Mucosal SARS-CoV-2 S1 adenovirus-based vaccine elicits robust systemic and mucosal immunity and protects against disease in animals. mBio . 12/04 2024:e0217024. doi:10.1128/mbio.02170-2429. Raffael N, Krammer F. Universal influenza virus vaccines and therapeutic antibodies. Clinical Microbiology and Infection . 04/01 2017;23(4):222-228. doi:10.1016/j.cmi.2017.02.00930. Mills CD, Kincaid K, Alt JM, Heilman MJ, Hill AM. M-1/M-2 Macrophages and the Th1/Th2 Paradigm. J Immunol . 6/15 2000;164(12):6166-6173. doi:10.4049/jimmunol.164.12.616631. Gunay G, Hamsici S, Lang GA, Lang ML, Kovats S, Acar H. Peptide Aggregation Induced Immunogenic Rupture (PAIIR). Advanced Science . 7 2022;9(21):e2105868. doi:10.1002/advs.20210586832. Neutra MR, Kozlowski PA. Mucosal vaccines: the promise and the challenge. Nature Reviews Immunology . 2006/02/01 2006;6(2):148-158. doi:10.1038/nri177733. Lamm ME. INTERACTION OF ANTIGENS AND ANTIBODIES AT MUCOSAL SURFACES. Annu Rev Microbiol . 10 1997;51(1):311-340. doi:10.1146/annurev.micro.51.1.31134. Joseph J. Harnessing Nasal Immunity with IgA to Prevent Respiratory Infections. Immuno . 10/17 2022;2(4):571-583. doi:10.3390/immuno204003635. Tianyang M, Benjamin I, Mario AP-H, et al. Unadjuvanted intranasal spike vaccine elicits protective mucosal immunity against sarbecoviruses. Science . 10/27 2022;378(6622)doi:10.1126/science.abo252336. Qian W, Chenchen Y, Li Y, et al. Intranasal booster using an Omicron vaccine confers broad mucosal and systemic immunity against SARS-CoV-2 variants. Sig Transduct Target Ther . 04/17 2023;8(1)doi:10.1038/s41392-023-01423-637. Tang J, Zeng C, Cox TM, et al. Respiratory mucosal immunity against SARS-CoV-2 after mRNA vaccination. Sci Immunol . 10/28 2022;7(76)doi:10.1126/sciimmunol.add485338. Topol E, Iwasaki A. Operation Nasal Vaccine—Lightning speed to counter COVID-19. Sci Immunol . 07/21 2022;7(74)doi:10.1126/sciimmunol.add994739. Wilkinson T, Li C, Cecilia C, et al. Preexisting influenza-specific CD4+ T cells correlate with disease protection against influenza challenge in humans. Nature Medicine . 01/29 2012;18(2):274-280. doi:10.1038/nm.261240. Deborah MB, Sarah L, Garcia-Hernandez M, Swain S. Multifunctional CD4 Cells Expressing Gamma Interferon and Perforin Mediate Protection against Lethal Influenza Virus Infection. J Virol . 04/04 2012;86(12):6792-6803. doi:10.1128/JVI.07172-1141. Lu IN, Farinelle S, Sausy A, Muller CP. Identification of a CD4 T-cell epitope in the hemagglutinin stalk domain of pandemic H1N1 influenza virus and its antigen-driven TCR usage signature in BALB/c mice. Cell Mol Immunol . 5/9 2016;14(6):511-520. doi:10.1038/cmi.2016.2042. Grant E, Quiñones-Parra S, Clemens EB, Kedzierska K. Human influenza viruses and CD8(+) T cell responses. Current Opinion in Virology . 02/01 2016;16:132-142. doi:10.1016/j.coviro.2016.01.01643. McMaster S, Jarad JW, Hong W, Kohlmeier J. Airway-Resident Memory CD8 T Cells Provide Antigen-Specific Protection against Respiratory Virus Challenge through Rapid IFN-γ Production. J Immunol . 07/01 2015;195(1):203-209. doi:10.4049/jimmunol.140297544. Knudson CJ, Weiss KA, Hartwig SM, Varga SM. The Pulmonary Localization of Virus-Specific T Lymphocytes Is Governed by the Tissue Tropism of Infection. J Virol . 8/15 2014;88(16):9010-9016. doi:10.1128/JVI.00329-1445. Swarnalekha N, Schreiner D, Litzler LC, et al. T resident helper cells promote humoral responses in the lung. Sci Immunol . 1/8 2021;6(55)doi:10.1126/sciimmunol.abb680846. Afkhami S, Yushi Y, Xing Z. Methods and clinical development of adenovirus-vectored vaccines against mucosal pathogens. Molecular Therapy - Methods & Clinical Development . 04/27 2016;3:16030. doi:10.1038/mtm.2016.3047. Barry M. Single-cycle adenovirus vectors in the current vaccine landscape. Expert Review of Vaccines . 01/18 2018;17:163-173. doi:10.1080/14760584.2018.141906748. Lasaro M, Ertl H. New insights on adenovirus as vaccine vectors. Molecular Therapy . 08/01 2009;17(8):1333-9. doi:10.1038/mt.2009.13049. Sumida SM, Truitt DM, Kishko MG, et al. Neutralizing antibodies and CD8+ T lymphocytes both contribute to immunity to adenovirus serotype 5 vaccine vectors. J Virol . 3 2004;78(6):2666-73. doi:10.1128/jvi.78.6.2666-2673.200450. Croyle MA, Patel A, Tran KN, et al. Nasal Delivery of an Adenovirus-Based Vaccine Bypasses Pre-Existing Immunity to the Vaccine Carrier and Improves the Immune Response in Mice. PLoS ONE . 10/29 2008;3(10):e3548. doi:10.1371/journal.pone.000354851. Jason SR, Pillet S, Bello A, Kobinger G. Airway Delivery of an Adenovirus-Based Ebola Virus Vaccine Bypasses Existing Immunity to Homologous Adenovirus in Nonhuman Primates. J Virol . 01/09 2013;87(7):3668-3677. doi:10.1128/JVI.02864-1252. Jinyi T, Zeng C, Cox TM, et al. Respiratory mucosal immunity against SARS-CoV-2 following mRNA vaccination. Sci Immunol . 07/19 2022;7(76)doi:10.1126/sciimmunol.add485353. Ilinykh PA, Periasamy S, Huang K, et al. A single intranasal dose of human parainfluenza virus type 3-vectored vaccine induces effective antibody and memory T cell response in the lungs and protects hamsters against SARS-CoV-2. npj Vaccines . 04/25 2022;7(1)doi:10.1038/s41541-022-00471-354. Mao T, Israelow B, Peña-Hernández MA, et al. Unadjuvanted intranasal spike vaccine elicits protective mucosal immunity against sarbecoviruses. Science . 378(6622):eabo2523. doi:10.1126/science.abo252355. Takano T, Sato T, Kotaki R, et al. Heterologous SARS-CoV-2 spike protein booster elicits durable and broad antibody responses against the receptor-binding domain. Nature Communications . 3/15 2023;14(1)doi:10.1038/s41467-023-37128-156. Li J, Hou L, Guo X, et al. Heterologous AD5-nCOV plus CoronaVac versus homologous CoronaVac vaccination: a randomized phase 4 trial. Nature Medicine . 01/27 2022;28(2):401-409. doi:10.1038/s41591-021-01677-z57. Pozzetto B, Legros V, Djebali S, et al. Immunogenicity and efficacy of heterologous ChAdOx1–BNT162b2 vaccination. Nature . 10/21 2021;600(7890):701-706. doi:10.1038/s41586-021-04120-y58. Liu J, Xu K, Xing M, et al. Heterologous prime-boost immunizations with chimpanzee adenoviral vectors elicit potent and protective immunity against SARS-CoV-2 infection. Cell Discovery . 12/18 2021;7(1):123. doi:10.1038/s41421-021-00360-459. Vierboom MPM, Chenine AL, Darrah PA, et al. Evaluation of heterologous prime-boost vaccination strategies using chimpanzee adenovirus and modified vaccinia virus for TB subunit vaccination in rhesus macaques. npj Vaccines . 5/14 2020;5(1)doi:10.1038/s41541-020-0189-2\nSupplementary Material\nFile (figure.zip)\n- Download\n- 211.75 MB\nInformation & Authors\nInformation\nVersion history\nPeer review timeline\nPublished\nJournal of Medical Virology\nVersion of Record27 Mar 2025Published\nCopyright\nThis work is licensed under a Non Exclusive No Reuse License.\nCollection\nKeywords\nAuthors\nMetrics & Citations\nMetrics\nArticle Usage\n347views\n227downloads\nCitations\nDownload citation\nJia Li, Tangqi Wang, Xiaojuan Guo, et al.\nBroad mucosal and systemic immunity in mice induced by intranasal booster with a novel recombinant adenoviral based vaccine protects against divergent influenza A virus. Authorea. 05 January 2025.\nDOI: https://doi.org/10.22541/au.173610176.62554351/v1\nDOI: https://doi.org/10.22541/au.173610176.62554351/v1\nIf you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.\nFor more information or tips please see 'Downloading to a citation manager' in the Help menu.","source_license":"CC-BY-4.0","license_restricted":false}