Disease ecology and zoonotic risk of clade 2.3.4.4b H5N1 high pathogenicity avian influenza in the sub-Antarctic region

Disease ecology and zoonotic risk of clade 2.3.4.4b H5N1 high pathogenicity avian influenza in the sub-Antarctic region | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Disease ecology and zoonotic risk of clade 2.3.4.4b H5N1 high pathogenicity avian influenza in the sub-Antarctic region Benjamin C. Mollett, Joshua G. Lynton-Jenkins, Samuel Richardson, and 27 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8029950/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract High pathogenicity avian influenza virus (HPAIV) has significantly impacted upon avian and mammalian populations across the Antarctic region. All viruses detected have been genotype B3.2 with phylogenetic analyses indicating multiple independent incursions from continental South America to, and transmission between, sub-Antarctic islands. From a zoonotic perspective, several isolates contained markers of mammalian adaptation in PB2 with functional characterisation of mutants demonstrating efficient replication in primary human airway epithelial cell cultures, demonstrating that these PB2-mutations alone contributed to enhanced polymerase activity in human cell lines. No mammalian-adaptive mutations were detected in the haemagglutinin or neuraminidase genes, with viruses retaining avian receptor binding preferences. Antigenic characterisation demonstrated cross-reactivity with existing pre-pandemic candidate vaccine strains and all viruses remained susceptible to licensed frontline antiviral therapeutics. We demonstrate a complex evolving viral ecology in the sub-Antarctic region involving both avian and marine mammal hosts, with significant implications for regional wildlife populations and zoonotic risk. Biological sciences/Genetics/Functional genomics/Mutagenesis Biological sciences/Microbiology/Virology/Influenza virus Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction The emergence of clade 2.3.4.4b H5N1 high pathogenicity avian influenza virus (HPAIV) on the sub-Antarctic islands 1-3 demonstrated the truly global dissemination of these viruses by wild species and the fragility of remotely located ecosystems to incursions of emerging viral pathogens. This panzootic began with the emergence of a novel H5N1 genotype belonging to clade 2.3.4.4b in Europe or central Asia in 2020 4 . In late 2021, this virus spread into North America through trans-Atlantic pathways via migratory birds 5,6 such as gull species. Since its arrival in North America, clade 2.3.4.4b H5N1 has undergone significant diversification through reassortment with low pathogenicity avian influenza viruses (LPAIV) 7 . By late 2022, a North American H5N1 genotype (B3.2) was detected in South America for the first time 8,9 , signalling the initial threat of this virus to the wider region 10,11 . The rapid expansion of the virus across South America significantly raised the risk of incursion into the sub-Antarctic and Antarctic regions 1,2,12 . The arrival of HPAIV in South American mammals, signalling increased zoonotic risk, was marked by an outbreak of H5N1 that caused extensive mortality in marine mammal population (mainly pinniped species), initially in South American sea lions ( Otaria flavescens ) reported in Chile in June 2023 13 , and by August 2023 HPAIV had been reported in mass mortalities in marine mammal colonies off the southern coast of Argentina 14 . The virus reached Uruguay 11 and southern Brazil 15 before the observation of a mass mortality event in southern elephant seals ( Mirounga leonina ), affecting over 17,000 animals in Argentina 16 . While detections in wild birds and marine mammal across South America were caused by the same B3.2 genotype, interestingly most viral sequences from South American marine mammals clustered more closely with one another than with most avian origin sequences, suggesting mammal-to-mammal spread within the continent 16 . Indeed, this ‘marine-mammal clade’ contains adaptations not seen in most avian-origin genotype B3.2 H5N1 HPAIVs including in the polymerase basic 2 (PB2) protein (D701N and/or Q591K) 13 , potentially indicating an increased zoonotic threat posed by these viruses. Following the spread of HPAIV though South America, two human infections with B3.2 H5N1 HPAIVs were confirmed in Chile and Ecuador in 2023 17,18 . The viral sequences from the Chilean human case also bore the PB2, Q591K and D701N mutations, and the individual had reported walking on a beach where there were dead sea lions present 19 . In October 2023, clade 2.3.4.4b H5N1 HPAIV genotype B3.2 was detected for the first time in the sub-Antarctic, marking the first ever confirmed incursion of HPAIV in this region 1 . This initial detection in a Southern Fulmar ( Fulmarus glacialoides ) on the Falkland Islands (FIs) was followed shortly by a detection of positive cases in wild birds across South Georgia (SG) 1,2 . Critically, alongside infection of avian species, mammalian species were also significantly hit by infection with large mortality events being observed in both southern elephant seals ( Mirounga leonina ) and Antarctic fur seals ( Arctocephalus gazella ) 1,2 . H5N1 HPAIV was subsequently detected in main-land Antarctica and on sub-Antarctic islands across the South Atlantic and South Indian Oceans in 2024 20 . Here we report on the disease events, genetic characterisation and likely introduction routes into these islands and assess the zoonotic and pandemic potential of these emerging H5N1 HPAIVs. Results Widespread detection of H5N1 HPAIV in wild avian and mammalian species across sub-Antarctic and Antarctic islands during the 2023/2024 austral summer. Following the earlier genetic characterisation of H5N1 HPAIV in both SG and the FIs 1,2 , sequencing efforts were undertaken on samples submitted over the 2023/24 austral summer (Supplementary Tables 1 and 2). HPAIV infection was confirmed in multiple mammalian and avian species, including southern elephant seals and Antarctic fur seals, as well as several bird species, including brown skua ( Stercorarius antarcticus ), snowy albatross ( Diomedea exulans ), gentoo penguin ( Pygoscelis papua ), king penguin ( Aptenodytes patagonicus ), southern rockhopper penguin ( Eudyptes chrysocome ), black-browed albatross ( Thalassarche melanophris ), and a variable hawk ( Geranoaetus polyosoma ) (Table 1). In total, between the 7 December 2023 and the 11 February 2024, 53% (n=27/51) of dead or moribund birds sampled on South Georgia (SG) tested positive for H5 HPAIV RNA. All H5-positive samples were also positive for N1 viral RNA (vRNA), except for one king penguin from Will Point, on the north coast of South Georgia, which was weakly positive by the M-gene and H5-HP RT-PCR only. H5 HPAIV RNA was detected in four of six bird species tested (wandering albatross, gentoo penguin, brown skua, and king penguin) across 13 locations and 11 timepoints between 16thDecember 2023 and 6th February 2024 (Table 1, Figure 1, Table S1). On the Falkland Islands (FIs), 69% (n=38/55) of dead or moribund birds sampled tested positive for H5 HPAIV vRNA. Again, all positive samples were also positive for N1 vRNA, with one exception (a brown skua from Saunders Island) which was weakly positive only for the M-gene and H5-HP vRNA. H5 HPAIV was detected in five of six bird species tested on the FIs (black-browed albatross, brown skua, gentoo penguin, southern rockhopper penguin, and variable hawk). These detections were made across seven locations and five timepoints between 19 January and 21 February 2024 (Table 1, Figure 1, Table S1). Across all avian samples, the highest percentage positivity for H5 HPAIV detections occurred in snowy albatrosses (100%; n=9/9), followed by black-browed albatrosses, with 92% (n=23/25); southern rockhopper penguins (86%; n=5/6), brown skuas (69%; n=6/13), king penguins (5/11; 46%); and gentoo penguins (45% (n=10/22); n=5/11 on SG and n=5/11 on FI). A single positive detection was recorded for a variable hawk (1/1; 100%). No HPAIV vRNA was detected in giant petrels ( Macronectes spp .) (0/6), snowy sheathbill ( Chionis albus ) (0/1), rock shag ( Leucocarbo magellanicus ) (0/1) or austral thrush ( Turdus falcklandii ) (0/1) (Table 1, Figure 1, Table S1). Among the positive samples, the strongest detection of vRNA was observed in brown skuas, particularly in brain and oropharyngeal samples (Table S1). Among mammalian species, H5N1 HPAIV vRNA was detected in 73% (n=51/72) dead or moribund individuals sampled on SG between 17 January and 11 February 2024, spanning 16 locations and 12 timepoints (Table 1, Figure 1, Table S2). The highest detection rate was observed in Antarctic fur seals, with 80% (n=40/50) individuals testing positive, the highest number of detections across all species tested. Southern elephant seals also tested positive, though at a lower frequency (55%; n=11/22). Multiple independent introductions of genotype B3.2 H5N1 HPAIV occurred into the Antarctic region from South America in the 2023/2024 austral summer. In total, whole genome sequences (WGSs) were generated for samples from 57 birds. This included 40 genomes from SG, comprising 12 from avian carcasses, (5 wandering albatrosses, 3 brown skuas, 3 gentoo penguins, and a single king penguin) and 28 from mammalian carcasses (22 fur seals and 6 elephant seals). An additional 17 avian-origin genomes were obtained from samples collected from the FIs (13 from black-browed albatrosses, 2 from gentoo penguins, 1 from a southern rockhopper penguin, and 1 from a variable hawk) (Table S1 and S2). Sequences were analysed alongside available H5N1 clade 2.3.4.4b full-genome sequences using the GenoFLU-multi tool 21 to assess genetic ancestry and segment-level reassortment patterns. These sequences were classified within the B3.2 genotype 21 , reflecting the continued southward expansion of this lineage from North America into South America during early 2022 1 . Novel sequences were analysed alongside representative H5N1 clade 2.3.4.4b HA sequences from North and South America in a time-resolved phylogenetic analysis (Figure 2). The maximum clade credibility (MCC) phylogeny of HA demonstrated five viral introductions into the FIs and SG, with each introduction numbered in chronological order (Figure 2). Each introduction corresponds to a distinct node containing groupings of FI and SG sequences. Groups 1, 3, 4 and 5 cluster with South American wild bird and mammalian viruses, with South American sequences (Argentina for groups 1, 3 and 4; Brazil for group 5) forming the basal node to those from SG and the FIs. This suggests that the virus arrived and where samples and sequences are available, appeared to diversify locally. Moreover, these introductions specifically cluster with marine mammal-origin viruses from Argentina, and Brazil, as well as mammalian and seabird viruses from Uruguay, indicating possible spillover events from South American marine mammals and seabirds to sub-Antarctic wildlife or at least species that range across both regions. Groups 1, 3 and 5 were comprised of only FI sequences each forming independent clusters consistent with separate introductions into the islands. Group 4, in contrast, consists of sequences from both SG and FI, with the majority being with SG with some FI interspersed. Group 5 detections include sequences from samples from the FIs alone with the basal node suggesting introduction from Brazil. The analysis revealed five possible viral introductions into SG and the FI. Temporal inference using time to most common ancestor (TMRCA) estimates revealed that the earliest introductions occurred into SG in August and September 2023 (Groups 2 and 4). The FIs experienced repeated introductions over a broader timeframe, with TMRCAs ranging from early July to November 2023 (Groups 1, 3 and 5). Although overlapping TMRCAs are observed, their distinct phylogenetic clustering supports their classification as independent introductions. To investigate these introductions, a discrete trait analysis with country-level sequence origin as the geographic state was used. Transition rates between locations were estimated from Markov jumps, and the mean rate with corresponding 95% highest posterior density (HPD) interval was calculated. This analysis indicated that introductions into FIs are primarily associated with Argentina with moderate mean transition rates with other South American sources (Figure 3, Table S3). The evolutionary emergence of mammalian-adaptive substitutions was examined using PB2 and PA gene sequences from H5N1 HPAIV Genotype B3.2. Amino acid residues at positions 591 and 701 previously linked to enhanced replication in mammalian cells 22,23 , were assessed. Among 57 viral genomes (29 avian (50.9%) / 28 Mammal (49.1%)) analysed, 94.7% (n=54/57) contained PB2 D701N and 98.2% (n=56/57) carried PB2 Q591K, with 94.7% (n=54/57) harbouring both mutations demonstrating strong co-occurrence of these adaptive markers in viruses detected in SG and FI. Most sequences clustered within the South American marine-mammal clade, which is partially defined by these substitutions 13 . Two viruses from FIs, one from a variable hawk ( Geranoaetus polyosoma , A/Variable_Hawk/Falkland_Islands/007272/2024) and one from a Black-browed albatross ( Thalassarche melanophris , A/Black−browed_albatross/Falkland_Islands/004745/2024) retained the avian-like PB2 701D residue while carrying the PB2 Q591K. Conversely, a brown skua ( Stercorarius antarcticus , A/Brown_Skua/Prion_Island/004878/2024) collected in SG retained both avian residues, PB2 701D and PB2 591Q and clustered clustered with the earlier South Georgia clade, derived from south American poultry (Figure 2). In the PA segment, most sequences fell within the same marine-mammal clade characterised by the M86I substitution, with the same brown skua representing the only exception (Figure 2, Figures S1–S2) Genotype B3.2 H5N1 HPAIVs possessing mammalian adaptive mutations have enhanced replicative fitness in human cells. Representative viral isolates were selected for functional assessment of mutations and compare the role of key residues within the polymerase complex via mini-replicon systems and live virus assays. Initially a mini-replicon system was used to assess mutations detected in the isolate A/southern_fulmar/Falkland Islands/133789/2023 (H5N1) (A Group 4 virus (Figure 2) denoted as mammalian-adapted B3.2 #1), namely the PB2 Q591K and D701N substitutions as well as the PA M86I mutation, previously associated with enhanced replication in mammalian hosts. A further non-mammalian adapted avian isolate, A/Brown_Skua/Hound_Bay/133947/2023 (A Group 2 virus (Figure 2) denoted from here as avian B3.2 #1), that retained the avian-like residues at these positions (PB2 591Q, PB2 701D and PA 86M), served as a comparative control. The polymerase assay assessment in human cells demonstrated that removing all three putative mammalian adaptations resulted in a polymerase that was unable to efficiently replicate in human cells (Figure 4A). Reverting the sequence back to PB2 Q591K or D701N resulted in a significant enhancement in polymerase activity and adding both together resulted in a further significant boost, showing these mutations have an additive effect. PA M86I had a far more modest impact, when combined with PB2 591K or 701N, suggesting it is less likely to contribute to the mammalian adaptation of this clade and in isolation did not affect polymerase activity (Figure 4A). Further, virus replication kinetics were assessed in primary human nasal epithelial cells (hNECs) from three independent human donors maintained as an air liquid interface (ALI). The panel used for the mini-replicon assay was expanded to include a further mammalian-adapted virus, A/Fur Seal/Salisbury_Plain/004762/2024 (A Group 4 virus (Figure 2) denoted as mammalian-adapted B3.2 #2) and a non-adapted avian isolate, A/ South_Georgia_Shag/King_Edward_Cove/141245/2023 (A Group 2 virus (Figure 2) denoted avian B3.2 #2). A further relevant isolate, A/dairy cattle/Texas/24-008749-001-original/2024 (denoted B3.13 Cattle/Texas) was used as a comparator that contains polymerase mutations associated with zoonotic risk and has demonstrated robust replication in human respiratory cells 24 . This panel enabled evaluation of which mammalian adaptive mutations influence viral replication competence in a human-relevant in vitro system 24 . Mammalian-adapted B3.2 isolates replicated to significantly higher titres than the avian isolates in hNECs across the different donors (Figure 4B). Notably, both B3.2 mammalian-adapted viruses achieved higher replication levels than the B3.13 Cattle/Texas H5N1 virus which possesses several established mammalian-adaptive polymerase mutations 24 (Figure 4B). The ability of representative viruses to bind different receptor moieties was also assessed. A pseudovirus-based virus entry assay was developed, using well described controls and demonstrated that the avian origin viruses, both genotype B3.13 and B3.2, preferentially bound α2,3-linked sialylated receptors (Figure 4C). In contrast, a human seasonal H1N1 derived from the 2009 pandemic (A/Rhode Island/04/2026; H1N1pdm09), and an H5N1 BB genotype HA modified to contain a well described receptor switching mutation L226Q 25 , had a strong preference for α2,6-linked sialylated receptors abundant in the human airway. The H7N9 strain bound both α2,3-linked and α2,6-linked sialylated receptors (Figure 4C). Finally, to confirm these results, reverse genetics approaches were developed for the mammalian-adapted B3.2 #1 isolate to generate a virus containing the H5 HA gene with the polybasic cleavage site removed and its six internal genes replaced with those of the attenuated laboratory strain PR8. This virus was rescued, propagated, purified and tested for its receptor binding using the biophysical assay, bio-layer interferometry, using two receptor binding analogues, the avian-like α2,3-linked and the human-like α2,6-linked sialic acid receptors. Consistent with the pseudovirus results, and in line analyses of European and North American clade 2.3.4.4b viruses 26 , the mammalian-adapted B3.2 #1 virus did not show any detectable binding to the human-like α2,6-linked sialic acid receptors, and showed strong binding to the avian-like α2,3-linked sialic acid receptor (Figure 4D). Antigenic assessment confirms good recognition by antisera raised to within-clade candidate vaccine viruses To further characterise the zoonotic risk posed by these viruses, a selection of five B3.2 H5N1 isolates; two mammalian (A/fur_seal/Miles_Bay/004783/2024, and A/fur_seal/Salisbury_Plain/004762/2024) and three avian (A/brown_skua/Hound_Bay/133949/2023, A/brown_skua/Hound_Bay/133947/2023 and A/kelp_gull/Moltke_Harbour/133754/2023) all from SG, were screened against ferret antisera raised towards clade 2.3.4.4b H5 pre-pandemic candidate vaccine viruses (CVVs) using the haemagglutination inhibition (HI) assay (Table 2). Antisera raised towards the A/Astrakhan/3212/2020 (H5N8) clade 2.3.4.4b CVV showed at least an 8-fold reduction in HI titre towards the B3.2 isolates from SG relative to the homologous virus. However, both of the H5N1 clade 2.3.4.4b CVVs (A/American Wigeon/South Carolina/22-000345-001/2021 and A/chicken/Ghana/AVL-76321VIR7050-39/2021) exhibited good recognition of the SG isolates with at most a 2-fold drop in HI titre relative to the homologous viruses. Sub-Antarctic B3.2 H5N1 HPAIV remain susceptible to antiviral therapeutics Further to assessing recognition by currently recommended H5 clade 2.3.4.4b CVVs, we also sought to phenotypically assess the susceptibility of the SG viruses to antiviral therapeutics. The five viruses from SG used to assess CVV recognition were also tested for susceptibility towards the neuraminidase inhibitors oseltamivir and zanamivir, alongside two human seasonal influenza H1N1 viruses, which are resistant (A/Alabama/03/2020) and susceptible (A/Illinois/45/2019) to oseltamivir. The five B3.2 H5N1 viruses had half maximal inhibition concentrations (IC 50 ) ranging between 1.37 nM and 4.70 nM for oseltamivir (Figure 5A), and between 0.40 nM and 0.51 nM for zanamivir (Figure 5B) demonstrating that they remain susceptible to these antivirals. Discussion Novel data generated here has demonstrated potential pathways for virus incursion from south America into the sub-Antarctic and Antarctic region, impacts of mammalian-associated mutations in PB2 and the current status of viruses in the sub-Antarctic region from an antigenic perspective. Following widespread circulation of H5N1 HPAIV across South America in late 2022 8,10,11 , the virus was first detected in SG in late 2023 1 . Surveillance during 2023/24 austral summer confirmed extensive detection in wild birds across SG and FIs, all belonging to the B3.2 genotype, commonly circulating on the South American Continent 27 1 8,10,11 . Phylogenetic analysis indicates up to five separate incursions into SG and FIs from continental South America before southward spread to the Antarctic Peninsula. This suggests that SG may act as a stepping stone for further dissemination to remote sub-Antarctic regions 31 , exemplified by the detection in March 2025 of genetically closely related H5N1 in three brown skuas on Gough Island 32 , suggesting transoceanic spread. These findings highlight the potential for seabird-mediated long-range transmission and raise concern that further spread to continental regions such as South Africa or Australia could enable wider dissemination via established flyways 31 . Marine mammals may represent an additional route of virus introduction and maintenance in sub-Antarctic ecosystems. Several incursions involved viruses carrying mammalian adaptive markers suggesting a secondary infection of avian species with mammalian-derived strains likely through scavenging of infected carcasses 31 . In SG, non-colonial bird species are likely exposed through scavenging, while dense marine mammal colonies provide an environment conducive to mammal-to-mammal-transmission. Cross-species interactions, including scavenging activities may facilitate bi-directional exchange between avian and mammalian hosts although the limited availability of samples from these remote ecosystems constrain our understanding of transmission pathways. Zoonotic potential is a key factor for these viruses with multiple factors being associated with increased risk 33 including: ( i ) polymerase activity in novel species ( ii ) HA receptor binding and ( iii ) HA stability. Avian-origin influenza viruses have restricted replication in mammalian cells due to poor usage of pro-viral host factors by the virus polymerase 33 although this can be overcome through mutations in PB2 such as E627K, Q591K or D701N 34 . Additionally avian influenza viruses generally have a strong binding preference towards sialylated glycans that contain an α2,3-linkage between the terminal sialic acid and the penultimate galactose 33 contrasting with human-transmissible viruses that preferentially bind α2,6-linked sialic acids. Several mutations have been described that allow a switch in receptor preference from α2,3- to α2,6-linked sialic acids in the H5 subtype, including Q226L 25,35,36 . Here, viruses from SG and FIs often contained the PB2 Q591K, D701N and PA M86I mutations. Interestingly, 74 human infections have been detected in the Americas between 2022 and February 2025 37 , mostly linked to genotype B3.13 circulating in US dairy cattle 37 . Two genotype B3.2 human infections have been reported, in Ecuador (January 2023) 18 and in Chile 17 (March 2023) with the latter involving PB2 Q591K and D701N 19 . The Chilean virus also caused fatal disease in ferrets, with animal-to-animal transmission in a direct contact setting, but not via respiratory droplet or fomites 38 . We demonstrated a marked difference in replication kinetics in primary human airway cultures between the avian and marine-mammal-clade viruses detected in SG. Furthermore, sequential reversion of putative adaptive mutations (PB2 591K and 701N, and PA 86I) to their avian equivalent (591Q, 701D, and PA 86M) reduced polymerase activity in human cells, with Q591K and D701N having the greatest effect. This suggests that mammalian adaptations in the A/Southern fulmar/Falkland Islands/133789/2023 isolate enhanced the ability of this virus to replicate in human cells, likely due to an accumulation of polymerase mutations following sustained mammal-to-mammal transmission. Phylogenetic analyses also supported the potential for mammal-to-mammal transmission as clades displayed continued circulation in marine mammal populations in SG. There were also detections from avian hosts interspersed with these clades demonstrating avian infection with these viruses. However, the following assessment of mutations in HA, the changes required to enable a shift in receptor binding properties has not occurred. The impact of such changes on avian replication dynamics in different avian species remain unclear. Both genotypes B3.13 and B3.2 have retained preferential binding to α2,3 sialic acid linked receptors. This is highly significant in risk assessment of emerging viruses from a zoonotic perspective 33 . Finally, whilst there was a reduction in antigenic-reactivity against the Astrakhan CVV, good reactivity was detected against two other CVVs namely, A/American Wigeon/South Carolina/22-000345-001/2021 and A/chicken/Ghana/AVL-76321VIR7050-39/2021 suggesting strong antigenic match. Whilst antigenic evolution requires constant assessment wherever viruses are emerging, these results indicate that the HAs have not yet evolved to escape neutralisation by key CVVs. The data generated here also indicates that these viruses remain susceptible to commonly licensed neuraminidase inhibitors. Sensitivity to neuraminidase inhibitors reveals that current preparedness methods are effective, although continual assessment is required. In conclusion, the data generated has demonstrated the need to maximise genetic characterisation of viruses wherever outbreaks occur. With the continued detection of H5N1 globally it is no longer sufficient to simply subtype through analysis of HA and NA sequences. Genotypes and markers of adaptation must be characterised and where risk increases, sequences made accessible to the global community. Our data has demonstrated some impact on replication of mammalian mutations, and further studies detailing the impact of these mutations on in-vivo infection need to be prioritised to support the observations here. Methods Ethics and biosafety HPAIV work is categorised as specified animal pathogens order (SAPO) 4 and Advisory Committee on Dangerous Pathogens (ACDP) hazard group 3 by United Kingdom (UK) regulations. Work with these viruses was undertaken in a licensed containment level (CL) 3 (ACDP-3/SAPO4) facility of The Animal and Plant Health Agency (APHA), The Pirbright Institute under GMRA (BAG-RA-226) and The Francis Crick Institute Isolation of virus in specific pathogen-free (SPF) embryonated fowls’ eggs (EFEs) was undertaken under Home Office license PP9307748. All virus risk assessments were approved by the appropriate internal committees, as well as the UK Health and Safety Executive (HSE) and, where necessary, the UK scientific advisory committee for genetic modification (SACGM). Ethics for the use of the primary human airway epithelial cells were as described previously 39 . Briefly, donors provided written consent and Ethics approval was given through the Living Airway Biobank, administered through the UCL Great Ormond Street Institute of Child Health (REC reference: 19/NW/0171, IRAS project ID: 261511, Northwest Liverpool East Research Ethics Committee). Nasal brushings were obtained by trained clinicians from adult (30–50 years) donors who reported no respiratory symptoms in the preceding 7-weeks. Brushings were taken from the inferior nasal concha zone using cytological brushes (Scientific Laboratory Supplies, CYT1050). All methods were performed following the relevant guidelines and regulations. Samples and collection We previously reported the detection of H5N1 HPAIV in samples collected from birds on SG between 8 October and 6 December 2023, and from mammals between 31 October and 9 December 2023, as well as from birds on the FI between 30 October and 10 December 2023 1 . Here, we present findings from a subsequent sampling period, encompassing birds and mammals on SG (7 December 2023 to 11 February 2024) and birds on the Falkland Islands (19 January to 21 February 2024) (Table S1). A total of 211 samples were collected from 106 individual dead birds: 116 samples from 51 individuals on SG, and 95 samples from 55 individuals on the FI. On SG, bird samples were obtained from 25 locations across 22 discrete timepoints between 7 December 2023 and 11 February 2024. On the FI, samples were collected from 14 locations over 9 timepoints between 19 January and 21 February 2024. Mammalian sampling was undertaken wherever mortalities were detected. A total of 224 samples were collected from 67 individual animals, including 49 Antarctic fur seals and 18 Southern elephant seals, across 23 locations and 16 timepoints between 14 December 2023 and 11 February 2024 (Table S2). Avian samples included oropharyngeal (OP), cloacal (C), brain tissue, and environmental faecal swabs. Mammalian samples included oral, nasal, rectal, ocular, and brain tissue samples, as well as environmental faecal swabs. All samples were collected from animals found dead in the field following suspicion of infection, under appropriate permits granted by the governments of the FI and SG, in collaboration with the British Antarctic Survey (BAS). Sampling was conducted with strict biosafety protocols in place to ensure the safety of personnel and to minimise the risk of pathogen transmission via fomites to healthy wildlife. Field personnel recorded location, species, and estimated age of each individual sampled. Molecular assessment of samples Swabs were immersed in 1 mL Leibovitz's L-15 Medium (L-15) (Gibco, USA). A section of each tissue (~1 gram) was transferred into L-15 medium and roughly homogenised. Supernatants from the processed swabs and tissues was used for RNA extraction as described previously 40 before testing extracted RNA for the presence or absence of viral nucleic acid by real-time reverse transcription polymerase chain reaction (RT-PCR) assays specific for the Matrix (M) gene 41,42 , the haemagglutinin (HA) gene 40 and the N1 neuraminidase (NA) gene 43 . The cut off for positivity for each of the three assays was ≤36.0 Cq. All amplifications were undertaken using an AriaMx qPCR System (Agilent, United Kingdom). Samples yielding Cq values <30 were used for attempted virus isolation. Genetic evaluation of H5N1 HPAIV sequences To generate sequence data for positive PCR samples, extracted viral RNA (vRNA) was converted to double-stranded cDNA and amplified using a one-step RT-PCR using SuperScript III One-Step RT-PCR kit (Thermo Fisher Scientific) 1 . Previously described approaches were used to generate genetic data 44,45 . With PCR products being processed using Agencourt AMPure XP beads (Beckman Coulter) and using the Native Barcoding Kit (Oxford Nanopore Technologies) GridION Mk1 (Oxford Nanopore Technologies) technologies as per manufacturer's instructions. All influenza sequences generated and used in this study are available through the GISAID EpiFlu Database (https://www.gisaid.org) and NCBI. The genotype of the viruses sequenced in this study was confirmed using the Genoflu-multi tool (https://github.com/moncla-lab/GenoFLU-multi). For the time-resolved phylogenetic analysis, all available hemagglutinin (HA) sequences (GISAID Accessed 28 May 2025) from South America and representative sequences from North America determined by PARNAS 46 were combined with those from SG and the FIs to infer phylogeny using BEAST version 1.10.5 47 , incorporating the BEAGLE library 48 . The analysis employed the HKYnucleotide substitution model 49 . A strict molecular clock and a coalescent constant size tree prior. Two independent Markov Chain Monte Carlo (MCMC) runs of 200 million steps each were conducted, with sampling every 20,000 steps ensuring all relevant effective sample sizes reached at least 200 as assessed in Tracer 1.7.2 50 . The initial 10% of each chain was discarded as burn-in, and the chains were combined using the LogCombiner tool in the BEAST package. Markov jump estimates of viral transition between discrete locations were obtained from the posterior distribution of the trees inferred by BEAST. For each transition the mean transition rate and 95% highest posterior density (HPD) interval were summarised. Post processing was performed using R (v4.5.0) using the ggplot2 51 and tidyverse packages 52 . Geographic visualisation of dispersal pathways was displayed using a custom python script utilising geopandas 53 , matplotlib 54 and cartopy 55 . Sequences were genotyped according to the USDA schema, using the GenoFLU tool (https://github.com/USDA-VS/GenoFLU) 21 Ancestral amino acid sequences were reconstructed with the ancestral module of TreeTime (v0.11.4) using the Tones-Taylor-Thornton (JTT) amino acid substitution model, based maximum likelihood phylogenies inferred for PB2 and PA segments. Phylogenetic trees were generated using maximum likelihood methods with 1,000 ultrafast bootstrap replicates and 1,000 SH-aLRT replicates to assess branch support. The input dataset included a subset of H5 subtype sequences representing over 97% of the known genetic diversity from North and South America, determined using PARNAS 46 . Virus isolation Virus isolation was undertaken using 200 µl of supernatant from clinical samples and assessment by undertaking two passages in 9-day-old specified pathogen free (SPF) embryonated fowl eggs (EFE) as previously described 56 . Successful isolation was confirmed by hemagglutination assay on harvested allantoic fluids using chicken red blood cells as previously described. Titres of >1/4 were considered positive for isolation. Isolates were further propagated in EFEs and titrated by plaque assays on MDCK cells as previously described 26 . Cell culture Human embryonic kidney 293T (HEK-293T) cells and the Madin-Darby canine kidney (MDCK) cells were acquired from Central Service Unites (CSU) of The Pirbright Institute (TPI). Both cell lines were cultured with Dulbecco’s Modified Eagle’s medium (DMEM), supplemented with 10% foetal calf serum (FCS), at 37°C with 5% CO 2 . Virus replication assessment Primary human airway cultures were differentiated as previously described 24,39 . Human nasal brushings from adult donors (n=3) were expanded on 3T3-J2 fibroblasts and frozen down as passage 1 (p1). Basal cells were then revived as needed, expanded in co-culture with 3T3-J2 fibroblasts until passage 2 and differentiated at an air-liquid interface as previously described 39 . Infection of the primary human airway cultures was undertaken as previously described 24 . Briefly, cells were washed with Dulbecco's phosphate-buffered saline supplemented with Calcium and Magnesium (DPBS++) to remove mucus and cell debris. Cultures were infected with 200µL of virus diluted in DPBS++ at an MOI of 0.01 and incubated at 37 o C for 1 hour. Virus-containing inoculum was aspirated, and wells were washed twice with DPBS++ to remove unbound virus. Time points were collected by adding 200µL of DPBS++ and incubated at 37 o C for 10 minutes before removal and storage. Infectious virus was determined by plaque assay on MDCK cells. The time course was undertaken at 37 o C, 5% CO 2 with samples being collected at 24, 48, 72 and 96 hours post-inoculation. Reverse genetics The HA with monobasic cleavage site and NA sequences of A/Southern fulmar/Falkland_Islands/133789/2023 (H5N1) were synthesised by GeneScript and cloned into the bidirectional pHW2000 vector. The virus rescue was undertaken as previously described using the internal genes from A/Puerto Rico/8/193 (H1N1) (PR8) 57 . The rescued viruses were propagated in 9- to 10-day old embryonated chicken eggs. Influenza minireplicon assay Expression plasmids for the B3.2 minigenome assays were synthesised by Geneart, mutated using site directed mutagenesis and subcloned into pCAGGS. HEK 293T cells were transfected in 24-well plates using lipofectamine 3000 transfection reagent (Thermo Fisher) with the following amount of pCAGGS-containing expression plasmids: 40ng of PB2, 40ng of PB1, 20ng of PA, 80ng of NP, 40ng of Renilla luciferase and 80ng of human-specific pol I vRNA Firefly luciferase. Cells were lysed 24 hours post-transfection using passive lysis buffer (Promega) and polymerase activity was measured using the Dual-Luciferase Reporter assay system (Promega) and a FLUOstar Omega plate reader (BMG Labtech). Firefly luciferase signal was normalised to Renilla luciferase signal to give relative luminescence units (RLU). Immunoblotting analysis HEK 293T cells were transfected in 6-well plates using lipofectamine 3000 transfection reagents (Thermo Fisher) with pCAGGS expression plasmids in the same ratio as described for the influenza minireplicon assay: 160ng of PB2, 160ng of PB1, 80ng of PA, 320ng of NP, 160ng of Renilla luciferase and 320ng of human-specific pol I vRNA Firefly luciferase. Cells were lysed 24 hours post-transfection in radioimmunoprecipitation assay buffer (10 mM Tris-HCl pH 8.0, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% SDS, 140 mM NaCl) supplemented with a protease inhibitor tablet (ThermoFisher). Lysates were incubated on ice for 1 h then centrifuged (16,000 g, 10 min). The supernatant was mixed with Laemmli buffer (Bio-Rad) supplemented with 10% beta-mercaptoethanol (Sigma Aldrich). Protein samples were resolved on a 4-20% Mini-PROTEAN polyacrylamide gel (Bio-Rad) and transferred to a low-fluorescence PVDF membrane (Merck). Membranes were blocked in TBS-5% milk for 1 h at room temperature and then washed with TBS. Primary antibodies (mouse α-tubulin, Abcam ab7291, 1:1250; rabbit α-PB2, Genetex GTX125926, 1:500; rabbit α-PA, Genetex GTX118991, 1:500) were diluted in TBS-T-2% milk and membranes were incubated overnight at 4 °C. Membranes were then washed 4x in TBS-T, incubated with secondary antibodies (goat α-mouse Alexa Fluor® 680, Abcam ab175775, 1:10,000; goat α-rabbit IRDye ® 800CW, Li-Cor 926-32211, 1:10,000) in TBS-T-2% milk for 45 minutes, washed 4x in TBS-T and washed 1x in TBS. Membranes were imaged using an Odyssey DLx (Li-Cor Biosciences) and Image Studio Lite software. Pseudovirus receptor binding preference assay The VSV-G expression plasmid, and lentiviral packaging genes and luciferase genome were used as previously described 58 . The following human codon optimised HA, NA, protease and sialyl-transferase expression plasmids were synthesised by Genscript in pcDNA3.1: H1HA - A/Rhode Island/04/2016(H1N1pdm09; ANM90381.1), H1HA - A/duck/Bavaria/1/1977(H1N1; ALG00691.1), N1NA - A/England/195/2009 (H1N1pdm09; ACR15618.1), H7HA and N9NA - A/Shanghai/02/2013 (H7N9; YP_009118475.1, YP_009118481.1), H5HA – A/dairy cattle/Texas/24-008749-001-original/2024 (H5N1; WPD27583.1), H5HA and N1NA - A/chicken/England/085598/2022 (H5N1; EPI2089022, EPI2089021), H5HA - A/southern fulmar/Falkland_Islands/133789/2023(H5N1; EPI2795786), human TMPRSS11D/HAT (EAX05559.1) human ST6 beta-galactosamide alpha-2,6-sialyltranferase 1(ST6; AAH31476.1) and CMP-N-acetylneuraminate-beta-galactosamide-alpha-2,3-sialyltransferase 4 (ST3; XP_047283377.1). Pseudovirus was produced in HEK 293T cells seeded in 6-well dishes. Cells were co-transfected using lipofectamine 3000 (Invitrogen) at ~70% confluency with 0.6µg of the luciferase reporter constructs (pCSFLW), 0.4µg of the HIV packaging plasmid pGAG-POL, and either 0.4µg of VSV-G (VSV-G only) or 0.4µg of the defined HA, 0.2µg of TMPRSS11D and 0.1µg of the defined NA. Pseudovirus containing supernatants were collected at 48 and 72 hours post-transfection, clarified by centrifugation, aliquoted and frozen at –80 o C. All H1N1 pseudoviruses were produced with the N1 NA from Eng/195, H7N9 HA was produced with the strain-matched NA, all H5N1 pseudoviruses were produced with the A/chicken/England/085598/2022 NA. HEK293 cells lacking sialyl-transferase activity (HEK 293 ΔST3GAL3/4/6 ΔST61/2; aka 293delST cells) were a kind gift from Professors Henrik Clausen and Yoshiki Narimatsu from the University of Copenhagen 59 . These cells were seeded in 6-well plates and transiently transfected at 70% confluency using lipofectamine 3000 with 500ng of an empty pcDNA3.1 vector, or either ST3 or ST6 expression constructs. Twenty-four hours post-transfection, cells were washed, resuspended in fresh media and seeded into 96-well plates at a density of 1x10 4 cells per well. Pseudovirus diluted in media containing oseltamivir (to a final concentration of 0.5 µM) was then added to each well. 48 hours post-transduction cells were lysed and read using Brightglo reagent (Promega) and a Glomax discover plate reader (Promega). Data for each pseudovirus was normalised to the empty vector control. Virus receptor binding by bio-layer interferometry Pelleted virus was purified through 30% and 60% sucrose gradients at 27,000rpm for 2h at 4℃. The purified virus was then diluted in HBS buffer containing 10μM oseltamivir carboxylate (Roche) and 10μM zanamivir (GSK). Binding affinity to avian-like sugar analogue 3SLN and human-like sugar analogue 6SLN was tested by Octet® R8 system (Sartorius) using streptavidin biosensors (Sartorius). Virus binding affinity was normalised to fractional saturation and the concentrations of sugar loadings 60 . Assessment of antigenicity Five virus isolates were tested in the hemagglutination inhibition (HI) assay using turkey erythrocytes and ferret antisera for antigenic characterisation as previously described 61 . Isolates were selected to be representative of Group 4 (2 isolates) and Group 3 (3 isolates) (Figure 1, Table 2). Ferret antisera had been previously raised against existing human pre-pandemic candidate vaccine strains of H5 clade 2.3.4.4b (Table 2). Neuraminidase inhibition assay Oseltamivir carboxylate and zanamivir were supplied by Hoffmann-La Roche and GlaxoSmithKline, respectively. MES assay buffer was prepared with 32.5mM MES (2-Morpholinoethanesulfonic acid, Sigma-Aldrich) and 4mM CaCl 2 in ddH 2 O, adjusted to pH 6.5. The MUNANA substrate solution consisted of 2’-(4-methylumbelliferyl)-α-D-N-acetylneuraminic acid (Sigma-Aldrich). The stop solution was prepared with 0.1 M glycine, 25% ethanol, 0.267% Super Q, and ddH 2 O. The susceptibility of virus isolates to oseltamivir and zanamivir was assessed using the MUNANA assay, which measures neuraminidase activity based on cleavage of the fluorogenic substrate and release of 4-methylumbelliferone (4-MU) as previously described 62 . For virus titration and inoculum normalization, test viruses were serially diluted in MES buffer into black, full-bottom 96-well microplates (Corning) and titrated in duplicate from 1:2 to 1:2048 to a final volume of 20µL. Subsequently, 30µL of 100µM MUNANA substrate was added to each well, and plates were incubated at 37°C with 5% CO₂ for 1h with shaking. Reactions were terminated with 150µL of stop solution, and fluorescence was measured using an Infinite 200 PRO plate reader (TECAN) with excitation at 360 nm and emission at 465 nm. Virus dilutions for the inhibition assay were determined using internationally defined methodologies 63 . For inhibition assays, oseltamivir carboxylate and zanamivir were serially diluted 4-fold in MES buffer to final concentrations of 4000–0.015nM in black, full-bottom 96-well microplates to a final volume of 20µL and incubated at 37°C with 5% CO₂ for 30 min with shaking. MUNANA substrate was then added and incubated for 1h, after which the reaction was stopped and fluorescence measured as described above. Declarations Acknowledgments The authors would like to thank Professors Henrik Clausen and Yoshiki Narimatsu of the University of Copenhagen for kindly sharing the HEK293 ΔST3GAL3/4/6 ΔST61/2 cells. Funding statement This research received no external funding. The testing and generation of the viral sequences was funded by the Department for Environment, Food and Rural Affairs (Defra, UK) and the Devolved Administrations of Scotland and Wales, through the following programmes: SV3400, SV3032, SV3006, SE2213 and SE2227. This work was also supported by the Biotechnology and Biological Sciences Research Council (BBSRC) and Department for Environment, Food and Rural Affairs (Defra, UK) research initiative ‘FluTrailMap’ [grant numbers BB/Y007271/1, BB/Y007298/1 ] and the Medical Research Council (MRC) and Defra research initiative ‘FluTrailMap-One Health’ [grant number MR/Y03368X/1]. Funded by the European Union under grant agreement (101084171) - (Kappa-Flu). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or REA. Neither the European Union nor the granting authority can be held responsible for them. Funding as also provided from Innovate UK (grant number 10085195). Work at Pirbright was supported by the BBSRC via the Pirbright Institute’s Strategic Programme Grants (ISPGs) [BBS/E/PI/230002A; BBS/E/PI/230002B], BBSRC National Bioscience Research Infrastructure: High Containment and Low Containment Services and Science Platforms grants [BBS/E/PI/23NB0004, BBS/E/PI/23NB0003]. Competing interests statement The authors declare no competing interests. Author contributions statement Conceptualisation: ACB, JJ, TPP; formal analysis: BM, JGLJ, SR, JQ, AMPB, RH, MB, SMR, HC, BC, JJ, ACB; investigation: BM, JGLJ, SR, JQ, RH, BM; resources: ACB, JJ, IHB, NL, ZF, EMF; writing—original draft, BM, JGLJ, TPP, JJ, ACB; writing—review and editing: ACB, JJ, APMB, AB, ZF, EMF, SMR, MB, IHB, JGLJ, BM, TPP. All authors have read and agreed to the final version of the manuscript. References Banyard, A. C. et al. Detection and spread of high pathogenicity avian influenza virus H5N1 in the Antarctic Region. Nature Communications 15 , 7433, doi:10.1038/s41467-024-51490-8 (2024). Bennison, A. et al. A case study of highly pathogenic avian influenza (HPAI) H5N1 at Bird Island, South Georgia: the first documented outbreak in the subantarctic region. Bird Study 71 , 380-391, doi:10.1080/00063657.2024.2396563 (2024). Lisovski, S. et al. Unexpected Delayed Incursion of Highly Pathogenic Avian Influenza H5N1 (Clade 2.3.4.4b) Into the Antarctic Region. Influenza Other Respir Viruses 18 , e70010, doi:10.1111/irv.70010 (2024). Xie, R. et al. The episodic resurgence of highly pathogenic avian influenza H5 virus. Nature 622 , 810-817, doi:10.1038/s41586-023-06631-2 (2023). Caliendo, V. et al. Transatlantic spread of highly pathogenic avian influenza H5N1 by wild birds from Europe to North America in 2021. Sci Rep 12 , 11729, doi:10.1038/s41598-022-13447-z (2022). Alkie, T. N. et al. Recurring Trans-Atlantic Incursion of Clade 2.3.4.4b H5N1 Viruses by Long Distance Migratory Birds from Northern Europe to Canada in 2022/2023. Viruses 15 , 1836 (2023). Kandeil, A. et al. Rapid evolution of A(H5N1) influenza viruses after intercontinental spread to North America. Nature Communications 14 , 3082, doi:10.1038/s41467-023-38415-7 (2023). World Organisation for Animal Health (WOAH). World Animal Health Information System (WAHIS) Events Management. Search: High pathogenicity avian influenza viruses (poultry) and South American countries. , (2025). Ruiz-Saenz, J., Martinez-Gutierrez, M. & Pujol, F. H. Multiple introductions of highly pathogenic avian influenza H5N1 clade 2.3.4.4b into South America. Travel Medicine and Infectious Disease 53 , 102591, doi:https://doi.org/10.1016/j.tmaid.2023.102591 (2023). Leguia, M. et al. Highly pathogenic avian influenza A (H5N1) in marine mammals and seabirds in Peru. Nat Commun 14 , 5489, doi:10.1038/s41467-023-41182-0 (2023). Marandino, A. et al. Spreading of the High-Pathogenicity Avian Influenza (H5N1) Virus of Clade 2.3.4.4b into Uruguay. Viruses 15 , doi:10.3390/v15091906 (2023). Kuiken, T. et al. Emergence, spread, and impact of high-pathogenicity avian influenza H5 in wild birds and mammals of South America and Antarctica. Conservation Biology n/a , e70052, doi:https://doi.org/10.1111/cobi.70052 (2025). Pardo-Roa, C. et al. Cross-species and mammal-to-mammal transmission of clade 2.3.4.4b highly pathogenic avian influenza A/H5N1 with PB2 adaptations. Nature Communications 16 , 2232, doi:10.1038/s41467-025-57338-z (2025). Plaza, P. I., Gamarra-Toledo, V., Rodríguez Euguí, J., Rosciano, N. & Lambertucci, S. A. Pacific and Atlantic sea lion mortality caused by highly pathogenic Avian Influenza A(H5N1) in South America. Travel Medicine and Infectious Disease 59 , 102712, doi:https://doi.org/10.1016/j.tmaid.2024.102712 (2024). de Araújo, A. C. et al. Incursion of Highly Pathogenic Avian Influenza A(H5N1) Clade 2.3.4.4b Virus, Brazil, 2023. Emerg Infect Dis 30 , 619-621, doi:10.3201/eid3003.231157 (2024). Uhart, M. M. et al. Epidemiological data of an influenza A/H5N1 outbreak in elephant seals in Argentina indicates mammal-to-mammal transmission. Nat Commun 15 , 9516, doi:10.1038/s41467-024-53766-5 (2024). Castillo, A. et al. The first case of human infection with H5N1 avian Influenza A virus in Chile. Journal of Travel Medicine 30 , taad083 (2023). Bruno, A. et al. First case of human infection with highly pathogenic H5 avian Influenza A virus in South America: A new zoonotic pandemic threat for 2023? Journal of Travel Medicine 30 , doi:10.1093/jtm/taad032 (2023). Pardo-Roa, C. et al. Cross-species transmission and PB2 mammalian adaptations of highly pathogenic avian influenza A/H5N1 viruses in Chile. bioRxiv , doi:10.1101/2023.06.30.547205 (2023). Bennett-Laso, B. et al. Confirmation of highly pathogenic avian influenza H5N1 in skuas, Antarctica 2024. Frontiers in Veterinary Science Volume 11 - 2024 , doi:10.3389/fvets.2024.1423404 (2024). Youk, S. et al. H5N1 highly pathogenic avian influenza clade 2.3.4.4b in wild and domestic birds: Introductions into the United States and reassortments, December 2021-April 2022. Virology 587 , 109860, doi:10.1016/j.virol.2023.109860 (2023). Li, Z. et al. Molecular Basis of Replication of Duck H5N1 Influenza Viruses in a Mammalian Mouse Model. Journal of Virology 79 , 12058-12064, doi:doi:10.1128/jvi.79.18.12058-12064.2005 (2005). Yamada, S. et al. Biological and structural characterization of a host-adapting amino acid in influenza virus. PLoS Pathog 6 , e1001034, doi:10.1371/journal.ppat.1001034 (2010). Dholakia, V. et al. Polymerase mutations underlie early adaptation of H5N1 influenza virus to dairy cattle and other mammals. bioRxiv , 2025.2001.2006.631435, doi:10.1101/2025.01.06.631435 (2025). Lin, T.-H. et al. A single mutation in bovine influenza H5N1 hemagglutinin switches specificity to human receptors. Science 386 , 1128-1134, doi:doi:10.1126/science.adt0180 (2024). Yang, J. et al. The haemagglutinin gene of bovine-origin H5N1 influenza viruses currently retains receptor-binding and pH-fusion characteristics of avian host phenotype. Emerg Microbes Infect 14 , 2451052, doi:10.1080/22221751.2025.2451052 (2025). Pan American Health Organization (PAHO). Epidemiological Update Avian Influenza A(H5N1) in the Americas Region , (2025). Phillips, R. A., Croxall, J. P., Silk, J. R. D. & Briggs, D. R. Foraging ecology of albatrosses and petrels from South Georgia: two decades of insights from tracking technologies. Aquatic Conservation: Marine and Freshwater Ecosystems 17 , S6-S21, doi:https://doi.org/10.1002/aqc.906 (2007). McCann, T. S. & Rothery, P. Population size and status of the southern elephant seal (Mirounga leonina) at South Georgia, 1951–1985. Polar Biology 8 , 305-309, doi:10.1007/BF00263179 (1988). Croxall, J. P., McCann, T. S., Prince, P. A. & Rothery, P. in Antarctic Ocean and Resources Variability. (ed Dietrich Sahrhage) 261-285 (Springer Berlin Heidelberg). Clessin, A. et al. Mass mortality events in the sub-Antarctic Indian Ocean caused by long-distance circumpolar spread of highly pathogenic avian influenza H5N1 clade 2.3.4.4b. bioRxiv , 2025.2002.2025.640068, doi:10.1101/2025.02.25.640068 (2025). Steinfurth, A. et al. Investigating high pathogenicity avian influenza virus incursions to remote islands: Detection of H5N1 on Gough Island in the South Atlantic Ocean. bioRxiv , 2025.2009.2006.674618, doi:10.1101/2025.09.06.674618 (2025). Long, J. S., Mistry, B., Haslam, S. M. & Barclay, W. S. Host and viral determinants of influenza A virus species specificity. Nature Reviews Microbiology 17 , 67-81, doi:10.1038/s41579-018-0115-z (2019). Carrique, L. et al. Host ANP32A mediates the assembly of the influenza virus replicase. Nature 587 , 638-643, doi:10.1038/s41586-020-2927-z (2020). Dadonaite, B. et al. Deep mutational scanning of H5 hemagglutinin to inform influenza virus surveillance. PLoS Biol 22 , e3002916, doi:10.1371/journal.pbio.3002916 (2024). Ríos Carrasco, M. et al. The Q226L mutation can convert a highly pathogenic H5 2.3.4.4e virus to bind human-type receptors. Proceedings of the National Academy of Sciences 122 , e2419800122, doi:doi:10.1073/pnas.2419800122 (2025). USA Centers for Disease Control and Prevention (CDC). H5 Bird Flu: Current Situation , (2025). Pulit-Penaloza, J. A. et al. Highly pathogenic avian influenza A(H5N1) virus of clade 2.3.4.4b isolated from a human case in Chile causes fatal disease and transmits between co-housed ferrets. Emerg Microbes Infect 13 , 2332667, doi:10.1080/22221751.2024.2332667 (2024). Woodall, M. N. J. et al. Age-specific nasal epithelial responses to SARS-CoV-2 infection. Nat Microbiol 9 , 1293-1311, doi:10.1038/s41564-024-01658-1 (2024). James, J. et al. Rapid and sensitive detection of high pathogenicity Eurasian clade 2.3.4.4b avian influenza viruses in wild birds and poultry. Journal of Virological Methods 301 , 114454, doi:https://doi.org/10.1016/j.jviromet.2022.114454 (2022). Nagy, A. et al. A universal RT-qPCR assay for "One Health" detection of influenza A viruses. PLoS One 16 , e0244669, doi:10.1371/journal.pone.0244669 (2021). Spackman, E. et al. Development of a real-time reverse transcriptase PCR assay for type A influenza virus and the avian H5 and H7 hemagglutinin subtypes. J Clin Microbiol 40 , 3256-3260, doi:10.1128/jcm.40.9.3256-3260.2002 (2002). Payungporn, S. et al. Single step multiplex real-time RT-PCR for H5N1 influenza A virus detection. Journal of Virological Methods 131 , 143-147, doi:https://doi.org/10.1016/j.jviromet.2005.08.004 (2006). Chrzastek, K. et al. Use of Sequence-Independent, Single-Primer-Amplification (SISPA) for rapid detection, identification, and characterization of avian RNA viruses. Virology 509 , 159-166, doi:10.1016/j.virol.2017.06.019 (2017). Leyson, C. et al. Pathogenicity and genomic changes of a 2016 European H5N8 highly pathogenic avian influenza virus (clade 2.3.4.4) in experimentally infected mallards and chickens. Virology 537 , 172-185, doi:10.1016/j.virol.2019.08.020 (2019). Markin, A. et al. PARNAS: Objectively Selecting the Most Representative Taxa on a Phylogeny. Syst Biol 72 , 1052-1063, doi:10.1093/sysbio/syad028 (2023). Baele, G. et al. BEAST X for Bayesian phylogenetic, phylogeographic and phylodynamic inference. Nature Methods 22 , 1653-1656, doi:10.1038/s41592-025-02751-x (2025). Ayres, D. L. et al. BEAGLE: An Application Programming Interface and High-Performance Computing Library for Statistical Phylogenetics. Systematic Biology 61 , 170-173, doi:10.1093/sysbio/syr100 (2011). Hasegawa, M., Kishino, H. & Yano, T. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. J Mol Evol 22 , 160-174, doi:10.1007/bf02101694 (1985). Rambaut, A., Drummond, A. J., Xie, D., Baele, G. & Suchard, M. A. Posterior Summarization in Bayesian Phylogenetics Using Tracer 1.7. Syst Biol 67 , 901-904, doi:10.1093/sysbio/syy032 (2018). Wilkinson, L. ggplot2: Elegant Graphics for Data Analysis by WICKHAM, H. Biometrics 67 , 678-679, doi:https://doi.org/10.1111/j.1541-0420.2011.01616.x (2011). Wickham, H. et al. Welcome to the Tidyverse. Journal of open source software 4 , 1686 (2019). Jordahl, K. et al. geopandas/geopandas: v0. 8.1 (Version v0. 8.1). Zenodo. Geneva, Switzerland (2020). Hunter, J. D. Matplotlib: A 2D Graphics Environment. Computing in Science & Engineering 9 , 90-95, doi:10.1109/MCSE.2007.55 (2007). Met Office. Cartopy: a cartographic python library with a matplotlib interface , (2010 - 2015). World Organisation for Animal Health (WOAH). Terrestrial Manual: Avian influenza (infection with avian influenza viruses) , 2019). Neumann, G. et al. Generation of influenza A viruses entirely from cloned cDNAs. Proceedings of the National Academy of Sciences 96 , 9345-9350, doi:doi:10.1073/pnas.96.16.9345 (1999). Peacock, T. P. et al. The furin cleavage site in the SARS-CoV-2 spike protein is required for transmission in ferrets. Nat Microbiol 6 , 899-909, doi:10.1038/s41564-021-00908-w (2021). Narimatsu, Y. et al. An Atlas of Human Glycosylation Pathways Enables Display of the Human Glycome by Gene Engineered Cells. Molecular Cell 75 , 394-407.e395, doi:https://doi.org/10.1016/j.molcel.2019.05.017 (2019). Xiong, X. et al. Receptor binding by a ferret-transmissible H5 avian influenza virus. Nature 497 , 392-396, doi:10.1038/nature12144 (2013). World Health Organisation (WHO). Recommended composition of influenza virus vaccines for use in the 2025-2026 northern hemisphere influenza season. (2025). Okomo-Adhiambo, M. et al. Neuraminidase inhibitor susceptibility surveillance of influenza viruses circulating worldwide during the 2011 Southern Hemisphere season. Influenza Other Respir Viruses 7 , 645-658, doi:10.1111/irv.12113 (2013). World Health Organisation (WHO). Laboratory methodologies for testing the antiviral susceptibility of influenza viruses: Neuraminidase inhibitor (NAI) , (2025). Tables Tables are available in the Supplementary Files section. Additional Declarations There is NO Competing Interest. Supplementary Files Table1speciesandresults.xlsx Table 1. Summary of H5 HPAIV Detections in the Falkland Islands and South Georgia for individuals, December 2023 to February 2024. Table2HIDatav2.docx Table 2. Hemagglutination inhibition (HI) assay results for five virus isolates from this study tested with ferret antisera raised against from human pre-pandemic candidate vaccine strains of H5 clade 2.3.4.4b. SupplementaryFigure1PB2AncestralReconstructionCladogram.pdf Supplementary Figure 1 SupplementaryFigure2PAAncestralReconstructionCladogram.pdf Supplementary Figure 2 TableS1Avian.pdf Table S1. Summary of date, location and species of avian samples collected from South Georgia and the Falkland Islands with RRT-PCR testing result for H5N1 HPAIV, virus isolation and references for associated sequence data. TableS2Mammalian.pdf Table S2. Summary of date, location and species of mammalian samples collected from South Georgia and the Falkland Islands with RRT-PCR testing result for H5N1 HPAIV, virus isolation and references for associated sequence data. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8029950","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":540336300,"identity":"ab979db3-966a-45e5-bfb8-ac11c4cab582","order_by":0,"name":"Benjamin C. Mollett","email":"","orcid":"","institution":"Department of Virology, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK","correspondingAuthor":false,"prefix":"","firstName":"Benjamin","middleName":"C.","lastName":"Mollett","suffix":""},{"id":540336301,"identity":"35cda1bf-aa55-4bb7-ba53-e7c49785d1de","order_by":1,"name":"Joshua G. Lynton-Jenkins","email":"","orcid":"","institution":"Department of Virology, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK; WOAH/FAO International Reference Laboratory for Avian Influenza, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK","correspondingAuthor":false,"prefix":"","firstName":"Joshua","middleName":"G.","lastName":"Lynton-Jenkins","suffix":""},{"id":540336302,"identity":"fbbd98a4-e49b-4dee-ad00-7b992f9d6ef8","order_by":2,"name":"Samuel Richardson","email":"","orcid":"https://orcid.org/0000-0001-7149-3832","institution":"The Pirbright Institute, Pirbright, Woking, United Kingdom, GU24 0NF","correspondingAuthor":false,"prefix":"","firstName":"Samuel","middleName":"","lastName":"Richardson","suffix":""},{"id":540336303,"identity":"3c1d58d7-6f48-4d56-b544-d18420dbf8a8","order_by":3,"name":"Jessica L. Quantrill","email":"","orcid":"https://orcid.org/0000-0002-2439-1822","institution":"Department of Infectious Disease, Imperial College London, London, UK, SW7 2AZ","correspondingAuthor":false,"prefix":"","firstName":"Jessica","middleName":"L.","lastName":"Quantrill","suffix":""},{"id":540336304,"identity":"1ee7dbef-0e6b-4e28-9b27-95d35275acb7","order_by":4,"name":"Alexander M.P. Byrne","email":"","orcid":"","institution":"Worldwide Influenza Centre, The Francis Crick Institute, 1 Midland Road, London NW1 1AT, UK","correspondingAuthor":false,"prefix":"","firstName":"Alexander","middleName":"M.P.","lastName":"Byrne","suffix":""},{"id":540336305,"identity":"edd53a6b-cc1d-4c23-9099-ca4e5ea919aa","order_by":5,"name":"Ruth Harvey","email":"","orcid":"","institution":"Worldwide Influenza Centre, The Francis Crick Institute, 1 Midland Road, London NW1 1AT, UK","correspondingAuthor":false,"prefix":"","firstName":"Ruth","middleName":"","lastName":"Harvey","suffix":""},{"id":540336306,"identity":"24841bbe-801b-4a82-9901-ffa45e537375","order_by":6,"name":"Lorin Adams","email":"","orcid":"","institution":"Worldwide Influenza Centre, The Francis Crick Institute, 1 Midland Road, London NW1 1AT, UK","correspondingAuthor":false,"prefix":"","firstName":"Lorin","middleName":"","lastName":"Adams","suffix":""},{"id":540336307,"identity":"6f12c990-03f6-4c10-af44-6324e8b4929a","order_by":7,"name":"Alizé Proust","email":"","orcid":"","institution":"Worldwide Influenza Centre, The Francis Crick Institute, 1 Midland Road, London NW1 1AT, UK","correspondingAuthor":false,"prefix":"","firstName":"Alizé","middleName":"","lastName":"Proust","suffix":""},{"id":540336308,"identity":"d0fd547c-a298-426a-bde3-3a56b10da52a","order_by":8,"name":"Maryn D. Brown","email":"","orcid":"https://orcid.org/0000-0001-6257-0672","institution":"Department of Infectious Disease, Imperial College London, London, UK, SW7 2AZ","correspondingAuthor":false,"prefix":"","firstName":"Maryn","middleName":"D.","lastName":"Brown","suffix":""},{"id":540336309,"identity":"234c80e2-451f-4db7-b88c-fbcf8164fcce","order_by":9,"name":"Jiayun Yang","email":"","orcid":"","institution":"The Pirbright Institute, Pirbright, Woking, United Kingdom, GU24 0NF","correspondingAuthor":false,"prefix":"","firstName":"Jiayun","middleName":"","lastName":"Yang","suffix":""},{"id":540336310,"identity":"625be8e3-5910-4187-bb90-64671e8eefc2","order_by":10,"name":"Scott M. Reid","email":"","orcid":"","institution":"Department of Virology, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK","correspondingAuthor":false,"prefix":"","firstName":"Scott","middleName":"M.","lastName":"Reid","suffix":""},{"id":540336311,"identity":"95215a86-cb48-470d-add5-52395732b5dd","order_by":11,"name":"Holly A. Coombes","email":"","orcid":"","institution":"Department of Virology, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK","correspondingAuthor":false,"prefix":"","firstName":"Holly","middleName":"A.","lastName":"Coombes","suffix":""},{"id":540336312,"identity":"ada964fc-a4a9-4f51-a63a-a0d2674d29e4","order_by":12,"name":"Ben Clifton","email":"","orcid":"","institution":"Department of Virology, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK","correspondingAuthor":false,"prefix":"","firstName":"Ben","middleName":"","lastName":"Clifton","suffix":""},{"id":540336313,"identity":"f91d8083-3852-45f9-adf6-4f3913d1af78","order_by":13,"name":"Ashley Bennison","email":"","orcid":"","institution":"British Antarctic Survey, Madingley Road, Cambridge CB3 0ET, UK","correspondingAuthor":false,"prefix":"","firstName":"Ashley","middleName":"","lastName":"Bennison","suffix":""},{"id":540336314,"identity":"23351b4f-977b-487d-b69f-743da944606d","order_by":14,"name":"Augustin Clessin","email":"","orcid":"","institution":"Centre d’Ecologie Fonctionnelle et Evolutive (CEFE), CNRS, Université Montpellier, EPHE, IRD, Montpellier, France","correspondingAuthor":false,"prefix":"","firstName":"Augustin","middleName":"","lastName":"Clessin","suffix":""},{"id":540336315,"identity":"4547c46e-ff09-4236-8c7e-24cbcb6b4569","order_by":15,"name":"Amandine Gamble","email":"","orcid":"","institution":"Department of Public \u0026 Ecosystem Health, K. Lisa Yang Center for Wildlife Health Cornell University, USA","correspondingAuthor":false,"prefix":"","firstName":"Amandine","middleName":"","lastName":"Gamble","suffix":""},{"id":540336316,"identity":"e6199c60-dbb9-408e-b2e9-28bb9d846255","order_by":16,"name":"Elaine M. Fitzcharles","email":"","orcid":"","institution":"British Antarctic Survey, Madingley Road, Cambridge CB3 0ET, UK","correspondingAuthor":false,"prefix":"","firstName":"Elaine","middleName":"M.","lastName":"Fitzcharles","suffix":""},{"id":540336317,"identity":"f6e89584-a46b-4163-a965-0bde50c3c227","order_by":17,"name":"Zoe Fowler","email":"","orcid":"","institution":"Department of Agriculture, Bypass Road, Stanley, Falkland Islands, FIQQ 1ZZ","correspondingAuthor":false,"prefix":"","firstName":"Zoe","middleName":"","lastName":"Fowler","suffix":""},{"id":540336318,"identity":"2f40f71b-0b49-4ffe-9bf7-d1c9ce41e96f","order_by":18,"name":"Tereza Masonou","email":"","orcid":"","institution":"Great Ormond Street UCL Institute of Child Health, London, UK, WC1N 1EH","correspondingAuthor":false,"prefix":"","firstName":"Tereza","middleName":"","lastName":"Masonou","suffix":""},{"id":540336319,"identity":"80df0159-758c-46d7-9de7-2c571f3162ce","order_by":19,"name":"Katie-Marie Case","email":"","orcid":"","institution":"Great Ormond Street UCL Institute of Child Health, London, UK, WC1N 1EH","correspondingAuthor":false,"prefix":"","firstName":"Katie-Marie","middleName":"","lastName":"Case","suffix":""},{"id":540336320,"identity":"ceacf535-f5b7-4236-8cc5-05960f67067d","order_by":20,"name":"Jila Aejian","email":"","orcid":"","institution":"Great Ormond Street UCL Institute of Child Health, London, UK, WC1N 1EH","correspondingAuthor":false,"prefix":"","firstName":"Jila","middleName":"","lastName":"Aejian","suffix":""},{"id":540336321,"identity":"7237f99c-d885-49b9-b47f-d199723d222f","order_by":21,"name":"Maximillian N.J. Woodall","email":"","orcid":"https://orcid.org/0000-0003-4511-9171","institution":"Great Ormond Street UCL Institute of Child Health, London, UK, WC1N 1EH","correspondingAuthor":false,"prefix":"","firstName":"Maximillian","middleName":"N.J.","lastName":"Woodall","suffix":""},{"id":540336322,"identity":"3695716d-3209-4087-a6e7-fd720daea116","order_by":22,"name":"Claire M. Smith","email":"","orcid":"https://orcid.org/0000-0002-8913-0009","institution":"Great Ormond Street UCL Institute of Child Health, London, UK, WC1N 1EH","correspondingAuthor":false,"prefix":"","firstName":"Claire","middleName":"M.","lastName":"Smith","suffix":""},{"id":540336323,"identity":"b93ff746-6022-451b-9c55-028bff672d7f","order_by":23,"name":"Munir Iqbal","email":"","orcid":"https://orcid.org/0000-0001-5165-5339","institution":"The Pirbright Institute, Pirbright, Woking, United Kingdom, GU24 0NF","correspondingAuthor":false,"prefix":"","firstName":"Munir","middleName":"","lastName":"Iqbal","suffix":""},{"id":540336324,"identity":"c55e9f1c-1b6e-4ae5-8352-7072a0a2e936","order_by":24,"name":"Ian H. Brown","email":"","orcid":"https://orcid.org/0000-0003-3909-1301","institution":"The Pirbright Institute, Pirbright, Woking, United Kingdom, GU24 0NF","correspondingAuthor":false,"prefix":"","firstName":"Ian","middleName":"H.","lastName":"Brown","suffix":""},{"id":540336325,"identity":"b03e6b3a-f0d1-45e1-a0cb-69cbaa089bd8","order_by":25,"name":"Nicola S. Lewis","email":"","orcid":"","institution":"Worldwide Influenza Centre, The Francis Crick Institute, 1 Midland Road, London NW1 1AT, UK","correspondingAuthor":false,"prefix":"","firstName":"Nicola","middleName":"S.","lastName":"Lewis","suffix":""},{"id":540336326,"identity":"0515b975-2c93-48f7-8d03-f6326513a881","order_by":26,"name":"Wendy S. Barclay","email":"","orcid":"https://orcid.org/0000-0002-6413-2454","institution":"Department of Infectious Disease, Imperial College London, London, UK, SW7 2AZ","correspondingAuthor":false,"prefix":"","firstName":"Wendy","middleName":"S.","lastName":"Barclay","suffix":""},{"id":540336327,"identity":"cd45a8c4-c34a-4744-9537-2b3752080172","order_by":27,"name":"Thomas P. Peacock","email":"","orcid":"https://orcid.org/0000-0001-7077-2928","institution":"The Pirbright Institute, Pirbright, Woking, United Kingdom, GU24 0NF; Department of Infectious Disease, Imperial College London, London, UK, SW7 2AZ","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"P.","lastName":"Peacock","suffix":""},{"id":540336328,"identity":"bb21f835-ed33-46e1-a6a3-77f11c0f6d72","order_by":28,"name":"Joe James","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYBACxgYgIQFisTcwMKOIEtbCc4BILQggkUCkFub25gMMFjXb5ORnPn72uKCGwa5fIoFNcgY+h/UcS2CQOHbb2OB2mrnxjGMMyTNnALVswKdlRo4BgwTb7cQN0glm0jxsDMkGZw6wST7AqyX/A4PEv9v182ce/ybN848oLTkMDJJttxMYbvCYSfO2MdgZHG8g4LCeYwYHJPtuG244k1NuzNsnkSDZ3thsic/7hu3NDx9LfLstL99+fNtjnm829vzMzAdv9uDT0sDAcFgCwmYDYonEBkIRKQ9y3AeEFgZ7vMpHwSgYBaNgRAIAAsRMAuf+oP8AAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-9673-3684","institution":"Department of Virology, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK; WOAH/FAO International Reference Laboratory for Avian Influenza, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK; Department of Infectious Disease, Imperial College London, London, UK, SW7 2AZ","correspondingAuthor":true,"prefix":"","firstName":"Joe","middleName":"","lastName":"James","suffix":""},{"id":540336299,"identity":"24036544-ec0f-4b78-b166-56bc2f5432ca","order_by":29,"name":"Ashley C. Banyard","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-1286-9825","institution":"Department of Virology, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK; WOAH/FAO International Reference Laboratory for Avian Influenza, Animal and Plant Health Agency (APHA-Weybridge), Woodham Lane, Addlestone, Surrey KT15 3NB, UK; School of Life Sciences, University of Sussex, Falmer, Brighton, BN1 9QG","correspondingAuthor":true,"prefix":"","firstName":"Ashley","middleName":"C.","lastName":"Banyard","suffix":""}],"badges":[],"createdAt":"2025-11-04 14:26:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8029950/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8029950/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95615024,"identity":"b3a080c7-3eb0-4b0d-a802-e55ab551e3f9","added_by":"auto","created_at":"2025-11-11 08:42:14","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":4344099,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGeographic distribution of H5 HPAIV detections in the Falklands and South Georgia. \u0026nbsp;\u003c/strong\u003eDetection of H5 HPAIV in marine mammals (squares) and avian species (circles) on the Falkland Islands (\u003cstrong\u003eA\u003c/strong\u003e, \u003cstrong\u003eC\u003c/strong\u003e, \u003cstrong\u003eE\u003c/strong\u003e) and South Georgia Island (\u003cstrong\u003eB\u003c/strong\u003e, \u003cstrong\u003eD\u003c/strong\u003e, \u003cstrong\u003eF\u003c/strong\u003e) collected in December 2023 (\u003cstrong\u003eA\u003c/strong\u003e \u0026amp; \u003cstrong\u003eB\u003c/strong\u003e), January 2024 (\u003cstrong\u003eC\u003c/strong\u003e\u0026amp; \u003cstrong\u003eD\u003c/strong\u003e), and February 2024 (\u003cstrong\u003eE\u003c/strong\u003e \u0026amp; \u003cstrong\u003eF\u003c/strong\u003e). Maps indicate the locations of positive detections based on RT-PCR testing.\u003c/p\u003e","description":"","filename":"Figure1mapofpositivesamples.png","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/492fd1ebbaf85a6715f874b4.png"},{"id":95615022,"identity":"6f7df36c-edf1-4326-819d-20435eeb92e1","added_by":"auto","created_at":"2025-11-11 08:42:14","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":5501713,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMaximum Clade credibility Phylogeny with PB2 mutations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMaximum clade credibility phylogeny of analysed viral genomes coloured by discrete trait representing inferred geographical transitions. PB2 mutations at key positions are displayed as a matrix aligned with the phylogeny. Highlighted clades denote distinct introduction events, numbered in order of their inferred time of introduction. Strains shown in bold within introduction groups 2 and 4 were selected for molecular risk assessment.\u003c/p\u003e","description":"","filename":"Figure2HAMCCwPB2Mutations.png","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/633324ea896f6f9f667c9602.png"},{"id":95657342,"identity":"37d8f9bb-0425-4a56-ad4c-6529463de910","added_by":"auto","created_at":"2025-11-11 16:20:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":598443,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGeographical map of inferred viral transitions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGeographical map of the southern hemisphere displaying the inferred viral transitions between countries. Transitions were inferred using Markov jumps with the transition lines being coloured by country of origin. Line width displays the strength of the mean rate, this is also displayed directly on the transition lines.\u003c/p\u003e","description":"","filename":"Figure3Mapofviraltransitions.png","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/3c3330657b68d98e79458546.png"},{"id":95615025,"identity":"358ab972-a282-4626-8d15-913f571b58d8","added_by":"auto","created_at":"2025-11-11 08:42:14","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":690014,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMolecular basis of mammalian adaptation of the mammal-associated B3.2 and risk assessment. (A)\u003c/strong\u003e Minireplicon assay of a reconstituted polymerase of the B3.2 H5N1 virus A/southern_fulmar/Falkland_Islands/133789/2023(H5N1) which naturally contains the putative mammalian adaptations PB2 Q591K, D701N and PA M86I. These putative adaptations were reverted back to the avian-like amino acid consensus and tested for polymerase activity in human HEK293T cells. Matched western blot showing equivalent PB2 and PA expression between the mutants. Data plotted as the mean + S. D. of N = 3 independent repeats and normalised to the avian-like ancestral virus. Statistics performed by one-way ANOVA with multiple comparisons comparing all conditions on log-transformed data. Only significant differences between avian precursor and other conditions (shown as asterisks above the bar chart) or between isogenic mutants (shown as asterisks above lines) plotted onto the graph. Log- normality confirmed by Shapiro Wilk test and QQ Plot. \u003cstrong\u003e(B)\u003c/strong\u003eVirus replication kinetics in primary human nasal epithelial cells maintained at air liquid interface from N = 3 independent donors. The two mammalian-adapted B3.2 isolates A/southern_fulmar/Falkland_Islands/133789/2023(H5N1) (mammalian-adapted B3.2 #1) and A/Fur Seal/Salisbury_Plain/004762/2024 (mammalian-adapted B3.2 #2) were compared against the two non-mammalian adapted avian isolates, A/Brown_Skua/Hound_Bay/133947/2023 (avian B3.2 #1) and A/South_Georgia_Shag/King_Edward_Cove/141245/2023 (avian B3.2 #2). A/dairy cattle/Texas/24-008749-001-original/2024 (cattle/Texas) was used as a comparator H5N1 with previous described mammalian-adaption. Cells were infected with an MOI of 0.01 and released infectious virus was titrated by plaque assay on MDCK. Data plotted as mean ± S.D. of N = 3 technical repeats. Statistics performed by-two-way ANOVA on log-transformed data with multiple comparisons (between different viruses at the same time point). Only significant differences between avian and mammalian-adapted B3.2 viruses annotated onto the graph. Log- normality confirmed by Shapiro Wilk test and QQ Plot. \u003cstrong\u003e(C)\u003c/strong\u003ePseudovirus receptor preference assay performed in HEK 293 cells with sialyl transferase activity knocked out by CRISPR and reconstituted by transiently transfecting the α2,3 sialyltransferase, ST3GAL4 (ST3) or the α2,6 sialyltransferase, ST6GAL1 (ST6). A pseudovirus expressing a human H1 HA (pandemic 2009), or avian H1N1 were used as controls for human and avian-adapted viruses, an H7N9 was used as a control for a dual receptor binder, and a panel of H5N1 or H5N1 mutants from the USA (B3.13), UK (BB), were used as comparators. BB L226Q was used as a control for a human-adapted H5N1. Entry of pseudoviruses was measured by firefly luciferase activity. Data plotted as mean + S. D. from N = 3 independent repeats. Statistics performed by two-way ANOVA on log-transformed data with multiple comparisons against the empty vector control for each virus. Log- normality confirmed by Shapiro Wilk test and QQ Plot. \u003cstrong\u003e(D)\u003c/strong\u003e Bio-layer interferometry of A/southern_fulmar/Falkland_Islands/133789/2023(H5N1), generated by reverse genetics by removing the polybasic cleavage site and rescuing the virus with the 6 internal genes from the attenuated laboratory strain A/Puerto Rico/8/1934(PR8). Virus was purified and run on a bio-layer interferometer against two receptor analogues, the avian-like α2,3-linked analogue 3′-sialyl-N-acetyllactosamine (3SLN) and the human-like α2,6-linked receptor analogue 6′-sialyl-N-acetyllactosamine (6SLN).\u003c/p\u003e","description":"","filename":"Figure4Functionalassessment.png","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/35a1fb5fd84806c529d371a3.png"},{"id":95615027,"identity":"e1246659-8217-4032-90a7-3b855e2b4213","added_by":"auto","created_at":"2025-11-11 08:42:14","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":296464,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure5BZanamivirimageonline.comerged.png","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/970e6ba68536eb79a5b52314.png"},{"id":100547527,"identity":"731a1c06-7d4b-40e7-81ba-0da633de5afe","added_by":"auto","created_at":"2026-01-19 08:15:57","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":10506213,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/d5ac745b-5214-40c6-8c2d-a4521648ce1c.pdf"},{"id":95615021,"identity":"7a31b818-c6e3-4176-ab27-e84c70ecb8a8","added_by":"auto","created_at":"2025-11-11 08:42:13","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":27878,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable 1. \u003c/strong\u003eSummary of H5 HPAIV Detections in the Falkland Islands and South Georgia for individuals, December 2023 to February 2024.\u003c/p\u003e","description":"","filename":"Table1speciesandresults.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/d4a401936ef089a72aa68d49.xlsx"},{"id":95615023,"identity":"6ec00beb-d248-417d-85bd-4816a6398e62","added_by":"auto","created_at":"2025-11-11 08:42:14","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":34025,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable 2. \u003c/strong\u003eHemagglutination inhibition (HI) assay results for five virus isolates from this study tested with ferret antisera raised against from human pre-pandemic candidate vaccine strains of H5 clade 2.3.4.4b.\u003c/p\u003e","description":"","filename":"Table2HIDatav2.docx","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/32f150d8bd32515142b66c72.docx"},{"id":95615028,"identity":"976f940a-fb9f-43bd-8126-c38b17f4b99f","added_by":"auto","created_at":"2025-11-11 08:42:14","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":391946,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Figure 1\u003c/p\u003e","description":"","filename":"SupplementaryFigure1PB2AncestralReconstructionCladogram.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/14a807de4154496087fbcc95.pdf"},{"id":95615030,"identity":"d40f2210-b626-4965-a06b-926b8fea894d","added_by":"auto","created_at":"2025-11-11 08:42:14","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":422851,"visible":true,"origin":"","legend":"Supplementary Figure 2","description":"","filename":"SupplementaryFigure2PAAncestralReconstructionCladogram.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/97ea1e1b59830f9532cc4333.pdf"},{"id":95615031,"identity":"133d04fa-15fd-4e7b-a45d-fbc43e0987d5","added_by":"auto","created_at":"2025-11-11 08:42:14","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":159855,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable S1. \u003c/strong\u003eSummary of date, location and species of avian samples collected from South Georgia and the Falkland Islands with RRT-PCR testing result for H5N1 HPAIV, virus isolation and references for associated sequence data.\u003c/p\u003e","description":"","filename":"TableS1Avian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/8bbadf5f16d633a70fd9c1ca.pdf"},{"id":95615029,"identity":"7e4f777f-9915-4322-b19e-c4415218d712","added_by":"auto","created_at":"2025-11-11 08:42:14","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":147461,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable S2. \u003c/strong\u003eSummary of date, location and species of mammalian samples collected from South Georgia and the Falkland Islands with RRT-PCR testing result for H5N1 HPAIV, virus isolation and references for associated sequence data.\u003c/p\u003e","description":"","filename":"TableS2Mammalian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8029950/v1/481513a0a5ae2e6f8c177d84.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Disease ecology and zoonotic risk of clade 2.3.4.4b H5N1 high pathogenicity avian influenza in the sub-Antarctic region","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe emergence of clade 2.3.4.4b H5N1 high pathogenicity avian influenza virus (HPAIV) on the sub-Antarctic islands\u003csup\u003e1-3\u003c/sup\u003e demonstrated the truly global dissemination of these viruses by wild species and the fragility of remotely located ecosystems to incursions of emerging viral pathogens. This panzootic began with the emergence of a novel H5N1 genotype belonging to clade 2.3.4.4b in Europe or central Asia in 2020\u003csup\u003e4\u003c/sup\u003e. In late 2021, this virus spread into North America through trans-Atlantic pathways via migratory birds\u003csup\u003e5,6\u003c/sup\u003e such as gull species. Since its arrival in North America, clade 2.3.4.4b H5N1 has undergone significant diversification through reassortment with low pathogenicity avian influenza viruses (LPAIV)\u003csup\u003e7\u003c/sup\u003e. By late 2022, a North American H5N1 genotype (B3.2) was detected in South America for the first time\u003csup\u003e8,9\u003c/sup\u003e, signalling the initial threat of this virus to the wider region\u003csup\u003e10,11\u003c/sup\u003e. The rapid expansion of the virus across South America significantly raised the risk of incursion into the sub-Antarctic and Antarctic regions\u003csup\u003e1,2,12\u003c/sup\u003e. The arrival of HPAIV in South American mammals, signalling increased zoonotic risk, was marked by an outbreak of H5N1 that caused extensive mortality in marine mammal population (mainly pinniped species), initially in South American sea lions (\u003cem\u003eOtaria flavescens\u003c/em\u003e) reported in Chile in June 2023\u003csup\u003e13\u003c/sup\u003e, and by August 2023 HPAIV had been reported in mass mortalities in marine mammal colonies off the southern coast of Argentina\u003csup\u003e14\u003c/sup\u003e. The virus reached Uruguay\u003csup\u003e11\u003c/sup\u003e and southern Brazil\u003csup\u003e15\u003c/sup\u003e before the observation of a mass mortality event in southern elephant seals (\u003cem\u003eMirounga leonina\u003c/em\u003e), affecting over 17,000 animals in Argentina\u003csup\u003e16\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile detections in wild birds and marine mammal across South America were caused by the same B3.2 genotype, interestingly most viral sequences from South American marine mammals clustered more closely with one another than with most avian origin sequences, suggesting mammal-to-mammal spread within the continent\u003csup\u003e16\u003c/sup\u003e. Indeed, this \u0026lsquo;marine-mammal clade\u0026rsquo; contains adaptations not seen in most avian-origin genotype B3.2 H5N1 HPAIVs including in the polymerase basic 2 (PB2) protein (D701N and/or Q591K)\u003csup\u003e13\u003c/sup\u003e, potentially indicating an increased zoonotic threat posed by these viruses. Following the spread of HPAIV though South America, two human infections with B3.2 H5N1 HPAIVs were confirmed in Chile and Ecuador in 2023\u003csup\u003e17,18\u003c/sup\u003e. The viral sequences from the Chilean human case also bore the PB2, Q591K and D701N mutations, and the individual had reported walking on a beach where there were dead sea lions present\u003csup\u003e19\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn October 2023, clade 2.3.4.4b H5N1 HPAIV genotype B3.2 was detected for the first time in the sub-Antarctic, marking the first ever confirmed incursion of HPAIV in this region\u003csup\u003e1\u003c/sup\u003e. This initial detection in a Southern Fulmar (\u003cem\u003eFulmarus glacialoides\u003c/em\u003e) on the Falkland Islands (FIs) was followed shortly by a detection of positive cases in wild birds across South Georgia (SG)\u003csup\u003e1,2\u003c/sup\u003e. Critically, alongside infection of avian species, mammalian species were also significantly hit by infection with large mortality events being observed in both southern elephant seals (\u003cem\u003eMirounga leonina\u003c/em\u003e) and Antarctic fur seals (\u003cem\u003eArctocephalus gazella\u003c/em\u003e)\u003csup\u003e1,2\u003c/sup\u003e. H5N1 HPAIV was subsequently detected in main-land Antarctica and on sub-Antarctic islands across the South Atlantic and South Indian Oceans in 2024\u003csup\u003e20\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eHere we report on the disease events, genetic characterisation and likely introduction routes into these islands and assess the zoonotic and pandemic potential of these emerging H5N1 HPAIVs.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eWidespread detection of H5N1 HPAIV in wild avian and mammalian species across sub-Antarctic and Antarctic islands during the 2023/2024 austral summer.\u003c/p\u003e\n\u003cp\u003eFollowing the earlier genetic characterisation of H5N1 HPAIV in both SG and the FIs\u003csup\u003e1,2\u003c/sup\u003e, sequencing efforts were undertaken on samples submitted over the 2023/24 austral summer (Supplementary Tables 1 and 2). HPAIV infection was confirmed in multiple mammalian and avian species, including southern elephant seals and Antarctic fur seals, as well as several bird species, including brown skua (\u003cem\u003eStercorarius antarcticus\u003c/em\u003e), snowy albatross (\u003cem\u003eDiomedea exulans\u003c/em\u003e), gentoo penguin (\u003cem\u003ePygoscelis papua\u003c/em\u003e), king penguin (\u003cem\u003eAptenodytes patagonicus\u003c/em\u003e), southern rockhopper penguin (\u003cem\u003eEudyptes chrysocome\u003c/em\u003e), black-browed albatross (\u003cem\u003eThalassarche melanophris\u003c/em\u003e), and a variable hawk (\u003cem\u003eGeranoaetus polyosoma\u003c/em\u003e) (Table 1).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn total, between the 7 December 2023 and the 11 February 2024, 53% (n=27/51) of dead or moribund birds sampled on South Georgia (SG) tested positive for H5 HPAIV RNA. All H5-positive samples were also positive for N1 viral RNA (vRNA), except for one king penguin from Will Point, on the north coast of South Georgia, which was weakly positive by the M-gene and H5-HP RT-PCR only. H5 HPAIV RNA was detected in four of six bird species tested (wandering albatross, gentoo penguin, brown skua, and king penguin) across 13 locations and 11 timepoints between 16thDecember 2023 and 6th February 2024 (Table 1, Figure 1, Table S1).\u003c/p\u003e\n\u003cp\u003eOn the Falkland Islands (FIs), 69% (n=38/55) of dead or moribund birds sampled tested positive for H5 HPAIV vRNA. Again, all positive samples were also positive for N1 vRNA, with one exception (a brown skua from Saunders Island) which was weakly positive only for the M-gene and H5-HP vRNA. H5 HPAIV was detected in five of six bird species tested on the FIs (black-browed albatross, brown skua, gentoo penguin, southern rockhopper penguin, and variable hawk). These detections were made across seven locations and five timepoints between 19 January and 21 February 2024 (Table 1, Figure 1, Table S1).\u003c/p\u003e\n\u003cp\u003eAcross all avian samples, the highest percentage positivity for H5 HPAIV detections occurred in snowy albatrosses (100%; n=9/9), followed by black-browed albatrosses, with 92% (n=23/25); southern rockhopper penguins (86%; n=5/6), brown skuas (69%; n=6/13), king penguins (5/11; 46%); and gentoo penguins (45% (n=10/22); n=5/11 on SG and n=5/11 on FI). A single positive detection was recorded for a variable hawk (1/1; 100%). No HPAIV vRNA was detected in giant petrels (\u003cem\u003eMacronectes spp\u003c/em\u003e.) (0/6), snowy sheathbill (\u003cem\u003eChionis albus\u003c/em\u003e) (0/1), rock shag (\u003cem\u003eLeucocarbo magellanicus\u003c/em\u003e) (0/1) or austral thrush (\u003cem\u003eTurdus falcklandii\u003c/em\u003e)\u0026nbsp;(0/1) (Table 1, Figure 1, Table S1). Among the positive samples, the strongest detection of vRNA was observed in brown skuas, particularly in brain and oropharyngeal samples (Table S1).\u003c/p\u003e\n\u003cp\u003eAmong mammalian species, H5N1 HPAIV vRNA was detected in 73% (n=51/72) dead or moribund individuals sampled on SG between 17 January and 11 February 2024, spanning 16 locations and 12 timepoints (Table 1, Figure 1, Table S2). The highest detection rate was observed in Antarctic fur seals, with 80% (n=40/50) individuals testing positive, the highest number of detections across all species tested. Southern elephant seals also tested positive, though at a lower frequency (55%; n=11/22).\u003c/p\u003e\n\u003cp\u003eMultiple independent introductions of genotype B3.2 H5N1 HPAIV occurred into the Antarctic region from South America in the 2023/2024 austral summer.\u003c/p\u003e\n\u003cp\u003eIn total, whole genome sequences (WGSs) were generated for samples from 57 birds. This included 40 genomes from SG, comprising 12 from avian carcasses, (5 wandering albatrosses, 3 brown skuas, 3 gentoo penguins, and a single king penguin) and 28 from mammalian carcasses (22 fur seals and 6 elephant seals). An additional 17 avian-origin genomes were obtained from samples collected from the FIs (13 from black-browed albatrosses, 2 from gentoo penguins, 1 from a southern rockhopper penguin, and 1 from a variable hawk) (Table S1 and S2).\u003c/p\u003e\n\u003cp\u003eSequences were analysed alongside available H5N1 clade 2.3.4.4b full-genome sequences using the GenoFLU-multi tool\u003csup\u003e21\u003c/sup\u003e to assess genetic ancestry and segment-level reassortment patterns. These sequences were classified within the B3.2 genotype\u003csup\u003e21\u003c/sup\u003e, reflecting the continued southward expansion of this lineage from North America into South America during early 2022\u003csup\u003e1\u003c/sup\u003e. Novel sequences were analysed alongside representative H5N1 clade 2.3.4.4b HA sequences from North and South America in a time-resolved phylogenetic analysis (Figure 2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe maximum clade credibility (MCC) phylogeny of HA demonstrated five viral introductions into the FIs and SG, with each introduction numbered in chronological order (Figure 2). Each introduction corresponds to a distinct node containing groupings of FI and SG sequences. Groups 1, 3, 4 and 5 cluster with South American wild bird and mammalian viruses, with South American sequences (Argentina for groups 1, 3 and 4; Brazil for group 5) forming the basal node to those from SG and the FIs. This suggests that the virus arrived and where samples and sequences are available, appeared to diversify locally. Moreover, these introductions specifically cluster with marine mammal-origin viruses from Argentina, and Brazil, as well as mammalian and seabird viruses from Uruguay, indicating possible spillover events from South American marine mammals and seabirds to sub-Antarctic wildlife or at least species that range across both regions. Groups 1, 3 and 5 were comprised of only FI sequences each forming independent clusters consistent with separate introductions into the islands. Group 4, in contrast, consists of sequences from both SG and FI, with the majority being with SG with some FI interspersed. Group 5 detections include sequences from samples from the FIs alone with the basal node suggesting introduction from Brazil.\u003c/p\u003e\n\u003cp\u003eThe analysis revealed five possible viral introductions into SG and the FI. Temporal inference using time to most common ancestor (TMRCA) estimates revealed that the earliest introductions occurred into SG in August and September 2023 (Groups 2 and 4). The FIs experienced repeated introductions over a broader timeframe, with TMRCAs ranging from early July to November 2023 (Groups 1, 3 and 5). Although overlapping TMRCAs are observed, their distinct phylogenetic clustering supports their classification as independent introductions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo investigate these introductions, a discrete trait analysis with country-level sequence origin as the geographic state was used. Transition rates between locations were estimated from Markov jumps, and the mean rate with corresponding 95% highest posterior density (HPD) interval was calculated. This analysis indicated that introductions into FIs are primarily associated with Argentina with moderate mean transition rates with other South American sources (Figure 3, Table S3).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe evolutionary emergence of mammalian-adaptive substitutions was examined using PB2 and PA gene sequences from H5N1 HPAIV Genotype B3.2. Amino acid residues at positions 591 and 701 previously linked to enhanced replication in mammalian cells\u003csup\u003e22,23\u003c/sup\u003e, were assessed. Among 57 viral genomes (29 avian (50.9%) / 28 Mammal (49.1%)) analysed, 94.7% (n=54/57) contained PB2 D701N and 98.2% (n=56/57) carried PB2 Q591K, with 94.7% (n=54/57) harbouring both mutations demonstrating strong co-occurrence of these adaptive markers in viruses detected in SG and FI. Most sequences clustered within the South American marine-mammal clade, which is partially defined by these substitutions\u003csup\u003e13\u003c/sup\u003e. Two viruses from FIs, one from a variable hawk (\u003cem\u003eGeranoaetus polyosoma\u003c/em\u003e, A/Variable_Hawk/Falkland_Islands/007272/2024) and one from a Black-browed albatross (\u003cem\u003eThalassarche melanophris\u003c/em\u003e, A/Black\u0026minus;browed_albatross/Falkland_Islands/004745/2024) retained the avian-like PB2 701D residue while carrying the PB2 Q591K. Conversely, a brown skua (\u003cem\u003eStercorarius antarcticus\u003c/em\u003e, A/Brown_Skua/Prion_Island/004878/2024) collected in SG retained both avian residues, PB2 701D and PB2 591Q and clustered clustered with the earlier South Georgia clade, derived from south American poultry (Figure 2). In the PA segment, most sequences fell within the same marine-mammal clade characterised by the M86I substitution, with the same brown skua representing the only exception (Figure 2, Figures S1\u0026ndash;S2)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGenotype B3.2 H5N1 HPAIVs possessing mammalian adaptive mutations have enhanced replicative fitness in human cells.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRepresentative viral isolates were selected for functional assessment of mutations and compare the role of key residues within the polymerase complex via mini-replicon systems and live virus assays. Initially a mini-replicon system was used to assess mutations detected in the isolate\u0026nbsp;A/southern_fulmar/Falkland Islands/133789/2023 (H5N1) (A Group 4 virus (Figure 2) denoted as mammalian-adapted B3.2 #1), namely the PB2 Q591K and D701N substitutions as well as the PA M86I mutation, previously associated with enhanced replication in mammalian hosts. A further\u0026nbsp;non-mammalian adapted avian isolate, A/Brown_Skua/Hound_Bay/133947/2023 (A Group 2 virus (Figure 2) denoted from here as avian B3.2 #1), that retained the avian-like residues at these positions (PB2 591Q, PB2 701D and PA 86M), served as a comparative control. The polymerase assay assessment in human cells demonstrated that removing all three putative mammalian adaptations resulted in a polymerase that was unable to efficiently replicate in human cells (Figure 4A). Reverting the sequence back to PB2 Q591K or D701N resulted in a significant enhancement in polymerase activity and adding both together resulted in a further significant boost, showing these mutations have an additive effect. PA M86I had a far more modest impact, when combined with PB2 591K or 701N, suggesting it is less likely to contribute to the mammalian adaptation of this clade and in isolation did not affect polymerase activity (Figure 4A).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFurther, virus replication kinetics were assessed in primary human nasal epithelial cells (hNECs) from three independent human donors maintained as an air liquid interface (ALI). The panel used for the mini-replicon assay was expanded to include a further mammalian-adapted virus, A/Fur Seal/Salisbury_Plain/004762/2024 (A Group 4 virus (Figure 2) denoted as mammalian-adapted B3.2 #2) and a non-adapted avian isolate, A/ South_Georgia_Shag/King_Edward_Cove/141245/2023 (A Group 2 virus (Figure 2) denoted avian B3.2 #2). A further relevant isolate, A/dairy cattle/Texas/24-008749-001-original/2024 (denoted B3.13 Cattle/Texas) was used as a comparator that contains\u0026nbsp;polymerase mutations associated with zoonotic risk and has demonstrated robust replication in human respiratory cells\u003csup\u003e24\u003c/sup\u003e. This panel enabled evaluation of which mammalian adaptive mutations influence viral replication competence in a human-relevant \u003cem\u003ein vitro\u003c/em\u003e system\u003csup\u003e24\u003c/sup\u003e.\u0026nbsp;Mammalian-adapted B3.2 isolates\u0026nbsp;replicated to significantly higher titres than the avian isolates in hNECs across the different donors (Figure 4B). Notably, both B3.2 mammalian-adapted viruses achieved higher replication levels than the B3.13 Cattle/Texas H5N1 virus which possesses several established mammalian-adaptive polymerase mutations\u003csup\u003e24\u003c/sup\u003e (Figure 4B).\u003c/p\u003e\n\u003cp\u003eThe ability of representative viruses to bind different receptor moieties was also assessed. A pseudovirus-based virus entry assay was developed, using well described controls and demonstrated that the avian origin viruses, both genotype B3.13 and B3.2, preferentially bound \u0026alpha;2,3-linked sialylated receptors (Figure 4C). In contrast, a human seasonal H1N1 derived from the 2009 pandemic (A/Rhode Island/04/2026; H1N1pdm09), and an H5N1 BB genotype HA modified to contain a well described receptor switching mutation L226Q\u003csup\u003e25\u003c/sup\u003e, had a strong preference for \u0026alpha;2,6-linked sialylated receptors abundant in the human airway. The H7N9 strain bound both\u0026nbsp;\u0026alpha;2,3-linked and \u0026alpha;2,6-linked sialylated receptors\u0026nbsp;(Figure 4C).\u003c/p\u003e\n\u003cp\u003eFinally, to confirm these results, reverse genetics approaches were developed\u0026nbsp;for the mammalian-adapted B3.2 #1 isolate\u0026nbsp;to generate a virus containing the H5 HA gene with the polybasic cleavage site removed and its six internal genes replaced with those of the attenuated laboratory strain PR8. This virus was rescued, propagated, purified and tested for its receptor binding using the biophysical assay, bio-layer interferometry, using two receptor binding analogues, the avian-like \u0026alpha;2,3-linked and the human-like \u0026alpha;2,6-linked sialic acid receptors. Consistent with the pseudovirus results, and in line analyses of \u0026nbsp;European and North American clade 2.3.4.4b viruses\u003csup\u003e26\u003c/sup\u003e, the\u0026nbsp;mammalian-adapted B3.2 #1\u0026nbsp;virus did not show any detectable binding to the human-like \u0026alpha;2,6-linked sialic acid receptors, and showed strong binding to the avian-like \u0026alpha;2,3-linked sialic acid receptor (Figure 4D).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAntigenic assessment confirms good recognition by antisera raised to within-clade candidate vaccine viruses\u003c/p\u003e\n\u003cp\u003eTo further characterise the zoonotic risk posed by these viruses, a selection of five B3.2 H5N1 isolates; two mammalian (A/fur_seal/Miles_Bay/004783/2024, and A/fur_seal/Salisbury_Plain/004762/2024) and three avian (A/brown_skua/Hound_Bay/133949/2023, A/brown_skua/Hound_Bay/133947/2023 and A/kelp_gull/Moltke_Harbour/133754/2023) all from SG, were screened against ferret antisera raised towards clade 2.3.4.4b H5 pre-pandemic candidate vaccine viruses (CVVs) using the haemagglutination inhibition (HI) assay (Table 2). Antisera raised towards the A/Astrakhan/3212/2020 (H5N8) clade 2.3.4.4b CVV showed at least an 8-fold reduction in HI titre towards the B3.2 isolates from SG relative to the homologous virus. However, both of the H5N1 clade 2.3.4.4b CVVs (A/American Wigeon/South Carolina/22-000345-001/2021 and A/chicken/Ghana/AVL-76321VIR7050-39/2021) exhibited good recognition of the SG isolates with at most a 2-fold drop in HI titre relative to the homologous viruses.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSub-Antarctic B3.2 H5N1 HPAIV remain susceptible to antiviral therapeutics\u003c/p\u003e\n\u003cp\u003eFurther to assessing recognition by currently recommended H5 clade 2.3.4.4b CVVs, we also sought to phenotypically assess the susceptibility of the SG viruses to antiviral therapeutics. The five viruses from SG used to assess CVV recognition were also tested for susceptibility towards the neuraminidase inhibitors oseltamivir and zanamivir, alongside two human seasonal influenza H1N1 viruses, which are resistant (A/Alabama/03/2020) and susceptible (A/Illinois/45/2019) to oseltamivir. The five B3.2 H5N1 viruses had half maximal inhibition concentrations (IC\u003csub\u003e50\u003c/sub\u003e) ranging between 1.37 nM and 4.70 nM for oseltamivir (Figure 5A), and between 0.40 nM and 0.51 nM for zanamivir (Figure 5B) demonstrating that they remain susceptible to these antivirals. \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eNovel data generated here has demonstrated potential pathways for virus incursion from south America into the sub-Antarctic and Antarctic region, impacts of mammalian-associated mutations in PB2 and the current status of viruses in the sub-Antarctic region from an antigenic perspective. Following widespread circulation of H5N1 HPAIV across South America in late 2022\u003csup\u003e8,10,11\u003c/sup\u003e, the virus was first detected in SG in late 2023\u003csup\u003e1\u003c/sup\u003e. Surveillance during 2023/24 austral summer confirmed extensive detection in wild birds across SG and FIs, all belonging to the B3.2 genotype, commonly circulating on the South American Continent\u003csup\u003e27\u003c/sup\u003e\u003csup\u003e1\u003c/sup\u003e\u003csup\u003e8,10,11\u003c/sup\u003e. Phylogenetic analysis indicates up to five separate incursions into SG and FIs from continental South America before southward spread to the Antarctic Peninsula. This suggests that SG may act as a stepping stone for further dissemination to remote sub-Antarctic regions\u003csup\u003e31\u003c/sup\u003e, exemplified by the detection in March 2025 of genetically closely related H5N1 in three brown skuas on Gough Island\u003csup\u003e32\u003c/sup\u003e, suggesting transoceanic spread. These findings highlight the potential for seabird-mediated long-range transmission and raise concern that further spread to continental regions such as South Africa or Australia could enable wider dissemination via established flyways \u003csup\u003e31\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMarine mammals may represent an additional route of virus introduction and maintenance in sub-Antarctic ecosystems. Several incursions involved viruses carrying mammalian adaptive markers suggesting a secondary infection of avian species with mammalian-derived strains likely through scavenging of infected carcasses\u003csup\u003e31\u003c/sup\u003e. In SG, non-colonial bird species are likely exposed through scavenging, while dense marine mammal colonies provide an environment conducive to mammal-to-mammal-transmission. Cross-species interactions, including scavenging activities may facilitate bi-directional exchange between avian and mammalian hosts although the limited availability of samples from these remote ecosystems constrain our understanding of transmission pathways.\u003c/p\u003e\n\u003cp\u003eZoonotic potential is a key factor for these viruses with multiple factors being associated with increased risk\u003csup\u003e33\u003c/sup\u003e including: (\u003cstrong\u003ei\u003c/strong\u003e) polymerase activity in novel species (\u003cstrong\u003eii\u003c/strong\u003e) HA receptor binding and (\u003cstrong\u003eiii\u003c/strong\u003e) \u0026nbsp;HA stability. Avian-origin influenza viruses have restricted replication in mammalian cells due to poor usage of pro-viral host factors by the virus polymerase\u003csup\u003e33\u003c/sup\u003e although this can be overcome through mutations in PB2 such as E627K, Q591K or D701N\u003csup\u003e34\u003c/sup\u003e. Additionally avian influenza viruses generally have a strong binding preference towards sialylated glycans that contain an\u0026nbsp;\u0026alpha;2,3-linkage between the terminal sialic acid and the penultimate galactose\u003csup\u003e33\u003c/sup\u003e contrasting with human-transmissible viruses that preferentially bind \u0026alpha;2,6-linked sialic acids. Several mutations have been described that allow a switch in receptor preference from \u0026alpha;2,3- to \u0026alpha;2,6-linked sialic acids in the H5 subtype, including Q226L\u003csup\u003e25,35,36\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eHere, viruses from SG and FIs often contained the PB2 Q591K, D701N and PA M86I mutations. Interestingly, 74 human infections have been detected in the Americas between 2022 and February 2025\u003csup\u003e37\u003c/sup\u003e, mostly linked to genotype B3.13 circulating in US dairy cattle\u003csup\u003e37\u003c/sup\u003e. Two genotype B3.2 human infections have been reported, in Ecuador (January 2023)\u003csup\u003e18\u003c/sup\u003e and in Chile\u003csup\u003e17\u003c/sup\u003e (March 2023) with the latter involving PB2 Q591K and D701N \u003csup\u003e19\u003c/sup\u003e. The Chilean virus also caused fatal disease in ferrets, with animal-to-animal transmission in a direct contact setting, but not via respiratory droplet or fomites\u003csup\u003e38\u003c/sup\u003e. We demonstrated a marked difference in replication kinetics in primary human airway cultures between the avian and marine-mammal-clade viruses detected in SG. Furthermore, sequential reversion of putative adaptive mutations (PB2 591K and 701N, and PA 86I) to their avian equivalent (591Q, 701D, and PA 86M) reduced polymerase activity in human cells, with Q591K and D701N having the greatest effect. This suggests that mammalian adaptations in the A/Southern fulmar/Falkland Islands/133789/2023 isolate enhanced the ability of this virus to replicate in human cells, likely due to an accumulation of polymerase mutations following sustained mammal-to-mammal transmission. Phylogenetic analyses also supported the potential for mammal-to-mammal transmission as clades displayed continued circulation in marine mammal populations in SG. There were also detections from avian hosts interspersed with these clades demonstrating avian infection with these viruses. However, the following assessment of mutations in HA, the changes required to enable a shift in receptor binding properties has not occurred. The impact of such changes on avian replication dynamics in different avian species remain unclear. Both genotypes B3.13 and B3.2 have retained preferential binding to\u0026nbsp;\u0026alpha;2,3 sialic acid\u0026nbsp;linked receptors. This is highly significant in risk assessment of emerging viruses from a zoonotic perspective\u003csup\u003e33\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eFinally, whilst there was a reduction in antigenic-reactivity against the Astrakhan CVV, good reactivity was detected against two other CVVs namely, A/American Wigeon/South Carolina/22-000345-001/2021 and A/chicken/Ghana/AVL-76321VIR7050-39/2021 suggesting strong antigenic match. Whilst antigenic evolution requires constant assessment wherever viruses are emerging, these results indicate that the HAs have not yet evolved to escape neutralisation by key CVVs. The data generated here also indicates that these viruses remain susceptible to commonly licensed neuraminidase inhibitors. Sensitivity to neuraminidase inhibitors reveals that current preparedness methods are effective, although continual assessment is required.\u003c/p\u003e\n\u003cp\u003eIn conclusion, the data generated has demonstrated the need to maximise genetic characterisation of viruses wherever outbreaks occur. With the continued detection of H5N1 globally it is no longer sufficient to simply subtype through analysis of HA and NA sequences. Genotypes and markers of adaptation must be characterised and where risk increases, sequences made accessible to the global community. Our data has demonstrated some impact on replication of mammalian mutations, and further studies detailing the impact of these mutations on in-vivo infection need to be prioritised to support the observations here.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003eEthics and biosafety\u003c/h2\u003e\n\u003cp\u003eHPAIV work is categorised as specified animal pathogens order (SAPO) 4 and Advisory Committee on Dangerous Pathogens (ACDP) hazard group 3 by United Kingdom (UK) regulations. Work with these viruses was undertaken in a licensed containment level (CL) 3 (ACDP-3/SAPO4) facility of The Animal and Plant Health Agency (APHA), The Pirbright Institute under GMRA (BAG-RA-226) and The Francis Crick Institute Isolation of virus in specific pathogen-free (SPF) embryonated fowls\u0026rsquo; eggs (EFEs) was undertaken under Home Office license PP9307748. All virus risk assessments were approved by the appropriate internal committees, as well as the UK Health and Safety Executive (HSE) and, where necessary, the UK scientific advisory committee for genetic modification (SACGM).\u003c/p\u003e\n\u003cp\u003eEthics for the use of the primary human airway epithelial cells were as described previously\u003csup\u003e39\u003c/sup\u003e. Briefly, donors provided written consent and Ethics approval was given through the Living Airway Biobank, administered through the UCL Great Ormond Street Institute of Child Health (REC reference: 19/NW/0171, IRAS project ID: 261511, Northwest Liverpool East Research Ethics Committee). Nasal brushings were obtained by trained clinicians from adult (30\u0026ndash;50 years) donors who reported no respiratory symptoms in the preceding 7-weeks. Brushings were taken from the inferior nasal concha zone using cytological brushes (Scientific Laboratory Supplies, CYT1050). All methods were performed following the relevant guidelines and regulations.\u003c/p\u003e\n\u003ch2\u003eSamples and collection\u003c/h2\u003e\n\u003cp\u003eWe previously reported the detection of H5N1 HPAIV in samples collected from birds on SG between 8 October and 6 December 2023, and from mammals between 31 October and 9 December 2023, as well as from birds on the FI between 30 October and 10 December 2023\u003csup\u003e1\u003c/sup\u003e. Here, we present findings from a subsequent sampling period, encompassing birds and mammals on SG (7 December 2023 to 11 February 2024) and birds on the Falkland Islands (19 January to 21 February 2024) (Table S1). A total of 211 samples were collected from 106 individual dead birds: 116 samples from 51 individuals on SG, and 95 samples from 55 individuals on the FI. On SG, bird samples were obtained from 25 locations across 22 discrete timepoints between 7 December 2023 and 11 February 2024. On the FI, samples were collected from 14 locations over 9 timepoints between 19 January and 21 February 2024. Mammalian sampling was undertaken wherever mortalities were detected. A total of 224 samples were collected from 67 individual animals, including 49 Antarctic fur seals and 18 Southern elephant seals, across 23 locations and 16 timepoints between 14 December 2023 and 11 February 2024 (Table S2). Avian samples included oropharyngeal (OP), cloacal (C), brain tissue, and environmental faecal swabs. Mammalian samples included oral, nasal, rectal, ocular, and brain tissue samples, as well as environmental faecal swabs. All samples were collected from animals found dead in the field following suspicion of infection, under appropriate permits granted by the governments of the FI and SG, in collaboration with the British Antarctic Survey (BAS). Sampling was conducted with strict biosafety protocols in place to ensure the safety of personnel and to minimise the risk of pathogen transmission via fomites to healthy wildlife. Field personnel recorded location, species, and estimated age of each individual sampled. \u003c/p\u003e\n\u003ch2\u003eMolecular assessment of samples\u003c/h2\u003e\n\u003cp\u003eSwabs were immersed in 1 mL Leibovitz\u0026apos;s L-15 Medium (L-15) (Gibco, USA). A section of each tissue (~1 gram) was transferred into L-15 medium and roughly homogenised. Supernatants from the processed swabs and tissues was used for RNA extraction as described previously\u003csup\u003e40\u003c/sup\u003e before testing extracted RNA for the presence or absence of viral nucleic acid by real-time reverse transcription polymerase chain reaction (RT-PCR) assays specific for the Matrix (M) gene\u003csup\u003e41,42\u003c/sup\u003e, the haemagglutinin (HA) gene\u003csup\u003e40\u003c/sup\u003e and the N1 neuraminidase (NA) gene\u003csup\u003e43\u003c/sup\u003e. The cut off for positivity for each of the three assays was \u0026le;36.0 Cq. All amplifications were undertaken using an AriaMx qPCR System (Agilent, United Kingdom). Samples yielding Cq values \u0026lt;30 were used for attempted virus isolation.\u003c/p\u003e\n\u003ch2\u003eGenetic evaluation of H5N1 HPAIV sequences\u003c/h2\u003e\n\u003cp\u003eTo generate sequence data for positive PCR samples, extracted viral RNA (vRNA) was converted to double-stranded cDNA and amplified using a one-step RT-PCR using SuperScript III One-Step RT-PCR kit (Thermo Fisher Scientific)\u003csup\u003e1\u003c/sup\u003e. Previously described approaches were used to generate genetic data\u003csup\u003e44,45\u003c/sup\u003e. With PCR products being processed using Agencourt AMPure XP beads (Beckman Coulter) and using the Native Barcoding Kit (Oxford Nanopore Technologies) GridION Mk1 (Oxford Nanopore Technologies) technologies as per manufacturer\u0026apos;s instructions. All influenza sequences generated and used in this study are available through the GISAID EpiFlu Database (https://www.gisaid.org) and NCBI. The genotype of the viruses sequenced in this study was confirmed using the Genoflu-multi tool (https://github.com/moncla-lab/GenoFLU-multi).\u003c/p\u003e\n\u003cp\u003eFor the time-resolved phylogenetic analysis, all available hemagglutinin (HA) sequences (GISAID Accessed 28 May 2025) from South America and representative sequences from North America determined by PARNAS\u003csup\u003e46\u003c/sup\u003e were combined with those from SG and the FIs to infer phylogeny using BEAST version 1.10.5\u003csup\u003e47\u003c/sup\u003e, incorporating the BEAGLE library\u003csup\u003e48\u003c/sup\u003e. The analysis employed the HKYnucleotide substitution model\u003csup\u003e49\u003c/sup\u003e. A strict molecular clock and a coalescent constant size tree prior. Two independent Markov Chain Monte Carlo (MCMC) runs of 200 million steps each were conducted, with sampling every 20,000 steps ensuring all relevant effective sample sizes reached at least 200 as assessed in Tracer 1.7.2\u003csup\u003e50\u003c/sup\u003e. The initial 10% of each chain was discarded as burn-in, and the chains were combined using the LogCombiner tool in the BEAST package. \u003c/p\u003e\n\u003cp\u003eMarkov jump estimates of viral transition between discrete locations were obtained from the posterior distribution of the trees inferred by BEAST. For each transition the mean transition rate and 95% highest posterior density (HPD) interval were summarised. Post processing was performed using R (v4.5.0) using the ggplot2\u003csup\u003e51\u003c/sup\u003eand tidyverse packages\u003csup\u003e52\u003c/sup\u003e. Geographic visualisation of dispersal pathways was displayed using a custom python script utilising geopandas\u003csup\u003e53\u003c/sup\u003e, matplotlib\u003csup\u003e54\u003c/sup\u003e and cartopy\u003csup\u003e55\u003c/sup\u003e. Sequences were genotyped according to the USDA schema, using the GenoFLU tool (https://github.com/USDA-VS/GenoFLU)\u003csup\u003e21\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eAncestral amino acid sequences were reconstructed with the ancestral module of TreeTime (v0.11.4) using the Tones-Taylor-Thornton (JTT) amino acid substitution model, based maximum likelihood phylogenies inferred for PB2 and PA segments. Phylogenetic trees were generated using maximum likelihood methods with 1,000 ultrafast bootstrap replicates and 1,000 SH-aLRT replicates to assess branch support. The input dataset included a subset of H5 subtype sequences representing over 97% of the known genetic diversity from North and South America, determined using PARNAS\u003csup\u003e46\u003c/sup\u003e. \u003c/p\u003e\n\u003ch2\u003eVirus isolation \u003c/h2\u003e\n\u003cp\u003eVirus isolation was undertaken using 200 \u0026micro;l of supernatant from clinical samples and assessment by undertaking two passages in 9-day-old specified pathogen free (SPF) embryonated fowl eggs (EFE) as previously described\u003csup\u003e56\u003c/sup\u003e. Successful isolation was confirmed by hemagglutination assay on harvested allantoic fluids using chicken red blood cells as previously described. Titres of \u0026gt;1/4 were considered positive for isolation. Isolates were further propagated in EFEs and titrated by plaque assays on MDCK cells as previously described\u003csup\u003e26\u003c/sup\u003e. \u003c/p\u003e\n\u003ch2\u003eCell culture\u003c/h2\u003e\n\u003cp\u003eHuman embryonic kidney 293T (HEK-293T) cells and the Madin-Darby canine kidney (MDCK) cells were acquired from Central Service Unites (CSU) of The Pirbright Institute (TPI). Both cell lines were cultured with Dulbecco\u0026rsquo;s Modified Eagle\u0026rsquo;s medium (DMEM), supplemented with 10% foetal calf serum (FCS), at 37\u0026deg;C with 5% CO\u003csub\u003e2\u003c/sub\u003e.\u003c/p\u003e\n\u003ch2\u003eVirus replication assessment\u003c/h2\u003e\n\u003cp\u003ePrimary human airway cultures were differentiated as previously described\u003csup\u003e24,39\u003c/sup\u003e. Human nasal brushings from adult donors (n=3) were expanded on 3T3-J2 fibroblasts and frozen down as passage 1 (p1). Basal cells were then revived as needed, expanded in co-culture with 3T3-J2 fibroblasts until passage 2 and differentiated at an air-liquid interface as previously described\u003csup\u003e39\u003c/sup\u003e. \u003c/p\u003e\n\u003cp\u003eInfection of the primary human airway cultures was undertaken as previously described\u003csup\u003e24\u003c/sup\u003e. Briefly, cells were washed with Dulbecco\u0026apos;s phosphate-buffered saline supplemented with Calcium and Magnesium (DPBS++) to remove mucus and cell debris. Cultures were infected with 200\u0026micro;L of virus diluted in DPBS++ at an MOI of 0.01 and incubated at 37\u003csup\u003eo\u003c/sup\u003eC for 1 hour. Virus-containing inoculum was aspirated, and wells were washed twice with DPBS++ to remove unbound virus. Time points were collected by adding 200\u0026micro;L of DPBS++ and incubated at 37\u003csup\u003eo\u003c/sup\u003eC for 10 minutes before removal and storage. Infectious virus was determined by plaque assay on MDCK cells. The time course was undertaken at 37\u003csup\u003eo\u003c/sup\u003eC, 5% CO\u003csub\u003e2 \u003c/sub\u003ewith samples being collected at 24, 48, 72 and 96 hours post-inoculation.\u003c/p\u003e\n\u003ch2\u003eReverse genetics \u003c/h2\u003e\n\u003cp\u003eThe HA with monobasic cleavage site and NA sequences of A/Southern fulmar/Falkland_Islands/133789/2023 (H5N1) were synthesised by GeneScript and cloned into the bidirectional pHW2000 vector. The virus rescue was undertaken as previously described using the internal genes from A/Puerto Rico/8/193 (H1N1) (PR8)\u003csup\u003e57\u003c/sup\u003e. The rescued viruses were propagated in 9- to 10-day old embryonated chicken eggs.\u003c/p\u003e\n\u003ch2\u003eInfluenza minireplicon assay\u003c/h2\u003e\n\u003cp\u003eExpression plasmids for the B3.2 minigenome assays were synthesised by Geneart, mutated using site directed mutagenesis and subcloned into pCAGGS. HEK 293T cells were transfected in 24-well plates using lipofectamine 3000 transfection reagent (Thermo Fisher) with the following amount of pCAGGS-containing expression plasmids: 40ng of PB2, 40ng of PB1, 20ng of PA, 80ng of NP, 40ng of Renilla luciferase and 80ng of human-specific pol I vRNA Firefly luciferase. Cells were lysed 24 hours post-transfection using passive lysis buffer (Promega) and polymerase activity was measured using the Dual-Luciferase Reporter assay system (Promega) and a FLUOstar Omega plate reader (BMG Labtech). Firefly luciferase signal was normalised to Renilla luciferase signal to give relative luminescence units (RLU).\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eImmunoblotting analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHEK 293T cells were transfected in 6-well plates using lipofectamine 3000 transfection reagents (Thermo Fisher) with pCAGGS expression plasmids in the same ratio as described for the influenza minireplicon assay: 160ng of PB2, 160ng of PB1, 80ng of PA, 320ng of NP, 160ng of Renilla luciferase and 320ng of human-specific pol I vRNA Firefly luciferase. Cells were lysed 24 hours post-transfection in radioimmunoprecipitation assay buffer (10 mM Tris-HCl pH 8.0, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% SDS, 140 mM NaCl) supplemented with a protease inhibitor tablet (ThermoFisher). Lysates were incubated on ice for 1 h then centrifuged (16,000 g, 10 min). The supernatant was mixed with Laemmli buffer (Bio-Rad) supplemented with 10% beta-mercaptoethanol (Sigma Aldrich). Protein samples were resolved on a 4-20% Mini-PROTEAN polyacrylamide gel (Bio-Rad) and transferred to a low-fluorescence PVDF membrane (Merck). Membranes were blocked in TBS-5% milk for 1 h at room temperature and then washed with TBS. Primary antibodies (mouse \u0026alpha;-tubulin, Abcam ab7291, 1:1250; rabbit \u0026alpha;-PB2, Genetex GTX125926, 1:500; rabbit \u0026alpha;-PA, Genetex GTX118991, 1:500) were diluted in TBS-T-2% milk and membranes were incubated overnight at 4 \u0026deg;C. Membranes were then washed 4x in TBS-T, incubated with secondary antibodies (goat \u0026alpha;-mouse Alexa Fluor\u0026reg; 680, Abcam ab175775, 1:10,000; goat \u0026alpha;-rabbit IRDye\u003csup\u003e\u0026reg;\u003c/sup\u003e 800CW, Li-Cor 926-32211, 1:10,000) in TBS-T-2% milk for 45 minutes, washed 4x in TBS-T and washed 1x in TBS. Membranes were imaged using an Odyssey DLx (Li-Cor Biosciences) and Image Studio Lite software.\u003c/p\u003e\n\n\u003ch2\u003ePseudovirus receptor binding preference assay\u003c/h2\u003e\n\u003cp\u003eThe VSV-G expression plasmid, and lentiviral packaging genes and luciferase genome were used as previously described\u003csup\u003e58\u003c/sup\u003e. The following human codon optimised HA, NA, protease and sialyl-transferase expression plasmids were synthesised by Genscript in pcDNA3.1: H1HA - A/Rhode Island/04/2016(H1N1pdm09; ANM90381.1), H1HA - A/duck/Bavaria/1/1977(H1N1; ALG00691.1), N1NA - A/England/195/2009 (H1N1pdm09; ACR15618.1), H7HA and N9NA - A/Shanghai/02/2013 (H7N9; YP_009118475.1, YP_009118481.1), H5HA \u0026ndash; A/dairy cattle/Texas/24-008749-001-original/2024 (H5N1; WPD27583.1), H5HA and N1NA - A/chicken/England/085598/2022 (H5N1; EPI2089022, EPI2089021), H5HA - A/southern fulmar/Falkland_Islands/133789/2023(H5N1; EPI2795786), human TMPRSS11D/HAT (EAX05559.1) human ST6 beta-galactosamide alpha-2,6-sialyltranferase 1(ST6; AAH31476.1) and CMP-N-acetylneuraminate-beta-galactosamide-alpha-2,3-sialyltransferase 4 (ST3; XP_047283377.1). Pseudovirus was produced in HEK 293T cells seeded in 6-well dishes. Cells were co-transfected using lipofectamine 3000 (Invitrogen) at ~70% confluency with 0.6\u0026micro;g of the luciferase reporter constructs (pCSFLW), 0.4\u0026micro;g of the HIV packaging plasmid pGAG-POL, and either 0.4\u0026micro;g of VSV-G (VSV-G only) or 0.4\u0026micro;g of the defined HA, 0.2\u0026micro;g of TMPRSS11D and 0.1\u0026micro;g of the defined NA. Pseudovirus containing supernatants were collected at 48 and 72 hours post-transfection, clarified by centrifugation, aliquoted and frozen at \u0026ndash;80\u003csup\u003eo\u003c/sup\u003eC. All H1N1 pseudoviruses were produced with the N1 NA from Eng/195, H7N9 HA was produced with the strain-matched NA, all H5N1 pseudoviruses were produced with the A/chicken/England/085598/2022 NA.\u003c/p\u003e\n\u003cp\u003eHEK293 cells lacking sialyl-transferase activity (HEK 293 \u0026Delta;ST3GAL3/4/6 \u0026Delta;ST61/2; aka 293delST cells) were a kind gift from Professors Henrik Clausen and Yoshiki Narimatsu from the University of Copenhagen\u003csup\u003e59\u003c/sup\u003e. These cells were seeded in 6-well plates and transiently transfected at 70% confluency using lipofectamine 3000 with 500ng of an empty pcDNA3.1 vector, or either ST3 or ST6 expression constructs. Twenty-four hours post-transfection, cells were washed, resuspended in fresh media and seeded into 96-well plates at a density of 1x10\u003csup\u003e4\u003c/sup\u003e cells per well. Pseudovirus diluted in media containing oseltamivir (to a final concentration of 0.5 \u0026micro;M) was then added to each well. 48 hours post-transduction cells were lysed and read using Brightglo reagent (Promega) and a Glomax discover plate reader (Promega). Data for each pseudovirus was normalised to the empty vector control.\u003c/p\u003e\n\u003ch2\u003eVirus receptor binding by bio-layer interferometry\u003c/h2\u003e\n\u003cp\u003ePelleted virus was purified through 30% and 60% sucrose gradients at 27,000rpm for 2h at 4℃. The purified virus was then diluted in HBS buffer containing 10\u0026mu;M oseltamivir carboxylate (Roche) and 10\u0026mu;M zanamivir (GSK). Binding affinity to avian-like sugar analogue 3SLN and human-like sugar analogue 6SLN was tested by Octet\u0026reg; R8 system (Sartorius) using streptavidin biosensors (Sartorius). Virus binding affinity was normalised to fractional saturation and the concentrations of sugar loadings\u003csup\u003e60\u003c/sup\u003e.\u003c/p\u003e\n\u003ch2\u003eAssessment of antigenicity\u003c/h2\u003e\n\u003cp\u003eFive virus isolates were tested in the hemagglutination inhibition (HI) assay using turkey erythrocytes and ferret antisera for antigenic characterisation as previously described\u003csup\u003e61\u003c/sup\u003e. Isolates were selected to be representative of Group 4 (2 isolates) and Group 3 (3 isolates) (Figure 1, Table 2). Ferret antisera had been previously raised against existing human pre-pandemic candidate vaccine strains of H5 clade 2.3.4.4b (Table 2). \u003c/p\u003e\n\u003ch2\u003eNeuraminidase inhibition assay\u003c/h2\u003e\n\u003cp\u003eOseltamivir carboxylate and zanamivir were supplied by Hoffmann-La Roche and GlaxoSmithKline, respectively. MES assay buffer was prepared with 32.5mM MES (2-Morpholinoethanesulfonic acid, Sigma-Aldrich) and 4mM CaCl\u003csub\u003e2\u003c/sub\u003e in ddH\u003csub\u003e2\u003c/sub\u003eO, adjusted to pH 6.5. The MUNANA substrate solution consisted of 2\u0026rsquo;-(4-methylumbelliferyl)-\u0026alpha;-D-N-acetylneuraminic acid (Sigma-Aldrich). The stop solution was prepared with 0.1 M glycine, 25% ethanol, 0.267% Super Q, and ddH\u003csub\u003e2\u003c/sub\u003eO. The susceptibility of virus isolates to oseltamivir and zanamivir was assessed using the MUNANA assay, which measures neuraminidase activity based on cleavage of the fluorogenic substrate and release of 4-methylumbelliferone (4-MU) as previously described\u003csup\u003e62\u003c/sup\u003e. For virus titration and inoculum normalization, test viruses were serially diluted in MES buffer into black, full-bottom 96-well microplates (Corning) and titrated in duplicate from 1:2 to 1:2048 to a final volume of 20\u0026micro;L. Subsequently, 30\u0026micro;L of 100\u0026micro;M MUNANA substrate was added to each well, and plates were incubated at 37\u0026deg;C with 5% CO₂ for 1h with shaking. Reactions were terminated with 150\u0026micro;L of stop solution, and fluorescence was measured using an Infinite 200 PRO plate reader (TECAN) with excitation at 360 nm and emission at 465 nm. Virus dilutions for the inhibition assay were determined using internationally defined methodologies\u003csup\u003e63\u003c/sup\u003e. For inhibition assays, oseltamivir carboxylate and zanamivir were serially diluted 4-fold in MES buffer to final concentrations of 4000\u0026ndash;0.015nM in black, full-bottom 96-well microplates to a final volume of 20\u0026micro;L and incubated at 37\u0026deg;C with 5% CO₂ for 30 min with shaking. MUNANA substrate was then added and incubated for 1h, after which the reaction was stopped and fluorescence measured as described above. \u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAcknowledgments\u003c/h2\u003e\n\u003cp\u003eThe authors would like to thank Professors Henrik Clausen and Yoshiki Narimatsu of the University of Copenhagen for kindly sharing the HEK293 \u0026Delta;ST3GAL3/4/6 \u0026Delta;ST61/2 cells.\u003c/p\u003e\n\u003ch2\u003eFunding statement\u003c/h2\u003e\n\u003cp\u003eThis research received no external funding. The testing and generation of the viral sequences was funded by the Department for Environment, Food and Rural Affairs (Defra, UK) and the Devolved Administrations of Scotland and Wales, through the following programmes: SV3400, SV3032, SV3006, SE2213 and SE2227. This work was also supported by the Biotechnology and Biological Sciences Research Council (BBSRC) and Department for Environment, Food and Rural Affairs (Defra, UK) research initiative \u0026lsquo;FluTrailMap\u0026rsquo; [grant numbers BB/Y007271/1,\u0026nbsp;BB/Y007298/1\u003c/p\u003e\n\u003cp\u003e]\u0026nbsp;and the Medical Research Council (MRC) and Defra research initiative \u0026lsquo;FluTrailMap-One Health\u0026rsquo; [grant number MR/Y03368X/1]. Funded by the European Union under grant agreement (101084171) - (Kappa-Flu). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or REA. Neither the European Union nor the granting authority can be held responsible for them. Funding as also provided from Innovate UK (grant number 10085195). Work at Pirbright was supported\u0026nbsp;by the BBSRC via the Pirbright Institute\u0026rsquo;s Strategic Programme Grants (ISPGs) [BBS/E/PI/230002A; BBS/E/PI/230002B], BBSRC National Bioscience Research Infrastructure: High Containment and Low Containment Services and Science Platforms grants [BBS/E/PI/23NB0004, BBS/E/PI/23NB0003].\u003c/p\u003e\n\u003ch2\u003eCompeting interests statement\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eAuthor contributions statement\u003c/h2\u003e\n\u003cp\u003eConceptualisation: ACB, JJ, TPP; formal analysis: BM, JGLJ, SR, JQ, AMPB, RH, MB, SMR, HC, BC, JJ, ACB; investigation: BM, JGLJ, SR, JQ, RH, BM; resources: ACB, JJ, IHB, NL, ZF, EMF; writing\u0026mdash;original draft, BM, JGLJ, TPP, JJ, ACB; writing\u0026mdash;review and editing: ACB, JJ, APMB, AB, ZF, EMF, SMR, MB, IHB, JGLJ, BM, TPP. All authors have read and agreed to the final version of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBanyard, A. C.\u003cem\u003e et al.\u003c/em\u003e Detection and spread of high pathogenicity avian influenza virus H5N1 in the Antarctic Region. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 7433, doi:10.1038/s41467-024-51490-8 (2024).\u003c/li\u003e\n\u003cli\u003eBennison, A.\u003cem\u003e et al.\u003c/em\u003e A case study of highly pathogenic avian influenza (HPAI) H5N1 at Bird Island, South Georgia: the first documented outbreak in the subantarctic region. \u003cem\u003eBird Study\u003c/em\u003e \u003cstrong\u003e71\u003c/strong\u003e, 380-391, doi:10.1080/00063657.2024.2396563 (2024).\u003c/li\u003e\n\u003cli\u003eLisovski, S.\u003cem\u003e et al.\u003c/em\u003e Unexpected Delayed Incursion of Highly Pathogenic Avian Influenza H5N1 (Clade 2.3.4.4b) Into the Antarctic Region. \u003cem\u003eInfluenza Other Respir Viruses\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, e70010, doi:10.1111/irv.70010 (2024).\u003c/li\u003e\n\u003cli\u003eXie, R.\u003cem\u003e et al.\u003c/em\u003e The episodic resurgence of highly pathogenic avian influenza H5 virus. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e622\u003c/strong\u003e, 810-817, doi:10.1038/s41586-023-06631-2 (2023).\u003c/li\u003e\n\u003cli\u003eCaliendo, V.\u003cem\u003e et al.\u003c/em\u003e Transatlantic spread of highly pathogenic avian influenza H5N1 by wild birds from Europe to North America in 2021. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 11729, doi:10.1038/s41598-022-13447-z (2022).\u003c/li\u003e\n\u003cli\u003eAlkie, T. N.\u003cem\u003e et al.\u003c/em\u003e Recurring Trans-Atlantic Incursion of Clade 2.3.4.4b H5N1 Viruses by Long Distance Migratory Birds from Northern Europe to Canada in 2022/2023. \u003cem\u003eViruses\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1836 (2023).\u003c/li\u003e\n\u003cli\u003eKandeil, A.\u003cem\u003e et al.\u003c/em\u003e Rapid evolution of A(H5N1) influenza viruses after intercontinental spread to North America. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 3082, doi:10.1038/s41467-023-38415-7 (2023).\u003c/li\u003e\n\u003cli\u003eWorld Organisation for Animal Health (WOAH). \u003cem\u003eWorld Animal Health Information System (WAHIS) Events Management. Search: High pathogenicity avian influenza viruses (poultry) and South American countries.\u003c/em\u003e, \u0026lt;https://wahis.woah.org/#/event-management\u0026gt; (2025).\u003c/li\u003e\n\u003cli\u003eRuiz-Saenz, J., Martinez-Gutierrez, M. \u0026amp; Pujol, F. H. Multiple introductions of highly pathogenic avian influenza H5N1 clade 2.3.4.4b into South America. \u003cem\u003eTravel Medicine and Infectious Disease\u003c/em\u003e \u003cstrong\u003e53\u003c/strong\u003e, 102591, doi:https://doi.org/10.1016/j.tmaid.2023.102591 (2023).\u003c/li\u003e\n\u003cli\u003eLeguia, M.\u003cem\u003e et al.\u003c/em\u003e Highly pathogenic avian influenza A (H5N1) in marine mammals and seabirds in Peru. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 5489, doi:10.1038/s41467-023-41182-0 (2023).\u003c/li\u003e\n\u003cli\u003eMarandino, A.\u003cem\u003e et al.\u003c/em\u003e Spreading of the High-Pathogenicity Avian Influenza (H5N1) Virus of Clade 2.3.4.4b into Uruguay. \u003cem\u003eViruses\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, doi:10.3390/v15091906 (2023).\u003c/li\u003e\n\u003cli\u003eKuiken, T.\u003cem\u003e et al.\u003c/em\u003e Emergence, spread, and impact of high-pathogenicity avian influenza H5 in wild birds and mammals of South America and Antarctica. \u003cem\u003eConservation Biology\u003c/em\u003e \u003cstrong\u003en/a\u003c/strong\u003e, e70052, doi:https://doi.org/10.1111/cobi.70052 (2025).\u003c/li\u003e\n\u003cli\u003ePardo-Roa, C.\u003cem\u003e et al.\u003c/em\u003e Cross-species and mammal-to-mammal transmission of clade 2.3.4.4b highly pathogenic avian influenza A/H5N1 with PB2 adaptations. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 2232, doi:10.1038/s41467-025-57338-z (2025).\u003c/li\u003e\n\u003cli\u003ePlaza, P. I., Gamarra-Toledo, V., Rodr\u0026iacute;guez Eugu\u0026iacute;, J., Rosciano, N. \u0026amp; Lambertucci, S. A. Pacific and Atlantic sea lion mortality caused by highly pathogenic Avian Influenza A(H5N1) in South America. \u003cem\u003eTravel Medicine and Infectious Disease\u003c/em\u003e \u003cstrong\u003e59\u003c/strong\u003e, 102712, doi:https://doi.org/10.1016/j.tmaid.2024.102712 (2024).\u003c/li\u003e\n\u003cli\u003ede Ara\u0026uacute;jo, A. C.\u003cem\u003e et al.\u003c/em\u003e Incursion of Highly Pathogenic Avian Influenza A(H5N1) Clade 2.3.4.4b Virus, Brazil, 2023. \u003cem\u003eEmerg Infect Dis\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 619-621, doi:10.3201/eid3003.231157 (2024).\u003c/li\u003e\n\u003cli\u003eUhart, M. M.\u003cem\u003e et al.\u003c/em\u003e Epidemiological data of an influenza A/H5N1 outbreak in elephant seals in Argentina indicates mammal-to-mammal transmission. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 9516, doi:10.1038/s41467-024-53766-5 (2024).\u003c/li\u003e\n\u003cli\u003eCastillo, A.\u003cem\u003e et al.\u003c/em\u003e The first case of human infection with H5N1 avian Influenza A virus in Chile. \u003cem\u003eJournal of Travel Medicine\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, taad083 (2023).\u003c/li\u003e\n\u003cli\u003eBruno, A.\u003cem\u003e et al.\u003c/em\u003e First case of human infection with highly pathogenic H5 avian Influenza A virus in South America: A new zoonotic pandemic threat for 2023? \u003cem\u003eJournal of Travel Medicine\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, doi:10.1093/jtm/taad032 (2023).\u003c/li\u003e\n\u003cli\u003ePardo-Roa, C.\u003cem\u003e et al.\u003c/em\u003e Cross-species transmission and PB2 mammalian adaptations of highly pathogenic avian influenza A/H5N1 viruses in Chile. \u003cem\u003ebioRxiv\u003c/em\u003e, doi:10.1101/2023.06.30.547205 (2023).\u003c/li\u003e\n\u003cli\u003eBennett-Laso, B.\u003cem\u003e et al.\u003c/em\u003e Confirmation of highly pathogenic avian influenza H5N1 in skuas, Antarctica 2024. \u003cem\u003eFrontiers in Veterinary Science\u003c/em\u003e \u003cstrong\u003eVolume 11 - 2024\u003c/strong\u003e, doi:10.3389/fvets.2024.1423404 (2024).\u003c/li\u003e\n\u003cli\u003eYouk, S.\u003cem\u003e et al.\u003c/em\u003e H5N1 highly pathogenic avian influenza clade 2.3.4.4b in wild and domestic birds: Introductions into the United States and reassortments, December 2021-April 2022. \u003cem\u003eVirology\u003c/em\u003e \u003cstrong\u003e587\u003c/strong\u003e, 109860, doi:10.1016/j.virol.2023.109860 (2023).\u003c/li\u003e\n\u003cli\u003eLi, Z.\u003cem\u003e et al.\u003c/em\u003e Molecular Basis of Replication of Duck H5N1 Influenza Viruses in a Mammalian Mouse Model. \u003cem\u003eJournal of Virology\u003c/em\u003e \u003cstrong\u003e79\u003c/strong\u003e, 12058-12064, doi:doi:10.1128/jvi.79.18.12058-12064.2005 (2005).\u003c/li\u003e\n\u003cli\u003eYamada, S.\u003cem\u003e et al.\u003c/em\u003e Biological and structural characterization of a host-adapting amino acid in influenza virus. \u003cem\u003ePLoS Pathog\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, e1001034, doi:10.1371/journal.ppat.1001034 (2010).\u003c/li\u003e\n\u003cli\u003eDholakia, V.\u003cem\u003e et al.\u003c/em\u003e Polymerase mutations underlie early adaptation of H5N1 influenza virus to dairy cattle and other mammals. \u003cem\u003ebioRxiv\u003c/em\u003e, 2025.2001.2006.631435, doi:10.1101/2025.01.06.631435 (2025).\u003c/li\u003e\n\u003cli\u003eLin, T.-H.\u003cem\u003e et al.\u003c/em\u003e A single mutation in bovine influenza H5N1 hemagglutinin switches specificity to human receptors. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e386\u003c/strong\u003e, 1128-1134, doi:doi:10.1126/science.adt0180 (2024).\u003c/li\u003e\n\u003cli\u003eYang, J.\u003cem\u003e et al.\u003c/em\u003e The haemagglutinin gene of bovine-origin H5N1 influenza viruses currently retains receptor-binding and pH-fusion characteristics of avian host phenotype. \u003cem\u003eEmerg Microbes Infect\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 2451052, doi:10.1080/22221751.2025.2451052 (2025).\u003c/li\u003e\n\u003cli\u003ePan American Health Organization (PAHO). \u003cem\u003eEpidemiological Update Avian Influenza A(H5N1) in the Americas Region\u003c/em\u003e, \u0026lt;https://www.paho.org/sites/default/files/2025-03/2025-mar-4-phe-epidupdate-avianinfluenza-eng-final.pdf\u0026gt; (2025).\u003c/li\u003e\n\u003cli\u003ePhillips, R. A., Croxall, J. P., Silk, J. R. D. \u0026amp; Briggs, D. R. Foraging ecology of albatrosses and petrels from South Georgia: two decades of insights from tracking technologies. \u003cem\u003eAquatic Conservation: Marine and Freshwater Ecosystems\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, S6-S21, doi:https://doi.org/10.1002/aqc.906 (2007).\u003c/li\u003e\n\u003cli\u003eMcCann, T. S. \u0026amp; Rothery, P. Population size and status of the southern elephant seal (Mirounga leonina) at South Georgia, 1951\u0026ndash;1985. \u003cem\u003ePolar Biology\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 305-309, doi:10.1007/BF00263179 (1988).\u003c/li\u003e\n\u003cli\u003eCroxall, J. P., McCann, T. S., Prince, P. A. \u0026amp; Rothery, P. in \u003cem\u003eAntarctic Ocean and Resources Variability.\u003c/em\u003e (ed Dietrich Sahrhage) 261-285 (Springer Berlin Heidelberg).\u003c/li\u003e\n\u003cli\u003eClessin, A.\u003cem\u003e et al.\u003c/em\u003e Mass mortality events in the sub-Antarctic Indian Ocean caused by long-distance circumpolar spread of highly pathogenic avian influenza H5N1 clade 2.3.4.4b. \u003cem\u003ebioRxiv\u003c/em\u003e, 2025.2002.2025.640068, doi:10.1101/2025.02.25.640068 (2025).\u003c/li\u003e\n\u003cli\u003eSteinfurth, A.\u003cem\u003e et al.\u003c/em\u003e Investigating high pathogenicity avian influenza virus incursions to remote islands: Detection of H5N1 on Gough Island in the South Atlantic Ocean. \u003cem\u003ebioRxiv\u003c/em\u003e, 2025.2009.2006.674618, doi:10.1101/2025.09.06.674618 (2025).\u003c/li\u003e\n\u003cli\u003eLong, J. S., Mistry, B., Haslam, S. M. \u0026amp; Barclay, W. S. Host and viral determinants of influenza A virus species specificity. \u003cem\u003eNature Reviews Microbiology\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 67-81, doi:10.1038/s41579-018-0115-z (2019).\u003c/li\u003e\n\u003cli\u003eCarrique, L.\u003cem\u003e et al.\u003c/em\u003e Host ANP32A mediates the assembly of the influenza virus replicase. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e587\u003c/strong\u003e, 638-643, doi:10.1038/s41586-020-2927-z (2020).\u003c/li\u003e\n\u003cli\u003eDadonaite, B.\u003cem\u003e et al.\u003c/em\u003e Deep mutational scanning of H5 hemagglutinin to inform influenza virus surveillance. \u003cem\u003ePLoS Biol\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, e3002916, doi:10.1371/journal.pbio.3002916 (2024).\u003c/li\u003e\n\u003cli\u003eR\u0026iacute;os Carrasco, M.\u003cem\u003e et al.\u003c/em\u003e The Q226L mutation can convert a highly pathogenic H5 2.3.4.4e virus to bind human-type receptors. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, e2419800122, doi:doi:10.1073/pnas.2419800122 (2025).\u003c/li\u003e\n\u003cli\u003eUSA Centers for Disease Control and Prevention (CDC). \u003cem\u003eH5 Bird Flu: Current Situation\u003c/em\u003e, \u0026lt;https://www.cdc.gov/bird-flu/situation-summary/index.html\u0026gt; (2025).\u003c/li\u003e\n\u003cli\u003ePulit-Penaloza, J. A.\u003cem\u003e et al.\u003c/em\u003e Highly pathogenic avian influenza A(H5N1) virus of clade 2.3.4.4b isolated from a human case in Chile causes fatal disease and transmits between co-housed ferrets. \u003cem\u003eEmerg Microbes Infect\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 2332667, doi:10.1080/22221751.2024.2332667 (2024).\u003c/li\u003e\n\u003cli\u003eWoodall, M. N. J.\u003cem\u003e et al.\u003c/em\u003e Age-specific nasal epithelial responses to SARS-CoV-2 infection. \u003cem\u003eNat Microbiol\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 1293-1311, doi:10.1038/s41564-024-01658-1 (2024).\u003c/li\u003e\n\u003cli\u003eJames, J.\u003cem\u003e et al.\u003c/em\u003e Rapid and sensitive detection of high pathogenicity Eurasian clade 2.3.4.4b avian influenza viruses in wild birds and poultry. \u003cem\u003eJournal of Virological Methods\u003c/em\u003e \u003cstrong\u003e301\u003c/strong\u003e, 114454, doi:https://doi.org/10.1016/j.jviromet.2022.114454 (2022).\u003c/li\u003e\n\u003cli\u003eNagy, A.\u003cem\u003e et al.\u003c/em\u003e A universal RT-qPCR assay for \u0026quot;One Health\u0026quot; detection of influenza A viruses. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, e0244669, doi:10.1371/journal.pone.0244669 (2021).\u003c/li\u003e\n\u003cli\u003eSpackman, E.\u003cem\u003e et al.\u003c/em\u003e Development of a real-time reverse transcriptase PCR assay for type A influenza virus and the avian H5 and H7 hemagglutinin subtypes. \u003cem\u003eJ Clin Microbiol\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 3256-3260, doi:10.1128/jcm.40.9.3256-3260.2002 (2002).\u003c/li\u003e\n\u003cli\u003ePayungporn, S.\u003cem\u003e et al.\u003c/em\u003e Single step multiplex real-time RT-PCR for H5N1 influenza A virus detection. \u003cem\u003eJournal of Virological Methods\u003c/em\u003e \u003cstrong\u003e131\u003c/strong\u003e, 143-147, doi:https://doi.org/10.1016/j.jviromet.2005.08.004 (2006).\u003c/li\u003e\n\u003cli\u003eChrzastek, K.\u003cem\u003e et al.\u003c/em\u003e Use of Sequence-Independent, Single-Primer-Amplification (SISPA) for rapid detection, identification, and characterization of avian RNA viruses. \u003cem\u003eVirology\u003c/em\u003e \u003cstrong\u003e509\u003c/strong\u003e, 159-166, doi:10.1016/j.virol.2017.06.019 (2017).\u003c/li\u003e\n\u003cli\u003eLeyson, C.\u003cem\u003e et al.\u003c/em\u003e Pathogenicity and genomic changes of a 2016 European H5N8 highly pathogenic avian influenza virus (clade 2.3.4.4) in experimentally infected mallards and chickens. \u003cem\u003eVirology\u003c/em\u003e \u003cstrong\u003e537\u003c/strong\u003e, 172-185, doi:10.1016/j.virol.2019.08.020 (2019).\u003c/li\u003e\n\u003cli\u003eMarkin, A.\u003cem\u003e et al.\u003c/em\u003e PARNAS: Objectively Selecting the Most Representative Taxa on a Phylogeny. \u003cem\u003eSyst Biol\u003c/em\u003e \u003cstrong\u003e72\u003c/strong\u003e, 1052-1063, doi:10.1093/sysbio/syad028 (2023).\u003c/li\u003e\n\u003cli\u003eBaele, G.\u003cem\u003e et al.\u003c/em\u003e BEAST X for Bayesian phylogenetic, phylogeographic and phylodynamic inference. \u003cem\u003eNature Methods\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 1653-1656, doi:10.1038/s41592-025-02751-x (2025).\u003c/li\u003e\n\u003cli\u003eAyres, D. L.\u003cem\u003e et al.\u003c/em\u003e BEAGLE: An Application Programming Interface and High-Performance Computing Library for Statistical Phylogenetics. \u003cem\u003eSystematic Biology\u003c/em\u003e \u003cstrong\u003e61\u003c/strong\u003e, 170-173, doi:10.1093/sysbio/syr100 (2011).\u003c/li\u003e\n\u003cli\u003eHasegawa, M., Kishino, H. \u0026amp; Yano, T. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. \u003cem\u003eJ Mol Evol\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 160-174, doi:10.1007/bf02101694 (1985).\u003c/li\u003e\n\u003cli\u003eRambaut, A., Drummond, A. J., Xie, D., Baele, G. \u0026amp; Suchard, M. A. Posterior Summarization in Bayesian Phylogenetics Using Tracer 1.7. \u003cem\u003eSyst Biol\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 901-904, doi:10.1093/sysbio/syy032 (2018).\u003c/li\u003e\n\u003cli\u003eWilkinson, L. ggplot2: Elegant Graphics for Data Analysis by WICKHAM, H. \u003cem\u003eBiometrics\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 678-679, doi:https://doi.org/10.1111/j.1541-0420.2011.01616.x (2011).\u003c/li\u003e\n\u003cli\u003eWickham, H.\u003cem\u003e et al.\u003c/em\u003e Welcome to the Tidyverse. \u003cem\u003eJournal of open source software\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1686 (2019).\u003c/li\u003e\n\u003cli\u003eJordahl, K.\u003cem\u003e et al.\u003c/em\u003e geopandas/geopandas: v0. 8.1 (Version v0. 8.1). Zenodo. \u003cem\u003eGeneva, Switzerland\u003c/em\u003e (2020).\u003c/li\u003e\n\u003cli\u003eHunter, J. D. Matplotlib: A 2D Graphics Environment. \u003cem\u003eComputing in Science \u0026amp; Engineering\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 90-95, doi:10.1109/MCSE.2007.55 (2007).\u003c/li\u003e\n\u003cli\u003eMet Office. \u003cem\u003eCartopy: a cartographic python library with a matplotlib interface\u003c/em\u003e, \u0026lt;https://scitools.org.uk/cartopy\u0026gt; (2010 - 2015).\u003c/li\u003e\n\u003cli\u003eWorld Organisation for Animal Health (WOAH). \u003cem\u003eTerrestrial Manual: Avian influenza (infection with avian influenza viruses)\u003c/em\u003e, 2019).\u003c/li\u003e\n\u003cli\u003eNeumann, G.\u003cem\u003e et al.\u003c/em\u003e Generation of influenza A viruses entirely from cloned cDNAs. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e96\u003c/strong\u003e, 9345-9350, doi:doi:10.1073/pnas.96.16.9345 (1999).\u003c/li\u003e\n\u003cli\u003ePeacock, T. P.\u003cem\u003e et al.\u003c/em\u003e The furin cleavage site in the SARS-CoV-2 spike protein is required for transmission in ferrets. \u003cem\u003eNat Microbiol\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 899-909, doi:10.1038/s41564-021-00908-w (2021).\u003c/li\u003e\n\u003cli\u003eNarimatsu, Y.\u003cem\u003e et al.\u003c/em\u003e An Atlas of Human Glycosylation Pathways Enables Display of the Human Glycome by Gene Engineered Cells. \u003cem\u003eMolecular Cell\u003c/em\u003e \u003cstrong\u003e75\u003c/strong\u003e, 394-407.e395, doi:https://doi.org/10.1016/j.molcel.2019.05.017 (2019).\u003c/li\u003e\n\u003cli\u003eXiong, X.\u003cem\u003e et al.\u003c/em\u003e Receptor binding by a ferret-transmissible H5 avian influenza virus. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e497\u003c/strong\u003e, 392-396, doi:10.1038/nature12144 (2013).\u003c/li\u003e\n\u003cli\u003eWorld Health Organisation (WHO). Recommended composition of influenza virus vaccines for use in the 2025-2026 northern hemisphere influenza season. (2025).\u003c/li\u003e\n\u003cli\u003eOkomo-Adhiambo, M.\u003cem\u003e et al.\u003c/em\u003e Neuraminidase inhibitor susceptibility surveillance of influenza viruses circulating worldwide during the 2011 Southern Hemisphere season. \u003cem\u003eInfluenza Other Respir Viruses\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 645-658, doi:10.1111/irv.12113 (2013).\u003c/li\u003e\n\u003cli\u003eWorld Health Organisation (WHO). \u003cem\u003eLaboratory methodologies for testing the antiviral susceptibility of influenza viruses: Neuraminidase inhibitor (NAI)\u003c/em\u003e, \u0026lt;https://www.who.int/teams/global-influenza-programme/laboratory-network/quality-assurance/antiviral-susceptibility-influenza/neuraminidase-inhibitor\u0026gt; (2025).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8029950/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8029950/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"High pathogenicity avian influenza virus (HPAIV) has significantly impacted upon avian and mammalian populations across the Antarctic region. All viruses detected have been genotype B3.2 with phylogenetic analyses indicating multiple independent incursions from continental South America to, and transmission between, sub-Antarctic islands. From a zoonotic perspective, several isolates contained markers of mammalian adaptation in PB2 with functional characterisation of mutants demonstrating efficient replication in primary human airway epithelial cell cultures, demonstrating that these PB2-mutations alone contributed to enhanced polymerase activity in human cell lines. No mammalian-adaptive mutations were detected in the haemagglutinin or neuraminidase genes, with viruses retaining avian receptor binding preferences. Antigenic characterisation demonstrated cross-reactivity with existing pre-pandemic candidate vaccine strains and all viruses remained susceptible to licensed frontline antiviral therapeutics. We demonstrate a complex evolving viral ecology in the sub-Antarctic region involving both avian and marine mammal hosts, with significant implications for regional wildlife populations and zoonotic risk.","manuscriptTitle":"Disease ecology and zoonotic risk of clade 2.3.4.4b H5N1 high pathogenicity avian influenza in the sub-Antarctic region","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-11 08:42:09","doi":"10.21203/rs.3.rs-8029950/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3221232c-a675-4e92-9121-11c4133fd565","owner":[],"postedDate":"November 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":57475130,"name":"Biological sciences/Genetics/Functional genomics/Mutagenesis"},{"id":57475131,"name":"Biological sciences/Microbiology/Virology/Influenza virus"}],"tags":[],"updatedAt":"2026-01-16T21:15:23+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-11 08:42:09","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8029950","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8029950","identity":"rs-8029950","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}