In turkeys, unlike chickens, the non-structural NS1 protein does not play a significant role in the replication and tissue tropism of the H7N1 avian influenza virus.

The economic losses caused by high pathogenicity (HP) avian influenza viruses (AIV) in the poultry industry worldwide are enormous. Although chickens and turkeys are closely related Galliformes, turkeys are thought to be a bridging host for the adaptation of AIV from wild birds to poultry because of their high susceptibility to AIV infections. HPAIV evolve from low pathogenicity (LP) AIV after circulation in poultry through mutations in different viral proteins, including the non-structural protein (NS1), a major interferon (IFN) antagonist of AIV. At present, it is largely unknown whether the virulence determinants of HPAIV are the same in turkeys and chickens. Previously, we showed that mutations in the NS1 of HPAIV H7N1 significantly reduced viral replication in chickens in vitro and in vivo. Here, we investigated the effect of NS1 on the replication and virulence of HPAIV H7N1 in turkeys after inoculation with recombinant H7N1 carrying a naturally truncated wild-type NS1 (with 224 amino-acid "aa" in length) or an extended NS1 with 230-aa similar to the LP H7N1 ancestor. There were no significant differences in multiple-cycle viral replication or in the efficiency of NS1 in blocking IFN induction in the cell culture. Similarly, all viruses were highly virulent in turkeys and replicated at similar levels in various organs and swabs collected from the inoculated turkeys. These results suggest that NS1 does not play a role in the virulence or replication of HPAIV H7N1 in turkeys and further indicate that the genetic determinants of HPAIV differ in these two closely related galliform species.

Inactivation of Avian Influenza Virus Inoculated into Ground Beef Patties Cooked on a Commercial Open-Flame Gas Grill.

With the emergence of clade 2.3.4.4b H5N1 highly pathogenic avian influenza virus (AIV) infection of dairy cattle and its subsequent detection in raw milk, coupled with recent AIV infections affecting dairy farm workers, experiments were conducted to affirm the safety of cooked ground beef related to AIV because such meat is often derived from cull dairy cows. Specifically, retail ground beef (percent lean:fat = ca. 80:20) was inoculated with a low pathogenic AIV (LPAIV) isolate to an initial level of 5.6 log10 50% egg infectious doses (EID50)  per 300 g patty. The inoculated meat was pressed into patties (ca. 2.54 cm thick, ca. 300 g each) and then held at 4 °C for up to 60 min. In each of the two trials, two patties for each of the following three treatments were cooked on a commercial open-flame gas grill to internal instantaneous temperatures of 48.9 °C (120°F), 62.8 °C (145°F), or 71.1 °C (160°F), but without any dwell time. Cooking inoculated ground beef patties to 48.9 °C (ave. cooking time of ca. 15 min) resulted in a mean reduction of ≥2.5 ± 0.9 log10 EID50 per 300 g of ground beef as assessed via quantification of virus in embryonating chicken eggs (ECEs). Likewise, cooking patties on a gas grill to 62.8 °C (ave. cooking time of ca. 21 min) or to the USDA FSIS recommended minimum internal temperature for ground beef of 71.1 °C (ave. cooking time of ca. 24 min) resulted in a reduction to nondetectable levels from initial levels of ≥5.6 log10 EID50 per 300 g. These data establish that levels of infectious AIV are substantially reduced within inoculated ground beef patties (20% fat) using recommended cooking procedures.

High Prevalence of Highly Pathogenic Avian Influenza: A Virus in Vietnam's Live Bird Markets.

Background: In recent years, Vietnam has suffered multiple epizootics of influenza in poultry. Methods: From 10 January 2019 to 26 April 2021, we employed a One Health influenza surveillance approach at live bird markets (LBMs) and swine farms in Northern Vietnam. When the COVID-19 pandemic permitted, each month, field teams collected oral secretion samples from poultry and pigs, animal facility bioaerosol and fecal samples, and animal worker nasal washes at 4 LBMs and 5 swine farms across 5 sites. Initially samples were screened with molecular assays followed by culture in embryonated eggs (poultry swabs) or Madin-Darby canine kidney cells (human or swine swabs). Results: Many of the 3493 samples collected had either molecular or culture evidence for influenza A virus, including 314 (37.5%) of the 837 poultry oropharyngeal swabs, 144 (25.1%) of the 574 bioaerosol samples, 438 (34.9%) of the 1257 poultry fecal swab samples, and 16 (1.9%) of the 828 human nasal washes. Culturing poultry samples yielded 454 influenza A isolates, 83 of which were H5, and 70 (84.3%) of these were highly pathogenic. Additionally, a positive human sample had a H9N2 avian-like PB1 gene. In contrast, the prevalence of influenza A in the swine farms was much lower with only 6 (0.4%) of the 1700 total swine farm samples studied, having molecular evidence for influenza A virus. Conclusions: This study suggests that Vietnam's LBMs continue to harbor high prevalences of avian influenza A viruses, including many highly pathogenic H5N6 strains, which will continue to threaten poultry and humans.

An avian-origin internal backbone effectively increases the H5 subtype avian influenza vaccine candidate yield in both chicken embryonated eggs and MDCK cells.

Inactivated vaccines play an important role in preventing and controlling the epidemic caused by the H5 subtype avian influenza virus. The vaccine strains are updated in response to alterations in surface protein antigens, while an avian-derived vaccine internal backbone with a high replicative capacity in chicken embryonated eggs and MDCK cells is essential for vaccine development. In this study, we constructed recombinant viruses using the clade 2.3.4.4d A/chicken/Jiangsu/GY5/2017(H5N6, CkG) strain as the surface protein donor and the clade 2.3.4.4b A/duck/Jiangsu/84512/2017(H5N6, Dk8) strain with high replicative ability as an internal donor. After optimization, the integration of the M gene from the CkG into the internal genes from Dk8 (8GM) was selected as the high-yield vaccine internal backbone, as the combination improved the hemagglutinin1/nucleoprotein (HA1/NP) ratio in recombinant viruses. The r8GMΔG with attenuated hemagglutinin and neuraminidase from the CkG exhibited high-growth capacity in both chicken embryos and MDCK cell cultures. The inactivated r8GMΔG vaccine candidate also induced a higher hemagglutination inhibition antibody titer and microneutralization titer than the vaccine strain using PR8 as the internal backbone. Further, the inactivated r8GMΔG vaccine candidate provided complete protection against wild-type strain challenge. Therefore, our study provides a high-yield, easy-to-cultivate candidate donor as an internal gene backbone for vaccine development.

Genotypic and phenotypic susceptibility of emerging avian influenza A viruses to neuraminidase and cap-dependent endonuclease inhibitors.

Avian influenza outbreaks, including ones caused by highly pathogenic A(H5N1) clade 2.3.4.4b viruses, have devastated animal populations and remain a threat to humans. Risk elements assessed for emerging influenza viruses include their susceptibility to approved antivirals. Here, we screened >20,000 neuraminidase (NA) or polymerase acidic (PA) protein sequences of potentially pandemic A(H5Nx), A(H7Nx), and A(H9N2) viruses that circulated globally in 2010-2023. The frequencies of NA or PA substitutions associated with reduced inhibition (RI) or highly reduced inhibition (HRI) by NA inhibitors (NAIs) (oseltamivir, zanamivir) or a cap-dependent endonuclease inhibitor (baloxavir) were low: 0.60% (137/22,713) and 0.62% (126/20,347), respectively. All tested subtypes were susceptible to NAIs and baloxavir at sub-nanomolar concentrations. A(H9N2) viruses were the most susceptible to oseltamivir, with IC50s 3- to 4-fold lower than for other subtypes (median IC50: 0.18 nM; n = 22). NA-I222M conferred RI of A(H5N1) viruses by oseltamivir (with a 26-fold IC50 increase), but NA-S246N did not reduce inhibition. PA-E23G, PA-K34R, PA-I38M/T, and the previously unreported PA-A36T caused RI by baloxavir in all subtypes tested. Avian A(H9N2) viruses endemic in Egyptian poultry predominantly acquired PA-I38V, which causes only a <3-fold decrease in the baloxavir EC50 and fails to meet the RI criteria. PA-E199A/D in A(H7Nx) and A(H9N2) viruses caused a 2- to 4-fold decrease in EC50 (close to the borderline for RI) and should be closely monitored. Our data indicate antiviral susceptibility is high among avian influenza A viruses with pandemic potential and present novel markers of resistance to existing antiviral interventions.

Phylogenetic and mutational analysis of H10N3 avian influenza A virus in China: potential threats to human health.

In recent years, the avian influenza virus has emerged as a significant threat to both human and public health. This study focuses on a patient infected with the H10N3 subtype of avian influenza virus, admitted to the Third People's Hospital of Kunming City on March 6, 2024. Metagenomic RNA sequencing and polymerase chain reaction (PCR) analysis were conducted on the patient's sputum, confirming the H10N3 infection. The patient presented severe pneumonia symptoms such as fever, expectoration, chest tightness, shortness of breath, and cough. Phylogenetic analysis of the Haemagglutinin (HA) and neuraminidase (NA) genes of the virus showed that the virus was most closely related to a case of human infection with the H10N3 subtype of avian influenza virus found in Zhejiang Province, China. Analysis of amino acid mutation sites identified four mutations potentially hazardous to human health. Consequently, this underscores the importance of continuous and vigilant monitoring of the dynamics surrounding the H10N3 subtype of avian influenza virus, utilizing advanced genomic surveillance techniques.

Multiple transatlantic incursions of highly pathogenic avian influenza clade 2.3.4.4b A(H5N5) virus into North America and spillover to mammals.

Highly pathogenic avian influenza (HPAI) viruses have spread at an unprecedented scale, leading to mass mortalities in birds and mammals. In 2023, a transatlantic incursion of HPAI A(H5N5) viruses into North America was detected, followed shortly thereafter by a mammalian detection. As these A(H5N5) viruses were similar to contemporary viruses described in Eurasia, the transatlantic spread of A(H5N5) viruses was most likely facilitated by pelagic seabirds. Some of the Canadian A(H5N5) viruses from birds and mammals possessed the PB2-E627K substitution known to facilitate adaptation to mammals. Ferrets inoculated with A(H5N5) viruses showed rapid, severe disease onset, with some evidence of direct contact transmission. However, these viruses have maintained receptor binding traits of avian influenza viruses and were susceptible to oseltamivir and zanamivir. Understanding the factors influencing the virulence and transmission of A(H5N5) in migratory birds and mammals is critical to minimize impacts on wildlife and public health.

The Alarming Situation of Highly Pathogenic Avian Influenza Viruses in 2019-2023.

Avian influenza viruses (AIVs) have the potential to cause severe illness in wild birds, domestic poultry, and humans. The ongoing circulation of highly pathogenic avian influenza viruses (HPAIVs) has presented significant challenges to global poultry industry and public health in recent years. This study aimed to elucidate the circulation of HPAIVs during 2019 to 2023. Specifically, we assess the alarming global spread and continuous evolution of HPAIVs. Moreover, we discuss their transmission and prevention strategies to provide valuable references for future prevention and control measures against AIVs.

Mapping the risk of introduction of highly pathogenic avian influenza to Swedish poultry.

Outbreaks of highly pathogenic avian influenza (HPAI) have resulted in severe economic impact for national governments and poultry industries globally and in Sweden in recent years. Veterinary authorities can enforce prevention measures, e.g. mandatory indoor housing of poultry, in HPAI high-risk areas. The aim of this study was to conduct a spatiotemporal mapping of the risk of introduction of highly pathogenic avian influenza virus (HPAIV) to Swedish poultry from wild birds, utilising existing data sources. A raster calculation method was used to assess the spatiotemporal risk of introduction of HPAIV to Swedish poultry. The environmental infectious pressure of HPAIV was first calculated in each 5 km by 5 km cell using four risk factors: density of selected species of wild birds, air temperature, presence of agriculture as land cover and presence of HPAI in wild birds based on data from October 2016-September 2021. The relative importance of each risk factor was weighted based on opinion of experts. The estimated environmental infectious pressure was then multiplied with poultry population density to obtain risk values for risk of introduction of HPAIV to poultry. The results showed a large variation in risk both on national and local level. The counties of Skåne and Östergötland particularly stood out regarding environmental infectious pressure, risk of introduction to poultry and detected outbreaks of HPAI. On the other hand, there were counties, identified as having higher risk of introduction to poultry which never experienced any outbreaks. A possible explanation is the variation in poultry production types present in different areas of Sweden. These results indicate that the national and local variation in risk for HPAIV introduction to poultry in Sweden is high, and this would support more targeted compulsory prevention measures than what has previously been employed in Sweden. With the current and evolving HPAI situation in Europe and on the global level, there is a need for continuous updates to the risk map as the virus evolves and circulates in different wild bird species. The study also identified areas of improvement, in relation to data use and data availability, e.g. improvements to poultry registers, inclusion of citizen reported mortality in wild birds, data from standardised wild bird surveys, wild bird migration data as well as results from ongoing risk-factor studies.

Novel Genotypes of Highly Pathogenic Avian Influenza H5N1 Clade 2.3.4.4b Viruses, Germany, November 2023.

Several subtypes and many different genotypes of highly pathogenic avian influenza viruses of subtype H5 clade 2.3.4.4b have repeatedly caused outbreaks in Germany. Four new highly pathogenic avian influenza genotypes emerged in November 2023 after reassortment with low pathogenicity precursors, replacing genotype BB, which had dominated in Europe since 2022.

Multiplex Dual-Target Reverse Transcription PCR for Subtyping Avian Influenza A(H5) Virus.

An increased risk for human infection with avian influenza A(H5N1) viruses is of concern. We developed an internally controlled, dual-target reverse transcription PCR for influenza A(H5) subtyping. This test could be used to detect influenza A(H5) in clinical samples.

Evaluating the epizootic and zoonotic threat of an H7N9 low-pathogenicity avian influenza virus (LPAIV) variant associated with enhanced pathogenicity in turkeys.

Between 2013 and 2017, the A/Anhui/1/13-lineage (H7N9) low-pathogenicity avian influenza virus (LPAIV) was epizootic in chickens in China, causing mild disease, with 616 fatal human cases. Despite poultry vaccination, H7N9 has not been eradicated. Previously, we demonstrated increased pathogenesis in turkeys infected with H7N9, correlating with the emergence of the L217Q (L226Q H3 numbering) polymorphism in the haemagglutinin (HA) protein. A Q217-containing virus also arose and is now dominant in China following vaccination. We compared infection and transmission of this Q217-containing 'turkey-adapted' (ty-ad) isolate alongside the H7N9 (L217) wild-type (wt) virus in different poultry species and investigated the zoonotic potential in the ferret model. Both wt and ty-ad viruses demonstrated similar shedding and transmission in turkeys and chickens. However, the ty-ad virus was significantly more pathogenic than the wt virus in turkeys but not in chickens, causing 100 and 33% mortality in turkeys respectively. Expanded tissue tropism was seen for the ty-ad virus in turkeys but not in chickens, yet the viral cell receptor distribution was broadly similar in the visceral organs of both species. The ty-ad virus required exogenous trypsin for in vitro replication yet had increased replication in primary avian cells. Replication was comparable in mammalian cells, and the ty-ad virus replicated successfully in ferrets. The L217Q polymorphism also affected antigenicity. Therefore, H7N9 infection in turkeys can generate novel variants with increased risk through altered pathogenicity and potential HA antigenic escape. These findings emphasize the requirement for enhanced surveillance and understanding of A/Anhui/1/13-lineage viruses and their risk to different species.

Co-infection of H9N2 subtype avian influenza virus and QX genotype live attenuated infectious bronchitis virus increase the pathogenicity in SPF chickens.

Avian influenza virus (AIV) infection and vaccination against live attenuated infectious bronchitis virus (aIBV) are frequent in poultry worldwide. Here, we evaluated the clinical effect of H9N2 subtype AIV and QX genotype aIBV co-infection in specific-pathogen-free (SPF) white leghorn chickens and explored the potential mechanisms underlying the observed effects using by 4D-FastDIA-based proteomics. The results showed that co-infection of H9N2 AIV and QX aIBV increased mortality and suppressed the growth of SPF chickens. In particular, severe lesions in the kidneys and slight respiratory signs similar to the symptoms of virulent QX IBV infection were observed in some co-infected chickens, with no such clinical signs observed in single-infected chickens. The replication of H9N2 AIV was significantly enhanced in both the trachea and kidneys, whereas there was only a slight effect on the replication of the QX aIBV. Proteomics analysis showed that the IL-17 signaling pathway was one of the unique pathways enriched in co-infected chickens compared to single infected-chickens. A series of metabolism and immune response-related pathways linked with co-infection were also significantly enriched. Moreover, co-infection of the two pathogens resulted in the enrichment of the negative regulation of telomerase activity. Collectively, our study supports the synergistic effect of the two pathogens, and pointed out that aIBV vaccines might increased IBV-associated lesions due to pathogenic co-infections. Exacerbation of the pathogenicity and mortality in H9N2 AIV and QX aIBV co-infected chickens possibly occurred because of an increase in H9N2 AIV replication, the regulation of telomerase activity, and the disturbance of cell metabolism and the immune system.

Avian influenza viruses in wild birds in Canada following incursions of highly pathogenic H5N1 virus from Eurasia in 2021-2022.

Following the detection of novel highly pathogenic avian influenza virus (HPAIV) H5N1 clade 2.3.4.4b in Newfoundland, Canada, in late 2021, avian influenza virus (AIV) surveillance in wild birds was scaled up across Canada. Herein, we present the results of Canada's Interagency Surveillance Program for Avian Influenza in Wild Birds during the first year (November 2021-November 2022) following the incursions of HPAIV from Eurasia. The key objectives of the surveillance program were to (i) identify the presence, distribution, and spread of HPAIV and other AIVs; (ii) identify wild bird morbidity and mortality associated with HPAIV; (iii) identify the range of wild bird species infected by HPAIV; and (iv) genetically characterize detected AIV. A total of 6,246 sick and dead wild birds were tested, of which 27.4% were HPAIV positive across 12 taxonomic orders and 80 species. Geographically, HPAIV detections occurred in all Canadian provinces and territories, with the highest numbers in the Atlantic and Central Flyways. Temporally, peak detections differed across flyways, though the national peak occurred in April 2022. In an additional 11,295 asymptomatic harvested or live-captured wild birds, 5.2% were HPAIV positive across 3 taxonomic orders and 19 species. Whole-genome sequencing identified HPAIV of Eurasian origin as most prevalent in the Atlantic Flyway, along with multiple reassortants of mixed Eurasian and North American origins distributed across Canada, with moderate structuring at the flyway scale. Wild birds were victims and reservoirs of HPAIV H5N1 2.3.4.4b, underscoring the importance of surveillance encompassing samples from sick and dead, as well as live and harvested birds, to provide insights into the dynamics and potential impacts of the HPAIV H5N1 outbreak. This dramatic shift in the presence and distribution of HPAIV in wild birds in Canada highlights a need for sustained investment in wild bird surveillance and collaboration across interagency partners. IMPORTANCE: We present the results of Canada's Interagency Surveillance Program for Avian Influenza in Wild Birds in the year following the first detection of highly pathogenic avian influenza virus (HPAIV) H5N1 on the continent. The surveillance program tested over 17,000 wild birds, both sick and apparently healthy, which revealed spatiotemporal and taxonomic patterns in HPAIV prevalence and mortality across Canada. The significant shift in the presence and distribution of HPAIV in Canada's wild birds underscores the need for sustained investment in wild bird surveillance and collaboration across One Health partners.

Detection and characterization of H5N1 HPAIV in environmental samples from a dairy farm.

The recent expansion of HPAIV H5N1 infections in terrestrial mammals in the Americas, most recently including the outbreak in dairy cattle, emphasizes the critical need for better epidemiological monitoring of zoonotic diseases. In this work, we detected, isolated, and characterized the HPAIV H5N1 from environmental swab samples collected from a dairy farm in the state of Kansas, USA. Genomic sequencing of these samples uncovered two distinctive substitutions in the PB2 (E249G) and NS1 (R21Q) genes which are rare and absent in recent 2024 isolates of H5N1 circulating in the mammalian and avian species. Additionally, approximately 1.7% of the sequence reads indicated a PB2 (E627K) substitution, commonly associated with virus adaptation to mammalian hosts. Phylogenetic analyses of the PB2 and NS genes demonstrated more genetic identity between this environmental isolate and the 2024 human isolate (A/Texas/37/2024) of H5N1. Conversely, HA and NA gene analyses revealed a closer relationship between our isolate and those found in other dairy cattle with almost 100% identity, sharing a common phylogenetic subtree. These findings underscore the rapid evolutionary progression of HPAIV H5N1 among dairy cattle and reinforces the need for more epidemiological monitoring which can be done using environmental sampling.

Genomic Characterization of Highly Pathogenic Avian Influenza A H5N1 Virus Newly Emerged in Dairy Cattle.

In March 2024, the emergence of highly pathogenic avian influenza (HPAI) A (H5N1) infections in dairy cattle was detected in United States for the first time. We genetically characterize HPAI viruses from dairy cattle showing an abrupt drop in milk production, as well as from two cats, six wild birds, and one skunk. They share nearly identical genome sequences, forming a new genotype B3.13 within the 2.3.4.4b clade. B3.13 viruses underwent two reassortment events since 2023 and exhibit critical mutations in HA, M1, and NS genes but lack critical mutations in PB2 and PB1 genes, which enhance virulence or adaptation to mammals. The PB2 E627 K mutation in a human case associated with cattle underscores the potential for rapid evolution post-infection, highlighting the need for continued surveillance to monitor public health threats.

Mortality in sea lions is associated with the introduction of the H5N1 clade 2.3.4.4b virus in Brazil October 2023: whole genome sequencing and phylogenetic analysis.

Clade 2.3.4.4b highly pathogenic avian influenza (HPAI) H5N1 virus was detected in the South American sea lions found dead in Santa Catarina, Brazil, in October 2023. Whole genome sequencing and comparative phylogenetic analysis were conducted to investigate the origin, genetic diversity, and zoonotic potentials of the H5N1 viruses. The H5N1 viruses belonged to the genotype B3.2 of clade 2.3.4.4b H5N1 virus, which was identified in North America and disseminated to South America. They have acquired new amino acid substitutions related to mammalian host affinity. Our study provides insights into the genetic landscape of HPAI H5N1 viruses in Brazil, highlighting the continuous evolutionary processes contributing to their possible adaptation to mammalian hosts.

Red knots in Europe: a dead end host species or a new niche for highly pathogenic avian influenza?

The 2020/2021 epidemic in Europe of highly pathogenic avian influenza virus (HPAIV) of subtype H5 surpassed all previously recorded European outbreaks in size, genotype constellations and reassortment frequency and continued into 2022 and 2023. The causative 2.3.4.4b viral lineage proved to be highly proficient with respect to reassortment with cocirculating low pathogenic avian influenza viruses and seems to establish an endemic status in northern Europe. A specific HPAIV reassortant of the subtype H5N3 was detected almost exclusively in red knots (Calidris canutus islandica) in December 2020. It caused systemic and rapidly fatal disease leading to a singular and self-limiting mass mortality affecting about 3500 birds in the German Wadden Sea, roughly 1 % of the entire flyway population of islandica red knots. Phylogenetic analyses revealed that the H5N3 reassortant very likely had formed in red knots and remained confined to this species. While mechanisms of virus circulation in potential reservoir species, dynamics of spill-over and reassortment events and the roles of environmental virus sources remain to be identified, the year-round infection pressure poses severe threats to endangered avian species and prompts adaptation of habitat and species conservation practices.

Highly pathogenic avian influenza A (H5N1) virus outbreak in Peru in 2022-2023.

Background: An epizootic of highly pathogenic avian influenza A (H5N1) has spread worldwide since 2022. Even though this virus has been extensively studied for many decades, little is known about its evolution in South America. Methods: Here, we describe the sequencing and characterization of 13 H5N1 genomes collected from wild birds, poultry, and wild mammals in Peru during the genomic surveillance of this outbreak. Results: The samples belonged to the highly pathogenic avian influenza (H5N1) 2.3.4.4b clade. Chilean and Peruvian samples clustered in the same group and therefore share a common ancestor. An analysis of the hemagglutinin and neuraminidase genes detected new mutations, some dependent upon the host type. Conclusions: The genomic surveillance of highly pathogenic avian influenza is necessary to promote the One Health policy and to overcome the new problems entailed by climate change, which may alter the habitats of resident and migratory birds.

Strong and consistent effects of waterbird composition on HPAI H5 occurrences across Europe.

Since 2014, highly pathogenic avian influenza (HPAI) H5 viruses of clade 2.3.4.4 have been dominating the outbreaks across Europe, causing massive deaths among poultry and wild birds. However, the factors shaping these broad-scale outbreak patterns, especially those related to waterbird community composition, remain unclear. In particular, we do not know whether these risk factors differ from those of other H5 clades. Addressing this knowledge gap is important for predicting and preventing future HPAI outbreaks. Using extensive waterbird survey datasets from about 6883 sites, we here explored the effect of waterbird community composition on HPAI H5Nx (clade 2.3.4.4) spatial patterns in the 2016/2017 and 2020/2021 epidemics in Europe, and compared it with the 2005/2006 HPAI H5N1 (clade 2.2) epidemic. We showed that HPAI H5 occurrences in wild birds in the three epidemics were strongly associated with very similar waterbird community attributes, which suggested that, in nature, similar interspecific transmission processes operate between the HPAI H5 subtypes or clades. Importantly, community phylogenetic diversity consistently showed a negative association with H5 occurrence in all three epidemics, suggesting a dilution effect of phylogenetic diversity. In contrast, waterbird community variables showed much weaker associations with HPAI H5Nx occurrence in poultry. Our results demonstrate that models based on previous epidemics can predict future HPAI H5 patterns in wild birds, implying that it is important to include waterbird community factors in future HPAI studies to predict outbreaks and improve surveillance activities.