Skip to main content
NCBI home page
Science Advances logoLink to Science Advances
. 2025 Jan 22;11(4):eado9140. doi: 10.1126/sciadv.ado9140

Association of poultry vaccination with interspecies transmission and molecular evolution of H5 subtype avian influenza virus

Bingying Li 1,†, Jayna Raghwani 2,†, Sarah C Hill 2, Sarah François 3,4, Noémie Lefrancq 5, Yilin Liang 1, Zengmiao Wang 1, Lu Dong 6, Phillipe Lemey 7, Oliver G Pybus 2,3,*, Huaiyu Tian 1,*
PMCID: PMC11753422  PMID: 39841843

Abstract

The effectiveness of poultry vaccination in preventing the transmission of highly pathogenic avian influenza viruses (AIVs) has been debated, and its impact on wild birds remains uncertain. Here, we reconstruct the movements of H5 subtype AIV lineages among vaccinated poultry, unvaccinated poultry, and wild birds, worldwide, from 1996 to 2023. We find that there is a time lag in viral transmission among different host populations and that movements from wild birds to unvaccinated poultry were more frequent than those from wild birds to vaccinated poultry. Furthermore, our findings suggest that the HA (hemagglutinin) gene of the AIV lineage that circulated predominately in Chinese poultry experienced greater nonsynonymous divergence and adaptive fixation than other lineages. Our results indicate that the epidemiological, ecological, and evolutionary consequences of widespread AIV vaccination in poultry may be linked in complex ways and that much work is needed to better understand how such interventions may affect AIV transmission to, within, and from wild birds.

Mass poultry vaccination affects H5 avian influenza virus transmission and evolution, revealing complex host-virus dynamics.

INTRODUCTION

During the summer of 2022, seabirds in many European, North American, and African countries suffered unprecedented mortality from avian influenza virus (AIV) (1). The causative virus is highly pathogenic H5 subtype AIV (HPAIV) belonging to clade 2.3.4.4b and has been detected at high incidences in wild birds (2). Wild birds, the natural reservoir of AIVs, can acquire and transmit viruses to poultry or mammals (3, 4) and play a major role in the maintenance and global dissemination of AIVs (5–8). In addition, transmission within and between species of genetically diverse AIVs can facilitate genomic reassortment (9, 10) and the emergence of previously unidentified HPAIV lineages (11–16). Consequently, HPAIV pose a severe threat to human, wild, and domestic animal health.

To protect poultry from HPAIV infection, some countries, mainly in Asia and Africa, have implemented vaccination programs in poultry for H5 subtype HPAIV (Fig. 1A) (17–20). On the basis of national vaccination data from 2010, Egypt had the highest vaccination coverage in poultry (82%), followed by China [73% in 2009 and 93% in layer and slow-growing meat chickens, 14% in fast-growing meat chickens, and <30% in ducks in 2018 (21)], Vietnam (31%), and Indonesia (12%) (22). In 2013, the vaccination coverage of commercial layer flocks in two districts of Bangladesh was reported to be 32 and 54% (23). France and the Netherlands have also vaccinated poultry; however, their overall vaccination coverages are very low (<0.1%). Since 2005, China has implemented a nationwide vaccination program (24) and is responsible for >90% of the global consumption of H5 AIV vaccines. Because of the continuing evolution of H5 AIV, vaccine strains are frequently updated to ensure their effectiveness (table S1) (25, 26). Several studies have suggested that extensive vaccination of Chinese poultry against H5 AIV has suppressed outbreaks effectively and decreased substantially the prevalence of H5 AIV in live bird markets (23, 26–28).

Fig. 1. Global dynamics of H5 AIV.

Fig. 1.

Open in a new tab

(A) Vaccination coverage and number of H5 AIV HA gene sequences across different continents and countries. Countries are shaded according to the vaccination coverage in poultry in 2010. The pie charts show the total number of H5 AIV HA gene sequences sampled from poultry, wild birds, and environment since 1996. Viral sequences are categorized according to their host: wild birds (purple), poultry (yellow), and other avian/environment (white). (B) Number of H5 AIV HA gene sequences sampled in poultry and wild birds, per year, per continent. (C) Maximum clade credibility tree of H5 AIV HA gene sequences sampled from 1996 to 2023. Tree tips are colored according to the host from which the sequence was sampled, whereas internal branches represent ancestral host states inferred using the asymmetric discrete phylogenetic model (dark blue, wild birds; orange, Chinese poultry; light green, European poultry; purple, Indonesian poultry; pink, Japanese poultry; light blue, Korean poultry; dark green, Vietnamese poultry; and yellow, Bangladeshi poultry). HPAIV, highly pathogenic influenza virus; LPAIV, lowly pathogenic influenza virus. Inset: a root-to-tip regression of genetic divergence against dates of sample collection. (D) Cumulative number of host population changes (Markov jumps) on lineages in the HA gene phylogeny. The lineage transitions between hosts (wild birds, WB; Chinese poultry, CP; and European poultry, EP) were summarized from a posterior sample of trees from the asymmetric discrete phylogenetic model. (E) Time series of the annual mean number of HA gene lineage transitions between wild birds, Chinese poultry, and European poultry.

However, there are concerns that mass vaccination against H5 AIV could affect the molecular evolution of the virus (29). For example, after the introduction of mass vaccination against H5N1 AIV in China, a notable increase in the evolutionary rate of H5N1 AIV was observed during 2005 to 2010 (30). Similarly, in North America, the evolutionary rate of H5N2 AIV in Mexico was inferred to be higher between 1993 and 2002 after a period of mass vaccination in this region, compared to the evolutionary rate of H5 AIV in the United States, where vaccination was not used (31). Furthermore, the AIV lineages that circulate in Egypt and Indonesia, where vaccination against H5N1 is prevalent, are characterized by higher evolutionary rates and a greater number of positively selected sites in the hemagglutinin (HA) gene compared to countries where vaccination is absent, such as Nigeria, Turkey, and Thailand (32, 33). These studies have concentrated mainly on viruses sampled from poultry and lack data from wild birds and at the wild bird–poultry interface, which hinders our understanding of the underlying mechanisms of AIV evolution.

Here, we investigate the evolutionary history of H5 subtype AIV lineages among vaccinated poultry, unvaccinated poultry, and wild birds, worldwide, from 1996 to 2023 and compare the evolutionary dynamics of H5 AIV in different host populations that vary in vaccination status. Our results can be used to inform the development of effective strategies for AIV vaccination and immunization.

RESULTS

Interspecies transmission of H5 AIV at the interface of poultry and wild birds

We collated a total of 22,606 HA gene sequences belonging to H5 AIV, sampled from each continent since 1996 (Fig. 1, A and B). The virus sequences were unevenly distributed over time and space across host populations (table S2). Therefore, for downstream analyses, we focused on Europe and Asia, from where sufficient viral genetic data were available from long-term sampling of poultry and wild birds to quantify the dynamics of virus transmission and evolution within and among host populations. By taking into account the implementation of vaccination programs and the availability of viral genetic sequences (fig. S1), we categorized sequences into eight groups based on the country of sampling, host species, and vaccination status: wild birds (without vaccination), European poultry (without vaccination), Japanese poultry (without vaccination), Korean poultry (without vaccination), Indonesian poultry (with vaccination), Bangladeshi poultry (with vaccination), Vietnamese poultry (with vaccination), and Chinese poultry (with vaccination). For the remainder of the analysis, we use the terms with vaccination and without vaccination as a simplified binary classification while acknowledging that nationally reported vaccination coverages can range continuously from 0 to 100% and coverage can vary within each country.

To investigate H5 AIV transmission and evolution between and within these eight groups since 1996, we used 2407 H5 HA gene sequences to reconstruct time scaled phylogenies under an asymmetric discrete trait model (Fig. 1C, fig. S2, and table S3). Interspecies movements of H5 AIV lineages from Chinese poultry to wild birds and from wild birds to European poultry were frequently observed (Fig. 1D). Bayesian reconstruction of the host status of H5 AIV lineages indicates that there were three waves of spillovers from Chinese poultry to wild birds (CPtoWB) and from wild birds to European poultry (WBtoEP) (Fig. 1E). To investigate lineage transitions among these three host types before 2020, we conducted an extended convergent cross mapping (CCM) analysis on the time series of mean annual lineage transitions (Markov jumps) from Chinese poultry to wild birds and from wild birds to European poultry, between 1996 and 2019. We find that CPtoWB transitions preceded WBtoEP transitions, with a time lag of ~2 to 3 years (Fig. 1E and fig. S3). The observed pattern is robust to different genomic sampling strategies (figs. S4 to S7).

Overall, more viral lineage transitions were observed from wild birds to unvaccinated poultry populations than to vaccinated poultry populations (Fig. 2A, fig. S8, and table S4). Interspecies transitions from unvaccinated poultry to wild birds were frequent after 2016 (fig. S8). Virus lineage movements between poultry populations were not common. This observation was also robust to diverse genomic sampling strategies (fig. S9 and table S5).

Fig. 2. Transmission and evolution of the HA gene of H5 AIV between and within different host populations.

Fig. 2.

Open in a new tab

(A) Interspecies lineage transmission between wild birds and poultry populations with different vaccination statues. The plots were summarized from a posterior sample of trees from the asymmetric discrete phylogenetic model. Chord diagrams show the mean cumulative lineage transitions between different groups of sequences (dark blue, wild birds; orange, Chinese poultry; light green, European poultry; purple, Indonesian poultry; pink, Japanese poultry; light blue, Korean poultry; dark green, Vietnamese poultry; and yellow, Bangladeshi poultry). Animal silhouettes are from PhyloPic.org. (B) Time-resolved phylogenies of H5 AIV lineages in wild birds and vaccinated poultry. The dashed line represents the date when each country started implementing avian influenza vaccination for poultry. The density plot shows the estimated most recent common ancestor (tMRCA) for each lineage. (C) The estimated substitution rates of the HA gene vary among host-specific H5 AIV lineages. Lineages in unvaccinated host populations (the early- and late-wild bird lineages) are shown below the line, and those in vaccinated host populations (the Chinese, Bangladeshi, and Indonesian poultry lineages) are shown above the line. The highlighted region shows 95% CIs. The bottom panel shows a scatterplot of the vaccination coverage (VC%) associated with each lineage versus the inferred evolutionary rate for that lineage. (D) Scatterplot of the variance in sampling date versus estimated evolutionary rate, for each host-specific lineage. The error bars show the 95% CIs for estimated evolutionary rates. A regression analysis (excluding the Chinese poultry lineage data point) was conducted to show the negative relationship between expected due to the time dependency of inferred evolutionary rates (r = −0.92; P < 0.05) and highlight that the Chinese poultry data point is an outlier. The inset panel shows a scatterplot of vaccination coverage (VC%) associated with each lineage against the residual of the fitted regression.

Evolutionary dynamics of H5 AIV in different host populations

We next investigated the evolutionary dynamics of H5 AIV in wild birds and poultry populations with different vaccination levels. Poultry-dominated viral lineages were identified only for China (vaccinated since 2005), Bangladesh (vaccinated since 2012), and Indonesia (vaccinated since 2004) (Fig. 2B). Our results indicated that the Chinese poultry lineage had a higher substitution rate (mean rate = 5.38 × 10−3 sub per site per year; 95% highest posterior density (HPD): 5.02 × 10−3 to 5.76 × 10−3; P < 0.05) than the early-wild bird lineage (3.39 × 10−3 sub per site per year; 95% HPD: 2.97 × 10−3 to 3.83 × 10−3) (Fig. 2C). Notably, this result is robust to the time dependence of virus evolutionary rates (34) because both lineages have similarly high variation in sequence sampling dates (Fig. 2D). The substitution rates of the poultry-associated viral lineages in Bangladesh and Indonesia were also faster than those of the early-wild bird lineage (P < 0.05). However, there is no clear relationship between the substitution rate and vaccination coverage (Fig. 2C). In addition, the substitution rate of the Chinese poultry lineage during the vaccination period (5.12 × 10−3 sub per site per year; 95% HPD: 4.71 × 10−3 to 5.55 × 10−3) was faster than before vaccination (4.79 × 10−3 sub per site per year; 95% HPD: 4.38 × 10−3 to 5.18 × 10−3; P < 0.05; fig. S10). Furthermore, the substitution rate of the late-wild bird lineage was faster compared to the Chinese poultry lineage from which it emerged (5.91 × 10−3 sub per site per year; 95% HPD: 5.03 × 10−3 to 6.56 × 10−3). However, this result should be interpreted with caution as the late-wild bird lineage is associated with limited temporal sampling and therefore may harbor mildly deleterious mutations.

If the higher evolutionary rate in the HA gene in vaccinated poultry lineages is simply due to increased viral transmission facilitated by high poultry densities, then we should also observe higher evolutionary rates in other gene segments (35). In contrast, we estimate a faster rate of evolution for PB2 in the wild bird lineage (3.62 × 10−3 sub per site per year; 95% HPD: 3.23 × 10−3 to 3.92 × 10−3; P < 0.05) than in the Chinese poultry lineage (3.44 × 10−3 sub per site per year; 95% HPD: 3.08 × 10−3 to 3.78 × 10−3; figs. S11 and S12). These results suggest that the elevated evolutionary rate in the HA gene in Chinese poultry may not be attributable to higher transmission rates alone and instead could result from vaccine-driven selection in the HA gene.

In addition, we subdivided the poultry population into different species groups: chicken, duck, goose, and other (unclassified poultry). We found that domestic ducks maintained the spread of H5 AIV in poultry in China and Bangladesh (fig. S13, A and B), whereas domestic chickens were more important in H5 AIV transmission in Indonesian poultry (fig. S13, C and D). However, frequent interspecies transmission makes it difficult to identify specific poultry lineages.

Viral adaptive evolution in host populations with different vaccination status

We then investigated whether the higher rate of molecular evolution observed in the Chinese poultry lineage, compared to the early-wild bird lineage, could be explained by a greater viral adaptive evolution in poultry. For each lineage, we calculated the nonsynonymous and synonymous divergence of the HA gene over time, from a known reference sequence [National Center for Biotechnology Information (NCBI) Reference Sequence: AF144305; Fig. 3, A to C]. Our results indicate higher nonsynonymous divergence for the Chinese poultry lineage than for other lineages, such as wild bird lineages and poultry lineages in Bangladesh and Indonesia, suggesting that the overall higher rate for Chinese lineage is associated with adaption rather than merely faster transmission (Chinese poultry lineage gradient = 0.0027, P < 0.05; early-wild bird lineage = 0.0005, P < 0.05; late-wild bird lineage = 0.0007, P < 0.05; Bangladeshi poultry lineage = 0.0009, P < 0.05; Indonesian poultry lineage I = 0.0011, P < 0.05; Indonesian poultry lineage II = 0.0021, P < 0.05). Furthermore, the nonsynonymous divergence rate increased in the Chinese poultry lineage after mass vaccination [gradient before 2005 = 0.0017, 95% confidence interval (CI) = 0.0011 to 0.0023; gradient between 2005 and 2010 = 0.0046, 95% CI = 0.0041 to 0.0052; table S6]. To test for the potential impact of different sampling intensities among countries, the Chinese poultry lineage was resampled and the results remained consistent (fig. S14). This indicates that the Chinese poultry lineage exhibited a greater nonsynonymous divergence, especially after vaccination, compared to the wild bird lineages and the poultry lineages that experienced lower vaccination coverage.

Fig. 3. Temporal dynamics in divergence and adaptive fixation of the H5 AIV HA gene, in different host-specific lineages.

Fig. 3.

Open in a new tab

(A to C) Nonsynonymous (solid lines) and synonymous (dashed lines) divergence of the HA gene through time. The host-specific lineages were classified into three groups: (A) early-wild bird lineage (e-WB; only the top lineage was retained) and late-wild bird linage (l-WB); (B) Indonesian poultry lineage I (IP-I), Indonesian poultry lineage II (IP-II), and Bangladeshi poultry lineage (BP); and (C) Chinese poultry lineage (CP). Divergences were computed using 1-year sliding windows. Shaded regions show 95% CIs. The dashed line represents the date when each country started implementing avian influenza vaccination for poultry. (D) Accumulation of viral adaptive substitutions in the Chinese poultry lineage. Two regression lines were estimated, before (r1) and after (r2) 2005. The first sequence, sampled in 1996, was used as the ancestral sequence, and those sampled from 1997 to 1999 were excluded due to insufficient sample size.

Because nonsynonymous divergence can result from either positive selection or random genetic drift, we also used an independent population genetic method to estimate the rate of adaptive substitutions in the HA gene of the Chinese poultry lineage. The estimated rate of adaptive fixation in that lineage was higher after the initiation of vaccination in Chinese poultry (gradient before 2005 = 0.24 adaptive fixations per codon per year; P < 0.05; gradient after 2005 = 0.78 adaptive fixations per codon per year; P < 0.05; Fig. 3D). These estimated adaptation fixation rates are rapid but lower than those previously inferred for the HA gene of human influenza subtype H3N2 (1.52) and subtype H1N1 (1.02) (36). Next, we estimated the dN/dS values for each host-associated lineage using the renaissance counting method (37). The mean dN/dS value is higher in the Chinese poultry lineage (0.24) than in the early-wild bird lineage (0.16) and the late-wild bird lineage (0.18) (table S7). The mean dN/dS values of the Bangladeshi (0.17) and Indonesian poultry I and II lineages (0.18 for both) were also comparatively lower.

Furthermore, dN/dS values were calculated for individual codons in the HA gene, for each host-specific H5 AIV lineage. We used multiple methods to detect sites under positive selection (Table 1). Our results indicate that the number of amino acid sites under positive selection in the Chinese poultry lineage was higher compared to the two wild bird lineages, as well as the Bangladeshi and Indonesian poultry lineages (Table 1). A higher number of positively selected sites were also observed in resampled Chinese poultry lineage (table S8), indicating that the finding is robust to diverse sampling strategies (tables S9 and S10).

Table 1. Positively selected sites in the HA gene among different host-specific H5 AIV lineages.

FEL, Fixed Effects Likelihood; SLAC, Single Likelihood Ancestor Counting; FUBAR, Fast Unconstrained Bayesian AppRoximation; RC, renaissance counting method implemented in BEAST.

Host-specific lineages Vaccination coverage Livestock density in 2015 (birds/km2)† Methods
Chicken Duck RC FEL (P < 0.1) SLAC (P < 0.1) FUBAR (PP > 0.9)
Chinese poultry lineage 73% (22) 860 166 3, 61, 142, 156, 157, 171, 172, 178, 185, 205, 242, 285, 289, 527 87, 142*, 154*, 156, 157*, 172*, 185, 285*, 291*, 325* 3, 142*, 154*, 157*, 172*, 185*, 285* 142*, 154*, 157*, 171, 172*, 285*
Bangladesh poultry lineage 43% (23) 1917 387 170, 204, 205 204*, 205* 204, 205 170, 204*, 205*
Indonesia poultry lineage I 12% (22) 1512 121 170, 204, 211 205*, 554 205 170*, 205*
Indonesia poultry lineage II 10, 156, 529 3, 11, 156*, 529 156* 5*, 11*, 156*
Early-wild bird lineage – – 10, 102, 170, 171 102*, 170*, 171 102*, 170 102*, 170*, 171
Late-wild bird lineage – – 10 185, 507 - 252
Open in a new tab

*Sites inferred to be positively selected with a posterior probability (PP) of >0.95 or a P value (P) of <0.05.

†Mean livestock density was calculated using data from the Gridded Livestock of the World (96) website, hosted by the Food and Agriculture Organization (https://dataverse.harvard.edu/dataverse/glw). Regions with <10 birds/km2 were excluded from the calculation.

Most of the positively selected sites were associated with known immune epitopes. In the Chinese poultry lineage, these were mainly located in the receptor binding subdomain (fig. S15). For example, positive selection was observed at sites 156, 157, and 242, which are recognized as CD8+ T cell epitopes (38), and at site 87, which is associated with antibody epitope 65C6 (39). Within the late-wild bird lineage, positively selected site 252 is associated with the H5246-260 epitope, which can induce T cell activation in chickens immunized against H5 HA antigen (40). The positive selection analysis also revealed mutations (mostly in the receptor binding subdomain) that are associated with changes within the Chinese poultry lineage (D142E, H154Q/L/N, R156V/T/M/K/N/A, S157P/A, S171D/N, Q185R/K/S/G, and L285V/M/I) and with the transition of the virus from the Chinese poultry lineage to the late-wild bird lineage (E142, Q154, A/T156, P157, D171, R185, and V285). The amino acid states of these sites in the early-wild bird lineage (D142, N154, R156, S157, N171, Q185, and L285) are mostly different from those in the late-wild bird lineage, indicating that these changes are not reversion mutations.

DISCUSSION

Since 2020, H5 subtype AIVs have caused outbreaks in European and Asian countries, posing a serious real threat to the poultry industry and a potential threat to public health. We reconstructed the interspecies transmission history of H5 subtype AIVs among wild birds and poultry populations with different vaccination levels. Our analysis reveals a shift from a lineage circulating within Chinese poultry to one circulating among wild birds. The wild bird lineage has been transmitted frequently to unvaccinated European poultry, whereas the spillover from wild birds to vaccinated poultry appears to be impeded. Furthermore, the virus lineage circulating in Chinese poultry exhibits evidence of more nonsynonymous and adaptive molecular evolution in the HA gene after the date of introduction of mass poultry vaccination. The Chinese poultry lineage may have experienced more vaccine-driven selection compared to other lineages. Although this pattern is consistent with the hypothesis that vaccination may have affected upon HA evolution in these poultry lineages, our finding does not establish a causal relationship and further virological work will be needed to test this hypothesis directly. In addition, more research is also needed to assess whether these mutations in HA have had an impact on the late-wild bird lineage.

Because of the higher virus prevalence, Chinese poultry, as well as Southeast Asian poultry, have been regarded as the main reservoirs of H5 HPAIV (41). H5 AIV spread to other regions through bird migration (42) and poultry trade (43), and our results reveal a 2- to 3-year delay between the peak of lineage dissemination from Chinese poultry and the peak of lineage introduction into European poultry. Although it has been suggested that the intercontinental spread of H5 AIV can occur within a single avian migratory cycle (5, 7, 42), the delay we observe may be attributed in part to preexisting immunity in wild birds, which may provide partial protection against infection and disease (44–46), leading to reduced circulation and potentially dampening large-scale outbreaks in wild birds. However, the duration of protection conferred by previous AIV infection and its impact on the epidemiology of AIV have not yet been fully elucidated (47). Furthermore, the relatively short generation length of many wild birds (~3 years) may result in higher turnover (48) of immunologically naïve wild birds in nature. As the reservoir of H5 AIV shifted from Chinese poultry to wild birds, frequent migration and large-scale spatial distribution of wild birds probably facilitated interspecies transmission between wild birds and poultry populations (42). It is crucial that countries and regions enhance regular surveillance of AIVs in wild birds and actively and closely monitor the dynamics of virus transmission.

Following the mass poultry vaccination strategy implemented in China since 2005, the spread of H5 AIV there has been relatively well controlled. Interspecies transmission of these viruses from or to Chinese poultry seems to be limited. However, the Chinese poultry lineage may have experienced more antigenic evolution compared to other lineages. Mutations at amino acid positions 136, 142, 157, 172, 201, and 205 in the H5 AIV HA gene have been shown to reduce reactivity to specific antibodies (49). Amino acids at positions 142, 172, and 205 also appear to function as immunodominant epitopes in H5 viruses (49). We detected positive selection at positions 142, 157, 172, and 205 in the Chinese poultry lineage but not in the wild bird lineages or the other poultry lineages. The H5N6 virus now circulating in Chinese poultry is antigenically divergent from the strains included in the commercial vaccine in China (50, 51), which may result in reduced vaccine effectiveness. Our study indicates that, when vaccination is used, regular monitoring and refinement of vaccines to target emerging escape variants are necessary to respond to the emergence of previously unidentified viruses. Although contemporary H5 HPAIVs are considered unlikely to acquire the ability to infect and circulate stably among the human population (52), there is still an urgent need to control the spread of the virus among wild birds, not only for the preservation of wildlife but also for ensuring the safety of poultry (53).

Because H5 AIV has become increasingly widespread in wild bird populations (54, 55), many unvaccinated countries may be considering large-scale vaccination of poultry populations to prevent the transmission of H5 AIV from wild birds to poultry. Achieving high vaccine coverage and efficacy will be critical for reducing the infection rates of H5 AIV in vaccinated populations (56–61). Studies have shown that inadequate vaccination coverage and suboptimal vaccine efficacy can lead to reemergence of H5 AIV outbreaks in vaccinated poultry (62–64), potentially allowing virulent strains to emerge and circulate within the population by keeping hosts alive while still allowing transmission (65, 66). However, among the countries now vaccinating poultry against H5 AIV, China is, to the best of our knowledge, the only country that updates vaccine strains every 1 to 2 years. In contrast, other countries do not update their vaccine strains as frequently, which may lead to vaccine mismatch and larger epizootics. China initiated poultry vaccination against HPAI H7 in 2017. Although the prevalence of H5/H7 AIV in Chinese poultry was reduced by mass vaccination (26, 28), stronger immune selection may increase the rate of viral adaptation (67, 68). Vaccine-driven evolution may also increase infection risk for immunologically naive individuals, such as unvaccinated poultry and wild birds (66). Continued molecular epidemiological surveillance in vaccinated poultry and at the wild bird–poultry interface is vital for monitoring virus adaptive evolution and characterizing viral mutations (69, 70).

Our study has several limitations. First, although we obtained robust results supported by different datasets, the heterogeneous sampling rates of infected wild birds and poultry may bias ancestral reconstructions. It is challenging to accurately identify ancestral host states in AIV lineages for countries with limited AIV surveillance in wild bird populations. Second, extracting viral lineages that circulate exclusively in vaccinated poultry was not possible due to the unknown vaccination status of the sampled hosts. Vaccination coverage can vary among different poultry species, even within the same country, and may exhibit heterogeneity among regions and over time. Third, the mixed distribution of various poultry species and the heterogeneity in sampling intensity across different regions complicate the inference of the specific impact of each poultry species on viral evolution and transmission. Fourth, the uncertainty and absence of viral genetic and surveillance data from many countries, both with and without vaccination, make it difficult at present to undertake a more quantitative comparative analysis of the association between vaccination levels and virus evolutionary dynamics. Last, although amino acid changes at certain positively selected sites on the HA gene were observed during the transition from the Chinese poultry lineage to the late-wild bird lineage, future experiments should be conducted to determine whether these mutations affect the transmissibility of H5 AIV in poultry and wild birds.

In conclusion, we find that vaccination in Asian poultry likely reduced the interspecies transmission of these viruses. We observe elevated rates of HA molecular evolution and adaptation in H5 AIV in Chinese poultry after the introduction of vaccination. Such circumstances may have increased the probability of infection in wild species at the wild bird–domestic bird interface and, in turn, altered the selective pressure acting upon the virus. As avian influenza continues to pose considerable challenges to wild and domestic animals, our research can help inform the development of preventive measures against AIV, such as global vaccination policies.

MATERIALS AND METHODS

Sequence data

We collated 12,818 publicly available HA gene sequences of H5 AIV, with a minimum length of 1200 nucleotides, sampled in Asia and Europe from January 1996 to February 2023 from the Global Initiative on Sharing All Influenza Data (GISAID). Only sequences with available information on date of collection, host species, and sampling location were retained for further analysis. We aligned the sequences using MAFFT v7.487 (71) with the autoselect algorithm (72). Recombinant sequences were detected using RDP4 package (73). Seven methods from the RDP package were used—RDP, GENECONV, Bootscan, Maxchi, Chimaera, SiScan, and 3seq—with default parameters, allowing a maximum acceptable P value of 0.05. A sequence was considered a potential recombinant if at least three of the seven detection methods showed a statistically significant difference (P < 0.05) (74, 75). Using this criterion, 13 recombinant sequences were identified and removed. The maximum likelihood tree was estimated using FastTree v2.1.11 (76) under the automatically determined best-fit substitution model, followed by a root-to-tip regression analysis performed in TempEst v1.5.3 (77). We manually identified and removed 57 sequences with unexpectedly high or low levels of genetic divergence, given their sampling time. Last, 10,225 sequences were retained for further analysis.

On the basis of the sustained sampling efforts and the appropriate viral sample size (>500 sequences; fig. S1), six Asian countries (China, Bangladesh, Japan, Korea, Indonesia, and Vietnam) were selected. All European countries were collectively treated as a single group. To focus on the viral lineages that circulate in wild birds and poultry, the sequences were categorized into eight groups based on both geographic information and host type. These groups include wild birds (3118 sequences), European poultry (1939 sequences, without vaccination against H5 AIV), Japanese poultry (521 sequences, without vaccination against H5 AIV), Korean poultry (494 sequences, without vaccination against H5 AIV), Bangladeshi poultry (506 sequences, with vaccination against H5 AIV), Indonesian poultry (787 sequences, with vaccination against H5 AIV), Vietnamese poultry (1329 sequences, with vaccination against H5 AIV), and Chinese poultry (1531 sequences, with vaccination against H5 AIV).

We then downsampled these datasets in a stratified manner to create a more equitable distribution of AIV sequences between wild birds and poultry (main dataset): (i) wild bird dataset: randomly selected at most two sequences per month per country outside China and per month per province in China, comprising 1087 gene sequences from January 1999 to January 2023; (ii) European poultry dataset: randomly selected at most one sequence per month per country, including 338 gene sequences from January 1997 to January 2023; (iii) Japanese poultry dataset: randomly selected at most two sequences per month, including 72 HA gene sequences from January 2000 to January 2023; (iv) Korean poultry dataset: randomly selected at most two sequences per month, including 72 gene sequences from October 2008 to October 2022; (v) Bangladeshi poultry dataset: randomly selected at most one sequence per month, including 106 HA gene sequences from January 2007 to August 2022; (vi) Indonesian poultry dataset: randomly selected at most one sequence per month, including 121 HA gene sequences from January 2003 to March 2022; (vii) Vietnamese poultry dataset: randomly selected at most one sequence per month, including 151 HA gene sequences from 2003 to December 2021; and (viii) Chinese poultry dataset: randomly selected at most one sequence per month per province, including 462 gene sequences from January 1996 to March 2022.

As the ancestral reconstruction of discrete traits depends on sample availability, over- or undersampling of sequences from a particular trait can substantially affect the estimated ancestral states (78, 79). Two different subsampling strategies were used to test the sensitivity of the model to sampling bias: (i) random sampling to reduced biased sampling between periods and groups and (ii) random sampling for an equivalent representation of different groups. Further details on the subsampling procedure and dataset composition can be found in the Supplementary Materials.

Considering that substantial mutations have also accumulated in the PB2 gene (80, 81), we used a similar method to collate PB2 gene sequences from H5 AIV sampled from Asia and Europe for sensitivity analysis (see the Supplementary Materials).

Phylogenetic inference

Evolutionary histories were estimated with the Bayesian phylogenetic package BEAST v1.10.4 (82), using the BEAGLE (83) library to improve computational speed. Specifically, we used an SRD06 substitution model (42), an uncorrelated lognormal relaxed clock (42), and a Gaussian Markov random field (GMRF) Bayesian Skygrid coalescent model (84). We subsequently used an eight-state discrete trait analysis (DTA) implemented in BEAST 1.10.4 to infer ancestral node hosts on empirical distributions of 500 time-calibrated trees sampled from the posterior tree distributions (85). An asymmetric model was used for the host discrete trait, which allows different rates of lineage movement between each pair of host states (86). Three independent Markov chain Monte Carlo (MCMC) runs were performed for 400 million steps and logged every 20,000 steps. The first 10% of each chain was discarded as burn-in. We confirmed the convergence of all chains in Tracer v1.7.1 (87), ensuring that the effective sample size (ESS) was > 200 for all parameters. A maximum clade credibility tree was estimated using TreeAnnotator v1.10.4 and subsequently visualized using FigTree v1.4.4 (http://tree.bio.ed.ac.uk/software/figtree) along with the R package ggtree v2.4.1 (88). The transition histories between host discrete states were summarized using the TreeMarkovJumpHistoryAnalyzer tool in BEAST, which extracts Markov jumps and their timings from a posterior tree distribution annotated with Markov jump histories (89). We used the BaTS 2.0 software (90) to investigate the uncertainty arising from phylogenetic error (grouped by sampling location and host), which was compared to a null hypothesis that there was no association between the phylogenetic structure and traits by performing tip randomization with 1000 replicates (table S11).

Evolutionary analysis within host-specific populations

To infer virus dissemination within host-specific populations, we removed AIV phylogenetic clades that were determined to represent transmissions between different species (43). We extracted sequences for each host-specific lineage based on the aforementioned DTA and independently estimated the temporal phylogenies using the same substitution, clock, and tree models as described above. A strong phylogenetic temporal structure was observed in all host-specific lineages (fig. S16).

Divergence analysis

To estimate site-specific synonymous and nonsynonymous substitutions in the different host-specific lineages, we applied a renaissance counting (37) procedure implementing a codon position–specific HKY nucleotide substitution model along with an uncorrelated lognormal relaxed clock (42) and the GMRF Bayesian Skygrid coalescent model (84). Sites under positive selection were also identified using complementary approaches. Specifically, FUBAR (Fast Unconstrained Bayesian AppRoximation) (91), SLAC (Single Likelihood Ancestor Counting) methods (92), and FEL (Fixed Effects Likelihood) methods (92) as implemented in Hyphy v2.5 were used (93).

The synonymous and nonsynonymous divergence for each host-specific lineage were calculated as the average Hamming distance between each sequenced isolate in that lineage and a reference sequence (NCBI Reference Sequence: AF144305). The estimated divergence was calculated by dividing the total number of observed differences between the isolate and reference nucleotide sequence that resulted in a substitution(nonsynonymous or synonymous) by the number of possible nucleotide mutations that could result in a substitution, weighted by kappa (94), the transition/transversion rate ratio, which was inferred from host-specific analysis using BEAST.

Time series analysis

The extended CCM was applied to detect the time lags between the annual number of Markov jumps from Chinese poultry to wild birds (CPtoWB) and from wild birds to European poultry (WBtoEP) during 1996 to 2019 (95). First, we randomly selected 20 inferred Markov jump states among host populations from the converged MCMC chains. These 20 separate time series of Markov jumps were concatenated into the long-term chain for the extended CCM analysis. Then, we determined the optimal embedding dimension, which is optimal or very close to optimal for both variables. Last, the optimal value of the cross-mapping lag was determined by testing different values. The optimal cross-mapping lags are negative and with the magnitude of the lag roughly equal to the time delay between two time series (95). This analysis was performed using the R package rEDM v1.14.0 (95).

Estimation of rates of adaptive substitution

We used an established population genetic method related to the McDonald-Kreitman test (36, 68) to estimate the number of adaptive substitutions per codon per year in the HA H5 gene from the Chinese poultry lineage. We used a consensus of HA sequences from the earliest time point (sampled in 1996) as an outgroup to estimate ancestral and derived site frequencies at subsequent time points. A bootstrap analysis with 1000 replicates was performed to assess statistical uncertainty.

Acknowledgments

We gratefully acknowledge the authors of the originating and submitting laboratories for crucial contributions to the generation and sharing of genome sequences and associated metadata through the GISAID.

Funding: H.T. acknowledges support from the National Key Research and Development Program of China (2022YFC2303803), the National Natural Science Foundation of China (82073616 and 82204160), the Beijing Natural Science Foundation (L232014), the Fundamental Research Funds for the Central Universities (2233300001), the Beijing Research Center for Respiratory Infectious Diseases Project (BJRID2024-002), and the Research on Key Technologies of Plague Prevention and Control in Inner Mongolia Autonomous Region (2021ZD0006). J.R. and O.G.P. were supported by the UKRI GCRF One Health Poultry Hub (grant no. B/S011269/1). S.C.H. was supported by a Sir Henry Wellcome Postdoctoral Fellowship from the Wellcome Trust (220414/Z/20/Z) (https://wellcome.org/). P.L. acknowledges support from the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 725422-ReservoirDOCS), from the European Union’s Horizon 2020 project MOOD (grant agreement no. 874850), and from the Wellcome Trust through project 206298/Z/17/Z. The funders had no role in study design, data collection, and analysis; the decision to publish; or in preparation of the manuscript.

Author contributions: Conceptualization: H.T. and O.G.P. Investigation: B.L., H.T., J.R., and Y.L. Methodology: H.T., O.G.P., P.L., B.L., and S.C.H. Data curation: B.L. and H.T. Validation: B.L., J.R., H.T., S.F., and L.D. Formal analysis: B.L., J.R., H.T., and L.D. Funding acquisition: H.T. Project administration: H.T. Resources: B.L., J.R., and H.T. Software: B.L. and J.R. Supervision: H.T. Visualization: B.L. and H.T. Writing—original draft: B.L., H.T., and O.G.P. Writing—review and editing: B.L., H.T., O.G.P., J.R., N.L., S.C.H., P.L., S.F., Z.W., and L.D. All authors read and approved the manuscript.

Competing interests: The authors declare that they have no competing interests.

Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. The gene segment sequences are available in the GenBank (www.ncbi.nlm.nih.gov/genome/viruses/variation/flu/) and GISAID (platform.gisaid.org/) databases. All original code and data are available in Zenodo (https://doi.org/10.5281/zenodo.14174518) or GitHub (https://github.com/bingyinglee/paper-poultry-vaccination-of-H5-AIV).

Supplementary Materials

This PDF file includes:

Supplementary Information Text

Figs. S1 to S16

Tables S1 to S11

References

sciadv.ado9140_sm.pdf (1.8MB, pdf)

REFERENCES AND NOTES

  • 1.Kuiken T., Cromie R., Protect wildlife from livestock diseases. Science 378, 5 (2022). [DOI] [PubMed] [Google Scholar]
  • 2.Safety E. F., Aznar I., Baldinelli F., Stoicescu A., Kohnle L., Annual report on surveillance for avian influenza in poultry and wild birds in Member States of the European Union in 2021. EFSA J. 20, e07554 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Lebarbenchon C., Feare C. J., Renaud F., Thomas F., Gauthier-Clerc M., Persistence of highly pathogenic avian influenza viruses in natural ecosystems. Emerging Infect. Dis. 16, 1057–1062 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Alexander D. J., An overview of the epidemiology of avian influenza. Vaccine 25, 5637–5644 (2007). [DOI] [PubMed] [Google Scholar]
  • 5.Olsen B., Munster V. J., Wallensten A., Waldenstrom J., Osterhaus A., Fouchier R. A. M., Global patterns of influenza A virus in wild birds. Science 312, 384–388 (2006). [DOI] [PubMed] [Google Scholar]
  • 6.Tian H. Y., Zhou S., Dong L., Van Boeckel T. P., Cui Y. J., Wu Y. R., Cazelles B., Huang S. Q., Yang R. F., Grenfell B. T., Xu B., Avian influenza H5N1 viral and bird migration networks in Asia. Proc. Natl. Acad. Sci. U.S.A. 112, 172–177 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Zhang G. G., Li B. Y., Raghwani J., Vrancken B., Jia R., Hill S. C., Fournie G., Cheng Y. C., Yang Q. Q., Wang Y. X., Wang Z. M., Dong L., Pybus O. G., Tian H. Y., Bidirectional movement of emerging H5N8 avian influenza viruses between Europe and Asia via migratory birds since early 2020. Mol. Biol. Evol. 40, msad019 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Tian H. Y., Cui Y. J., Dong L., Zhou S., Li X. W., Huang S. Q., Yang R. F., Xu B., Spatial, temporal and genetic dynamics of highly pathogenic avian influenza A (H5N1) virus in China. BMC Infect. Dis. 15, 54 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Taubenberger J. K., Kash J. C., Influenza virus evolution, host adaptation, and pandemic formation. Cell Host Microbe 7, 440–451 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Dugan V. G., Chen R., Spiro D. J., Sengamalay N., Zaborsky J., Ghedin E., Nolting J., Swayne D. E., Runstadler J. A., Happ G. M., Senne D. A., Wang R. X., Slemons R. D., Holmes E. C., Taubenberger J. K., The evolutionary genetics and emergence of avian influenza viruses in wild birds. PLOS Pathog. 4, e1000076 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Peiris J. S. M., de Jong M. D., Guan Y., Avian influenza virus (H5N1): A threat to human health. Clin. Microbiol. Rev. 20, 243–267 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Pasick J., Berhane Y., Joseph T., Bowes V., Hisanaga T., Handel K., Alexandersen S., Reassortant highly pathogenic influenza A H5N2 virus containing gene segments related to Eurasian H5N8 in British Columbia, Canada, 2014. Sci. Rep. 5, 9484 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Bertran K., Swayne D. E., Pantin-Jackwood M. J., Kapczynski D. R., Spackman E., Suarez D. L., Lack of chicken adaptation of newly emergent Eurasian H5N8 and reassortant H5N2 high pathogenicity avian influenza viruses in the US is consistent with restricted poultry outbreaks in the Pacific flyway during 2014-2015. Virology 494, 190–197 (2016). [DOI] [PubMed] [Google Scholar]
  • 14.Li X. Y., Cui P. F., Zeng X. Y., Jiang Y. P., Li Y. B., Yang J. X., Pan Y. D., Gao X. X., Zhao C. H., Wang J. H., Wang K., Deng G. H., Guo J., Characterization of avian influenza H5N3 reassortants isolated from migratory waterfowl and domestic ducks in China from 2015 to 2018. Transbound. Emerging. Dis. 66, 2605–2610 (2019). [DOI] [PubMed] [Google Scholar]
  • 15.Yeo S. J., Hoang V. T., Duong T. B., Nguyen N. M., Tuong H. T., Azam M., Sung H. W., Park H., Emergence of a novel reassortant H5N3 avian influenza virus in Korean mallard ducks in 2018. Intervirology 65, 1–16 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Gu M., Liu W., Cao Y., Peng D., Wang X., Wan H., Zhao G., Xu Q., Zhang W., Song Q., Li Y., Liu X., Novel reassortant highly pathogenic avian influenza (H5N5) viruses in domestic ducks, China. Emerg. Infect. Dis. 17, 1060–1063 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Kang Y. M., Cho H. K., Kim H. M., Lee C. H., Kim D. Y., Choi S. H., Lee M. H., Kang H. M., Protection of layers and breeders against homologous or heterologous HPAIv by vaccines from Korean national antigen bank. Sci. Rep. 10, 9436 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Villanueva-Cabezas J. P., Coppo M. J. C., Durr P. A., McVernon J., Vaccine efficacy against Indonesian Highly Pathogenic Avian Influenza H5N1: Systematic review and meta-analysis. Vaccine 35, 4859–4869 (2017). [DOI] [PubMed] [Google Scholar]
  • 19.Hill E. M., House T., Dhingra M. S., Kalpravidh W., Morzaria S., Osmani M. G., Brum E., Yamage M., Kalam M. A., Prosser D. J., Takekawa J. Y., Xiao X. M., Gilbert M., Tildesley M. J., The impact of surveillance and control on highly pathogenic avian influenza outbreaks in poultry in Dhaka division, Bangladesh. PLOS Comput. Biol. 14, e1006439 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hoang H. T. T., Nguyen C. H., Nguyen N. T. T., Pham A. D., Nguyen H. T. T., Le T. H., Tran H. X., Chu H. H., Nguyen N. T., Immunization with the H5N1 recombinant vaccine candidate induces high protection in chickens against Vietnamese highly pathogenic avian influenza virus strains. Vaccines 8, 159 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Zeng X., Tian G., Shi J., Deng G., Li C., Chen H., Vaccination of poultry successfully eliminated human infection with H7N9 virus in China. Sci. China Life Sci. 61, 1465–1473 (2018). [DOI] [PubMed] [Google Scholar]
  • 22.Swayne D. E., Impact of vaccines and vaccination on global control of avian influenza. Avian Dis. 56, 818–828 (2012). [DOI] [PubMed] [Google Scholar]
  • 23.Parvin R., Nooruzzaman M., Kabiraj C. K., Begum J. A., Chowdhury E. H., Islam M. R., Harder T., Controlling avian influenza virus in Bangladesh: Challenges and recommendations. Viruses 12, 751 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Liu S., Zhuang Q. Y., Wang S. C., Jiang W. M., Jin J. H., Peng C., Hou G. Y., Li J. P., Yu J. M., Yu X. H., Liu H. L., Sun S. F., Yuan L. P., Chen J. M., Control of avian influenza in China: Strategies and lessons. Transbound. Emerg. Dis. 67, 1463–1471 (2020). [DOI] [PubMed] [Google Scholar]
  • 25.Sun Z., Wang J., Huang Z., Assessment of China’s H5N1 routine vaccination strategy. Sci. Rep. 7, 46441 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Wu J., Ke C., Lau E. H. Y., Song Y., Cheng K. L., Zou L., Kang M., Song T., Peiris M., Yen H.-L., Influenza H5/H7 virus vaccination in poultry and reduction of zoonotic infections, Guangdong Province, China, 2017–18. Emerg. Infect. Dis. 25, 116–118 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Liu L. L., Zeng X. Y., Chen P. C., Deng G. H., Li Y. B., Shi J. Z., Gu C. Y., Kong H. H., Suzuki Y., Jiang Y. P., Tian G. B., Chen H. L., Characterization of clade 7.2 H5 avian influenza viruses that continue to circulate in chickens in China. J. Virol. 90, 9797–9805 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Song W., Guo J., Ni X., Wu J., Xia W., He F., Wang X., Fan G., Zhou K., Wu Y., Chen S., Chen H., Changes of avian influenza virus subtypes before and after vaccination in live poultry in Nanchang, China from 2016 to 2019. Microbes Infect. 23, 104848 (2021). [DOI] [PubMed] [Google Scholar]
  • 29.Iwami S., Suzuki T., Takeuchi Y., Paradox of vaccination: Is vaccination really effective against avian flu epidemics? PLOS ONE 4, e4915 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Wang Z., Jiang W., Liu S., Hou G., Li J., Wang Z., Chen J., Increased substitution rate in H5N1 avian influenza viruses during mass vaccination of poultry. Chin. Sci. Bull. 57, 2419–2424 (2012). [Google Scholar]
  • 31.Lee C. W., Senne D. A., Suarez D. L., Effect of vaccine use in the evolution of Mexican lineage H5N2 avian influenza virus. J. Virol. 78, 8372–8381 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Cattoli G., Fusaro A., Monne I., Coven F., Joannis T., El-Hamid H. S. A., Hussein A. A., Cornelius C., Amarin N. M., Mancin M., Holmes E. C., Capua I., Evidence for differing evolutionary dynamics of A/H5N1 viruses among countries applying or not applying avian influenza vaccination in poultry. Vaccine 29, 9368–9375 (2011). [DOI] [PubMed] [Google Scholar]
  • 33.Kwon J. H., Lee D. H., Criado M. F., Killmaster L., Ali M. Z., Giasuddin M., Samad M. A., Karim M. R., Hasan M., Brum E., Nasrin T., Swayne D. E., Genetic evolution and transmission dynamics of clade 2.3.2.1a highly pathogenic avian influenza A/H5N1 viruses in Bangladesh. Virus Evol. 6, veaa046 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Ghafari M., du Plessis L., Raghwani J., Bhatt S., Xu B., Pybus O. G., Katzourakis A., Purifying selection determines the short-term time dependency of evolutionary rates in SARS-CoV-2 and pH1N1 influenza. Mol. Biol. Evol. 39, msac009 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Fourment M., Holmes E. C., Avian influenza virus exhibits distinct evolutionary dynamics in wild birds and poultry. BMC Evol. Biol. 15, 120 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Bhatt S., Holmes E. C., Pybus O. G., The genomic rate of molecular adaptation of the human influenza A virus. Mol. Biol. Evol. 28, 2443–2451 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Lemey P., Minin V. N., Bielejec F., Kosakovsky Pond S. L., Suchard M. A., A counting renaissance: Combining stochastic mapping and empirical Bayes to quickly detect amino acid sites under positive selection. Bioinformatics 28, 3248–3256 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Qiu X., Duvvuri V. R., Gubbay J. B., Webby R. J., Kayali G., Bahl J., Lineage-specific epitope profiles for HPAI H5 pre-pandemic vaccine selection and evaluation. Influenza Other Respi. Viruses 11, 445–456 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Qian M., Hu H., Zuo T., Wang G., Zhang L., Zhou P., Unraveling of a neutralization mechanism by two human antibodies against conserved epitopes in the globular head of H5 hemagglutinin. J. Virol. 87, 3571–3577 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Haghighi H. R., Read L. R., Haeryfar S. M., Behboudi S., Sharif S., Identification of a dual-specific T cell epitope of the hemagglutinin antigen of an H5 avian influenza virus in chickens. PLOS ONE 4, e7772 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.King J., Harder T., Conraths F. J., Beer M., Pohlmann A., The genetics of highly pathogenic avian influenza viruses of subtype H5 in Germany, 2006–2020. Transbound. Emerg. Dis. 68, 1136–1150 (2021). [DOI] [PubMed] [Google Scholar]
  • 42.The Global Consortium for H5N8 and Related Influenza Viruses , Role for migratory wild birds in the global spread of avian influenza H5N8. Science 354, 213–217 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Yang Q. Q., Zhao X., Lemey P., Suchard M. A., Bi Y. H., Shi W. F., Liu D., Qi W. B., Zhang G. G., Stenseth N. C., Pybus O. G., Tian H. Y., Assessing the role of live poultry trade in community-structured transmission of avian influenza in China. Proc. Natl. Acad. Sci. U.S.A. 117, 5949–5954 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Caliendo V., Leijten L., van de Bildt M. W. G., Poen M. J., Kok A., Bestebroer T., Richard M., Fouchier R. A. M., Kuiken T., Long-term protective effect of serial infections with H5N8 highly pathogenic avian influenza virus in wild ducks. J. Virol. 96, e0123322 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Koethe S., Ulrich L., Ulrich R., Amler S., Graaf A., Harder T. C., Grund C., Mettenleiter T. C., Conraths F. J., Beer M., Globig A., Modulation of lethal HPAIV H5N8 clade 2.3.4.4B infection in AIV pre-exposed mallards. Emerg. Microbes Infect. 9, 180–193 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Hill S. C., Manvell R. J., Schulenburg B., Shell W., Wikramaratna P. S., Perrins C., Sheldon B. C., Brown I. H., Pybus O. G., Antibody responses to avian influenza viruses in wild birds broaden with age. Proc. R. Soc. London Ser. B 283, 20162159 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Verhagen J. H., Hofle U., van Amerongen G., van de Bildt M., Majoor F., Fouchier R. A. M., Kuiken T., Long-term effect of serial infections with H13 and H16 low-pathogenic avian influenza viruses in black-headed gulls. J. Virol. 89, 11507–11522 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Bird J. P., Martin R., Akcakaya H. R., Gilroy J., Burfield I. J., Garnett S. T., Symes A., Taylor J., Sekercioglu C. H., Butchart S. H. M., Generation lengths of the world’s birds and their implications for extinction risk. Conserv. Biol. 34, 1252–1261 (2020). [DOI] [PubMed] [Google Scholar]
  • 49.Zhang Y., Cui P., Shi J., Chen Y., Zeng X., Jiang Y., Tian G., Li C., Chen H., Kong H., Deng G., Key amino acid residues that determine the antigenic properties of highly pathogenic H5 influenza viruses bearing the clade 2.3.4.4 hemagglutinin gene. Viruses 15, 2249 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Jiang W. M., Dong C. X., Liu S., Peng C., Yin X., Liang S. B., Zhang L., Li J. P., Yu X. H., Li Y., Wang J. J., Hou G. Y., Zeng Z., Liu H. L., Emerging novel reassortant influenza A (H5N6) viruses in poultry and humans, china, 2021. Emerg. Infect. Dis. 28, 1064–1066 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Zhang J., Ye H., Liu Y., Liao M., Qi W., Resurgence of H5N6 avian influenza virus in 2021 poses new threat to public health. Lancet Microbe 3, e558 (2022). [DOI] [PubMed] [Google Scholar]
  • 52.Ciminski K., Chase G., Schwemmle M., Beer M., Advocating a watch-and-prepare approach with avian influenza. Nat. Microbiol. 8, 1603–1605 (2023). [DOI] [PubMed] [Google Scholar]
  • 53.Kozlov M., US will vaccinate birds against avian flu for first time. Nature 618, 220–221 (2023). [DOI] [PubMed] [Google Scholar]
  • 54.Klaassen M., Wille M., The plight and role of wild birds in the current bird flu panzootic. Nat. Ecol. Evol. 7, 1541–1542 (2023). [DOI] [PubMed] [Google Scholar]
  • 55.Wille M., Barr I. G., Resurgence of avian influenza virus. Science 376, 459–460 (2022). [DOI] [PubMed] [Google Scholar]
  • 56.Sitaras I., Rousou X., Kalthoff D., Beer M., Peeters B., de Jong M. C. M., Role of vaccination-induced immunity and antigenic distance in the transmission dynamics of highly pathogenic avian influenza H5N1. J. R. Soc. Interface 13, 20150976 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Walker P. G. T., Cauchemez S., Metras R., Dung D. H., Pfeiffer D., Ghani A. C., A Bayesian approach to quantifying the effects of mass poultry vaccination upon the spatial and temporal dynamics of H5N1 in Northern Vietnam. PLOS Comput. Biol. 6, e1000683 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Cui P. F., Zeng X. Y., Li X. Y., Li Y. B., Shi J. Z., Zhao C. H., Qu Z. Y., Wang Y. W., Guo J., Gu W. L., Ma Q., Zhang Y. C., Lin W. P., Li M. H., Tian J. M., Wang D. X., Xing X., Liu Y. J., Pan S. X., Zhang Y. P., Bao H. M., Liu L. L., Tian G. B., Li C. J., Deng G. H., Chen H. L., Genetic and biological characteristics of the globally circulating H5N8 avian influenza viruses and the protective efficacy offered by the poultry vaccine currently used in China. Sci. China Life Sci. 65, 795–808 (2022). [DOI] [PubMed] [Google Scholar]
  • 59.Poetri O. N., Bouma A., Murtini S., Claassen I., Koch G., Soejoedono R. D., Stegeman J. A., van Boven M., An inactivated H5N2 vaccine reduces transmission of highly pathogenic H5N1 avian influenza virus among native chickens. Vaccine 27, 2864–2869 (2009). [DOI] [PubMed] [Google Scholar]
  • 60.Terregino C., Toffan A., Cilloni F., Monne I., Bertoli E., Castellanos L., Amarin N., Mancin M., Capua I., Evaluation of the protection induced by avian influenza vaccines containing a 1994 Mexican H5N2 LPAI seed strain against a 2008 Egyptian H5N1 HPAI virus belonging to clade 2.2.1 by means of serological and in vivo tests. Avian Pathol. 39, 215–222 (2010). [DOI] [PubMed] [Google Scholar]
  • 61.Ellis T. M., Leung C., Chow M. K. W., Bissett L. A., Wong W., Guan Y., Peiris J. S. M., Vaccination of chickens against H5N1 avian influenza in the face of an outbreak interrupts virus transmission. Avian Pathol. 33, 405–412 (2004). [DOI] [PubMed] [Google Scholar]
  • 62.Magalhaes R. J. S., Pfeiffer D. U., Otte J., Evaluating the control of HPAIV H5N1 in Vietnam: Virus transmission within infected flocks reported before and after vaccination. BMC Vet. Res. 6, 31 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Islam A., Munro S., Hassan M. M., Epstein J. H., Klaassen M., The role of vaccination and environmental factors on outbreaks of high pathogenicity avian influenza H5N1 in Bangladesh. One Health 17, 100655 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64.Poetri O., Bouma A., Claassen I., Koch G., Soejoedono R., Stegeman A., van Boven M., A single vaccination of commercial broilers does not reduce transmission of H5N1 highly pathogenic avian influenza. Vet. Res. 42, 74 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Read A. F., Baigent S. J., Powers C., Kgosana L. B., Blackwell L., Smith L. P., Kennedy D. A., Walkden-Brown S. W., Nair V. K., Imperfect vaccination can enhance the transmission of highly virulent pathogens. PLOS Biol. 13, e1002198 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Gandon S., Mackinnon M. J., Nee S., Read A. F., Imperfect vaccines and the evolution of pathogen virulence. Nature 414, 751–756 (2001). [DOI] [PubMed] [Google Scholar]
  • 67.Grenfell B. T., Pybus O. G., Gog J. R., Wood J. L. N., Daly J. M., Mumford J. A., Holmes E. C., Unifying the epidemiological and evolutionary dynamics of pathogens. Science 303, 327–332 (2004). [DOI] [PubMed] [Google Scholar]
  • 68.Raghwani J., Bhatt S., Pybus O. G., Faster adaptation in smaller populations: Counterintuitive evolution of HIV during childhood infection. PLOS Comput. Biol. 12, e1004694 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Guo J., Song W. T., Ni X. S., Liu W., Wu J. W., Xia W., Zhou X. F., Wang W., He F. L., Wang X., Fan G. Y., Zhou K., Chen H. Y., Chen S. G., Pathogen change of avian influenza virus in the live poultry market before and after vaccination of poultry in southern China. Virol. J. 18, 213 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Cheng W., Chong K. C., Lau T., Wang X. X., Yu Z., Liu S. L., Wang M., Pan J. R., Chen E. F., Comparison of avian influenza virus contamination in the environment before and after massive poultry H5/H7 vaccination in Zhejiang Province, China. Open Forum Infect. Dis. 6, ofz197 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Katoh K., Standley D. M., MAFFT multiple sequence alignment software Version 7: Improvements in performance and usability. Mol. Biol. Evol. 30, 772–780 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Baker B. J., De Anda V., Seitz K. W., Dombrowski N., Santoro A. E., Lloyd K. G., Diversity, ecology and evolution of Archaea. Nat. Microbiol. 5, 887–900 (2020). [DOI] [PubMed] [Google Scholar]
  • 73.Martin D. P., Murrell B., Golden M., Khoosal A., Muhire B., RDP4: Detection and analysis of recombination patterns in virus genomes. Virus Evol. 1, vev003 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74.Kiwelu I. E., Novitsky V., Margolin L., Baca J., Manongi R., Sam N., Shao J., McLane M. F., Kapiga S. H., Essex M., Frequent intra-subtype recombination among HIV-1 circulating in Tanzania. PLOS ONE 8, e71131 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 75.Koh E. Y., Ong J., Wang Y. F., Toh X., Fernandez C. J., Huangfu T. Q., Hall R. N., Toh S., Lim K., Sng W., Lim H. P., Ho K. L., Chang S. F., Yap H. H., Rabbit haemorrhagic disease virus 2 from Singapore 2020 outbreak revealed an Australian recombinant variant. Virus Evol. 9, vead029 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 76.Price M. N., Dehal P. S., Arkin A. P., FastTree 2—Approximately maximum-likelihood trees for large alignments. PLOS ONE 5, e9490 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.Rambaut A., Lam T. T., Carvalho L. M., Pybus O. G., Exploring the temporal structure of heterochronous sequences using TempEst (formerly Path-O-Gen). Virus Evol. 2, vew007 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 78.Kalkauskas A., Perron U., Sun Y. X., Goldman N., Baele G., Guindon S., De Maio N., Sampling bias and model choice in continuous phylogeography: Getting lost on a random walk. PLOS Comput. Biol. 17, e1008561 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 79.Lemey P., Hong S. L., Hill V., Baele G., Poletto C., Colizza V., O’Toole A., McCrone J. T., Andersen K. G., Worobey M., Nelson M. I., Rambaut A., Suchard M. A., Accommodating individual travel history and unsampled diversity in Bayesian phylogeographic inference of SARS-CoV-2. Nat. Commun. 11, 5110 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 80.Escalera-Zamudio M., Golden M., Gutiérrez B., Thézé J., Keown J. R., Carrique L., Bowden T. A., Pybus O. G., Parallel evolution in the emergence of highly pathogenic avian influenza A viruses. Nat. Commun. 11, 5511 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 81.He W.-T., Wang L., Zhao Y., Wang N., Li G., Veit M., Bi Y., Gao G. F., Su S., Adaption and parallel evolution of human-isolated H5 avian influenza viruses. J. Infect. 80, 630–638 (2020). [DOI] [PubMed] [Google Scholar]
  • 82.Suchard M. A., Lemey P., Baele G., Ayres D. L., Drummond A. J., Rambaut A., Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10. Virus Evol. 4, vey016 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 83.Ayres D. L., Darling A., Zwickl D. J., Beerli P., Holder M. T., Lewis P. O., Huelsenbeck J. P., Ronquist F., Swofford D. L., Cummings M. P., Rambaut A., Suchard M. A., BEAGLE: An application programming interface and high-performance computing library for statistical phylogenetics. Syst. Biol. 61, 170–173 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 84.Hill V., Baele G., Bayesian estimation of past population dynamics in BEAST 1.10 using the Skygrid coalescent model. Mol. Biol. Evol. 36, 2620–2628 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 85.Lemey P., Rambaut A., Bedford T., Faria N., Bielejec F., Baele G., Russell C. A., Smith D. J., Pybus O. G., Brockmann D., Suchard M. A., Unifying viral genetics and human transportation data to predict the global transmission dynamics of human influenza H3N2. PLOS Pathog. 10, e1003932 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 86.Edwards C. J., Suchard M. A., Lemey P., Welch J. J., Barnes I., Fulton T. L., Barnett R., O’Connell T. C., Coxon P., Monaghan N., Valdiosera C. E., Lorenzen E. D., Willerslev E., Baryshnikov G. F., Rambaut A., Thomas M. G., Bradley D. G., Shapiro B., Ancient hybridization and an Irish origin for the modern polar bear matriline. Curr. Biol. 21, 1251–1258 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 87.Dudas G., Carvalho L. M., Rambaut A., Bedford T., MERS-CoV spillover at the camel-human interface. eLife 7, e31257 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 88.Yu G. C., Smith D. K., Zhu H. C., Guan Y., Lam T. T. Y., GGTREE: An R package for visualization and annotation of phylogenetic trees with their covariates and other associated data. Methods Ecol. Evol. 8, 28–36 (2017). [Google Scholar]
  • 89.Nguyen P. T., Kant R., van den Broeck F., Suvanto M. T., Alburkat H., Virtanen J., Ahvenainen E., Castren R., Hong S. L., Baele G., Ahava M. J., Jarva H., Jokiranta S. T., Kallio-Kokko H., Kekaelaeinen E., Kirjavainen V., Kortela E., Kurkela S., Lappalainen M., Liimatainen H., Suchard M. A., Hannula S., Ellonen P., Sironen T., Lemey P., Vapalahti O., Smura T., The phylodynamics of SARS-CoV-2 during 2020 in Finland. Commun. Med. 2, 65 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 90.Parker J., Rambaut A., Pybus O. G., Correlating viral phenotypes with phylogeny: Accounting for phylogenetic uncertainty. Infect. Genet. Evol. 8, 239–246 (2008). [DOI] [PubMed] [Google Scholar]
  • 91.Murrell B., Moola S., Mabona A., Weighill T., Sheward D., Kosakovsky Pond S. L., Scheffler K., FUBAR: A fast, unconstrained bayesian approximation for inferring selection. Mol. Biol. Evol. 30, 1196–1205 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 92.Kosakovsky Pond S. L., Frost S. D., Not so different after all: A comparison of methods for detecting amino acid sites under selection. Mol. Biol. Evol. 22, 1208–1222 (2005). [DOI] [PubMed] [Google Scholar]
  • 93.Pond S. L. K., Poon A. F. Y., Velazquez R., Weaver S., Hepler N. L., Murrell B., Shank S. D., Magalis B. R., Bouvier D., Nekrutenko A., Wisotsky S., Spielman S. J., Frost S. D. W., Muse S. V., HyPhy 2.5—A customizable platform for evolutionary hypothesis testing using phylogenies. Mol. Biol. Evol. 37, 295–299 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 94.Kistler K. E., Bedford T., Evidence for adaptive evolution in the receptor-binding domain of seasonal coronaviruses OC43 and 229e. eLife 10, e64509 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 95.Ye H., Deyle E. R., Gilarranz L. J., Sugihara G., Distinguishing time-delayed causal interactions using convergent cross mapping. Sci. Rep. 5, 14750 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 96.Gilbert M., Nicolas G., Cinardi G., Van Boeckel T. P., Vanwambeke S. O., Wint G., Robinson T. P., Global distribution data for cattle, buffaloes, horses, sheep, goats, pigs, chickens and ducks in 2010. PLOS Comput. Biol. 5, 1–11 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 97.Shi J., Zeng X., Cui P., Yan C., Chen H., Alarming situation of emerging H5 and H7 avian influenza and effective control strategies. Emerg. Microbes Infect. 12, 2155072 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 98.Zeng X., Tian G., Chen H., Progress in development and application of H5/H7 avian influenza vaccines in China. Sci. Sin. 53, 1700–1712 (2023). [Google Scholar]
  • 99.Tran C. C., Yanagida J. F., Saksena S., Fox J., An alternative vaccination approach for the prevention of highly pathogenic avian influenza subtype H5N1 in the Red River Delta, Vietnam—A geospatial-based cost-effectiveness analysis. Vet. Sci. 3, 6 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 100.D. H. Chu, in Vietnam Country Report on Vaccination for Avian Influenza (AI) (World Organisation for Animal Health, 2017).
  • 101.Tarigan S., Wibowo M. H., Indriani R., Sumarningsih S., Artanto S., Idris S., Durr P. A., Asmara W., Ebrahimie E., Stevenson M. A., Ignjatovic J., Field effectiveness of highly pathogenic avian influenza H5N1 vaccination in commercial layers in Indonesia. PLOS ONE 13, e0190947 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 102.Swayne D. E., Suarez D. L., Spackman E., Jadhao S., Dauphin G., Kim-Torchetti M., McGrane J., Weaver J., Daniels P., Wong F., Selleck P., Wiyono A., Indriani R., Yupiana Y., Siregar E. S., Prajitno T., Smith D., Fouchier R., Antibody titer has positive predictive value for vaccine protection against challenge with natural antigenic-drift variants of H5N1 high-pathogenicity avian influenza viruses from indonesia. J. Virol. 89, 3746–3762 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 103.Yang Q., Wang B., Lemey P., Dong L., Mu T., Wiebe R. A., Guo F., Trovao N. S., Park S. W., Lewis N., Tsui J. L. H., Bajaj S., Cheng Y., Yang L., Haba Y., Li B., Zhang G., Pybus O. G., Tian H., Grenfell B., Synchrony of bird migration with global dispersal of avian influenza reveals exposed bird orders. Nat. Commun. 15, 1126 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information Text

Figs. S1 to S16

Tables S1 to S11

References

sciadv.ado9140_sm.pdf (1.8MB, pdf)

Articles from Science Advances are provided here courtesy of American Association for the Advancement of Science

RESOURCES