Characterization and pathogenicity of a novel triple-reassortant H6N6 avian influenza virus associated with reproductive failure in breeding ducks Authors: Puduo Zhu, Liguo Gao, Zhihong Luo, Jiajia Liu, Xudong He, Qi Zhou, Qunhui Li, Zhuanqiang Yan, Wencheng Lin, Feng Chen Published: 2026 ABSTRACT AbstractH6N6 avian influenza virus (AIV) has emerged as a predominant subtype in recent years, with frequent isolations reported in China, Vietnam, and South Korea. As critical reservoirs, waterfowl play a pivotal role in the maintenance and transmission of H6N6 AIV. In this study, serological surveillance and virus isolation were conducted on affected breeding duck farms resulting in the successful isolation of a novel H6N6 strain. Phylogenetic analysis revealed that this isolate is a triple-reassortant virus, with its genome originating from three distinct genetic clades. Notably, key amino acid substitutions A138S and A222V were identified in the hemagglutinin (HA) protein. In vivo challenge experiments demonstrated that the virus could infect multiple species, including mice, specific-pathogen-free (SPF) chickens, and young Muscovy ducks, inducing mild clinical lesions. Crucially, infection in breeding ducks led to significant reproductive tract damage and a decline in egg production. These findings underscore the potential of low-pathogenic H6N6 AIV to cross species barriers and impair the productivity of breeding flocks, highlighting the urgent need for sustained epidemiological surveillance.Keywords: H6N6 avian influenza virus, Reproductive failure, Breeding ducks, Pathogenicity Introduction Avian influenza (AI) is a highly contagious disease in poultry caused by influenza A viruses (AIVs). These viruses exhibit an extensive host range, infecting various avian species and several mammals, including humans, pigs, horses, and dogs. Transmission primarily occurs through respiratory droplets, direct contact with secretions, or the fecal-oral route. AIVs maintain a global distribution, with wild waterfowl serving as the primary natural reservoirs that facilitate the viral persistence and evolution (Huang et al., 2010). Clinical manifestations in infected birds are highly variable, ranging from asymptomatic cases to respiratory distress-including coughing and dyspnea-as well as diarrhea, reduced egg production, and weight loss. In cases of highly pathogenic avian influenza (HPAI), severe signs such as cyanosis of the combs and wattles and facial edema are commonly observed. Historically, the first AIV isolation from turkeys was documented in 1901, following the earlier identification of the disease in Italy in 1878 (Perroncito, 1878; Schäfer, 1955). Furthermore, four major influenza pandemics have emerged since the 20th century: the 1918 H1N1, 1957 H2N2, 1968 H3N2, and the 2009 H1N1 pandemics (Paules and Subbarao, 2017), each resulting in substantial global morbidity and mortality. Molecular evolution, particularly mutations in the hemagglutinin (HA) gene, can significantly alter viral pathogenicity or host tropism. For instance, substitutions such as A222V, E190Q/G, Q226L, and G228S (H3 numbering) have been shown to shift viral receptor specificity, increasing affinity for human-type α-2,6-linked sialic acid receptors and thereby facilitating interspecies transmission (Bialy et al., 2024; Chen et al., 2005; Kim and Park, 2012). The H6 subtype AIVs are characterized by high genomic plasticity, frequently undergoing reassortment with other subtypes including H5, H7, H9, and H3 (Yan et al., 2023). Notable reassortment events involving H6N1, H5N1, and H9N2 were documented in Hong Kong in 1997 (Chin et al., 2002). Since then, H6N6 viruses of swine origin were identified in southern China in 2011 (Zhang et al., 2011), and human H6N1 infections were reported in Taiwan in 2013. Furthermore, H6N6 strains have been isolated from companion animals such as dogs (Lin et al., 2015). Beyond HA mutations, internal gene exchanges also play a critical role in virulence; for example, the PA-I38M and PB2-E627K mutations are known to enhance the pathogenicity of H6N6 AIVs in mammalian models (Bussey et al., 2010; Lin et al., 2015; Tan et al., 2014; Zhu et al., 2015). In poultry populations, H6N6 AIV infection is often subclinical or manifests as mild respiratory disease, leading to frequent underdiagnosis in field conditions. However, in breeding ducks, the virus can cause significant reproductive tract impairment and a marked decline in egg production (Armin R W Elbers, 2021). As one of the most prevalent subtypes worldwide, H6 AIVs have circulated extensively in southern China over the past two decades, exhibiting high host diversity (Xu et al., 2023). Between 2009 and 2021, dynamic reassortment between H6 and co-circulating subtypes has generated novel genotypes (Lin et al., 2020; Yoon et al., 2014). Given that some H6N6 viruses can infect and even cause lethality in mice without prior adaptation, continuous epidemiological surveillance is imperative to monitor their zoonotic potential. Materials and methods Serological survey in duck flocks Blood samples (approximately 1 mL each) were collected from the wing vein using sterile syringes. The samples were allowed to clot at room temperature for 1 h and then stored at 4 °C overnight to facilitate serum separation. Following complete separation from the cellular fraction, the serum was harvested and aliquoted into 8-tube strips, categorized by flock. In October 2021, a total of 1455 serum samples were collected from ducks of various age groups across four breeding duck farms in eastern and southern China. Hemagglutination inhibition (HI) assays were conducted in accordance with the national diagnostic standards for avian influenza. Hemagglutination inhibition assays were performed using standardized antigens for H3, H6, and H9 subtypes to evaluate exposure to multiple AIV subtypes. Standardized antigens for H3, H9, and H6 AIV were adjusted to a titer of 4 hemagglutinating units (HAU). Serum samples underwent twofold serial dilutions in 96-well V-bottom microplates (wells 1 to 10). Subsequently, 25 μL of the 4 HAU antigen was added to wells 1 through 11, with well 11 serving as the antigen control and well 12 as the blank (negative) control. After the addition of a 1% (v/v) chicken erythrocyte suspension, the plates were incubated at 37 °C for 25 min. The HI titers were then determined by visually assessing the hemagglutination patterns, with the highest dilution inhibiting complete hemagglutination recorded as the titer. Virus isolation Oropharyngeal and cloacal swabs were collected from Muscovy ducks on the sampled farms using sterile swabs. A total of 529 samples were obtained and immersed in phosphate-buffered saline (PBS) supplemented with 2000 U/mL penicillin and streptomycin. Following vortexing and centrifugation, the resulting supernatant was inoculated into the allantoic cavity of 9- to 11-day-old embryonated chicken eggs (ECEs). The eggs were incubated at 37 °C for 72 h, after which the allantoic fluid was harvested and screened for viral activity using hemagglutination (HA) assays. Samples exhibiting HA titers ≥ 2⁴ were subjected to sequencing for identification. Successfully identified isolates were subsequently stored at −80 °C for further analysis. Phylogenetic analysis Viral RNA was extracted from HA-positive allantoic fluid using the DNA/RNA Extraction Kit (Bioer Technology, Hangzhou, China) according to the manufacturer’s instructions. The full-length segments of the viral genome were amplified via RT-PCR using eight pairs of specific primers and the PrimeScript One Step RT-PCR Kit (TaKaRa, Beijing, China). The resulting PCR products were purified and ligated into the pMD19-T vector (TaKaRa). Positive clones were selected and sequenced by Sangon Biotech Co., Ltd. (Shanghai, China). Nucleotide sequences were assembled and manually edited using the SeqMan module of the DNAStar Lasergene 11.0 software package. Phylogenetic trees for all eight genomic segments were constructed using the neighbor-joining method with 1000 bootstrap replicates in MEGA software (version 7.0. Reference sequences were retrieved from the GISAID and NCBI databases to analyze the genetic relationships and evolutionary origins of the isolate. Receptor binding assays To assess receptor binding specificity, 100 μL of 10% goose red blood cells (GRBCs) was incubated with 1.25 U of α2,3-sialidase (TaKaRa, Kusatsu, Japan) at 37 °C for 1 h. The treated GRBCs were washed twice, adjusted to 0.5% in PBS, and mixed with serially diluted virus (50 μL) in a 96-well plate. After 15 min at room temperature, hemagglutination titers were determined. A(H1N1)pdm2009 (A/California/04/2009, CA/04) and H5N1 (A/mallard/Huadong/S/2005, HD/05) served as controls. For solid-phase direct binding, receptor analogs 3′SLN-PAA-biotin and Neu5Acα2-6GalNAcα-sp3-PAA-biotin (Glyconz Tech) were washed with PBS, added equally to a streptavidin-coated 96-well plate, and blocked overnight with 5% BSA at 4 °C. After washing with 0.05% PBST, 128 HA units of virus and AIV universal monoclonal antibody were added per well. Binding was detected using HRP-conjugated mouse IgG secondary antibody and TMB substrate. The reaction was stopped with 2 M H₂SO₄, and absorbance was read at 450 nm. Each sample was tested in triplicate. Animal experiments Mice challenge Twenty-two 6-week-old female BALB/c mice (Guangdong Medical Laboratory Animal Center, China) were randomly allocated into two groups. After light anesthesia with Zoletil 50, mice in the challenge group were inoculated intranasally with 106 EID50 of the virus isolate in a 50 µL volume, while control mice received an equal volume of sterile PBS. At 3 and 5 days post-inoculation (dpi), three mice per group were euthanized to harvest the spleen, lung, kidney, brain, turbinate, and trachea. Viral titers in these tissues were determined in SPF embryonated chicken eggs (ECEs) and calculated using the Reed-Muench method. The remaining five mice in each group were monitored daily for 14 days to record body weight changes and survival rates. SPF chicken challenge Four-week-old SPF chickens were selected as a standard age group to allow direct comparison with previous H6 AIV pathogenicity studies (Lin et al., 2020; Xu et al., 2023). Sixteen 4-week-old SPF chickens (Wens Da hua nong Biotechnology Co., Ltd., Guangdong, China) were randomly assigned to two groups (n = 8 per group). Prior to the challenge, all chickens were confirmed to be seronegative for AIV antibodies by hemagglutination inhibition (HI) assays. Each chicken was inoculated intranasally with 106 EID50 of the virus in a 100 µL volume. Oropharyngeal and cloacal swabs were collected at 2, 4, and 6 dpi. At 5 dpi, three chickens per group were necropsied, and tissues (heart, liver, spleen, lung, kidney, duodenum, and trachea) were collected for virus titration in ECEs. Young Muscovy ducks challenge Twenty 14-day-old healthy Muscovy ducks (Wens Food Group Co., Ltd., Guangdong, China) were randomly divided into two groups (n = 10 per group) and housed in negative-pressure isolators. Ducks were inoculated with 106 EID50​of the virus (200 µL) via the ocular-nasal route. Oropharyngeal and cloacal swabs were collected at 1, 3, 5, and 7 dpi. At 3 and 5 dpi, three ducks per group were necropsied to harvest tissues (heart, liver, spleen, lung, kidney, duodenum, and trachea) for virus titration. Breeding Muscovy ducks (reproductive performance study) To evaluate the impact of H6N6 AIV on reproductive performance, 80 30-week-old breeding Muscovy ducks were divided into two groups (n = 40 per group) and housed in separate isolation rooms. Ducks were inoculated with 106 EID50 of the virus (200 µL) via combined ocular and intranasal routes. Egg production was recorded daily for 54 days. Oropharyngeal and cloacal swabs were collected from 10 ducks per group at 3, 5, and 7 dpi for virus titration. At 10 dpi, three ducks per group were necropsied to evaluate pathological lesions in the ovaries and oviducts. Statistical analysis Data were analyzed using GraphPad Prism 8.0. Differences between groups were assessed by Student's t-test. A P-value < 0.05 was considered statistically significant. GenBank accession numbers The nucleotide sequences of the H6N6 AIV isolates obtained in this study have been deposited in GenBank (Table 1). Table 1Gene number.Table 1 dummy alt textGene nameGene NumberHAOQ842388NAOQ852021PAOQ861202PB1OQ862274PB2OQ861231NPOQ861139NSOQ861180MOQ861119∗Gene accession numbers and the highest homology strains of eight genes of thevirus. Gene number. ∗Gene accession numbers and the highest homology strains of eight genes of thevirus. Results Serological survey Hemagglutination inhibition (HI) assays revealed that H6 AIV antibody titers were significantly higher than those of H9 and H3 across all four surveyed breeding duck farms (HY, DP, ZS, and JL; P < 0.05). Specifically, farm HY exhibited the highest H6 antibody titers, followed by DP and ZS, while farm JL recorded the lowest (Fig. 1A). Dynamic monitoring of breeding Muscovy ducks from 30 to 47 weeks of age demonstrated that H6 antibody titers initially declined, reaching a nadir at 39 weeks, before subsequently increasing. This decline at week 39 indicates a critical immune gap period (Fig. 1B). Aggregated analysis further confirmed that H6 antibody titers were significantly more robust than those for H3 and H9, establishing H6 AIV as the predominant circulating subtype within the investigated breeding duck farms (Fig. 1C). Determination of hemagglutination titers. Hylogenetic analysis and molecular characterization From the affected farm, only one H6N6 isolate (designated DK/AH/21HY) was obtained. No other AIV subtypes were detected. The complete genome of the isolate A/duck/Anhui/21HY/2022 (designated as DK/AH/21HY) was sequenced and analyzed. Phylogenetic analysis revealed that DK/AH/21HY is a reassortant virus of the Eurasian lineage. Specifically, the HA, M, and NP genes were clustered within the ST2853-like clade, whereas the NA gene was affiliated with the ST192 clade. The remaining internal genes (PB2, PB1, PA, and NS) belonged to the ST339-like clade, indicating a complex triple-reassortant origin (Fig. 2A and 2B) (Figs. 3A-H). Genotype of the H6 subtype avian influenza virus. Phylogenetic analysis of the viral strain. Molecular characterization identified two critical amino acid substitutions, A138S and A222V (H3 numbering), within the HA receptor-binding domain (RBD) (Table 2). Sequence identity analysis demonstrated that all eight genomic segments shared high nucleotide similarity (98.29% to 99.76%) with recently circulating AIV strains in China. Notably, the HA, NA, and PB1 segments exhibited >98.82% identity with the A/chicken/Zhejiang/49/2021 (H6N6) strain, while the NP, NS, and PA segments shared >99% identity with A/duck/Anhui/SHT230213-1/2023 (H6). Furthermore, the M and PB2 genes showed high homology with recent isolates from Fujian and Jiangxi provinces, respectively (Figs. 3A–H). Table 2Key molecular marker of the isolated H6N6 virus.Table 2 dummy alt textViral proteinAmino AcidDK/AH/21HYPhenotypeReferenceHACleavage Sites 340-348PQIETR↓GLFTypical characteristics of low pathogenic avian influenza(Huang et al., 2010; Jallow et al., 2023)A138SSIncreased preference for human receptors(Jiao et al., 2008)E190QEDetermination of binding ability to mouse and human lung epithelial cells(Quan et al., 2019)A222VVIncreased preference for human receptorsQ226-G228QRGNAStalk Region DeletionNOEnhancement of mammalian virulence(Miao et al., 2022)E119VESensitive to the neuroaminidase inhibitor (Oseltamivir)(Munster et al., 2007)H275THPB2E627KEEnhancement of polymerase activity, thereby increasing the virulence and transmission of the virus in mammals(Bussey et al., 2010; Oh et al., 2023; Zhu et al., 2015)D701NDT271ATM2S31NSM2 channel blocker antiviral drugs(such as amantadine)develop resistance(Zu et al., 2009)NSD92EDEnhance (Jiao et al., 2008; Tan et al., 2014) ion of host antiviral cellular factors(Jiao et al., 2008)P42SPPAI38MIEnhancement of virulence in mice(Tan et al., 2014)∗H3 numbering. Bold indicate smutations at amino acid sites. Key molecular marker of the isolated H6N6 virus. ∗H3 numbering. Bold indicate smutations at amino acid sites. Receptor-binding specificity The receptor-binding preference of the DK/AH/21HY strain was evaluated using modified hemagglutination (HA) assays (Table 3). The isolate exhibited hemagglutinating activity toward both untreated and α−2,3-sialidase-treated goose erythrocytes, suggesting a dual affinity for both avian-type (α−2,3-SA) and human-type (α−2,6-SA) sialic acid receptors. This dual-binding profile was further validated and quantified by solid-phase binding ELISA (Fig. 4A–C). Table 3Hemagglutination assays of H6N6 subtype viruses using goose red blood cells with treatment.Table 3 dummy alt textVirus stainHA titers (log2)Untreated GRBCsTreated GRBCsCA/0466H5/Re-57-DK/AH/21HY77“-”:No blood clotting. Hemagglutination assays of H6N6 subtype viruses using goose red blood cells with treatment. “-”:No blood clotting. Solid-phase receptor-binding assay of DK/AH/21HY (4A), human isolate A/California/04/2009 (4B), and poultry isolate A/mallard/Huadong/S/2005 (4C). Pathogenicity in mice Following viral challenge, the clinical status and body weight of mice were monitored daily. At 2 days post-inoculation (dpi), a transient decline in body weight to 90% of the initial mass was observed, which was likely associated with anesthesia-induced stress rather than viral pathogenesis. Although weight slightly recovered by 4 dpi, a second reduction to 96% of the initial weight occurred at 5 dpi, followed by a gradual and steady recovery (Fig. 5A). No mortality was recorded throughout the 14-day experimental period (Fig. 5B), indicating that the DK/AH/21HY strain exhibits low virulence in the BALB/c mouse model. To evaluate tissue tropism, viral titers were quantified in aseptically collected organs. The results demonstrated that viral loads were predominantly localized in the nasal turbinates and lungs, confirming that the virus primarily targets the respiratory system in mice (Table 4) (Fig. 3c and d). Pathogenicity of DK/AH/21HY in BALB/c mice. Mice were challenged with 106 EID50 of virus. Body weight (5A), survival (5B), and egg production rate in breeding ducks were monitored for 14 d post-infection. Table 4Replication of H6N6 viruses in mice.Table 4 dummy alt textVirusViral titers in organs (Log10 EID50/mL, mean±SD)Days post infectionKidneySpleenBrainturbinateLungTracheaMock3dpi//////5dpi//////DK/AH/21HY3dpi///3.2 ± 0.352.83 ± 0.850.72 ± 1.255dpi///3.05 ± 0.583.59 ± 1.32.07 ± 1.27“/”:The virus was not detected in the undiluted sample; the virus titers were shown as mean±SEM. Replication of H6N6 viruses in mice. “/”:The virus was not detected in the undiluted sample; the virus titers were shown as mean±SEM. Infection dynamics in SPF chickens To evaluate the replication and shedding kinetics of the isolate, oropharyngeal and cloacal swabs were collected at multiple time points post-inoculation (dpi). Viral shedding was found to peak at 2 dpi. Although the virus remained detectable in both oropharyngeal and cloacal swabs at 4 dpi, it was no longer recoverable by 6 dpi (Table 5). Tissue distribution analysis conducted at 5 dpi showed that the virus was isolated from only a single sample each of the spleen and duodenum, while all other organs (heart, liver, lung, kidney, and trachea) tested negative. These results suggest that the DK/AH/21HY strain exhibits low infectivity and restricted replication capacity in SPF chickens (Table 6). Table 5Virus titers at 2,4 and 6 dpi in chickens.Table 5 dummy alt textVirusMeasure the number of positive swabs/total number of collected swabs2dpi4dpi6dpiCloacalOropharyngealCloacalOropharyngealCloacalOropharyngealMock//////DK/AH/21HY2.5(1/3)1(1/3)1(1/3)///SPF chickens were intranasally inoculated with 106EID50 virus, and oropharyngeal and cloacal swabs were collected on the indicated days. The virus titers were shown as mean±SEM, Abbreviations:dpi,daypostinfection.“/”:The virus was not detected in the undiluted sample. Virus titers at 2,4 and 6 dpi in chickens. SPF chickens were intranasally inoculated with 106EID50 virus, and oropharyngeal and cloacal swabs were collected on the indicated days. The virus titers were shown as mean±SEM, Abbreviations:dpi,daypostinfection.“/”:The virus was not detected in the undiluted sample. Table 6Tissue distribution of the H6N6 virus in chickens at 5 dpi.Table 6 dummy alt textVirusViral titers in organs (Log10 EID50/mL, mean±SD)HeartLiverSpleenLungKidneyDuodenumTracheaMock//////DK/AH/21HY//1.3(1/3)//2.4(1/3)2.85 ± 2.48“/”:The virus was not detected in the undiluted sample; the virus titers were shown as mean±SEM. Tissue distribution of the H6N6 virus in chickens at 5 dpi. “/”:The virus was not detected in the undiluted sample; the virus titers were shown as mean±SEM. Infection dynamics in young Muscovy ducks In young Muscovy ducks, oropharyngeal viral shedding remained relatively low from 1 to 3 days post-inoculation (dpi), exhibiting a gradual upward trend before falling below the limit of detection from 5 dpi onward. In contrast, viral titers in cloacal swabs peaked early at 1 dpi, followed by a progressive decline through 7 dpi (Table 7). Systemic replication was markedly restricted, with low viral titers detected exclusively in the spleen, while other harvested tissues remained negative (Table 8). Table 7Virus titers at 1,3,5, and 7 dpi in young ducks.Table 7 dummy alt textVirusVirus titers at1,3,5, and 7dpi in ducks(Log10 EID50/mL, mean±SD)1dpi3dpi5dpi7dpiCloacalOropharyngealCloacalOropharyngealCloacalOropharyngealCloacalOropharyngealMock////////DK/AH/21HY2.5 ± 0.252(1/4)2.4 ± 1.041.1 ± 0.131.92 ± 0.44/1.3 ± 0.34/Young ducks were intranasally inoculated with 106EID50 virus, and oropharyngeal and cloacal swabs were collected on the indicated days. The virus titers were shown as mean±SEM, Abbreviations:dpi,day post infection.“/”:The virus was not detected in the undiluted sample. Virus titers at 1,3,5, and 7 dpi in young ducks. Young ducks were intranasally inoculated with 106EID50 virus, and oropharyngeal and cloacal swabs were collected on the indicated days. The virus titers were shown as mean±SEM, Abbreviations:dpi,day post infection.“/”:The virus was not detected in the undiluted sample. Table 8Tissue distribution of H6N6 virus in ducks at 3 and 5 dpi.Table 8 dummy alt textVirusViral titers in organs(Log10 EID50/mL, mean±SD)Days post infectionHeartLiverSpleenLungKidneyduodenumTracheaMock3dpi///////5dpi///////DK/AH/21HY3dpi//0.92 ± 0.53///2.17 ± 0.355dpi//1.5(1/3)///1.44 ± 0.23“/”:The virus was not detected in the undiluted sample; the virus titers were shown as mean±SEM. Tissue distribution of H6N6 virus in ducks at 3 and 5 dpi. “/”:The virus was not detected in the undiluted sample; the virus titers were shown as mean±SEM. Impact on the reproductive performance of breeding Muscovy ducks To evaluate the impact of H6N6 AIV on the productivity of breeding Muscovy ducks, an experimental infection model was utilized. Viral excretion peaked at 3 days post-inoculation (dpi), at which point 100% of the challenged ducks were positive for oropharyngeal shedding, and 80% were positive for cloacal shedding. The proportion of birds shedding the virus declined to 40% by 5 dpi, and no virus was recoverable from either site by 7 dpi. These results demonstrate that breeding Muscovy ducks are highly susceptible to H6N6 AIV infection, which is characterized by efficient viral replication and a shedding duration of at least five days (Table 9). Table 9Tissue distribution of H6N6 virus in breeding ducks at 3,5 and 7 dpi.Table 9 dummy alt textVirusVirus shedding detection in breeding ducks via swabs3dpi5dpi7dpiCloacalOropharyngealCloacalOropharyngealCloacalOropharyngealMock//////DK/AH/21HY10/108/104/104/10//Breeding ducks were intranasally inoculated with 106EID50 virus, and oropharyngeal and cloacal swabs were collected on the indicated days. The virus titers were shown as mean±SEM, Abbreviations:dpi,daypostinfection.“/”:The virus was not detected in the undiluted sample. Tissue distribution of H6N6 virus in breeding ducks at 3,5 and 7 dpi. Breeding ducks were intranasally inoculated with 106EID50 virus, and oropharyngeal and cloacal swabs were collected on the indicated days. The virus titers were shown as mean±SEM, Abbreviations:dpi,daypostinfection.“/”:The virus was not detected in the undiluted sample. Daily egg production was monitored over a 54-day period to evaluate the long-term impact of H6N6 AIV infection. In the challenged group, egg production began to decline as early as 2 days post-inoculation (dpi). The maximum discrepancy in the lay rate between the challenged and control groups reached 24.15% at 37 dpi. While the control group exhibited a standard physiological production curve—characterized by a steady increase to peak followed by a gradual decline—the egg production rate of the challenged group remained persistently lower throughout the study, during the 54-day observation period, egg production in the challenged group did not return to control levels(Reproductive failure was defined as a sustained decline in daily egg production exceeding 20% compared to controls, accompanied by gross and histopathological lesions in the ovaries and oviducts (follicular liquefaction, oviductal edema, and lymphocytic infiltration), suggesting potential long-term or permanent reproductive impairment. Necropsy at 10 dpi revealed mild ovarian edema and marked liquefaction of the follicles. These findings demonstrate that H6N6 AIV infection induces significant reproductive tract impairment, leading to a substantial and sustained reduction in the egg production of breeding ducks (Fig. 5C). Histopathological findings Mice: At 3 and 5 dpi, pulmonary lesions were characterized by consolidation, thickened alveolar septa, congestion, and focal hemorrhage, accompanied by mild inflammatory cell infiltration. Bronchiolar epithelial cells exhibited desquamation and attenuation. Other systemic tissues remained histologically unremarkable (6A:a–d). Pathogenicity of the H6N6 isolate in mice and poultry. Mice were challenged with 106 EID50 of virus. Histopathological changes (H&E staining) at 3 and 5 dpi in infected mice (6A), at 5 dpi in infected chickens (6B), at 3 and 5 dpi in infected ducks (6C), and at 10 dpi in infected breeding ducks (6D). Magnification, ×200. SPF Chickens: The lungs displayed only mild thickening of the alveolar septa and vascular congestion, with occasional focal inflammatory cell infiltration. No significant histopathological lesions were observed in the trachea or other visceral organs (6B:a-h). Young Muscovy Ducks: No significant histopathological changes were identified in any of the examined tissues, including the heart, liver, spleen, lungs, kidneys, duodenum, and trachea (6c:a-l). Breeding Muscovy Ducks: In the reproductive tract, the oviduct mucosa exhibited multifocal lymphocytic infiltration. The ovaries showed follicular degeneration and liquefaction, which were consistent with the previously described gross observations (6c:a-h). Discussion This study provides a comprehensive assessment of the epidemiological status, molecular evolution, and pathogenicity of the H6N6 avian influenza virus (AIV) in breeding ducks. By integrating serological surveillance with multi-species infection trials, we systematically characterized a novel H6N6 isolate (DK/AH/21HY), providing a scientific basis for the refined monitoring and control of H6 subtypes. Our serological analysis identifies the H6 subtype as the dominant circulating AIV in the surveyed breeding duck farms, aligning with recent findings by Yan et al. (2023) and Lin et al. (2023). A notable observation was the nadir in H6 antibody titers at 39 weeks of age, which suggests a critical "immune gap." This period of vulnerability likely results from the waning of earlier vaccine-induced or naturally acquired immunity, highlighting the need for optimized immunization protocols during the mid-to-late laying period. Based on the identified immune gap at 39 weeks of age, combined with the antibody decay pattern documented in the national immunization guidelines and studies on avian influenza vaccination schedules in ducks(Yang et al., 2014), we recommend the following adjustments for commercial breeding farms: shorten the booster interval during the laying period to 3–4 months; add an H6-specific booster at 36–38 weeks of age; and implement antibody-monitoring-based precision immunization, with a booster triggered when the mean flock antibody titer falls below 6 Log2. Given that waterfowl serve as primary natural reservoirs, such immune gaps significantly facilitate viral persistence and inter-farm transmission (Quan et al., 2019). Genomic analysis revealed that the DK/AH/21HY strain is a triple-reassortant virus within the Eurasian lineage, incorporating genetic components from ST2853-like, ST192, and ST339-like clades (Bui et al., 2025; Cui et al., 2022). This complex reassortment pattern underscores the high genetic plasticity of H6 AIVs and suggests frequent genetic exchange between co-circulating clades in eastern and southern China (Li et al., 2026; Xu et al., 2023; Yingying Du, 2024; Zou et al., 2016). The high nucleotide identity with other regional strains further supports the notion of localized circulation and endemic persistence of these genotypes. The identified HA mutations, A138S and A222V, are situated within the receptor-binding domain (RBD) and have been previously associated with expanded host tropism. Our receptor-binding assays confirmed that these substitutions facilitate dual affinity for both avian-type α−2,3 and human-type α−2,6 sialic acid receptors (Wang et al., 2015; Yang Zhenyu, 2025). Such a shift in binding preference is a key prerequisite for crossing the species barrier (Gillim-Ross et al., 2008; Kim and Park, 2012). Although the virus shows detectable binding to human-type receptors, the preference remains avian-dominant, indicating that cross-species transmission is still biologically inefficient at this stage. Nevertheless, as demonstrated by Arai and Watanabe (2021), Yang et al. (2025), H6N6 viruses with dual-receptor binding properties can replicate in human respiratory tract cells without prior adaptation, underscoring the potential public health risk. However, the isolate maintained a low-pathogenicity phenotype in mice and chickens, likely due to the absence of key mammalian-adaptive virulence markers such as PB2-E627K and PA-I38M, which are typically required for severe systemic disease (Mostafa et al., 2018; Zhu et al., 2015), The shorter shedding duration in SPF chickens compared to breeding ducks likely reflects species-specific differences in host permissiveness, as ducks are natural reservoirs for H6N6 AIV and support more efficient viral replication (Munster et al., 2007). Continued circulation in ducks and occasional spillover into mammals may select for such substitutions over time, as demonstrated during serial passaging of related avian influenza viruses in mouse models (Peng et al., 2018). A primary finding of this study is the significant 24.15% decline in egg production in breeding ducks infected with H6N6 AIV. While previous reports have documented severe economic losses (up to 74% mortality or culling) during H6N1 outbreaks in chickens(Landman Wjm, 2019; Mcmenamy et al., 2024), data regarding H6N6-induced reproductive impairment in ducks have been relatively sparse. Our results demonstrate that H6N6 AIV specifically targets the reproductive system, as evidenced by follicular liquefaction and lymphocytic salpingitis. Although the clinical signs remained mild, the sustained reduction in productivity poses a substantial economic threat to the breeding duck industry. The reproductive tract tropism of this low-pathogenic H6N6 strain may involve several factors. First, LPAIV tropism is host maturation-dependent. de Wit et al. (2020) reported that an H3N1 LPAIV caused 100% egg drop and high viral loads in the oviduct of adult layers, while young birds showed only mild signs. Our 30-week-old breeding ducks were at peak lay, with active reproductive physiology that may increase susceptibility. Second, LPAIV can reach the reproductive tract via non-respiratory routes. Pantin-Jackwood et al. (2012) demonstrated that intracloacal inoculation of H6N2 LPAIV directly infected the oviduct and caused egg drop. Our isolate may reach the oviduct via coelomic spread from the respiratory tract. Third, the oviduct is a known target organ for LPAIV (Pantin-Jackwood and Swayne, 2009). Therefore, the affinity of this H6N6 strain for the reproductive tract is not unique but a recognized feature of LPAIV in adult laying birds. Despite these insights, several limitations warrant consideration. First, this study utilized a single isolate from a restricted geographic area, so the findings may not fully represent the diversity of circulating H6N6 strains. Second, direct comparison with parental clade viruses was not performed, although available data suggest the isolate retains a low-pathogenicity phenotype with enhanced reproductive tract tropism (Huang et al., 2010; Li et al., 2026; Yuan et al., 2016). Third, egg quality parameters (e.g., shell thickness, fertility rates) were not evaluated, despite evidence that AIV infection can affect egg quality (Qi et al., 2016; Stephens et al., 2020; Swayne et al., 2012). Fourth, environmental samples (water, bedding) were not tested, limiting our understanding of viral persistence outside the host. Finally, while we demonstrated mammalian adaptation potential, further trials in pigs or dogs would provide a more comprehensive assessment of zoonotic risk. Future research should prioritize large-scale genomic surveillance, investigate the molecular mechanisms of H6N6-induced ovarian damage, and systematically evaluate egg quality and environmental transmission. In conclusion, H6N6 AIV has emerged as the predominant subtype in breeding duck farms in eastern and southern China, with a critical immune gap occurring at 39 weeks of age. The identified DK/AH/21HY strain is a multi-clade reassortant harboring HA mutations that confer interspecies transmission potential. Although classified as low-pathogenic, the virus induces significant reproductive tract lesions and impairs egg production in breeding ducks. These findings emphasize the necessity for continuous epidemiological surveillance and the development of targeted control strategies to safeguard both poultry health and public safety. Ethics statement This study was approved by the Animal Care Committee of South China Agricultural University (approval ID: SYXK-2019-0136). All animal experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of the People’s Republic of China. The study is reported in compliance with the ARRIVE guidelines. Funding statement This work was supported by the China Postdoctoral Science Foundation [grant numbers 2021M692455]. Author contributions The study was conceived and designed by Feng Chen and Puduo Zhu. Puduo Zhu performed the experiments and wrote the first draft. Liguo Gao and Xudong He established the experimental protocols and conducted the data collection. Zhihong Luo and Wencheng Lin organized and validated the experimental data. Jiajia Liu and Qi Zhou supplied the technical resources and participated in the experimental validation. Qunhui Li interpreted the data and provided scientific supervision. Zhuanqiang Yan coordinated the project management and logistics. Feng Chen supervised the study, acquired the funding, and revised the manuscript. All authors reviewed and approved the final submission. Disclosures The authors declare no competing financial, professional, or personal interests that could have influenced the work reported in this manuscript. Signed by all authors on behalf of the research team. Corresponding author: Feng Chen (chenfeng@scau.edu.cn) Acknowledgments Special thanks to the technical department of the Guangdong Wenshi Group for providing the platform and appreciation to Dr. Qunhui Li for experimental guidance. References