Highly Pathogenic Avian Influenza Virus Exposure and Infection in Free-Ranging Bobcats (Lynx rufus) in New York, USA Open Access

The introduction of novel pathogens can threaten multiple species simultaneously with uncertain and potentially serious outcomes. One such example, highly pathogenic (HP) H5N1 influenza A virus (IAV), is a zoonotic pathogen that poses a threat to wildlife populations and the poultry industry, with the potential to result in population declines (Falchieri et al. 2022) and severe economic losses (Sonaiya 2007). Since December 2021, an outbreak of HP H5N1 of the Eurasian lineage goose/Guangdong clade 2.3.4.4b has been spreading throughout the US (Caliendo et al. 2022). Historically, HP IAV has been characterized by high morbidity and mortality in poultry, with rare spillover into mammals, including humans (Uyeki 2008). The current outbreak has affected an unprecedented number of wild mammals (Puryear and Rundstadler 2024). Large-scale outbreaks of the currently circulating HP H5N1 in mammals have particularly affected pinnipeds (Puryear and Rundstadler 2024), with one of the deadliest outbreaks having occurred in an elephant seal (Mirounga leonine) population in Argentina: seal pup mortality reached 96% because of H5N1 infection during the outbreak, compared with 0.8% in a previous year without high disease prevalence (Campagna et al. 2024). Such outbreaks could have devastating effects on vulnerable populations, and reports have indicated that HP H5N1 has already affected at least 67 different species of mammals (Elsmo et al. 2023; FAO 2024).

The order Carnivora contains the highest number of mammal species affected by HP H5N1; this is probably a result of increased exposure from consumption of infected carrion and avian prey (Chestakova et al. 2023). Some carnivore species have already experienced declines due to human persecution and natural resource exploitation (Di Marco et al. 2014; Ripple et al. 2014), making emerging wildlife diseases with high morbidity and mortality rates particularly concerning. For example, black-footed ferrets (Mustela nigripes) were endangered in the US by the 1980s because of habitat loss and mortalities resultant from poisonings that targeted prairie dogs (Cynomys spp.; Santymire et al. 2014). Despite conservation efforts to recover the species, canine distemper virus and sylvatic plague nearly extirpated the remaining black-footed ferrets (Santymire et al. 2014). Cases like this emphasize the need to track such disease effects before they can devastate populations.

Bobcat populations in New York State have been recovering from large-scale historical declines due to persecution as a perceived nuisance species from the 1800s through the late 1900s (Fox and Brocke 1983), with recent estimates suggesting that occurrence and abundance remain low throughout the state (Twining et al. 2024a, b). To elucidate potential threats to the population, we conducted health assessments on bobcats captured during our ongoing work across New York. This work uses camera trapping and GPS collar deployment to estimate bobcat abundance. As part of the health assessments, we tested for the prevalence of antibodies to H5 and N1 in free-ranging, presumably healthy bobcats, and we tested for HP H5N1 infection in two individuals that died during the study. Bobcats (n=16) were live captured from January through March 2024 (Cornell Institutional Animal Care and Use Committee permit 2023-0163) in seven counties across New York (Table 1; Fig. 1). We anesthetized trapped bobcats using a combination of either 10 mg/kg ketamine hydrochloride and 1.5 mg/kg xylazine hydrochloride or 5 mg/kg ketamine hydrochloride and 0.05 mg/kg dexmedetomidine hydrochloride. We weighed and sexed bobcats while they were under anesthesia and performed a brief physical examination. Animal-borne GPS collars (LiteTrack Iridium, Lotek Wireless Inc., Newmarket, Ontario, Canada) were deployed on 14 of the bobcats while they were anesthetized. Each collar weighed either 195 g or 295 g and was always <3% of the individual bobcat’s body mass (Wilson et al. 2021). Standard precautions were taken to prevent contamination and infection, including using disposable surgical masks and nitrile gloves and cleaning equipment between captures.

We collected blood from the medial saphenous or cephalic vein and stored it chilled until same-day centrifugation for serum separation. Serum was stored at −80 C until testing for antibodies to IAV using a commercial blocking ELISA (bELISA; IDEXX AI MultiS-Screen Ab test, IDEXX Laboratories, Westbrook, Maine, USA); a sample to negative absorbance value of <0.7 was classified as positive. All samples were further tested by hemagglutination inhibition (HI), virus neutralization (VN), and enzyme-linked lectin assay (ELLA) for antibodies to H5 and N1 as described in Stallknecht et al. (2020, 2022). Two attenuated viruses produced by reverse genetics (rg) were used as antigens for HI and VN: Specifically, rg blue-winged teal (BWT) containing the hemagglutinin (HA) and neuraminidase (NA) from low-pathogenic (LP) A/BWT/AI12-4150/Texas/2012 (H5N2) and rg Astrakhan (AST) IDCDC-RG71A (H5N8) containing a modified HA and an NA from HP 2.3.4.4b A/AST/3212/2020 (H5N8); remaining gene segments for both viruses were from A/PuertoRico/8/34. For ELLA, A/ruddy turnstone/New Jersey/AI13-2948/2013 (H10N1) was used as an N1 antigen. Samples were considered positive for antibodies to H5 if they tested positive on HI or VN for either antigen. For HI, VN, and ELLA, a titer of ≥32, ≥20, or ≥80 was considered positive, respectively.

During this study, two collared animals were confirmed to have died (Table 1). Both carcasses were recovered and examined by a board-certified anatomic pathologist (G.R.H.) at the Cornell Animal Health Diagnostic Center (AHDC; Ithaca, New York, USA). Paired tissue sections from all major organs and lesions were fixed in 10% neutral buffered formalin and routinely processed and stained with H&E. Brain tissue was submitted for rabies virus fluorescent antibody testing and real-time reverse transcription PCR (rRT-PCR) targeting the IAV matrix and H5 genes. Lung tissue from one animal was also submitted for rRT-PCR. The AHDC follows the National Veterinary Services (NVSL) Laboratory standard operating procedure for avian influenza virus PCR analysis (National Animal Health Laboratory Network 2023).

A total of 9/16 individuals (56%) tested positive for antibodies to IAV as determined by bELISA (Table 1). Of these, 4/9 (44%) tested positive for antibodies to H5 as determined by both HI and VN. These four also tested seropositive for antibodies to N1. Therefore, 4/16 bobcats (25%) had antibodies to both H5 and N1. One bELISA-positive sample tested positive only for antibodies to N1, and four samples were seronegative for antibodies to both H5 and N1. Felids are susceptible to LP IAV, and these five animals may have been naturally infected before capture with IAV representing non-H5 subtypes (Driskell et al. 2013). All bELISA-negative samples also tested negative for antibodies to both H5 and N1. Higher antibody titers were observed with the rg AST HP 2.3.4.4 representative antigen compared with the rg BWT LP North American antigen in both the HI and VN assays (Table 1). Movement data on the four bobcats that tested positive for antibodies to H5 and N1 showed that bobcats #2 and #14 remained alive as of June 2024, when their GPS collars stopped communicating, and bobcats #1 and #16 remained alive as of November 2024.

Of the two bobcats that were confirmed to have died during this study, one adult female (bobcat #4) was found dead on 23 May 2024, 119 d after capture. Serology indicated this bobcat was exposed to IAV before capture, but IAV was not detected via rRT-PCR of brain tissue at the time of necropsy. Rabies virus was also not detected. At the time of capture, bobcat #4 weighed 10.2 kg and was in good body condition; at necropsy, it weighed 6.7 kg, a loss of approximately 34% of its body weight. Examination during necropsy revealed that this animal was lactating. A cause of death was not determined. One adult male bobcat (bobcat #11) was found dead on 15 March 2024, 34 d after capture. This individual tested negative for IAV antibodies upon initial capture. At the time of capture, bobcat #11 weighed 13.7 kg and was in good body condition; at necropsy, the animal was 11.9 kg, a loss of approximately 13% of its body weight. Rabies virus was not detected. Notable findings included a moderate cestode burden in the small intestine and dark-red-to-black luminal content (presumably melena) in the large intestine. Liver tissue was negative for anticoagulant rodenticides that may have explained the melena. Histopathology of the lungs revealed a moderate, locally extensive, chronic, granulomatous pneumonia with intralesional larval lungworms and eggs. The cerebrum had a moderate, multifocal-to-generalized, subacute, lymphoplasmacytic and neutrophilic meningoencephalitis with individual neuronal necrosis and lymphocytic perivascular cuff formation (Fig. 2), consistent with IAV cases in mesocarnivores that have been reported by Elsmo et al. (2023). Gastrointestinal parasitism and hemorrhage have also been reported in a subset of mesocarnivores with IAV infection (Elsmo et al. 2023).

We detected IAV in both the brain and lung tissues of bobcat #11 via rRT-PCR; cycle threshold (Ct) values for the matrix gene were 11.5 and 21.4, respectively, and for the H5 gene were 13.7 and 23.9, respectively. We expected a lower Ct value in brain tissue than in lung tissue; in other mesocarnivores infected with HPAI clade 2.3.4.4b, brain has been the most common tissue to have the strongest amplification signal on rRT-PCR, consistent with meningoencephalitis being a primary histologic lesion (Elsmo et al. 2023). Strain identification using whole-genome sequencing at the NVSL revealed that this was EA/AM 2.3.4.4b H5N1 (C2.1), a reassortant of the H5 goose/Guangdong and North American wild bird lineage. Sections of brain were sent to the University of Georgia Veterinary Diagnostic Laboratory, Athens, Georgia for immunohistochemical staining using a monoclonal antibody to IAV, and there was robust (neuronal) immunoreactivity throughout the cerebral cortex (Fig. 3). Both EA (Eurasian) H5N1 and EA/AM (H5 goose/Guangdong and North American wild bird lineage) H5N1 strains have previously been reported in bobcats in California, Colorado, Vermont, Washington, and Wisconsin (USDA APHIS 2024).

The detection of H5 and N1 antibodies in 25% of tested bobcats in our study suggests survival from previous HP H5N1 infection and is consistent with previously reported results from the Netherlands, in which antibodies to H5 were detected in 29% of the wild, culled beech martens (Martes foina) that were tested (Chestakova et al. 2023). It is unclear why some individuals experience morbidity and mortality and others do not. Bobcat #11, which died of HP H5N1 infection, was naïve for antibodies to IAV at the time of capture and had no indication of underlying health issues. Virulence of individual North American HP H5N1 genotypes may be a factor, as has been demonstrated in previous studies (Kandeil et al. 2023); thus, future research could benefit from further analysis of the virus sequenced from bobcat #11.

On the basis of our sample of bobcats distributed across New York State, exposure of bobcats to HP H5N1 appeared considerable (25%). Currently, risk factors for HP H5N1 infection in mammals are not well defined; however, they might include infection and mortality in wild and domestic birds, behavioral traits such as scavenging carrion or feeding on birds, and inadequate biosecurity measures for disposal of livestock carcasses (ENETWILD Consortium 2024). Further research could benefit from comparing the movement behavior (from GPS data) of bobcats infected with HPAI and those without infection. Several other carnivore species recovering from historical declines in the US also have been reported with HP H5N1 infections, including the grizzly bear (Ursus arctos horribilis) and American marten (Martes americana, USDA APHIS 2025), highlighting the importance of monitoring disease risks and pathogen prevalence to prevent declines in both known vulnerable carnivore populations and carnivore populations with unknown risk factors. Furthermore, although spillover of HP H5N1 to humans is rare and has primarily occurred through contact with infected poultry and livestock, the risk of viral adaptations that may occur during spillover into wild mammal populations could affect the pandemic potential of HP H5N1 (Peiris 2009). Thus, it is also critical for human populations that the virus is closely monitored.

This research was funded by the New York State Department of Environmental Conservation (DEC) and the Federal Aid in Wildlife Restoration (W-173-G grant F22AF02393-00). Serologic work was funded in part by the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, under contract 75N93021C00016. We thank the DEC staff, licensed trappers, and Janet L. Swanson Wildlife Hospital staff who assisted in the bobcat capture effort. We also thank Kevin Hynes for reviewing our manuscript. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government. All data analyzed in the study are presented in Table 1.