-- title: "Avian Influenza H5N1 in Dairy Cattle: Cross-Species Transmission, Clinical Signs, and Diagnostic Challenges" category: "livestock-viruses" metaDescription: "A comprehensive veterinary review of H5N1 avian influenza spillover into dairy cattle, covering virology, pathogenesis, clinical presentation, diagnostic methods, and surveillance challenges." primaryKeyword: "H5N1 dairy cattle" secondaryKeywords: ["avian influenza in cattle", "cross-species transmission", "bovine influenza diagnostics", "H5N1 clinical signs cattle", "molecular detection milk"]

Avian Influenza H5N1 in Dairy Cattle: Cross-Species Transmission, Clinical Signs, and Diagnostic Challenges

Introduction

The emergence of highly pathogenic avian influenza (HPAI) H5N1 virus (clade 2.3.4.4b) in North American dairy cattle represents a significant departure from the established host range of influenza A viruses in livestock. Historically, cattle were considered resistant to productive infection with avian influenza viruses due to the absence of compatible sialic acid receptors in the bovine respiratory tract. The documented spillover and sustained within-herd transmission of H5N1 in dairy herds has compelled a re-evaluation of interspecies barrier mechanisms, viral tropism, and diagnostic preparedness in bovine populations. This reference article synthesizes the current understanding of H5N1 infection in dairy cattle with a focus on cross-species transmission dynamics, clinical presentation, and the specific diagnostic challenges encountered in field and laboratory settings.

Virological Basis and Host Adaptation

Receptor Binding and Cellular Entry

Influenza A virus host tropism is primarily governed by the binding preference of the hemagglutinin (HA) protein for specific sialic acid (SA) linkages. Avian influenza viruses typically bind to alpha-2,3-linked SA receptors, which are abundant in the avian intestinal and respiratory tracts. Human-adapted strains preferentially bind alpha-2,6-linked SA receptors. The bovine respiratory tract was traditionally thought to express predominantly alpha-2,6 linkages, rendering cattle refractory to avian strain infection.

Ding and Ding [7] provided a comprehensive analysis of the molecular adaptations facilitating H5N1 entry into bovine cells. The HA of clade 2.3.4.4b viruses exhibits a glutamine-to-leucine substitution at position 226 (H3 numbering) that enhances binding to alpha-2,6 SA receptors. This single amino acid change, combined with the loss of a glycosylation site near the receptor binding domain, broadens the host range sufficiently to permit infection of bovine mammary epithelial cells. The mammary gland, rather than the respiratory tract, appears to be the primary target organ in cattle, a finding that is unprecedented for influenza A viruses in ruminants.

Replication Kinetics and Tissue Tropism

Experimental inoculation studies have confirmed that dairy cows are susceptible to H5N1 infection at remarkably low infectious doses. Lee et al. [4] demonstrated that a dose as low as 10 plaque-forming units administered via the teat canal resulted in productive infection, with viral titers in milk reaching up to 10^6 TCID50/mL within 48 hours. The virus replicated to high titers in mammary epithelial cells but was largely undetectable in nasal swabs or lung tissue, indicating a tissue-restricted tropism distinct from classical respiratory influenza infection.

The molecular basis for mammary gland tropism involves the high expression of α2,6 SA receptors on mammary alveolar epithelial cells during lactation. The lactating mammary gland also presents a unique immunological microenvironment characterized by local immune modulation to prevent rejection of milk proteins. Singh et al. [1] characterized the compartmentalized cytokine response in H5N1-infected bovine mammary tissue, revealing a dysregulated network of pro-inflammatory cytokines (IL-6, TNF-α, CXCL8) and a suppression of interferon-stimulated genes. This localized immune remodeling facilitates viral replication while limiting systemic inflammation, explaining the paradoxical absence of severe respiratory disease in infected animals.

Clinical Presentation in Dairy Cattle

Acute Phase Signs

The clinical syndrome of H5N1 infection in dairy cattle is distinct from classical influenza in other livestock species. The onset is acute, occurring within 2 to 4 days post-exposure. The predominant clinical sign is a sudden drop in milk production, often exceeding 50 percent within 24 to 48 hours. Affected cows produce a thick, colostrum-like secretion that is yellow to yellow-brown in color. This altered milk consistency is associated with the massive sloughing of infected mammary epithelial cells and the influx of inflammatory cells into the gland lumen.

Fever, typically ranging from 39.5°C to 41.5°C, accompanies the early phase of infection but is often transient, resolving within 24 to 48 hours. Other clinical signs include decreased feed intake, mild dehydration, and ruminal stasis. Respiratory signs such as coughing, nasal discharge, or dyspnea are conspicuously absent in most naturally infected animals, distinguishing this syndrome from respiratory influenza infections in pigs or horses.

The clinical presentation in lactating dairy cows is summarized in Table 1.

Table 1. Clinical Signs of H5N1 Infection in Lactating Dairy Cattle

Subclinical Infection and Reservoir Potential

A substantial proportion of infected animals, particularly those in early lactation or with lower milk production, may exhibit subclinical infection. Stenkamp-Strahm et al. [10] conducted a longitudinal surveillance study in California dairy herds and detected influenza A viral RNA in bulk tank milk from pens without clinical signs, indicating that asymptomatic shedding can sustain within-herd transmission. The proportion of subclinical infections was estimated at 25 to 40 percent based on serological surveys. Shittu et al. [2] detected anti-H5 antibodies in naturally infected cattle for over one year following exposure, confirming that recovered animals maintain a durable antibody response. The presence of a long-term serological memory in cattle has implications for interpreting serosurveillance data and for differentiating recent from historical infections.

Diagnostic Methods

Molecular Detection

Reverse transcription quantitative polymerase chain reaction (RT-qPCR) is the primary diagnostic modality for H5N1 detection in cattle. The matrix (M) gene assay is used for initial influenza A screening, followed by H5-specific subtyping assays. Giersch et al. [15] validated a high-throughput RT-qPCR subtyping panel that simultaneously detects H1pdm09, H3, and H5 in bovine clinical samples. This panel demonstrated 100 percent sensitivity for H5 detection in spiked milk samples at a limit of detection of 10 RNA copies per reaction.

Sample selection is critical for molecular diagnostic accuracy. In dairy cattle, milk is the optimal specimen due to the high viral loads present in mammary secretions. Nasal swabs, traditionally preferred for influenza diagnosis in other species, yield false-negative results in the majority of infected cows due to the absence of respiratory tract replication. Lee et al. [4] reported that milk samples from infected quarters had viral RNA concentrations that were 4 to 6 orders of magnitude higher than concurrent nasal swabs.

Bulk tank milk (BTM) sampling has emerged as a practical herd-level surveillance tool. Stenkamp-Strahm et al. [10] demonstrated that RT-qPCR testing of BTM can detect H5N1 in herds with a within-herd prevalence as low as 5 percent. However, the sensitivity of BTM testing is influenced by dilution effects, milk volume, and the distribution of infected animals across milking groups.

Interlaboratory Variability and Method-Dependent Sensitivity

Miller et al. [8] conducted an interlaboratory comparison study evaluating 12 different RT-qPCR protocols for H5N1 detection in milk. The results revealed significant variability in analytical sensitivity, with cycle threshold (Ct) values differing by up to 8 cycles between laboratories for identical samples. Protocols that included a sample concentration step, such as ultrafiltration or polyethylene glycol precipitation, had lower Ct values and higher detection rates for low-viral-load samples. Standardization of RNA extraction methods and primer-probe sets across diagnostic laboratories is an urgent requirement for consistent surveillance.

Serological Testing

Serological assays for H5N1 in cattle include hemagglutination inhibition (HI) tests, virus neutralization tests (VNT), and commercial enzyme-linked immunosorbent assays (ELISA) developed for avian species, with modifications for bovine serum. Shittu et al. [2] demonstrated that HI titers in naturally infected cattle persisted above 1:40 for 12 to 15 months post-infection. The specificity of HI testing in cattle is high, but cross-reactivity with other influenza A subtypes (particularly H1 and H3, which circulate in swine and may occasionally infect cattle) requires careful assay validation.

The detection of antibodies in milk, using either indirect ELISA or HI, has been proposed as a non-invasive surveillance tool. Antibody titers in milk correlate with serum titers but are approximately 2- to 4-fold lower. The presence of antibodies in bulk tank milk can indicate herd-level exposure, but the timing of seroconversion (7 to 14 days post-infection) limits the utility of serology for acute outbreak detection.

Next-Generation Sequencing and Genomic Surveillance

Frederick et al. [11] described the application of next-generation sequencing (NGS) strategies during the H5N1 emergency response in the United States. Whole-genome sequencing of H5N1 isolates from bovine clinical samples provided crucial data on viral evolution, including the identification of mammalian adaptation markers in the HA, polymerase basic 2 (PB2), and neuraminidase (NA) genes. The PB2 E627K substitution, a canonical mammalian adaptation marker, was identified in a subset of bovine isolates, although its presence was not universal. Whole-genome sequencing also enabled phylogenetic tracking of transmission chains between farms and from cattle to other species.

Diagnostic Workflow

The diagnostic decision tree for investigating suspected H5N1 in dairy cattle is presented in Figure 1.

flowchart TD
    A[Suspected H5N1 outbreak\nin dairy herd], > B{Clinical signs present?\nDecreased milk production,\nfever, thickened milk}
    B, >|Yes| C[Collect individual\nmilk samples from\naffected cows]
    B, >|No| D[Collect bulk tank\nmilk sample for\nherd-level screening]
    C, > E[RT-qPCR for\ninfluenza A M gene]
    D, > E
    E, > F{Influenza A positive?\nCt value < 40}
    F, >|No| G[Consider alternative\ndiagnoses: mastitis,\nmetabolic disease]
    F, >|Yes| H[H5 subtyping\nRT-qPCR]
    H, > I{H5 positive?}
    I, >|Yes| J[Confirmatory testing:\nWhole genome sequencing,\nVirus isolation]
    I, >|No| K[Subtype identification\nfor non-H5 influenza A\n(e.g., H1, H3)]
    J, > L[Report to veterinary\nauthorities, implement\nquarantine and biosecurity]
    K, > L
    G, > M[Repeat testing if\nclinical signs persist\nin 48-72 hours]
    M, > E

Diagnostic Challenges

Sample Type and Quality

The atypical tissue tropism of H5N1 in cattle creates a diagnostic challenge for veterinarians accustomed to collecting respiratory samples for influenza testing. The high diagnostic yield from milk samples is not widely recognized, and many initial diagnostic attempts relied on nasal swabs, leading to false-negative results and delayed outbreak detection. Stall-side sample handling is also critical. Milk samples should be collected aseptically from individual quarters or composite samples from affected cows. Samples must be refrigerated at 4°C and shipped on cold packs within 24 hours. Freezing at -20°C is acceptable for longer storage but may reduce viral RNA recovery.

Viral Stability in Milk

Schafers et al. [9] characterized the stability of H5N1 in bovine milk under various storage and processing conditions. The virus remained infectious in raw milk for at least 5 days at 4°C and for over 2 days at 22°C. High-temperature short-time pasteurization (72°C for 15 seconds) effectively inactivated the virus, but the detection of viral RNA by RT-qPCR in pasteurized milk samples raised concerns about interpreting positive results from processed milk. RNA detection in pasteurized milk does not indicate infectious virus, but it complicates surveillance efforts that rely solely on molecular testing without viability assessment.

Diagnostic Window and Sampling Timing

The window for molecular detection of H5N1 in milk is relatively narrow. Viral RNA concentrations peak at 2 to 4 days post-infection and decline rapidly after 7 to 10 days. Stenkamp-Strahm et al. [10] reported that individual milk samples become RT-qPCR-negative in most animals by 14 to 21 days post-infection. Herd-level testing using bulk tank milk may remain positive for longer periods (3 to 5 weeks) as new infections occur in previously unexposed animals. Delays in sample collection after the onset of clinical signs can result in false-negative molecular test results, emphasizing the need for rapid diagnostic response during suspected outbreaks.

Serological Interpretation

The longevity of anti-H5 antibodies in cattle [2] presents a challenge for distinguishing active from historical infections. A single positive serological result does not confirm active viral shedding. Paired serology (acute and convalescent samples collected 14 to 28 days apart) is required to demonstrate seroconversion or a four-fold rise in antibody titer. The application of serological testing to bulk tank milk for herd-level surveillance is further complicated by the dilution of antibodies in pooled samples and the variable antibody levels among individual cows.

One Health Surveillance and Epidemiological Implications

The emergence of H5N1 in dairy cattle has significant implications for One Health surveillance, bridging livestock production, wildlife ecology, and agricultural biosecurity. Adiga et al. [3] developed a high-resolution digital simulation model integrating livestock movement data, wild bird migration patterns, and human population density to predict multi-host epidemic spread. The model identified dairy farm density, proximity to waterfowl habitats, and the frequency of cattle transport as key risk factors for H5N1 introduction and propagation.

The European Food Safety Authority (EFSA) overview [13] documented the continued circulation of H5N1 clade 2.3.4.4b in wild bird populations across Europe, with spillover events into mammals including foxes, seals, and now cattle. The virus has demonstrated the capacity for onward mammalian transmission without full adaptation to mammalian hosts, a property that Quirk et al. [6] recapitulated in ferret transmission studies. Variable transmission efficiency among mammalian-origin H5N1 strains was observed, with some isolates transmitting via direct contact while others did not transmit at all. This heterogeneity underscores the unpredictable nature of H5N1 mammalian adaptation.

Control Strategies and Biosecurity

The compartmentalized nature of H5N1 infection in the bovine mammary gland has implications for control strategies. Traditional influenza control measures, such as respiratory isolation and ventilation management, may be insufficient to prevent transmission. Lee et al. [4] demonstrated that transmission between cows occurs primarily through contaminated milking equipment and milk splash, rather than through aerosol or direct contact. This finding suggests that targeted biosecurity measures, including the disinfection of milking clusters between cows, segregation of infected and uninfected milking groups, and the use of personal protective equipment by milkers, are more effective than respiratory barrier measures.

Malladi et al. [5] used an approximate Bayesian computation approach to estimate the within-herd transmission rate of H5N1 in a dairy herd. The basic reproduction number (R0) was estimated at 1.5 to 2.5 under standard management conditions, indicating that the virus can sustain transmission but is not explosively contagious. The implementation of prompt quarantine of affected pens, cessation of milk collection from infected groups, and enhanced milking hygiene reduced the effective reproduction number (Re) below 1 within 7 days.

Vaccination strategies for H5N1 in cattle are under investigation. Wiggins et al. [14] described a dual-route vaccination approach combining intramuscular and intranasal administration of an H5N1 vaccine in murine and bovine models. The intramuscular component induced systemic IgG responses, while the intranasal component elicited mucosal IgA and local cellular immunity in the respiratory and mammary tissues. Although the vaccine is not yet licensed for bovine use, the approach demonstrates the feasibility of inducing protective immunity in the target organ system. Mohammad et al. [12] provided a broader review of HPAI vaccination strategies, emphasizing the molecular basis of pathogenicity and the need for vaccines that account for the antigenic drift of circulating H5N1 strains.

Conclusions

The spillover of H5N1 avian influenza into dairy cattle represents a paradigm shift in the understanding of influenza host range. The virus has adapted to exploit the lactating bovine mammary gland as a primary replication site, a tissue tropism that distinguishes bovine H5N1 from all previously known influenza A infections in mammals. Clinical signs are predominantly mammary in origin, with acute drops in milk production and altered milk composition as cardinal features. Diagnosis relies on RT-qPCR testing of milk samples, as nasal swabs lack sensitivity. The rapid evolution of the virus and its capacity for onward transmission among cattle and to other mammals necessitates continued genomic surveillance, rigorous biosecurity, and the development of species-specific control tools.

The diagnostic challenges described in this review, including interlaboratory assay variability, sample type selection, and the narrow window for molecular detection, highlight the need for standardized protocols and clinician education. The integration of diagnostic data with computational modeling, as described by Adiga et al. [3] and Malladi et al. [5], will be essential for predicting and mitigating the impact of H5N1 on dairy production systems.

References

1. Singh G, Cool K, Ghimire S, et al. Compartmentalized cytokine networks and systemic immune remodelling in bovine mammary H5N1 infection. Emerg Microbes Infect. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42249592/

2. Shittu I, Rodriguez J, Oguzie JU, et al. Detection of antibodies to avian influenza virus H5N1 clade 2.3.4.4b in naturally infected cattle for more than a year. Sci Rep. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42209625/

3. Adiga A, Chopra A, Wilson ML, et al. A high-resolution, US-scale digital similar of interacting livestock, wild birds, and human ecosystems for multihost epidemic spread. Proc Natl Acad Sci U S A. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42207910/

4. Lee C, Tarbuck NN, Cochran HJ, et al. Dairy cows infected with influenza A(H5N1) reveals low infectious dose and transmission barriers. Nat Commun. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42177170/

5. Malladi S, Carestia A, Seys SA, et al. Estimating the Within-herd Transmission Rate of Highly Pathogenic Avian Influenza H5N1 Virus in a Dairy Herd using an Approximate Bayesian Computation Approach. J Dairy Sci. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42155711/

6. Quirk GE, Vu MN, Le Sage V, et al. Variable transmission efficiency of mammalian origin HPAI D1.1 H5N1 strains in ferrets. bioRxiv. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42146553/

7. Ding K, Ding Y. H5N1 avian influenza in dairy cattle: Molecular adaptation, transmission mechanisms, and control strategies. Virology. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42107281/

8. Miller MR, Frost K, Smith EL, et al. Evaluation of PCR-Based H5N1 Influenza Detection Methods in Milk from an Interlaboratory Comparison Study Demonstrating Method-Dependent Sensitivity Variability. J Food Prot. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42103287/

9. Schafers J, Warren CJ, Yang J, et al. Stability of influenza viruses in the milk of cows and sheep. J Gen Virol. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42095438/

10. Stenkamp-Strahm C, Melody B, Brinson P, et al. A longitudinal study of influenza A viral detection in bulk tank and pen-level milk collected from dairy farms in California affected by Highly Pathogenic Avian Influenza H5N1. J Dairy Sci. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42092544/

11. Frederick JC, Lacek KA, Wersebe MJ, et al. Next-Generation Sequencing Strategies During the 2024-2025 Avian Influenza A(H5N1) Emergency Response in the U.S. Viruses. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42043271/

12. Mohammad I, Hajelbashir MI, El-Bidawy MH, et al. Deciphering HPAI Influenza A Virus (H5N1): Molecular Basis of Pathogenicity, Zoonotic Potential, and Advances in Vaccination Strategies. Viruses. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42043199/

13. European Food Safety Authority (EFSA), European Centre for Disease Prevention and Control (ECDC), European Union Reference Laboratory for Avian Influenza (EURL), et al. Avian influenza overview December 2025-February 2026. EFSA J. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42016297/

14. Wiggins J, Madapong A, Weaver EA. Dual-route H5N1 vaccination induces systemic and mucosal immunity in murine and bovine models. NPJ Vaccines. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42014782/

15. Giersch K, Nörz D, Grunwald M, et al. Adaptation and validation of an influenza a subtyping panel for detection of H1pdm09, H3 and H5 on a high-throughput RT-qPCR system. Sci Rep. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42009717/