Postmortem Examination, Tissue Sampling, and Histopathologic Investigation

We collected a thorough clinical history from the rehabilitation center. We subjected the carcasses of 3 of the 5 swans, the fox, and 3 of the 5 seals (1 gray seal and 2 common seals) full postmortem examination (PME). Two swans were submitted frozen and did not undergo PME, but we took oral and cloacal swab samples from these carcasses for virologic testing. We examined plain swab samples from the oropharynx and cloaca of each of the swans, and similarly from the nasal cavity and rectum of each of the 3 seals, for preliminary influenza virologic testing. After receipt of the avian influenza virus results, we sampled the 3 freshest seal carcasses because they were more likely to yield meaningful results and have suitable samples for histopathologic examination. We took tissue samples of brain, liver, kidney, spleen, lymph nodes, and lung from the seals and the fox and stored them at −80°C until required.

For histologic examination, we took samples of heart, lung, liver, spleen, and brain tissues from the swans, fox, and seals, as well as lymph nodes of the mammalian species, and fixed them in 10% neutral buffered formalin before processing for hematoxylin and eosin staining and immunohistochemical (IHC) examination. We performed IHC examination by using anti–influenza A nucleoprotein primary antibody (Statens Serum Institute, https://en.ssi.dk), as previously described (2), and anti–canine distemper virus nucleoprotein mouse monoclonal primary antibody (Bio-Rad, https://www.bio-rad.com) in selected fox tissues.

Virologic Investigation

We assessed swabs taken from the swans and seals and tissue samples taken from the seals and fox for influenza A nucleic acid by using a screening real-time reverse transcription PCR (rRT-PCR) assay (3), followed by an H5 subtype–specific rRT-PCR assay (4–6). Where the subtype-specific rRT-PCR assay detected positive samples, we determined the hemagglutinin cleavage-site sequence (7).

We performed virus isolation on PCR-positive samples from the swans, seals, and fox by using specified pathogen-free embryonated fowls’ eggs (8,9). We used the 3 viral isolates recovered to generate whole-genome sequence (WGS) data by using a MiSeq platform (Illumina, https://www.illumina.com) (10). For comparative genetic analysis, we downloaded recent H5 2.3.4.4b virus hemagglutinin sequences from the GISAID EpiFlu database (https://platform.gisaid.org). We deposited sequences in the GISAID database (accession nos. EPI_ISL_1123360, EPI_ISL_2081527, and EPI_ISL_2081528).

Differential diagnosis for the fox samples included molecular testing for rabies virus and canine distemper virus (11,12). We assessed samples from all 3 seals for Leptospira infection (13). We subjected RNA obtained from the fox samples to sequence-independent single-primer amplification (14) to enable sequence generation. We then used the sequence data obtained from these samples to exclude other viral agents by removing reads that aligned with the Vulpes vulpes genome (15,16) and then undertaking de novo assembly to produce contiguous sequences by using SPAdes (17). We then screened these sequences by using custom viral databases (18) against the Bornaviridae, Circoviridae, Flaviviridae, Herpesviridae, Paramyxoviridae, Parvoviridae, Pheuiviridae, and Rhabdoviridae viral families.

Clinical Setting

The swans, gray seal, and fox were housed within the quarantine unit of the center, a suite of 17 individual cubicles accessed by a central corridor (Appendix Figure 1). This facility was designed to minimize transmission of microorganisms among residents; animals in separate cubicles had no direct contact. Basic biosecurity practices were in place, such as decontamination steps between cubicles and staff using respiratory protective equipment (e.g., N95 masks) and dedicated personal protective equipment for each cubicle.

The 5 juvenile mute swans were rescued from different locations and brought to the center for treatment during October 23–November 20, 2020 (Figure 1). These birds were admitted for various reasons, including trauma and being underweight and weak. Infection with avian influenza viruses was not suspected at admission. The juvenile mute swans were housed indoors within the center’s isolation unit, mixed in groups of up to 4 with other rescued adult mute swans (Appendix Figure 1). Each of the juvenile swans had been recovering uneventfully until the sudden onset of lethargy and death or euthanasia during November 25–29 (Figure 1). Clinical signs were not observed in the remaining adult mute swans (n = 6) within the isolation unit. The isolation unit also contained 30 mallards (Anas platyrhynchos) housed across 3 rooms in groups of 10. No clinical disease or deaths were recorded in these birds.

The 4 common seals were estimated to be 5–6 months of age and arrived individually at the facility 1–2 months before the disease episode (Figure 1), admitted for various reasons, including poor body condition, superficial bite wounds, and lungworm. In each case, the animals had been responding well to treatment and supportive care up until the sudden onset of seizures and subsequent death or euthanasia. All 4 seals that died had been in the isolation unit (cubicles 6, 8, and 11) (Appendix Figure 1) at the time the swans (cubicles 15 and 17) were affected (Appendix Figure 1) but had subsequently been moved into another area of the facility. The affected seals had previously had close contact with other common seals in other areas of the facility; none of those animals became ill.

The gray seal was a 2-week-old pup admitted for care after maternal abandonment 2 weeks earlier (Figure 1). It was housed in the isolation unit in a cubicle opposite 1 of the swans (cubicle 1) (Appendix Figure 1). The seal was in good bodily condition and progressing well until the sudden onset of fever, facial twitching, and stupor that led to euthanasia on welfare grounds.

The fox had been in the isolation unit of the center for 2 weeks (Figure 1) in cubicle 10 (Appendix Figure 1). It had been brought to the facility with large areas of alopecia and skin crusts over the body and limbs, consistent with mange. It had been receiving treatment and was reported to be progressing well until the sudden onset of malaise and inappetence and was found dead the following morning.

Pathologic Investigation

The body condition of the 3 swans examined ranged from good to poor. Gross findings among the carcasses included petechiae in the liver and epicardium in 1 bird and opacity of the air sacs in another. Microscopic examination of tissues from 1 bird revealed multifocal, necrotizing, nonsuppurative myocarditis, hepatitis, splenitis, nephritis, and encephalitis, along with intralesional presence of influenza A virus antigen, observed during IHC examination.

Figure 2

Figure 2. Histopathologic and immunohistochemical examination of the gray seal (Halichoerus grypus), common seal (Phoca vitulina), and red fox (Vulpes vulpes) infected with highly pathogenic avian...

The body condition of the 3 examined seals was judged to be fair. Gross examination of 2 common seals revealed generalized lymphadenomegaly and multiple pale foci in the lungs. One common seal also showed congested meninges. The gray seal pup showed generalized lymphadenomegaly but no other gross changes. Microscopic examination of the seal tissues revealed mild, eosinophilic, interstitial pneumonia in the 2 common seals, consistent with lungworm infection, and severe, necrotizing, nonsuppurative polioencephalitis in the 2 common seals and the gray seal. The lymph node sections showed nonspecific reactive hyperplasia, accounting for the gross enlargement of these organs. IHC examination for influenza A virus revealed multifocal immunolabelling in the neurons within the gray matter of the brain in all 3 seals, in close association with the inflammatory lesions (Figure 2, panel A), although virus antigen was absent in the lung, liver, kidney, and lymph node tissues of the seals.

The body condition of the fox was judged to be poor. The animal was visibly underweight, and gross examination revealed large areas of alopecia and crusts affecting the skin of the body and limbs, mild splenomegaly, and generalized reddening of the lungs. Microscopic examination of the brain, lung, and heart revealed severe, acute, nonsuppurative, polioencephalitis (Figure 2, panel B) and ventriculitis (Figure 2, panel C); severe, acute, necrotizing, nonsuppurative interstitial pneumonia (Figure 2, panel D); and mild, acute, nonsuppurative myocarditis (Figure 2, panel E). Within these lesions, IHC examination confirmed the presence of influenza virus antigens among neurons in the cerebrum (Figure 2, panel B) and cerebellum (Figure 2, panel F) and in ependymal cells (Figure 2, panel C), alveolar type I pneumocytes (Figure 2, panel D), and cardiomyocytes (Figure 2, panel E). IHC examination performed on sections of brain and lung showed no canine distemper virus.

Virologic Assessment

Oropharyngeal and cloacal swabs from all 5 swans tested positive for H5N8 (Appendix Table). WGS data were generated for 1 of the swans submitted (A/mute swan/England/234135/H5N8 2020/12/01).

Nasal and rectal swab samples taken during PME of the 3 seals tested negative by PCR for influenza A virus. However, we retrospectively detected influenza A(H5N8) virus nucleic acid in the brain and lung of the gray seal and the brain of 2 common seals (Appendix Table). We performed virus isolation on pooled samples and successfully isolated virus from the pooled brain tissue of seals (A/seal/England/AVP-031141/2020 [H5N8]).

We detected influenza A virus RNA in each of the fox samples (brain, liver, kidney, spleen, and lung), and PCR subtyping determined the influenza A virus subtype as H5N8 (Appendix Table). We isolated virus and generated WGS from the fox brain tissue (A/red fox/England/AVP-M1-21-01/2020 (H5N8)). The hemagglutinin cleavage site (CS) motif from selected fox and seal samples had the amino acid sequence PLREKRRKRGLF, consistent with 99% of CSs characterized across highly pathogenic avian influenza (HPAI) H5N8 virus sequences from the United Kingdom during autumn and winter 2020–2021. Analysis of WGS data generated from the fox, seal, and swan samples demonstrated high sequence similarity (>99.9% nucleotide identity across all gene segments). Alignment of the viral hemagglutinin (Appendix Figure 2, panel A) and neuraminidase genes (Appendix Figure 2, panel B) demonstrated that virus from the fox and seals clustered closely with the viruses detected in the swans from the same wildlife center. Comparison of WGS data from the fox, seal, and swan identified a total of 33 aa substitutions that have been associated with altered virulence reported through natural or experimental infections. Almost all substitutions have been identified in the H5N8 virus sequences generated during the 2020–2021 epizootic in avian species (Appendix Figure 2, panel C). The genomes of the fox, seal, and swan-derived viruses were homologous at the amino-acid level, with the exception of a single amino-acid substitution at position 701 in the polymerase basic protein 2 gene (D701N) that was only present in the fox and seal sequences. Analysis of WGS data obtained from the fox samples demonstrated a lack of other viral agents, including canine distemper virus.

At the time of the outbreak, the rehabilitation center fell within a government-mandated protection zone because of previously confirmed detection of HPAI virus infections; consequently, moving birds off the center was restricted around the time of the disease episode. Restrictions were also applied specifically to the rehabilitation center when suspicion of notifiable disease in the swans was reported. The cases investigated in this outbreak were limited to animals held within the quarantine ward, and over the subsequent months, no further cases of unusual or unexplained clinical disease or death occurred among the mammals and birds at the center. After the disease event, but before diagnosis of H5N8 in the seals and fox, the center was visited as part of surveillance activities within the protection zone. Resident bird species were examined for evidence of clinical disease, and 38 birds (6 swans, 30 mallards, 1 pigeon, and 1 widgeon) were sampled for virologic testing by PCR. All birds tested negative. Blood was not taken for analysis.