Abstract
The European hedgehog (Erinaceus europaeus) is a protected species of conservation concern in the UK. In recent years, there have been multiple incidents of fatal encephalitis in captive hedgehogs in wildlife rescue centers associated with the molecular detection of a hedgehog arterivirus (HhAV-1). However, it remains unclear whether the virus is the causative agent of the central nervous system (CNS) lesions. In a retrospective investigation using postmortem material from 7 captive hedgehogs with neurological disease, and a single hedgehog with previously identified meningoencephalitis, histologic examination was conducted in tandem with viral RNA in situ hybridization (ISH) to appraise tissue distribution of HhAV-1 and the colocalization with histologic lesions. ISH revealed multicellular tropism of HhAV-1 involving monocyte-macrophage and vascular endothelial cells, with viral RNA detected in multiple organs, likely due to endotheliotropism and viremia. In the CNS, encephalomyelitis was mild whilst viral RNA was abundant and widely distributed, particularly in the microglial population and localized to areas with glial nodules. Splenic lymphoid depletion was generally mild but was moderate to severe in 2 septicemic animals. Brain samples from 13 control hedgehogs, found dead in the wild due to predation/trauma, were also screened for HhAV-1, of which 8 tested positive by real-time reverse transcription polymerase chain reaction (RT-PCR) with a low viral load. No CNS lesions or ISH labeling was observed in 2 of these control hedgehogs that could be examined histologically. Combined, these findings indicate that HhAV-1 infections in captive hedgehogs in English wildlife rescue centers may be associated with histopathologic alterations and clinical neurological disease.
Arteriviruses are enveloped, single-stranded positive-sense RNA viruses in the order Nidovirales, family Arteriviridae, which are classified into 6 subfamilies (Crocarterivirinae, Equarterivirinae, Heroarterivirinae, Simarterivirinae, Variarterivirinae, and Zealarterivirinae).3,37 Following an investigation that excluded other viral, fungal, or protozoal etiologies, a novel hedgehog arterivirus (HhAV-1) was detected in juvenile and adult European hedgehogs (Erinaceus europaeus), which developed neurological clinical signs in care at a wildlife rescue center in Gloucestershire, South West England, in autumn/winter of 2019–2020. 8 Preliminary histological findings included moderate-to-severe multifocal gliosis, predominantly in the forebrain and hindbrain, and mononuclear meningitis, renal tubular proteinosis, and splenic lymphoid depletion. 8 Additionally, HhAV-1 was retrospectively detected in hedgehogs from 2 further incidents from wildlife rescue centers in Gloucestershire and Buckinghamshire, during the autumn-winter seasons of 2012–2013 and 2017–2018, respectively. HhAV-1 is a newly identified virus that has only been reported in England and the viral pathogenesis remains unclear.
In this retrospective case-control study, the virological and histologic investigation focused on the aforementioned 3 incidents in England (2012–2013, 2017–2018, and 2019–2020). This included 7 captive hedgehogs from 2 wildlife rescue centers with histories of neurological clinical signs, and a single previously reported case 36 with histologic evidence of meningoencephalitis. Additionally, 13 free-living hedgehogs found dead as a result of predation or trauma were included as controls. To elucidate the potential link between HhAV-1 and the manifestation of neuropathology, and to understand the systemic virus distribution and other possible viral-associated lesions in hedgehogs, correlative histologic examination with in situ hybridization (ISH) using RNAScope was conducted on hedgehogs presenting with and without clinical evidence of neurological disease.
Materials and Methods
Case Selection and Postmortem Examinations
These cases were retrieved from the archives of a national scanning surveillance program (www.gardenwildlifehealth.org), which collates reports of free-living hedgehog morbidity and mortality solicited from members of the public across Great Britain. Clinical cases comprised hedgehogs that died or were euthanized in captivity following hospitalization in wildlife rescue centers and were subsequently submitted for diagnostic investigation. These cases were further defined by the presence of neurological clinical signs or histologic evidence of meningoencephalitis of undetermined etiology, as documented in clinical and laboratory records,18,36 and subsequently tested positive for HhAV-1 by real-time reverse transcription polymerase chain reaction (real-time RT-PCR). A control group of 13 free-living hedgehogs was selected for the purpose of virus surveillance and comparison of infection status by real-time RT-PCR, for which the cause of death was established as predation/trauma, and no concurrent evidence of substantial infectious disease was detected on postmortem examination.
Hedgehog carcasses were subjected to comprehensive postmortem examinations,12,21 with systematic examination of body systems supported by microbiological and parasitological examinations as routine, and histologic examinations conducted in a subset of cases based on antemortem and gross findings. A suite of frozen and formalin-fixed tissues was archived for future study where the state of carcass preservation allowed.
Next-Generation Sequencing and Real-Time RT-PCR
Next-generation sequencing (NGS) and real-time RT-PCR were conducted as described previously. 8 Briefly, RNA was extracted from each sample of frozen brain tissue using TRIzol reagent (Thermo Fisher Scientific) and QIAamp viral RNA kit (Qiagen) and processed individually for sequencing on an Illumina NextSeq platform. NGS data assembly was performed using the HhAV-1 sequence (GenBank accession number MT415062.1) as referenced and also through de novo assembly using the SeqMan Pro 17.4 software of the DNASTAR Lasergene Core Suite (DNASTAR, Inc., https://www.dnastar.com). QuantiFast Pathogen RT-PCR kit (Qiagen) was used for the real-time RT-PCR as described previously. 8
Histology and Chromogenic ISH by RNAScope
Histologic examination combined with ISH was conducted on HhAV-1 PCR-positive hedgehogs with clinical or histologic evidence of neurological disease, using a range of available organs where the state of tissue preservation permitted meaningful interpretation. Where necessary to aid in the diagnosis, a selection of tissue sections was also stained using the Gram Twort method for bacteria. Histologic examination, combined with ISH, was conducted on brain tissue from 2 of the 13 control cases due to financial constraints and the state of carcass preservation. In addition, brain from a single PCR-negative control case was included to screen for non-specific labeling.
Three sets of 20-pair double Z proprietary RNA probes targeting open reading frame (ORF) 6 and ORF7 of HhAV-1 were designed and produced by Advanced Cell Diagnostics, Inc. The RNA probes were designated as probe A (nucleotide position 12701–13707, HhAV-1 genome accession number MT415062.1), probe B (nucleotide position 12523–13529), and probe C (nucleotide position 12523–13529) in this report. These RNAscope probes target ORF6 and ORF7 of HhAV-1 to maximize the sensitivity of RNAScope detection, as these ORFs encoding membrane and nucleocapsid proteins, respectively, are relatively conserved, and are the most abundantly expressed among arteriviruses. 28 Additionally, the HhAV-1 genome underwent in silico evaluation through alignment with the European hedgehog reference genome (mEriEur2.1, GCA_950295305.1). There was no nucleotide alignment detected between the virus genome and the host genome.
ISH was performed using the RNAScope 2.5 HD Brown Detection Kit (Advanced Cell Diagnostics, Inc.) as per the manufacturer’s instructions, which were previously reported. 22 Formalin-fixed, paraffin-embedded tissue sections of 4 μm were collected onto positively-charged slides and were dewaxed and hydrated through xylene and alcohol, respectively. This was followed by treatment with RNAscope hydrogen peroxide for 10 minutes at room temperature, and heat-mediated retrieval using Target Retrieval Solution for 15 minutes at 100°C and Protease Plus for 30 minutes at 40°C. RNA probes were then added to sections to allow hybridization for 2 hours at 40°C followed by 6 rounds of amplification with Hybridize Amp at 40°C and at room temperature, alternating between 30 and 15 minutes incubation, in the HybEZ oven. Slides were washed with 2× wash buffer for 2 minutes at room temperature between incubations. The signal was detected using 3,3’-diaminobenzidine chromogen. Sections were counter-stained with Mayer’s hematoxylin (Surgipath), dehydrated in ethanol and xylene, and glass coverslips were mounted with DPX mounting medium (TCS Biosciences Ltd.). Contiguous serial sections from the same tissues were also stained with hematoxylin and eosin according to standard protocols.
Results
NGS and Probe Design
Consensus HhAV-1 genome sequences obtained through NGS were utilized to develop probes for ISH (Supplemental Table S1). Three sets of probes (A, B, and C), designed to target the diversity of HhAV-1 genome sequences identified through NGS, ensured maximum nucleotide identity with their targets. The HhAV-1 ORF6 and ORF7 sequences, used for probe sets B and C, had nucleotide identities of 81% and 86%, respectively, with those of the published HhAV-1 sequence (accession number MT415062.1) or probe set A. The nucleotide sequences of the 3 ISH probes were compared to the sequences obtained from each animal to identify the highest percentage of nucleotide identity. Subsequently, only the set of RNA probes with the highest nucleotide identity to the cognate virus sequence, ranging between 86% and 100% nucleotide identity, was used for each affected animal. Probe B was applied to control animals, as it has a wide range of nucleotide identity (81%–99%) to the HhAV-1 detected in this study.
History and Clinical Findings
Eight captive hedgehogs with neurological disease that tested PCR positive for HhAV-1 were included in this study. These animals were from 3 morbidity and mortality incidents reported from 2 wildlife rehabilitation centers in Gloucestershire (center A) and Buckinghamshire (center B) across separate years: autumn-winter 2012/2013 (incident 1, center A), 2017/2018 (incident 2, center B), and 2019/2020 (incident 3, center A) 8 (Supplemental Table S2). Neurological clinical signs were a consistent feature of affected animals in incidents 2 and 3, affecting between 100 and 200 animals, including both juveniles and adults. In incident 2, cases 3, 4, and 5 (juveniles, 2 female and 1 male; Table 1) developed neurological signs approximately 4 weeks following hospitalization. Clinical signs noted included paddling movements in all limbs, an inability to stand or walk, and an inability to feed. The clinical course of animal 2 (adult male) from incident 2 also involved the development of paddling and spasmodic movements prior to succumbing to its condition. In incident 3, cases 6, 7, and 8 (juveniles) developed clinical disease between 1 week (case 7, female) and 7 weeks (cases 6 and 8, male) following hospitalization, deteriorated rapidly over the course of 3–4 days, and subsequently developed neurological signs, including tremors, twitching, ataxia/paresis, falling to the sides, paddling when laterally recumbent, and hyperesthesia. The animals were inappetent, dehydrated, and had lost weight. In contrast, clinical signs were varied and non-specific in incident 1, from which the female juvenile hedgehog with histologic evidence of encephalitis was retrospectively identified (Supplemental Table S2). This incident involved multiple mortalities of juvenile hedgehogs, with animal 1 being euthanized after minimal improvements in its condition despite 2 to 3 months of hospitalization. Real-time RT-PCR conducted on brain samples yielded cycle threshold (Ct) values ranging from 19.13 to 24.43.
European hedgehogs (Erinaceus europaeus) evaluated in this study.
Previously tested positive for cryptosporidiosis 36 and gammaherpesvirus 18 infection with histologic evidence of non-suppurative meningoencephalitis.
Previously tested positive for HhAV. 8
N/A = not applicable.
Control cases were included in this study (total n=13); 9 from road traffic accidents, 3 with predation as the proximate cause of death, and 1 with an undetermined cause of death, out of which 8 tested positive for HhAV-1 by PCR with Ct values ranging from 29.8 to 36.2 (Supplemental Table S3). Histologic examination and ISH were conducted on 2 of the PCR-positive control cases (cases 9 and 10) where formalin-fixed brain tissue was available (Table 1).
Histologic Findings in Hedgehogs With Neurological Disease
Central nervous system
In the cerebrum (Table 2) of cases that were not complicated by bacteremia (cases 1, 2, and 6–8; n=5), there was mild to moderate meningoencephalitis (n=5/5), characterized by multifocal glial nodules (n=4/5) (Fig. 1a), scattered lymphocytes and neutrophils in the neuropil (n=2/5), and infrequent lymphocytic perivascular cuffing (n=2/5). These changes were randomly distributed across both gray and white matter. There was no evidence of neuronal necrosis, neuronophagia, myelin degeneration, or spongiosis. Where choroid plexus and meninges were present (cases 3 and 7), the stroma was infiltrated by a small population of lymphocytes and a subset of these cells were labeled for viral RNA.
Summary of histopathology and hedgehog arterivirus 1 RNA in situ hybridization (ISH) labeling in European hedgehogs (Erinaceus europaeus).
The hedgehogs examined include those with (cases 1–8) and without (cases 9 and 10) evidence of neurological disease.
Empty cell indicates tissue not available for examination.
Histopathology score (left column): –, absent; +, mild; ++, moderate; +++, severe.
Abundance of viral RNA labeling (right column): –, absent; +, rare; ++, scattered; +++, moderate multifocal; ++++, abundant.
Abbreviations: ND, viral RNA in situ hybridization was not performed on the cerebrum of animal 2 as the material was exhausted; Ct, cycle threshold.
Histologic evidence of bacteremia, with Klebsiella pneumoniae spp pneumoniae isolated from animal 3.
Neuropathology associated with hedgehog arterivirus 1 (HhAV-1) infection. (a, b) Cerebrum. (a) There are multifocal glial nodules in the neuroparenchyma with (b) localization of viral RNA frequently in the glial cells (solid arrow) and occasionally within the vascular endothelium (open arrow). (a) Hematoxylin and eosin (HE), (b) HhAV-1 in situ hybridization (ISH). (c, d) Cerebellum. (c) Necrosis within the molecular layer characterized by clustering of cell debris (arrow). HE. (d) Serial sections reveal the presence of viral RNA in glial cells (solid arrow), vascular endothelium (open arrow), and meningeal cells. HhAV-1 ISH. (e, f) Spinal cord. (e) There are occasional shrunken and angular neurons (solid arrow) and vacuolated neuropil (open arrow). HE. Inset: viral RNA is multifocally distributed in both gray and white matter. (f) Viral RNA is labeled in glial cells. HhAV-1 ISH. Viral probes A (b) and B (d–f) were used for labeling.
In 3 animals where bacteremia was suspected (case 5; no other visceral tissues were available for examination) or confirmed in other organs (cases 3 and 4; also see description within the sections of lymphoid system, kidney, and liver), occasional vascular thrombosis was observed (n=3/3) in the cerebral neuroparenchyma, with 1 case (animal 3) also showing intralesional gram-negative coccobacillary emboli. Similarly, these animals also presented with multifocal glial nodules (n=2/3) and scattered lymphocytes and neutrophils in the neuropil (n=2/3).
ISH labeling performed on the brain sections (n=7, material was exhausted for case 2) revealed multifocal scattered to abundant viral RNA labeling in both glial and vascular endothelial cells (n=5/7), with a greater proportion of labeling present in glial than endothelial cells (Fig. 1b). The abundance of viral RNA labeling or cell tropism did not differ between bacteremic and non-bacteremic animals. In a small number of cases (n=2/7) only vascular labeling was detected, including a case that was complicated by bacteremia (case 3). Nevertheless, the viral RNA-positive vascular endothelial cells were histologically unremarkable on correlative evaluation. On the contrary, glial labeling was often colocalized with areas of gliosis but was also present in areas where gliosis was not apparent on hematoxylin and eosin-stained sections.
In the cerebellum of cases that were not complicated by bacteremia, there was rare glial nodule formation (n=2/5) and perivascular lymphocytic cuffing (n=2/5). In the bacteremic cases, there were scattered glial nodules, neutrophilic infiltrates, or necrotic cellular debris (Fig. 1c) in the molecular layer or cerebellar peduncles (n=2/3); endothelial hypertrophy (n=1/3); and lymphocytic or neutrophilic infiltration of the meninges (n=2/3). Concurrent glial and vascular endothelium labeling was observed in the majority of the cases (n=5/7), with a predominance of glial labeling over vascular endothelium (Fig. 1d). In isolated cases, there was either glial (case 6, n=1/7) or vascular endothelial (case 1, n=1/7) labeling.
Out of the 4 cases in which the brainstem was examined, there was only 1 case uncomplicated by bacteremia, in which histology revealed rare glial nodules or lymphocytic perivascular cuffing. Two of the 3 animals that were complicated by bacteremia also exhibited similar glial nodules and lymphocytic perivascular cuffing, and 1 animal was histologically unremarkable. Concurrent glial and vascular endothelial labeling was observed in these cases, with the abundance ranging between rare to multifocal and moderate.
In the 4 spinal cords examined, there was mild, multifocal gliosis in the gray matter, with occasional dispersed neutrophils within the neuropil (n=3/4) in both bacteremic (cases 3 and 4) and non-bacteremic (cases 2 and 5) cases (Fig. 1e). The viral RNA labeling was multifocal, with the involvement of both the gray and white matter (Fig. 1f). In 1 of the non-bacteremic animals (case 4), there was mild neuronal degeneration and vacuolated neuropil (Fig. 1e), which colocalized to areas with glial labeling (case 2, n=1/4; Fig. 1f). Hypertrophied vascular endothelium was infrequently detected in a bacteremic animal (n=1/4).
In summary, the brain and spinal cord consistently presented with glial nodules along with infrequent lymphocytic and neutrophilic infiltration in the neuropil, with similar labeled cell types and abundance of viral RNA, regardless of the presence of bacteremia.
Spleen
Of the spleens examined (n=7), non-bacteremic cases (n=5/7) exhibited mild to moderate lymphocytolysis (Fig. 2a). This was characterized by depletion of the white pulp, increased apoptotic bodies and tingible body macrophages in the germinal centers (Fig. 2b), and scattered fibrin deposits within the red pulp. In contrast, lymphocytolysis was marked in 2 bacteremic animals (cases 3 and 4; intralesional gram-negative coccobacillary emboli and isolation of Klebsiella pneumoniae spp. pneumoniae, case 3). No bacteria was isolated from the other hedgehog, possibly as a consequence of antimicrobial treatment in care.
Splenic, hepatic, and renal lesions associated with hedgehog arterivirus 1 (HhAV-1) infection. (a, b) Spleen. (a) Lymphoid depletion of the white pulp (arrow). Hematoxylin and eosin (HE). Inset: abundant viral RNA in the marginal zone and rarely in the T lymphocyte zone. HhAV-1 in situ hybridization (ISH). (b) Cells within the germinal center are sparse with widespread pyknosis and karyorrhexis (arrowheads), intermingled with large numbers of tingible body macrophages. HE. (c, d) Liver. (c) Multifocal fibrinohemorrhagic and necrotizing hepatitis (inset, HE) with dispersed viral RNA labeling in areas of necrosis (arrow) along with periportal labeling. HhAV-1 ISH. (d) Periportal viral labeling in histologically unremarkable liver. HhAV-1 ISH. (e, f) Kidney. (e) Renal cortical infarct (inset) with the edge of the infarct (dotted outline; inset, HE) exhibiting moderate viral RNA within the renal interstitium. HhAV-1 ISH. (f) Presence of pale eosinophilic fluid, neutrophils, histiocytes, and sloughed epithelium within a dilated renal calyx. HE. Inset: viral RNA localized within the cellular exudate. Viral probes A (d) and B (a, c, e, and f) were used for labeling.
In all spleen samples, viral RNA was generally abundant and densely labeled the marginal zone of the white pulp (n=7/7; Fig. 2a, inset). Additionally, a moderate amount of viral RNA was detected in histiocytes and dendritic cells within the parafollicular areas or scattered in the red pulp. Extramedullary hematopoiesis was observed in all of the spleen samples. Extramedullary hematopoiesis is common in hedgehogs as an incidental finding in the spleen 45 and was not colocalized with viral RNA labeling.
Lymph node
Three lymph nodes (location not specified) were available for examination from cases 1, 3, and 4. Moderate to severe necrotizing lymphadenitis with intralesional gram-negative coccobacillary emboli was detected in 2 bacteremic cases (cases 3 and 4). Viral RNA was present in histiocytes and/or dendritic cells in both the cortex and medulla and in the germinal centers with a higher density of labeling in the mantle region. The lymph node from the non-bacteremic animal was histologically unremarkable, and viral RNA was rarely detected in the capillaries of both the cortex and medulla.
Liver
Of the 6 livers examined, the 2 bacteremic animals (n=2/6, cases 3 and 4) exhibited moderate to severe, multifocal to coalescing, random, coagulative hepatic necrosis, with thrombosed sinusoids and intralesional gram-negative coccobacillary emboli. While there was a periportal labeling of Kupffer cells, there was also widespread viral RNA labeling colocalized to areas of necrosis (Fig. 2c).
In histologically unremarkable liver (n=4/6), there were small to moderate amounts of extramedullary hematopoiesis within the hepatic sinusoids (n=3/6; considered an incidental finding in hedgehogs 45 ), but this was not associated with viral RNA labeling. A moderate amount of viral RNA with a periportal distribution was detected (Fig. 2d), with most labeling present in Kupffer cells and rarely within the endothelium of central veins.
Kidney
Among the 6 kidneys examined, 4 cases were generally unremarkable histologically, with mild tubular lipidosis and proteinuria. Two bacteremic cases (cases 3 and 4, n=2/6) exhibited moderate to severe, multifocal to coalescing, acute, tubulointerstitial nephritis with cortical infarctions (Fig. 2e, inset). Additionally, gram-negative coccobacillary emboli were detected in 1 of these 2 animals (case 3). The amount of viral RNA labeling was moderate to abundant, present extensively within and at the edge of the infarcted areas, predominantly in the interstitial vessels and infrequently in renal tubules (Fig. 2e). One of the cases had dilated renal calices and within the lumen there were moderate amounts of degenerate neutrophils, macrophages, and sloughed renal tubular epithelia, within which a subset of cells was viral RNA positive (Fig. 2f). In non-bacteremic cases, viral RNA detection was rare to scattered, present in the capillaries of the intertubular spaces and rarely within the glomerular tufts.
Lung
Four of the lungs were histologically unremarkable (n=4/7). The other 3 animals exhibited moderate to severe interstitial pneumonia with concurrent bacterial emboli in 2 animals (cases 3 and 4, in which bacterial emboli were also found in the spleen, lymph node, liver, and kidney), and intrabronchial nematode infestation in 1 animal (case 4, presumptively Crenosoma striatum). The interstitial lesion comprised of moderate multifocal expansion of alveolar walls with edema, fibrin deposits, degenerating neutrophils, and erythrocytes, and occasionally the alveolar spaces were lined with hyaline membranes (Fig. 3a). In cases with intralesional gram-negative coccobacillary emboli, there were perivenular hemorrhages, detachment of endothelial cells, expansion of the tunica media of pulmonary arterioles with edema fluid, and lymphocytic cuffing. Viral RNA labeling was detected in all 7 animals with lung available for examination. This ranged from scattered to multifocal and was mostly present in the alveolar wall, with cellular morphologies suggestive of type I pneumocytes, capillary endothelium, or interstitial macrophages (Fig. 3a, inset). Infrequently, there was labeling in the bronchial submucosal capillaries.
Pulmonary, cardiac, and alimentary tract lesions and viral labeling associated with hedgehog arterivirus 1 (HhAV-1) infection. (a) Lung. Interstitial pneumonia. Hematoxylin and eosin (HE). Inset: viral RNA present in the alveolar wall. HhAV-1 in situ hybridization (ISH). (b) Heart. Necrotizing myocarditis (arrowhead) with a bacterial embolus (arrow). HE. Inset: localization of viral RNA within the capillaries and venules. HhAV-1 ISH. (c) Tongue. Histologically unremarkable tissue with rare multifocal submucosal labeling. HhAV-1 ISH. (d) Ileum. Depletion of Peyer’s patches (arrowheads). HE. Inset: viral RNA localized within the margin of the germinal centers and scattered labeling within the capillaries of the lamina propria. HhAV-1 ISH. Viral probe B (a-d) was used for labeling.
Heart
Most cases were histologically unremarkable (n=5/6) apart from 1 animal with severe multifocal necrotizing myocarditis with intralesional gram-negative coccobacillary emboli (case 3, Fig. 3b). These areas were colocalized with abundant viral RNA labeling of the capillaries. In other histologically normal hearts, viral RNA labeling was rare to scattered in the capillaries of the myocardium (n=4) or absent (n=1). Occasionally, bradyzoites were encountered (Sarcocystis spp., considered an incidental finding) in the heart but were not associated with reactive changes or viral RNA.
Digestive tract
The stomach and tongue (Fig. 3c) from 1 animal (case 2) were available for examination and were histologically unremarkable but contained rare viral RNA labeling in the submucosal capillaries. In the ileum of the same animal, there was moderate lymphoid depletion of the Peyer’s patches with the presence of tingible body macrophages (Fig. 3d). Strong viral RNA labeling was present in the mantle and marginal zone of the Peyer’s patches, in predominantly large mononuclear cells (presumptive dendritic or antigen-presenting cells) and scattered labeling was also present among the cellular debris within the germinal centers (Fig. 3d, inset). In the intestinal (case 2) and colonic (case 3) lamina propria, there was scattered viral RNA labeling of mononuclear cells. While the intestine of case 1 was not further investigated as the paraffin block was exhausted, the initial histology workup revealed localized jejunal cryptosporidiosis.
Adrenal gland
The adrenals evaluated (n=3) were histologically unremarkable. Viral RNA was rare to scattered and was randomly distributed in the stromal spaces of the cortex and medulla within the capillaries.
Histologic Findings in HhAV-1 PCR-Positive Control Hedgehogs
Histologic examination of brain identified no substantial lesions and ISH labeling did not reveal viral RNA in situ in the brain of the 2 control hedgehogs, animals 9 and 10 (Table 2, Supplemental Table S3).
Discussion
The in situ detection of HhAV-1 RNA labeling within areas of gliosis in the central nervous system (CNS) of multiple captive hedgehogs that developed neurological disease strongly supports an association between the virus and histologic lesions. While HhAV-1 RNA was also detected in other tissues, including the lung, liver, heart, kidney, spleen, lymph node, and digestive tract, consistent with a multisystemic viral infection, there was a lower abundance of viral RNA labeling and restricted cellular tropism in non-septicemic animals. In contrast, viral labeling in non-neural lesions in septicemic animals was more abundant, and typically more prominent in endothelium and necrotized areas. However, within the constraints of the small sample size, the potential for extraneural lesions to be induced directly by HhAV-1 alone cannot be fully elucidated. Experimental in vivo studies would be required to fulfill Koch’s postulates and definitively prove causation. However, the authors have currently been unable to obtain ex-vivo isolates of HhAV-1, and as a protected species in the United Kingdom experimental work to establish a colony of captive hedgehogs of consistent and previously known health status is unlikely to be feasible.
Primary neurological disease associated with arterivirus infection is relatively uncommon. The wobbly possum disease virus is the only virus known to consistently cause neurological disease in the Australian brushtail possum (Trichosurus vulpecula), both naturally and experimentally.13,23,29 The histological lesion in the CNS is characterized by mononuclear perivascular cuffing, but the viral pathogenesis remains unclear as the viral tropism has not yet been characterized.13,23,26,29 Lactate dehydrogenase elevating virus, another arterivirus, can cause encephalomyelitis or radiculoneuritis in laboratory mice by targeting the dorsal root ganglion neuronal cells.39,40 In large animal species, arterivirus-associated neurological diseases have been sporadically reported, such as highly pathogenic porcine reproductive and respiratory syndrome virus (PRRSV)-associated meningoencephalitis in pigs5,16,35 and aborted equine fetuses infected with equine arteritis virus (EAV). 19 PRRSV-associated meningoencephalitis in pigs demonstrates viral tropism for the monocyte-macrophage cell lineage within the brain,5,16,35 and the infection of microglial cells upregulates the expression of proinflammatory cytokines and chemokines. 6 Whilst the hedgehogs with observed disease included in this study only exhibited mild meningoencephalitis associated with HhAV-1 infection, cytokine release from microglial infection could potentially contribute to the bio- and/or neurochemical dysfunction of neurons and the apparent substantial neurological clinical impact.
Viral tropism for the monocyte-macrophage cell lineage, particularly in lymphoid tissues, is a common biological feature shared across arteriviruses including PRRSV, EAV, and simian hemorrhagic fever virus.15,34,42,44 Infection of the lymphoid tissues often results in viral-associated lymphoid depletion and/or necrosis.14,16,20,24,26,42 This can result in predisposition to secondary bacterial infection, which has been reported in pigs naturally or experimentally infected with PRRSV,31,38,41,46 and rhesus macaques (Macaca mulatta) experimentally inoculated with simian hemorrhagic fever virus. 20 In the current study, spleens from 7 animals were examined of which all presented with lymphocytolysis. Two further diseased hedgehogs (cases 3 and 4) that exhibited splenic lymphocytolysis had concurrent bacteremia (Klebsiella pneumoniae spp. pneumoniae was isolated from case 3), with ischemic and/or thrombotic lesions detected in the brain, heart, lung, kidney, spleen, and lymph node, consistent with septicemia. Given the small sample size in this study, definitive roles for viral or bacterial-mediated immunosuppression cannot be fully determined or excluded. Additionally, elevated corticosterone in captive hedgehogs 33 could be an additional predisposing factor to immunosuppression and/or activation of HhAV-1 infection. While the hypothesis of viral-mediated immunosuppression remains to be elucidated, the concurrent viral and bacterial infections highlight the need for strict biosecurity and sanitation measures to prevent the acquisition of other infections in the wildlife center environment.
Endotheliotropism was another common feature of HhAV-1 infection in the hedgehogs. Infected endothelial cells were generally unremarkable, rarely exhibited hypertrophy, and had no direct evidence of hemorrhage or edema. Contrastingly, in the subset of cases with concurrent bacteremia (cases 3 and 4) there was endothelial hypertrophy, thromboembolism, and infarctive coagulative necrosis in the brain, kidney, liver, spleen, and lymph node. Although the mechanism remains unclear, the prominent vascular labeling observed in bacteremic cases may be attributed to the recruitment of viral-infected immune cells or direct infection of the vascular endothelium. Based on this small number of cases, it seems unlikely that HhAV-1 contributes to a primary vasculopathy or vasculitis. Nevertheless, endotheliotropism remains an important feature in the broader pathogenesis of arteriviruses, facilitating viremia and the dissemination of the virus to various organs.
In the lung, RNA labeling was observed within the alveolar wall, with type I pneumocytes, capillary endothelial cells, and interstitial macrophages considered infected. Presumptive epitheliotropism of HhAV-1 was not detected in tissues other than the lung. While the potential for epitheliotropism cannot be conclusively disregarded, given the precedence of PRRSV and EAV in respiratory secretions,30,42 further confirmation will require multilabeling to elucidate the disease pathogenesis. As for the pathologic implication of epitheliotropism, 3 out of 7 animals demonstrated interstitial pneumonia, 2 of which had concurrent septicemia; therefore, such histologic lesions seem unlikely to be related to a primary HhAV-1 pulmonary infection.
In contrast to incidents 2 and 3, there were no consistent neurological signs in incident 1, with reported signs being non-specific (e.g., lethargy, inappetence) and varied (e.g., diarrhea, respiratory signs). Upon histologic examination of animal 1, non-suppurative meningoencephalitis was observed, which prompted its inclusion in the study to investigate for a potential viral etiology. ISH revealed the presence of viral RNA in the neuropil colocalized with areas of inflammation, and lymphoid depletion was observed in the spleen and lymph nodes in areas with viral RNA. Additional findings in animal 1, as well as in other hedgehogs that were investigated by gross postmortem examination from incident 1, included verminous pneumonia and intestinal cryptosporidiosis (Animal and Plant Health Agency, United Kingdom, unpublished data), which are common findings in hedgehogs admitted for care32,36 that may have further influenced the disease state. It was not possible to determine the extent to which HhAV-1 contributed to the ill-health of animal 1 in isolation, or whether HhAV-1 infection was a common feature affecting other hedgehogs in incident 1. No other animals from incident 1 had suitable material for inclusion in this study.
The mode of transmission of HhAV-1 remains unclear. Histologic analysis revealed the presence of viral RNA in cell exudates within the renal calyx and potentially type I pneumocytes. As virus-containing bodily fluids are known to enable transmission of other arteriviruses, respiratory secretions or urine can be hypothesized as media for HhAV-1 transmission.1,30 In addition, arteriviruses can cause persistent infections in some hosts7,11,25,43; therefore, it is possible that free-living hedgehogs may act as HhAV-1 carriers and the origin of infection. However, because the hedgehogs involved in disease outbreaks in the wildlife rescue centers were, at least on occasion, cohoused in the same air space as other species (e.g., birds, rabbits, rodents), 8 and the onset of neurological disease subsequent to casualty admission supported nosocomial infections in incidents 2 and 3, the possibility of viral origin in another species with spill-over of infection to hedgehogs in captivity cannot be excluded.
The large-scale hedgehog morbidity and mortality that occurred in the 3 incidents presented in this study prompted an extensive investigation to determine the etiology. There are multiple differential diagnoses for neurological disease in hedgehogs, which may result from septicemia or encephalitis, such as Salmonella Enteritidis, Listeria monocytogenes, or herpesvirus infection,17,18,21 as well as the so-called wobbly hedgehog syndrome 10 or idiopathic paralysis syndrome. 27 While gliosis was consistently detected microscopically in both uncomplicated and septicemia-complicated cases across the 3 separate incidents, the absence of macroscopic lesions and non-specific histologic findings in non-CNS tissues suggests a possible scenario where previous historical infections and associated diseases were not identified. Future investigations of neurological disease in European hedgehogs should comprise a suite of ancillary diagnostic tests to identify primary infections and potential comorbidities.
For hedgehogs under care in wildlife rescue centers, establishing clinical samples to use for virus detection would be useful. Since infected hedgehogs can be presumed viremic, as evidenced by endothelial and monocyte-macrophage infection, testing of blood samples may be considered. Additionally, in 1 of the HhAV-1 cases in this study, viral shedding into the renal calyx was observed. Therefore urine may be tested for HhAV-1 as has been suggested for EAV diagnosis. 9 While the possibility for viral shedding into the respiratory tract is yet to be determined, the potential of type I pneumocyte infection in hedgehogs, and understanding of PRRSV and EAV shedding in respiratory secretions, warrants the evaluation of HhAV-1 in nasopharyngeal swabs as a further diagnostic specimen. During postmortem examinations, brain, lung, liver, and spleen should be collected for virological testing to screen for HhAV-1 infection, based on the observed distribution of viral RNA by ISH in the diseased hedgehogs. Additionally, considering the potential for immunosuppression induced by HhAV-1 infection and/or concurrent bacterial infections, bacteriological testing of these tissues is recommended.
Virological testing of the control hedgehogs using real-time RT-PCR revealed a substantial proportion of animals identified as positive for HhAV-1 (n=8/13), with Ct values ≥29.8. While histologic examination of brain was only possible for 2 HhAV-PCR-positive control animals due to the lack of availability of well-preserved specimens, CNS lesions and viral RNA labeling were absent in these control animals. On the contrary, the hedgehogs with neurological disease had lower Ct values (19–24.4), suggesting higher viral loads in clinical cases. Although a detailed phylogenetic analysis will be reported separately, the arterivirus sequence obtained from neurological hedgehogs displayed wide nucleotide variation within ORF6 and ORF7 (Supplemental Table S1). Currently, it remains unclear whether the HhAV-1 detected in control hedgehogs is different from those found in hedgehogs with neurological disease. Further molecular characterization is required to address this information gap; however, the low viral RNA load in control animals presented a significant technical challenge for NGS and viral phylogenetic analysis. Given the occurrence of RT-PCR-positive control animals, further virological investigation along with correlative microscopic investigation with virus ISH is warranted to understand the infection epidemiology.
In this study, 3 animals from incident 2, specifically cases 3–5, exhibited histologic evidence of sepsis. Case 4 also had evidence of bacteriological infection, with isolation of Klebsiella pneumoniae. These findings may indicate that these animals acquired nosocomial infections. However, significant virus RNA labeling was detected, colocalizing with lesions within the liver, kidney, and spleen. The absence of similar lesions in other non-septicemic animals suggests that the increased viral RNA labeling may be attributed to the recruitment of infected immune cells, rather than a primary viral-mediated vascular disease.
The assessment of virus tropism represented an additional challenge in this study as the interpretation of infected cell populations relied on correlative assessment through conventional light microscopy. In cases of strong and abundant ISH signal, determining the specific cell population can be difficult. Future studies should incorporate colabeling with host cell markers for immunological confirmation of the identified cell populations or ultrastructure examination by transmission electron microscopy to better understand virus tropism.
In summary, this study supports HhAV-1 as a potential cause of outbreaks of fatal neurological disease in hedgehogs in English wildlife rehabilitation centers. Since hedgehogs are of current conservation concern in Great Britain and are admitted to such centers in large numbers each year, 2 there is an urgent need for further research to understand the viral epidemiology and impact. To mitigate the risks of inter- and intraspecific pathogen transmission, strict biosecurity measures (e.g., hygiene, quarantine) during rehabilitation are recommended as a routine, combined with husbandry protocols to minimize duration and stress in captive management.4,33
Supplemental Material
sj-pdf-1-vet-10.1177_03009858241300553 – Supplemental material for Colocalization of hedgehog arterivirus 1 (HhAV-1) and histologic lesions in the European hedgehog (Erinaceus europaeus) with neurological disease
Supplemental material, sj-pdf-1-vet-10.1177_03009858241300553 for Colocalization of hedgehog arterivirus 1 (HhAV-1) and histologic lesions in the European hedgehog (Erinaceus europaeus) with neurological disease by Fabian Z. X. Lean, Mark F. Stidworthy, Akbar Dastjerdi, Tim Partridge, Stephen Smith, Julie Gough, Alejandro Núñez, Becki Lawson and Katharina Seilern-Macpherson in Veterinary Pathology
Footnotes
Acknowledgements
We greatly appreciate assistance from members of the public who reported to the Garden Wildlife Health project and who took casualty hedgehogs to the wildlife centers for care. Our thanks extend to the wildlife rehabilitation center staff for their valuable assistance in hedgehog care and information sharing. We also extend our thanks to Alex Barlow for sharing information on historical hedgehog disease investigations conducted at APHA. We also express our gratitude to Audra Lynne-Schlachter for her assistance in editing the images. Finally, we would like to acknowledge the technical support provided by Hannah Davies and Nadia Inglese, pathology scientists at APHA, the microbiologists at IoZ, and the technical staff at IZVG.
Author Contributions
F.Z.X.L., M.F.S., B.L., and K.S.M. contributed to conceptualization. J.G and A.D. contributed to methodology. F.Z.X.L., M.F.S. K.S.M., T.P., and S.S. conducted formal analysis. B.L. and A.N. provided project leadership, financial, and laboratory resources. F.Z.X.L. wrote the original draft. All authors reviewed and edited the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was partly funded by the Department of Environment, Food and Rural Affairs (DEFRA) and The Welsh Government through the “Scanning Surveillance for Wildlife Diseases in England and Wales” (ED1600) project at APHA. Financial support for the Garden Wildlife Health project comes in part from the DEFRA, the Welsh Government, and the APHA Diseases of Wildlife Scheme Scanning Surveillance Program (ED1058); and from the Garfield Weston Foundation and the Universities Federation for Animal Welfare. BL receives financial support from Research England.
Ethical Approval
Sick and injured hedgehogs were taken into care for welfare reasons, and died or were euthanized under the Veterinary Surgeons Act 1966. Samples from wild animals were found dead in the wild prior to postmortem and, no ethical approval was required.
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
