Ferret Systemic Coronavirus in Alpha-1 Antitrypsin Knockout Ferrets

Abstract

Ferret systemic coronavirus (FRSCV) causes a highly fatal disease of ferrets (Mustela putorius furo). It is believed to be a mutated variant of ferret enteric coronavirus (FRECV) and has a clinical presentation similar to that of feline infectious peritonitis virus (FIPV) in cats. The interplay of infectious diseases and host genetics will become a greater issue in the research environment as genetically modified species other than rodents become available due to advances in gene editing technology. In this case series, we present the clinical and histopathologic features of a FRSCV outbreak that affected 5 out of 10 ferrets with α-1 antitrypsin knockout (AAT KO) over an approximately 1-y period. Clinical features varied, with the affected ferrets presenting with some combination of wasting, hind limb paralysis, incontinence or sudden death. Multiple ferrets had gross pathologic lesions consistent with FRSCV, but the lesions were typically mild. Microscopic pyogranulomatous inflammation was present in 4 ferrets. Immunohistochemistry using an anti-feline coronavirus antibody that cross reacts with ferret coronavirus confirmed infection of intralesional macrophages in 4 out of 5 animals with suspected FRSCV infection. PCR testing of formalin fixed tissue was negative for all ferrets. PCR testing of feces from healthy wild-type ferrets indicated that the endemic presence of FRECV genotype 2, while PCR surveillance testing of other in-house AAT KO ferrets revealed both enteric coronavirus genotypes 1 and 2. This case series highlights the potential for greater disease incidence in the future as genetically modified ferrets are used more often, and may support exclusion of FRECV and similar viruses from highly susceptible ferret genotypes.

Introduction

Ferrets are a common model for respiratory and infectious disease research due to their unique pulmonary anatomy and susceptibility to a variety of human pathogens.¹ With the advent of the first transgenic ferret in 2006 via somatic cell nuclear transfer and more recent use of CRISPR/Cas9- mediated genomic editing techniques, the importance of ferrets in research model has increased.^9,11 In particular, the recently developed α-1 antitrypsin knockout (AAT KO) ferrets have become a key model for chronic obstructive pulmonary disease and may prove a valuable resource for infectious respiratory disease research as well.^6,7

Ferrets are susceptible to a wide range of coronaviruses; this feature enhances their use as a model but also presents challenges for maintaining health in a research colony. Ubiquitous and endemic to many species, coronaviruses are a group of pleomorphic, large, enveloped, positive-stranded RNA viruses that are highly susceptible to mutation and are classified under the genus Coronavirus within the family Coronaviridae, order Nidovirales. Ferret systemic coronavirus (FRSCV) was first described in 2008 and is considered a group 1 coronavirus.⁵ FRSCV is histopathologically and pathophysiologically similar to feline infectious peritonitis virus (FIPV); both are characterized by pyogranulomatous inflammation.⁵ Ferret coronavirus genotype 1 has been more commonly associated with systemic FIP-like disease in ferrets and has a similar mutation in the spike protein that causes marked macrophage tropism, as reported for FIPV in cats, while ferret coronavirus genotype 2 has been found mainly in ferrets with enteric disease.^17,23 Ferret coronavirus genotype 1 has been found in healthy ferrets and both genotypes have been found in individual ferrets, but few ferrets with systemic disease are infected only by genotype 2.^19,20 Therefore, the relationship between genotype and pathotype of ferret coronaviruses and the mechanisms leading to systemic disease remains unclear.^14,16,19

To expand on the existing literature related to FRSCV, this report describes the clinical features, diagnosis, and significance of a FRSCV outbreak in AAT KO ferrets with a high degree of morbidity and mortality. With some unique clinical features and overall mild histopathologic lesions in most affected ferrets, this condition represents a departure from the more classic presentation described in the literature for wild-type ferrets.⁵

Case Series

Five male and 5 female AAT KO ferrets (4 to 4.5 mo of age) arrived at the facility in good health and were unloaded into a room with resident healthy wild type and Muc5b heterozygous ferrets. After a 1-wk acclimation period, these ferrets were enrolled in a 6-mo chronic smoking exposure experimental paradigm with serial monthly to bi-monthly bronchoalveolar lavage, µCT imaging, and FlexiVent™ measurements commonly in use by the pulmonary research group.¹⁸ Ferrets were observed daily by both veterinary and care staff. Husbandry conditions are described later in this report.

Case 1: Ferret 752.

Approximately 10 d after arrival, 2 naïve AAT KO ferrets (numbers 752 and 749) that were housed in different cages were reported for veterinary evaluation. On initial evaluation, ferret 752 had a poor prognosis due to severe clinical signs consisting of: body weight of 0.9 kg, a body condition score of 1 out of 5, paresis, dehydration, and bilateral white ocular discharge. Ferret 752 (male) was euthanized the same day with 1,170 mg pentobarbital sodium and 100 mg phenytoin (Euthasol, Vedco, Saint Joseph, MO) injected into the heart after sedation with 56.25 mg ketamine (Ketaset, Zoetis, Parsippany-Troy Hills, NJ) and 0.11 mg dexmedetomidine (Dexdomitor, Zoetis) intramuscularly. Terminal serum chemistry findings are shown in Table 1. Gross histology findings included a completely empty gastro-intestinal system and a significantly extended bladder. Histopathology revealed mild perivascular lymphocytic inflammation in the liver and gastrointestinal tract (data not shown). FRSCV immunohistochemistry testing on a formalin fixed mesenteric lymph node was negative.

Table 1.

Clinical chemistry values from 2 AAT KO ferrets with ferret systemic coronavirus infection.

Clinical Chemistry Values	Ferret 752	Ferret 749	Reference ranges
Albumin (g/dL)	4.2	4.6	2.9 ± 0.1
ALP (U/L)	59	45	174 ± 50
ALT (U/L)	189	488	217 ± 174
Amylase (U/L)	24	<5	30 ± 1
TBIL (mg/dL)	0.3	0.2	0.1 ± 0.1
BUN (mg/dL)	169	87	29 ± 5
Calcium (mg/dL)	11.3	10.6	10.0 ± 0.3
Phosphorus (mg/dL)	16.2	10.9	9.0 ± 1
Creatinine (mg/dL)	0.7	1.4	1.1 ± 1.5
Glucose (mg/dL)	232	102	105 ± 12
Sodium (mmol/L)	> 170	153	147 ± 2
Potassium (mmol/L)	8	4.3	4.8 ± 0.2
Total protein (g/dL)	6.3	6.7	5.4 ± 0.2
Globulin (g/dL)	2.1	2.2	2.4 ± 0.2

^**Abnormal values are in bold. Reference ranges (mean ± SD) based on reported data from Marshall Bioresources for Influenza-free Serum Chemistry Data (2016) for male ferrets between 12–20 wk of age (https://www.marshallbio.com/influenza-free-marshall-ferretr). No genotype specific serum chemistry normal ranges are available for AAT KO ferrets.

Case 2: Ferret 755.

Approximately 1 mo after arrival, ferret 755 (female) was noted to have a low body condition score (2/5). Ferret 755 was separated from the cage-mate, which had a high body condition but was otherwise within normal limits, to avoid disease in the healthy animal. Supplemental wet food was provided in the cage daily. Five days later, Ferret 755 was noted to have decreased activity, continued deterioration of body condition (1.5/5; weight 590 g), and mild dehydration. Supportive treatment of subcutaneous fluid administration as needed and daily syringe feeding was provided. Despite these treatments, the ferret continued to lose weight and developed additional clinical signs (urine staining with dermatitis and hyperemia and bilateral hind limb paresis with altered proprioceptive placing).

Due to overall poor prognosis, the ferret was euthanized after a final antemortem data collection procedure conducted 13 d after initial presentation. Lesions were less severe than in Ferret 749 but included granulomas and pyogranulomas in the spleen and mesenteric lymph nodes, panniculitis, and hepatic lipidosis.

Case 3: Ferret 756.

Approximately 3 mo after arrival, ferret 756 (female) was reported to veterinary staff for similar clinical signs: poor body condition score (2/5), weight of 0.5 kg, bilateral hind limb weakness, and peri-vaginal urine staining. Supportive care consisting of subcutaneous fluids as needed and daily syringe feedings was initiated. With no response to supportive care, the ferret was euthanized 12 d after initial presentation. This ferret had evidence of gross disease (not pictured) with the multifocal granulomas of the spleen and disseminated granulomatous and pyogranulomatous infiltrate throughout the upper gastrointestinal tract, pancreas, and mesenteric lymph nodes. FRSCV was confirmed later with immunohistochemistry.

Case 4: Ferret 753.

This ferret (male) was found dead in its cage by husbandry staff approximately 7 mo after arrival. Its cage mate (ET 751) was healthy on veterinary assessment that same day. Approximately 1 mo before death, ferret 753 weighed 1.4 kg. Similar to other cases, necropsy revealed microscopic granulomatous and pyogranulomatous lesions in the spleen, mesenteric lymph nodes, and upper gastrointestinal tract. In addition, like ferret 749 (Figure 1), ferret 753 had intimal arteritis, transmural fibrinoid necrosis, and intraluminal hemorrhage of the mesenteric arteries and stomach submucosa vasculature. FRSCV was confirmed later with immunohistochemistry.

Figure 1.

Histologic lesions indicative of vascular injury in a coronavirus infected α-1 antitrypsin deficient ferret. (A–B) Ferret 749 displayed histologic evidence of vascular inflammation and injury in the mesentery with both (A) intimal inflammation and proliferation and (B) transmural fibrinoid necrosis in small to medium caliber blood vessels in the mesentery.

Case 5: Ferret 749.

Approximately 10 d after arrival, ferret 749 was reported for veterinary evaluation. This male ferret had a similar but less severe presentation as compared with ferret 752. His exam consisted of: lethargy, body weight 1.1 kg, body condition score of 1 out of 5, dehydration (5% estimated), and bilateral white ocular discharge. Serum chemistry findings are shown in Table 1. This ferret received supportive treatment with daily subcutaneous fluids and syringe feeding of high calorie cat food and recovered within 10 d.

Ferret 749 was reported again to veterinary staff approximately 7 mo later for similar symptoms: lethargy, body condition score of 2 out of 5, dehydration, mild hind limb paresis, urine staining in the peri-anal area, mild diarrhea, and a palpable mass like structure in the lower abdomen. The ferret received 2 d of supportive care to allow final antemortem data collection for the primary study and was then euthanized as described above. Supportive care consisted of daily subcutaneous fluids and syringe feeding with high calorie cat food. Gross necropsy of this ferret revealed small irregular tan/red nodules protruding through the spleen (Figure 2A), mesentery and mesenteric lymph nodes (Figure 2B). Microscopic examination revealed multifocal to coalescing pyogranulomatous and granulomatous inflammation involving the spleen (Figure 2C), mesenteric lymph node (Figure 2D), pancreas (Figure 2E), and serosal aspects of the gastrointestinal tract (Figure 2F). FRSCV was confirmed with immunohistochemistry. Periarteritis and transmural fibrinoid necrosis of small to medium caliber vessels was seen in the mesentery (Figure 1A and 1B) in this ferret 749, indicating vascular involvement. Ferret 749 was the index case that prompted reevaluation of the other AAT KO ferret deaths and non-study related early euthanasia.

Figure 2.

Histologic lesions noted coronavirus infected in α1 anti-trypsin deficient ferrets. (A–B) Ferret 749 displayed grossly enlarged (A) spleen and (B) mesenteric lymph nodes with firm, irregular red to tan nodules extending from the capsular surfaces (arrows). (C–F) Histologic evidence of (C) pyogranulomas containing a central core of non-viable neutrophils surrounded by epithelioid macrophage as seen in this H+E image of spleen (40× magnification) or (D) foci of viable and necrotic macrophage containing amphophilic, vacuolated cytoplasm as seen in this H+E image of mesenteric lymph node (40× magnification). Similar inflammatory lesions were noted in the (E) pancreas, effacing normal acinar cell lobules with small numbers of residual acinar cells (arrows) or (F) extending into the small intestine through the serosa.

Summary of findings.

The most common clinical signs for all affected ferrets were decreased body condition score, dehydration, and hind limb paresis. Other common findings included ocular discharge and urinary incontinence. A summary of the clinical findings for all affected ferrets is presented in Table 2. The timeline from arrival to onset of disease ranged from less than 2 wk to 7 mo (Figure 3), and the age range at time of presentation ranged from 5 mo to 11 mo.

Table 2.

Summary of clinical symptoms for AAT KO ferrets with ferret systemic coronavirus infection.

Clinical Findings	Ferret(s) Affected	Number Affected of Total Number
Decreased body condition	749, 752, 755, 756	4 of 5
Dehydration	749, 752, 755, 756	4 of 5
Hindlimb paresis	749, 752, 755, 756	4 of 5
Lethargy	749, 752, 755	3 of 5
Urinary incontinence	749, 755, 756	3 of 5
Ocular discharge	749, 752	2 of 5
Palpable abdominal mass	749	1 of 5
Diarrhea	749	1 of 5
Found dead	753	1 of 5

Figure 3.

Timeline of Clinical Cases.

Gross postmortem findings in multiple ferrets include a spectrum of lesions, most notably splenomegaly and enlarged, mesenteric lymph nodes that on cut section were homogenously tan with irregular margins suggestive of capsular ulceration. All affected tissues (spleen, mesenteric lymph nodes, mesentery, pancreas, and intestinal serosa) had marked diffuse or segmental effacement of normal parenchyma, with replacement by pyogranulomatous to granulomatous inflammatory lesions varying from discrete to nodules comprised of a central necrotic core (neutrophils and cellular debris) surrounded by epithelioid macrophage containing abundant vacuolated cytoplasm and large oval indented to marginated nuclei. Granulomas and pyogranulomas were surrounded by dense foci of plasma cells and lymphocytes intermixed with activated fibroblasts and variably thick, dissecting bands of collagen. No infectious organisms were identified on PAS, Gram, and GMS stains. No evidence of central nervous system disease was seen that could account for the hind limb paresis and urinary incontinence seen clinically in the ferrets. FRSCV was confirmed with Immunohistochemistry in 4/5 ferrets with due to our increased suspicion of a colony outbreak of FRSCV. A summary of the pathology findings is presented in Table 3.

Table 3.

Summary of pathology findings for AAT KO ferrets with ferret systemic coronavirus

Pathology Findings	Ferret(s) Affected	Number Affected of Total Number
Gross evidence of pyogranulomas	749, 753, 755, 756	4 of 5
Gross splenomegaly and/or mesenteric lymphadenomegaly	749, 753, 755, 756	4 of 5
Histologic evidence of pyogranulomas	749, 753, 755, 756	4 of 5
Histologic evidence of vasculitis and fibrinoid necrosis	749, 752, 753	3 of 5
IHC (anti-FCoV 17monoclonal antibody FIPV3-70)	749, 753, 755, 756	4 of 5

For all IHC positive ferrets (749, 753, 755, and 756), PCR testing of paraffin-embedded mesenteric lymph node with granulomatous lesions was negative, perhaps due to low DNA yield from the sample. Survey PCR testing of pooled feces samples from 5 cages of wild type ferrets housed in the same room as the AAT KO ferrets revealed endemic Ferret Coronavirus Genotype 2, while fecal samples from each of 5 cages of a second cohort of AAT KO ferrets showed the presence of both Ferret Coronavirus genotype 1 and Ferret Coronavirus genotype 2. This test evaluates the presence of specific variants of the conserved spike protein (S). As the S protein is a major inducer of viral-neutralizing antibodies, it is important in antigenicity, with genotype 1 typically being associated with more severe disease.^9,22,23 However, both genotypes have been implicated in FRSCV.¹⁹

Remaining cohort.

Of the initial 10 ferrets in this shipment, all others died or were euthanized for causes considered to be unrelated to infection with FRSCV. These ferrets had been intended for use in a pilot study to characterize smoking paradigms for this genotype. All ferrets in this study underwent minimum µCT, FlexiVent™ measurements, bronchoalveolar lavage, and/or mucociliary clearance. Two ferrets in the cohort were exposed to cigarette smoke, which is a common procedure for this pulmonary research group.¹⁸ One female ferret developed a peri-osseous hematoma and underlying focally extensive hemorrhage of the cerebellar meninges and died approximately 5 mo after arrival; the lesions were thought to be secondary to an unobserved trauma. The remaining 4 ferrets (2 male and 2 female) died related to the project or were euthanized after reaching study endpoint. These ferrets did not present to veterinary staff for any additional examination during their time in the study.

Materials and Methods

Animal husbandry.

The University of Alabama at Birmingham (UAB) is an AAALAC-accredited facility, and animals are maintained in accordance with the Animal Welfare Act and the Guide for the Care and Use of Laboratory Animals.^2,8 The protocol for the experimental use of these ferrets was approved by the UAB Institutional Animal Care and Use Committee. Ferrets were obtained from Marshall Farms (North Rose, NY), which provides ferrets derived from an expansion of cloned founders generated at the University of Iowa.⁷ Marshall Farms tests for Aleutian disease, ferret enteric coronavirus, rotavirus A, 4 strains of human influenza virus, Bordetella bronchiseptica, Campylobacter spp., Helicobacter spp., Lawsonia intracellularis, Pasteurella multocida, Salmonella spp., Staphylococcus aureus, Yersinia enterocolitica, Coccidia, Dirofilaria immitis, Otodectes cynotis, Cryptosporidium parvum, and Giardia spp., and vaccinates against canine distemper virus and rabies virus. Ferrets in the cohort were pair-housed in modified rabbit caging at 72 ± 2 °F (22 ± 1 °C) on a 12:12-h light/dark cycle and provided with water (Birmingham city water) and food (LabDiet 5L14, St. Louis, MO) ad libitum.

Clinical chemistry and PCR testing.

All clinical chemistry measurements were performed inhouse using 0.1 mL serum on a VetScan VS2 Chemistry Analyzer using a Comprehensive Diagnostic Profile Rotor (Zoetis, Parsippany-Troy Hills, NJ). To rule out other potential underlying viral diseases, peri-mortem bronchoalveolar lavage samples from 753 and 755 were tested and found to be negative by PCR for influenza virus (H3N8, H5N1, and pan-influenza), Mycoplasma spp., Bordetella spp., Canine distemper virus, and SARS-CoV-2 (Antech Diagnostics, Fountain Valley, CA). Samples (paraffin-embedded lymph nodes) were submitted for Ferret Coronavirus PCR (MSU Veterinary Diagnostic Laboratory, Lansing, MI).

Histopathologic examination.

Full necropsies were performed on all 5 affected ferrets. Representative tissue specimens were fixed in 10% neutral buffered formalin, processed, and embedded in paraffin wax using standard methods. Unstained, 4-µm sections of spleen, liver, and mesenteric lymph nodes were prepared and stained with hematoxylin and eosin (H+E), periodic acid-Schiff (PAS), Brown-Hopps gram stain, and Grocott methenamine silver stain to characterize lesion morphology and examine for the presence of infectious agents (bacterial, fungal, and protozoal) by board certified veterinary anatomic pathologists (JBF and MK).

Immunohistochemistry.

For coronavirus immunohistochemistry, an automated staining system (Discovery Ultra, Ventana Medical Systems, Tucson, AZ) was used to label 5-μm sections of formalin fixed and paraffin embedded tissues (mesenteric lymph node or spleen) with an anti-FCoV monoclonal antibody FIPV3-70 at a dilution of 1:1,000 as the primary antibody (Custom Monoclonals, International, Sacramento, CA) and an alkaline phosphatase Ultraview red detection system (Ventana, Medical Systems, Tucson, AZ) after 52 min of antigen retrieval with retrieval solution CC1 (Ventana Medical Systems, Tucson, AZ) as previously described.³ Positive control tissue consisted of lymph node from a ferret with confirmed FRSCV based on histopathology, immunohistochemistry, and PCR that exhibited the appropriate immunohistochemical labeling pattern (cytoplasmic labeling of macrophages within pyogranulomatous infiltrates).

Discussion

Out of an initial cohort of 10 AAT KO ferrets, 4 (2 males,749 and 753; 2 females, 755 and 756) died or were euthanized due to FRSCV, as confirmed by immunohistochemistry. While not confirmed, FRSCV is highly suspected as the cause of death in another ferret (752, male). Based on immunohistochemistry and characteristic histopathology, this indicates a 40% to 50% disease incidence of presumptive FRSCV in our facility for this set of ferrets. Communications with the vendor and the academic institution at which the AAT KO ferrets were generated indicated that while FRSCV has been found sporadically in the colony, our high incidence of disease in such a small cohort was unusual. Our facility commonly houses between 30 and 60 ferrets at any one time, with individual ferrets maintained anywhere between several months and 2 y. Retrospective review of other ferret clinical and necropsy records over the past 5 y does not indicate prior instances of this disease in our facility.

The lack of positive PCR verification of FRSCV is a limitation of our report, as is our inability to completely rule out other causes of disease such as Aleutian disease. Because our index of suspicion for FRSCV was not initially high, freshly frozen tissue or feces was not saved for PCR analysis.

An interesting aspect of this case series is the distinctive clinical presentation and overall mild histopathology seen in AAT KO ferrets with FRSCV. A prior publication reviewed a case of FRSCV in a research ferret, but initial clinical signs included diarrhea or organomegaly, which are more consistent with the classic clinical signs associated with disease.³ AAT KO ferrets with FRSCV had similar presentations to a previous case in which the most distinguishing feature of the affected ferret was posterior paresis with or without urinary incontinence.¹³ In the seminal paper describing the clinicopathologic features of FRSCV, hind limb weakness and central nervous systems signs were considered less frequent findings; primary clinical signs were weight loss and palpable abdominal masses.⁵ In addition, the average age of the affected ferret in that paper was 11 mo, which is older than all but one of the ferrets affected in our report.⁵ Affected AAT KO ferrets showed no evidence of ocular involvement, as was reported in another case,¹² but at least 2 of our ferrets (752 and 749) showed clinically apparent ocular discharge antemortem. Ferret 749 showed the typical gross white nodules and severe splenomegaly with discrete nodules . Gross lesions for prior cases were mild, and our initial suspicion was that these lesions could be an uncharacterized phenotype of AAT KO ferrets with similar histopathologic findings to those of animal models of sarcoidosis. Currently, the literature characterizing AAT KO ferrets beyond pulmonary function is limited, which delayed our diagnosis.

The most common clinical chemistry finding in ferrets with FRSCV is reported to be hyperproteinemia (specifically hyperglobulinemia characterized by a polyclonal gammopathy); this finding is also observed in cats with FIPV.⁵ However, we did not find this abnormality in serum chemistry values from affected ferrets in our cohort, indicating an overall reduced immune response to infection. The clinical chemistries from ferrets 749 and 752 showed the following overlapping main abnormalities: a moderate to severe increase in serum urea nitrogen, mild increase in total protein characterized by increased albumin, and hyperphosphatemia. Total protein, BUN, and electrolytes were presumed elevated secondary to sample hemolysis, dehydration, poor body condition score, and muscle wasting that were seen in both ferrets

AAT KO ferrets are a genetic model of emphysema and were created using CRISPR/Cas9 technology.⁷ These ferrets develop lesions as early as 3 mo of age; the lesions are characterized by higher pulmonary compliance and limited expiratory airflow.⁶ However, AAT depletion is not restricted to lung tissue, but is rather expressed systemically, which may partially explain the high incidence of FRSCV that we saw in this population. Recent publications investigating the relationship between α-1 antitrypsin deficiency and SARS-CoV-2 infection in humans indicate an increased incidence of disease in ethnic populations with high carrier or homozygosity rates of the most common allele responsible for α-1 antitrypsin deficiency in humans (SERPINA1).^4,21 In addition, ongoing studies are currently evaluating the treatment efficacy for COVID-19 of FDA-approved drugs used to treat α-1 antitrypsin deficiency such as Nafamostat.²³ The proposed mechanism of action is that AAT inhibits TMPRSS2 (transmembrane protease serine 2), which prevents activation of the spike proteins on the outside of the coronavirus virion and thus reduces SARS-CoV-2 viral entry in airway tissues. This mechanism occurs in correlation with interaction with the ACE2 receptor for cell entry.¹⁵ Similar mechanisms could promote viral entry into the gastrointestinal tract of AAT KO ferrets and therefore increase disease incidence. The gastrointestinal system is becoming implicated as a key route of infection of COVID-19 in people due to cell receptor expression in various tissues.²⁴ Historically, the suspected primary means of transmission of FRECV in ferrets has been fecal-oral.⁵

Our case series highlights the value of the ferret AAT KO model in coronavirus research in general and potentially in SARS-Cov-2 research specifically. Facilitation of coronavirus infectious disease research using ferrets will require a better understanding of how to protect these animals from endemic circulating coronaviruses in vendor and research colonies.