Archives of https://doi.org/10.1007/s00705-018-3991-7

ORIGINAL ARTICLE

Characterization of deerpox in Florida white‑tailed fawns expands the known host and geographic range of this emerging pathogen

Katherine A. Sayler1 · Kuttichantran Subramaniam2 · Jessica M. Jacob2 · Julia C. Loeb3,4 · William F. Craft5 · Lisa L. Farina5 · Nicole I. Stacy6 · Nissin Moussatche7 · Laurie Cook8 · John A. Lednicky3,4 · Samantha M. Wisely1 · Thomas B. Waltzek2

Received: 2 January 2018 / Accepted: 6 June 2018 © Springer-Verlag GmbH Austria, part of Springer Nature 2018

Abstract Infections caused by mule deerpox virus (MDPV) have been sporadically reported in North American cervids. White-tailed deer ( virginianus) fawns from a farm located in South Central Florida presented with ulcerative and crusting lesions on the coronary band as well as the mucocutaneous tissues of the head. Evaluation of the crusted skin lesions was undertaken using microscopic pathology and molecular techniques. A crusted skin sample was processed for virus isolation in four mammalian cell lines. The resulting isolate was characterized by negative staining electron microscopy and deep sequencing. Histopathologic evaluation of the skin lesions from the fawns revealed a hyperplastic and proliferative epider- mis with ballooning degeneration of epidermal and follicular keratinocytes with intracytoplasmic eosinophilic inclusions. Electron microscopy of cell culture supernatant demonstrated numerous large brick-shaped particles typical of most poxvi- ruses. Polymerase chain reaction assays followed by Sanger sequencing revealed a poxvirus gene sequence nearly identical to that of previous strains of MDPV. The full was recovered by deep sequencing and genetic analyses supported the Florida white-tailed deer isolate (MDPV-F) as a strain of MDPV. Herein, we report the frst genome sequence of MDPV from a farmed white-tailed deer fawn in the South Central Florida, expanding the number of locations and geographic range in which MDPV has been identifed.

Introduction

The family includes a diverse group of envel- Handling Editor: YiMing Shao. oped, double-stranded DNA infecting a range of invertebrate (subfamily: Entomopoxvirinae) and Electronic supplementary material The online version of this (subfamily: ) hosts [1]. The International article (https​://doi.org/10.1007/s0070​5-018-3991-7) contains Committee on of Viruses currently recognizes supplementary material, which is available to authorized users.

* Thomas B. Waltzek 5 Department of Comparative, Diagnostic and Population [email protected] Medicine, College of Veterinary Medicine, University of Florida, Gainesville, FL, USA 1 Department of Wildlife Ecology and Conservation, 6 Department of Large Clinical Sciences, College University of Florida, Gainesville, FL, USA of Veterinary Medicine, University of Florida, Gainesville, 2 Department of Infectious Diseases and Immunology, College FL, USA of Veterinary Medicine, University of Florida, Bldg 1379, 7 Department of Molecular Genetics and , Mowry Road, Gainesville, FL 32611, USA University of Florida, Gainesville, FL, USA 3 Department of Environmental and Global Health, College 8 BDRL Whitetail Paradise Farm, Okeechobee, FL, USA of Public Health and Health Professions, University of Florida, Gainesville, FL, USA 4 Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA

Vol.:(0123456789)1 3 K. A. Sayler et al. nine chordopoxvirus genera which classify viruses that the head (e.g., eyelids, lips, ear tips, nose). Other gross infect (Capripoxvirus, Centapoxvirus, Cervid- lesions reported in some cases include keratoconjunctivi- poxvirus, Leporipoxvirus, Molluscipoxvirus, , tis, alopecia, ulceration of the upper alimentary tract (e.g., , Suipoxvirus, and ) and two that lips, mouth, tongue, esophagus, reticulum, rumen), and infect reptiles and birds (Crocodylidpoxvirus and Avipox- teats [19, 20, 23]. Microscopic lesions include a prolifera- virus) [1–3]. Poxvirus infections are diagnosed based on tive and ulcerative dermatitis with eosinophilic intracyto- clinical manifestation (e.g., induction of proliferative cuta- plasmic inclusions observed within afected keratinocytes neous lesions), microscopic pathology (e.g., proliferation, and an associated mixed infammatory infltrate [19, 21, ballooning degeneration and cytoplasmic inclusions), in 23, 24]. An experimental transmission study reproduced vitro growth characteristics, ultrastructural features [e.g., the mild cutaneous disease in fawns exposed large oval () or brick-shaped virions (other by intracutaneous and intravenous injection, while more poxviruses)] and molecular genetics [4]. Members of the severe disease was observed in a fawn commingled with family Poxviridae include highly contagious and pathogenic the infected fawns [25]. zoonotic agents, signifcant pathogens of livestock report- Genomic sequencing of two isolates from skin lesions able to the World Organization for Animal Health, and an of free-ranging mule deer, W-848-83 from Basin, Wyo- increasing number of viruses isolated from wildlife, includ- ming in 1983 and W-1170-84 from Burlington, ing cervids [5, 6]. in 1984, revealed they are genetically distinct poxviruses The Parapoxvirus contains four species: Bovine and represent isolates of a novel species, Mule deerpox papular stomatitis virus, virus (ORFV), Pseudocowpox- virus (MDPV), within the monotypic genus Cervidpox- virus (PCPV), and Parapoxvirus of in New Zealand virus [5]. Characterization of additional cervidpoxvirus (PVNZ), isolates of which are known to infect cases in from the 1990’s and early 2000’s [1]. Isolates of ORFV, PCPV, and PVNZ have been detected [22] suggest that this virus may be an emerging pathogen, in wild and farmed cervid populations including: red deer although it remains unclear whether its emergence is due ( elaphus) in New Zealand and Italy, (Rupi- to epidemiological factors or because of increased surveil- carpa rupicarpa) and ibex ( ibex) in Italy, and rein- lance. Phylogenetic analyses based on a conserved poxvirus deer (Rangifer tarandus) in Finland and Norway [7–10]. gene [e.g., IMV membrane protein (ORF A21L in typically present with pustular cutaneous lesions virus Copenhagen, GenBank accession no. M35027)] or the of the head (e.g., ears, face, lips, muzzle) and neck and in concatenation of multiple genes have demonstrated that cer- severe cases morbidity may approach 100% with afected vidpoxviruses belong to a clade of poxviruses (i.e., clade 2) animals succumbing to secondary infections or starvation. that also includes capri-, lepori-, sui-, yatapoxviruses, and Experimental transmission of ORFV by skin scarifcation Cotia virus [5, 22, 23, 26, 27]. Cervidpoxviruses encode a resulted in mild proliferative cutaneous disease in white- suite of proteins involved with immunomodulation includ- tailed deer (Odocoileus virginianus), mule deer (O. hemio- ing 10 of the 12 such protein families previously defned nus), (C. canadensis), , and (Alces alces) in other poxviruses [5, 27]. Additionally, MDPV encodes [11, 12]. Multiple reports of zoonotic transmission of cervid unique immunomodulatory proteins either absent in other parapoxvirus (CePPV) have also appeared [9, 13–17]. poxviruses or captured cellular orthologs present in other Poxviruses, ultrastructurally and genetically distinct poxviruses but likely independently captured by cervidpox- from parapoxvirus (PPV), have repeatedly been isolated viruses [5]. These unique cervidpoxvirus genes have been from North American wild and managed cervid popu- argued to infuence viral pathogenesis and encode a cellular lations [5, 18–24]. The ultrastructural features of these endothelin, an interleukin 1 receptor antagonist (IL-1Ra), cervid poxviruses isolated from a managed population of a c- lectin-like receptor (CTLR), a unique major his- reindeer in Canada [18], moribund wild mule deer in Wyo- tocompatibility complex-1 protein, and a distinct form of ming [19], a wild black-tailed deer (O. h. columbianus) in transforming growth factor beta 1 [5]. To date, the genes sufering from a fatal hemorrhagic adenovirus encoding the IL-1Ra and CTLR proteins have been detected disease [20], moribund managed pudu (Pudu puda) in St. in all examined cervidpoxvirus isolates [5, 22, 23]. Louis, California [21], moribund or dead wild cervids in Herein, we describe the first isolation and genomic the Pacifc Northwest [22], a dead white-tailed deer from sequencing of MDPV isolated from a Florida farmed white- a farmed population in Mississippi [22], and an inappar- tailed deer fawn (MDPV-F). A range of testing modalities ent infection in a managed (Gazella subgutturosa) were employed to support the diagnosis and the results are [23], all resemble (i.e., the virions are presented. This report is the second detection of MDPV brick shaped). Afected cervids typically display prolifera- in white-tailed deer, it extends the number of locations in tive cutaneous lesions that ulcerate and form crusts along which cervidpoxvirus has been detected, and is the second the coronary band, neck, and mucocutaneous tissues of report within the Southeastern United States.

1 3 Deerpox virus in Florida white-tailed deer

Materials and methods signs of disease. At birth, fawns were examined, weighed, administered Vision CDT, fawn paste, and B-complex, and Case history their navels were treated with iodine. No other treatments were performed until the onset of clinical signs. In July of 2016, at a deer farm in South Central Florida, a four week old moribund buck fawn with purulent exudate Clinical and anatomic pathology on the left dorsal ear tip, eyelids, and muzzle (Fig. 1A and B) was presented for medical evaluation. Severe lesions Prior to death, crusted skin lesions from both fawns’ eye- were treated topically with sulfonamide antibiotic cream lids, ear tips, and noses were removed with a sterile scal- (1% silver sulfadiazine, Monarch Pharmaceuticals, Bristol, pel and bisected laterally. One of the two resulting por- TN, USA). A moribund penmate, a four week-old doe fawn, tions was placed in 10% neutral bufered formalin, and the also presented with ulcerative lesions on the mucocutane- other in sterile tubes. The samples were then shipped on ous areas of the head as well as coronary bands (Fig. 1C) ice packs overnight to the University of Florida Cervidae and tongue. Three penmates of these two cases previously Health Research Initiative (CHeRI) Laboratory in Gaines- presented with ulcerative lesions and succumbed to disease, ville, Florida. Prior to removal of the crusting lesions from but specimens were not collected for evaluation. Grossly vis- the nose, imprints were taken from lesions in both animals. ible lesions progressed rapidly and both fawns succumbed Whole blood was submitted to a commercial laboratory for within fve days from the frst signs of disease. An additional a complete blood count and serum biochemical profle. At nine fawns from this herd presented with crusting lesions but the time of death (approximately four days after samples survived (number of afected fawns in herd: 14/27 = 52%). were initially taken), a feld necropsy was performed on the Four days prior to death, both animals were subjected to afected fawns. Portions of the liver, lung, spleen, kidney, phlebotomy and crusted skin lesions were sampled for cyto- skin, and tongue lesions were placed in 10% neutral buf- logic and histopathologic evaluations (see below). Fever and ered formalin for histopathology and separate sections of lameness were observed in both animals at the time of speci- each tissue were placed in sterile tubes and shipped to the men collection and both animals were again treated with CHeRI laboratory. Board certifed veterinary pathologists antibiotic cream as described above. performed the gross and histopathologic analyses. DNA extraction and polymerase chain reaction Animals (PCR) assays

Fawns from this herd were the product of a white-tailed deer DNA was extracted from post-mortem tissue homogenates breeding program for the purpose of producing trophy bucks (skin, liver, lung, spleen, and kidney), virions in the spent for . Fawns were born on a ranch and remained with cell culture medium (i.e., cell culture media from poxvirus- their mothers. Animals were maintained in one-acre pens. infected cells displaying virus-induced cytopathic efects Adult deer were a mixture of imported animals and animals [CPE]), and infected cells scraped of from inoculated cell that were born on the farm. Adults did not show clinical cultures (see cell culture section below) using a QIAamp

Fig. 1 Gross presentation of mule deerpox virus in a Florida white- an afected buck fawn. Panel C. Ulcerative lesions were present on tailed fawn displaying multiple purulent, ulcerative cutaneous lesions. the coronary bands of the afected doe fawn Panels A and B. Mucocutaneous lesions were visible on the head of

1 3 K. A. Sayler et al.

DNA Mini Kit (Qiagen, Valencia, CA, USA). The DNA USA) according to the manufacturer’s instructions. Multi- extracts were then subjected to conventional PCR assays plex RT-qPCR for BTV and EHDV was performed using a targeting the B2L gene that encodes immunogenic envelope primer and probe set specifc for each virus using previously protein of parapoxviruses (PPVs) and the DNA polymerase described protocols (Table 1) [31]. gene of herpesviruses (Table 1) as previously described [28, 29]. They were further evaluated against a poxvirus PCR Preparation of tissue homogenate assay that targets a portion of the insulin metalloproteinase- like protein and IMV membrane protein genes of low-GC Approximately 1 ml each of 10% w/v tissue homogenates chordopoxviruses [30] using a slightly modifed procedure. of a post-mortem crusted skin lesion, as well as liver, lung, Briefy, Promega Go Taq Flexi bufer and enzyme (Pro- and spleen specimens were generated in Advanced Dulbec- mega, Madison, WI, USA) were used, and Tween 20 was not co’s Modifed ’s Medium (Invitrogen Corp.) supple- included in the PCR mixture. Thermocycling was performed mented with 2 mM L-Alanyl-L-Glutamine (GlutaMAX™, in an Eppendorf Mastercycler Pro 384 as follows: 95 °C Invitrogen Corp.) and antibiotics (PSN; 50 μg/ml penicil- initial denaturation for 2 minutes followed by 25 cycles of lin, 50 μg/ml streptomycin, 100 μg/ml neomycin [Invitro- 95 °C for 30 s, 50 °C for 1 minute, 72 °C for 1 minute, and gen Corp.]) using a manual tissue grinder (Fisher Scientifc, an additional 15 cycles of 95 °C for 30 s, 53 °C for 30 s, Waltham, MA, USA). The tissue homogenates were than 72 °C for 1 minute, with a fnal elongation step of 72 °C fltered through a sterile 0.45 µm flter to remove particulates for 5 minutes. PCR products were then subjected to elec- and contaminating bacteria and fungi, and stored at –80 °C trophoresis in 1.5% agarose gels stained with ethidium bro- until they were used. mide and the amplicons of expected size were purifed using a QIAquick PCR Purifcation Kit (Qiagen, Valencia, CA, Isolation of virus in cultured cells USA). The concentration of purifed DNA was quantifed fuorometrically using a Qubit 3.0 Fluorometer and dsDNA After thawing on ice, aliquots (50 µl) of the fltered homoge- BR Assay Kit (Life Technologies, Carlsbad, CA, USA) and nates were inoculated onto nearly confluent monolayers the purifed amplicon sequenced in both directions on an of OHH1.K (Odocoileus hemionus hemionus [Columbian ABI 3139 platform (Applied Biosystems, Foster City, CA). black tail deer] kidney, ATCC CRL #6193), Vero E6 (Cer- As these fawns were from a farmed population in a loca- copithecus aethiops [African green monkey] kidney, ATCC tion where epizootic hemorrhagic disease virus (EHDV) and CRL #1586), BT ( taurus [] turbinate, ATCC CRL bluetongue virus (BTV) are endemic, RNA was extracted #1390), and LLC-MK2 (Macaca mulatta [Rhesus mon- from skin, liver, kidney, lung, and spleen homogenates key] kidney, ATCC CCL #7.1) cells that were subsequently using a QIAamp RNeasy Mini Kit (Qiagen, Valencia, CA, incubated at 35 °C in a 5% ­CO2 atmosphere. Four days

Table 1 Primers and probes used in this study Oligonucleotide ID Description Sequence Source

PPP-1 Parapoxviruses 5′-GTC GTC CAC GAT GAG CAG CT-3′ Inoshima et al., 2000 PPP-3 Parapoxviruses 5′-GCG AGT CCG AGA AGA ATA CG-3′ Inoshima et al., 2000 PPP-4 Parapoxviruses 5′-TAC GTG GGA AGC GCC TCG CT-3′ Inoshima et al., 2000 DFA Herpesviruses 5′-GAY TTY GCN AGY YTN TAY CC-3′ VanDevanter et al., 1996 ILK Herpesviruses 5′-TCC TGG ACA AGC AGC ARN YSG CNM TNA A-3′ VanDevanter et al., 1996 KG1 Herpesviruses 5′-GTC TTG CTC ACC AGN TCN CAN CCYT T-3′ VanDevanter et al., 1996 TGV Herpesviruses 5′-TGT AAC TCG GTG TAY GGN TTY CAN GG NGT-3′ VanDevanter et al., 1996 IYG Herpesviruses 5′-CAC AGA GTC CGT RTC NCC RTA DAT-3′ VanDevanter et al., 1996 Low-GC forward Low-GC content poxviruses 5′-ACA CCA AAA ACT CAT ATA ACT TCT-3′ Li et al., 2010 Low-GC reverse Low-GC content poxviruses 5′-CCT ATT TTA CTC CTT AGT AAA TGA T-3′ Li et al., 2010 BTV NS3 183F Bluetongue virus 5′AAA TMT TGG AYA AAG CRA TGT CAA A-3′ Wernike et al., 2015 BTV NS3 288R Bluetongue virus 5′-CTY ACR TCA TCA CGA AAC GCT-3′ Wernike et al., 2015 BTV NS3 242FAM Bluetongue virus 5′-FAM-AAR GCT GCA TTC GCA TCG TAC GC-BHQ1-3′ Wernike et al., 2015 EHD NS1 5F EHD virus 5′-AAA AAG TTC YTC GTC GAC TGC-3′ Wernike et al., 2015 EHD NS1 80R EHD virus 5′-ATT GGC RTA RTA ACT GTT CAT GTT-3′ Wernike et al., 2015 EHD NS1 27 TEX EHD virus 5′-Texas Red-ATC GAG ATG GAR CGC TTY TTG AGA Wernike et al., 2015 AAA T-BHQ2-3′

1 3 Deerpox virus in Florida white-tailed deer post-inoculation with the crusted skin and spleen homogen- Genome assembly and annotation ates, CPE were present in all inoculated cell lines, at which point the infected cells were scraped and collected along with De novo assembly of paired-end reads was performed using most of the spent cell growth medium and stored at −80 °C. the SPAdes 3.5.0 genome assembly algorithm [32] with default settings. The quality of the assembly was verifed by mapping reads against the fnal genome consensus using Negative stain electron microscopy Bowtie 2 2.1.0 [33] and visualized in Tablet 1.14.10.20 [34]. The genome of the poxvirus isolated from white-tailed deer An aliquot of spent cell growth media from infected was annotated using the Genome Annotation Transfer Utility OHH.1K cells was used for negative stain electron micros- (GATU) with mule deerpox virus isolate W-848-83 (RefSeq copy. Briefy, 1 mL was centrifuged at 175,000 × g for 1 h accession no. NC_006966) as the reference genome [35]. at 10 °C to pellet the virus. The concentrated virus was pro- cessed by resuspending the pellet in 10 mM Tris-HCl, pH Phylogenetic analyses 8.0, then layering the resuspended virus onto a Formvar/ carbon coated grid, followed by incubation for 10 minutes Maximum Likelihood (ML) analysis was performed on 36 at room temperature. Thereafter, the grid was washed in poxviruses (including the MDPV-F; Supplemental Table 1) 10 mM Tris-HCl, pH 8.0, then stained with Nano-W (Nan- using the concatenated amino acid (AA) alignment of 7 pox- oprobes, USA) for 5 minutes and examined in a Hitachi virus core proteins including the: RNA polymerase subunit H-7000 transmission electron microscope. RPO147, RNA polymerase subunit RPO132, RNA poly- merase-associated RAP94, mRNA capping enzyme large subunit, virion major core protein P4a, early transcription Virus concentration and DNA extraction factor VETFL, and NTPase. Sequences were aligned using MAFFT [36] and concatenated in Geneious v.7 (8110 AA Poxvirus virions present in the spent medium of OHH.1K characters including gaps) [37]. The ML analysis was per- cells inoculated with the crusted skin homogenate were formed in MEGA 7.0.18 using the LG model with the G + I concentrated using a three-step process: low-speed centrif- rate parameters and 1000 non-parametric bootstrap analyses ugation to pellet cellular debris, high-speed centrifugation to test clade robustness [38]. To elucidate the relationship to pellet the poxvirus virions, and exonuclease treatment of MDPV-F to other cervidpoxviruses for which partial to degrade DNA not protected within poxvirus virions. genomic sequences have been determined (Supplemental Briefy, 1 mL of the virus-containing material was thawed Table 1), an additional ML analysis was conducted using on ice and centrifuged at 5509 × g for 20 min in a Beckman representative Clade II poxviruses (genera Yatapoxvirus, J2-MI centrifuge at 4 °C to pellet debris. The supernatant Leporipoxvirus, Suipoxvirus, Cervidpoxvirus, and Capripox- was collected, centrifuged once more following the previ- virus) based on the AA amino acid sequences of the IMV ous parameters, and then the supernatant was centrifuged membrane protein (ORF A21L in vaccinia virus Copenha- at 101,360 × g for 1 hour, 15 min at 4 °C. The superna- gen, GenBank accession no. M35027). The sequences were tant was discarded and the pelleted virus resuspended in aligned using MAFFT (116 AA characters including gaps) resuspension bufer (10 mM Tris-HCl, pH 7.6, 10 mM KCl, and the ML analysis was conducted as described above. 1.5 mM MgCl­ 2), and treated with Baseline-ZERO DNase (Lucigen, Middleton, WI, USA), an exonuclease used to digest extraneous ds- and ssDNA to mononucleotides. Total Results DNA was extracted from the DNase-treated virus suspen- sion using a QIAamp DNA Mini Kit (Qiagen, Valencia, CA, Clinical and anatomic pathology USA) as described above. Hematology and biochemistry data indicated dehydration (increased packed cell volume) and suggested systemic Library construction and next‑generation inflammation (thrombocytosis, hyperfibrinogenemia) in sequencing both fawns. Tissue imprints of skin lesions from both ani- mals were consistent with suppurative dermatitis, suspected Purifed virus DNA was used to construct a DNA library bacterial infection (presence of mixed extracellular bacteria using a Nextera XT DNA Library Preparation Kit (Illumina, and degenerate neutrophils), and reactive fbroplasia based San Diego, CA, USA) as specifed by the manufacturer. on cytologic evaluation. Sequencing was performed using a v3 chemistry 600 cycle Multiple flat, thickened, cutaneous lesions were vis- kit on an Illumina MiSeq sequencer. ible on both animals particularly on the lips, muzzle, and

1 3 K. A. Sayler et al. coronary bands (doe fawn only) at necropsy. Multiple ulcera- from the homogenate generated from a crusted skin lesion. tive lesions were also observed on the tongue of the doe After removal of the primer sequences, BLASTN analy- fawn. Periocular alopecia and erosions were observed on sis of the sequence was identical to mule deerpox virus both animals. Polyarthritis and peripheral lymphadenopathy isolates W-848-83 and W-1170-84 (GenBank accession were observed in the buck fawn. The liver, lung, kidney, and nos. NC_006966 and AY689437). The same results were spleen appeared grossly normal in both fawns. obtained from DNA extracted from free virions in the spent Histopathologic evaluation of the skin lesions from the medium of cells inoculated with skin homogenate and spleen fawns revealed a hyperplastic and proliferative epidermis homogenate as well as scraped cells recovered from all inoc- with ballooning degeneration of epidermal and follicular ulated cell lines (see below). In contrast, no virus-specifc keratinocytes with intracytoplasmic eosinophilic inclusions amplicon was produced by the low-GC poxvirus PCR assay (Fig. 2). Crusting with mixed bacteria within the crusts, lym- using DNA extracted from the spent media of uninfected phohistiocytic and suppurative mural and luminal folliculitis negative control cells. and transdermal difuse interstitial lymphohistiocytic, neu- trophilic, and eosinophilic dermatitis were also observed. Virus isolation

PCR assays Virus-induced CPE were observed by light microscopy in all cell lines (OHH1.K, Vero E6, BT, and LLC-MK2) The samples tested negative for parapoxvirus, herpesvirus, inoculated with the crusted skin and spleen homogenates. BTV, and EHDV. The low-GC poxvirus PCR assay pro- Cytopathic efects initially consisted of enlarging of the duced the expected 220 bp amplicon from DNA extracted cytoplasm and development of frequent intracytoplasmic inclusion bodies, shown for Vero E6 and OHH1.K cells in Fig. 3, followed by detachment of the infected cells from the growing surface of the fask. Similar CPE were observed in BT and LLC-MK2 cells (data not shown).

Negative stain electron microscopy

Electron microscopy of the negative stain preparation revealed the presence of brick-shaped virus particles meas- uring approximately 270 × 210 nm when the mulberry- like structure of the virus was viewed, and approximately 290 × 240 nm when the enveloped mature virion was viewed (Fig. 4).

Sequence analyses

De novo assembly of 9,142,070 paired-end reads produced a contiguous consensus sequence of 163,340 bp which includes the 3458 bp inverted terminal repeats and a G+C content of 27.1%. Using Bowtie 2, a total of 2,626,445 reads (28.73%) aligned at an average coverage of 2654 reads/ nucleotide. Similar to the MDPV isolates W-848-83 and W-1170-84 that encoded 169 and 171 open reading frames (ORFs) respectively [5], the MDPV-F genome encoded Fig. 2 White-tailed deer epidermis. Panel A. At low magnifca- 170 proteins (Supplemental Figure 1). All 80 chordopox- tion there is a transition from normal epidermis at right (arrow) to a virus core proteins including the 49 poxvirus core proteins hyperplastic and proliferative epidermis with keratinocyte vacuolar were present in the MDPV-F genome based on a compari- and ballooning degeneration (star) to epidermal and dermal necrosis (lightning bolt). There is a difuse interstitial lymphohistiocytic, neu- son with vaccinia virus Copenhagen (GenBank accession trophilic, and eosinophilic dermatitis and mural and luminal suppura- no. M35027). The complete genome of MDPV-F has been tive folliculitis. 40× hematoxylin and eosin (H&E) staining. Panel B. deposited in GenBank under the accession no. MF966153. At higher magnifcation there is epidermal and follicular keratinocyte In comparison to MDPV isolate W-1170-84 and vacuolar and ballooning degeneration with many intracytoplasmic, variably-sized brightly eosinophilic inclusion bodies (arrows). 600× the MDPV-F, MDPV isolate W-848-83 is missing genes H&E staining DPV005 and DPV030, which encode a CD30-like protein

1 3 Deerpox virus in Florida white-tailed deer

Fig. 3 OHH1.K (top) and Vero E6 cells (bottom) inoculated with skin uninfected and infected, respectively. Cytopathic efects were not lesion homogenate 4 days post-infection. Panels A and B. OHH1.K observed in uninfected control cells but cell rounding and detachment cells uninfected and infected, respectively. Cytopathic efects were is evident in infected VeroE6 cells. Cellular inclusions were evident not observed in uninfected control cells but cell deterioration is in both cell lines. Original images taken at 400× magnifcation evident in infected OHH1.K cells. Panels C and D. Vero E6 cells

Fig. 4 Electron photomicrograph of typical poxvirus obtained from Panel B. The enveloped mature virion measuring approximately spent media from infected OHH.1K cells. Panel A. The mulberry- 290 × 240 nm. Scale bar = 0.1 µm like structure of the virus measuring approximately 270 × 210 nm. and a hypothetical protein, respectively (Supplemental other poxviruses including DPV006 which encodes an Table 2). The MDPV-F and MDPV isolate W-848-83 endothelin precursor, DPV054 which encodes an IL-1 encode hypothetical protein DPV029 which is absent in receptor antagonist, and DPV151 which encodes a MHC MDPV isolate W-1170-84. The MDPV isolate W-848- class I-like protein (Supplemental Table 2) [5]. Compari- 83 encodes a transforming growth factor 1 (TGF-1; son of the amino acid (AA) identities for two of these three DPV163) which is missing in the MDPV-F and MDPV unique cervidpoxvirus genes between the MDPV-F and isolate W-1170-84. Comparison of the MDPV-F and both mule deer poxvirus (which were identical to MDPV isolates W-848-83 and W-1170-84 revealed three one another) were among the lowest for any gene: DPV006 genes present in these cervidpoxviruses not present in (54.55%) and DPV151 (67.39%).

1 3 K. A. Sayler et al.

Overall, the MDPV-F and MDPV isolate W-1170-84 Discussion displayed more similar sized ORFs with higher AA identi- ties as compared to MDPV isolate W-848-83 (Supplemen- Our study provides the first detection of mule deer- tal Table 2). For example, the MDPV-F and MDPV isolate pox virus infection in Florida farmed white-tailed deer W-1170-84 displayed higher AA identities for four puta- (MDPV-F). The presenting complaint, an ulcerative der- tive virulence factors including: serpin-like protein (ORF matitis of the head and hooves in moribund white-tailed DPV003), MHC-like TNF binding protein (DPV008), deer fawns, resembled previous cervidpoxvirus infec- IL-1 receptor-like (DPV015), and IL-18 binding pro- tions in mule deer and black-tailed deer from the Pacifc teins (DPV021) (96.98- 99.56% identity, Supplemental Northwest [5, 19, 22] and white-tailed deer in Mississippi Table 2). By comparison, AA identities to MDPV isolate [23], as well as an unconfrmed cervidpoxvirus report in W-848-83 for these putative virulence factors ranged from pudu at the St. Louis Zoological Park [21]. Similar to our 64.40-79.41%. Finally, Maximum Likelihood phylogenetic case, natural and experimental cervidpoxvirus infections analyses supported MDPV-F as a member of the genus Cer- typically result in mild, self-limiting cutaneous disease vidpoxvirus most closely related to MDPV isolate W-1170- as has been reported for other related poxviruses such as 84 (Fig. 5 and 6).

Mule deerpox virus isolate MDPV-F 96 100 Mule deerpox virus isolate W-1170-84 Cervidpoxvirus

79 Mule deerpox virus isolate W-848-83 Swinepox virus Suipoxvirus 90 Goatpox virus Pellor

100 virus 17077-99 Capripoxvirus 99 virus NI-2490 fibroma virus 100 Leporipoxvirus isolate Belfast 100 Cotia virus SPAn232 virus isolate TPV-Kenya 100 100 Yatapoxvirus Eptesipox virus strain Washington Camelpox virus 81 99 Taterapox virus 100 Variola virus 76 virus Zaire-96-I-16 Orthopoxvirus Vaccinia virus 100 Ectromelia virus 100 virus Yokapox virus Centapoxvirus Orf 100 100 Pseudocowpox virus Parapoxvirus 100 Bovine papular stomatitis virus 94 virus subtype 1 Molluscipoxvirus strain Red UK 100 Pteropox virus Penguinpox virus isolate PSan92 100 100 Pigeonpox virus isolate FeP2 100 Avipoxvirus 100 Fowlpox virus

100 Canarypox virus Nile crocodilepox virus Crocodylidpoxvirus Salmon gillpox virus Amsacta moorei entomopoxvirus 100 100 Anomala cuprea entomopoxvirus Melanoplus sanguinipes entomopoxvirus

Fig. 5 Poxvirus Maximum Likelihood cladogram depicting the rela- subunit RPO132, RNA polymerase-associated RAP94, mRNA cap- tionship of the MDPV-F to representatives of the Poxviridae family ping enzyme large subunit, virion major core protein P4a, early tran- based on the concatenated amino acid (AA) sequences of seven con- scription factor VETFL, and NTPase (8110 AA characters including served genes: RNA polymerase subunit RPO147, RNA polymerase gaps). Nodes that are supported by bootstrap values > 70 are shown

1 3 Deerpox virus in Florida white-tailed deer

82 Mule deerpoxvirus Florida* 100 Cervidpoxvirus black-tailed deerCalifornia 1994 Cervidpoxvirus

Cervidpoxvirus mule deer Oregon 2005

Swinepox virus Suipoxvirus

Myxoma virusisolate Belfast Leporipoxvirus 100 Rabbitfibroma virus

Goatpox virusPellor

100 Lumpyskindisease virusNI-2490 Capripoxvirus

Sheeppoxvirus 17077-99

Cotiavirus SPAn232

Yaba monkey tumorvirus Yatapoxvirus 100 Tanapox virusisolate TPV-Kenya

Fig. 6 Poxvirus Maximum Likelihood cladogram depicting the rela- gen, GenBank accession no. M35027; 116 AA characters including tionship of the MDPV-F to representatives of the clade II poxviruses gaps). The asterisk (*) indicates mule deerpox virus isolates W-1170- (genera Yatapoxvirus, Leporipoxvirus, Suipoxvirus, Cervidpoxvirus, 84, W-848-83, Deerpox virus MS10-9055, and cervidpoxvirus rein- and Capripoxvirus) based on the partial amino acid (AA) sequences deer Ontario 2008 are identical to MDPV-F. Nodes that are supported of IMV membrane protein (ORF A21L in vaccinia virus Copenha- by bootstrap values > 70 are shown

swinepox virus [25]. The low incidence of MDPV cases isolated from the lesions of the two moribund white-tailed and the high seroprevalence in Oregon Odocoileus spp. deer fawns (data not shown), they were considered second- suggests MDPV is a pathogen of low virulence [39]. Like ary to the cervidpoxvirus infection as previously reported our cases, life-threatening illness involving the upper ali- [19, 21, 23]. mentary tract occurs in a proportion of afected young In this study, we took advantage of recent advances [22, 23, 25] or rarely, young and adult [21] deer presum- in high-throughput sequencing to generate the full ably due to an immunocompromised state [21, 23, 25]. genome of the MDPV-F isolate for comparison to pre- Fawns that succumbed to disease in our investigation were viously sequenced cervidpoxviruses. The phylogenetic previously exposed to large amounts of standing water in analysis based on the poxvirus IMV membrane protein their pens which may have limited space available for their (DPV122) confrmed MDPV-F as a member of the genus use leading to increased contact between animals or other Cervidpoxvirus with all previous cervid isolates [5, 22, stressors. 23]. BLASTN analysis of the IMV membrane protein The testing modalities reported here were instrumental in (DPV122) nucleotide sequence of MDPV-F revealed a confrming the MDPV diagnosis. Conventional PCR assays 100% nucleotide identity to isolates from: 1) Wyoming targeting a portion of the insulin metalloproteinase-like mule deer, Ontario reindeer, Mississippi white-tailed protein and IMV membrane protein genes facilitated rapid deer, 2) 99% (346/348) identity to black-tailed deer from confrmation of the cervidpoxvirus diagnosis and assisted California and Oregon, and 3) 98% (338/345) identity to in ruling out zoonoses (e.g., cervid parapoxviruses; CeP- Oregon mule deer (Fig. 6). Sequences from other alleged PVs) or viruses of regulatory concern (e.g., capripoxviruses cases of cervidpoxvirus infections, based on observed and malignant catarrhal fever). Similar to previous stud- gross lesions and virion ultrastructure, involving managed ies, examination of cell cultures by negative stain electron reindeer, managed pudu, and wild black-tailed deer were microscopy revealed large brick-shaped virions consistent not available for comparison but likely represent related with MDPV and unlike those reported in CePPVs [19–23]. cervidpoxviruses [18, 20, 21]. Although MDPV infections Although sampling methods and tissues collected have var- have typically been reported in cervids, partial DNA pol- ied among studies, the microscopic lesions reported here in ymerase (DPV047) and DNA topoisomerase (DPV086) Florida white-tailed deer are similar to previous cervidpox- sequences obtained from an inapparent poxvirus infection virus cases including epidermal hyperplasia with intracel- in a (family ) were found to be lular swelling and intracytoplasmic inclusions observed in nearly identical to previous MDPV isolates from Wyo- afected keratinocytes accompanied by a mixed infamma- ming mule deer [5, 24] and MDPV-F (data not shown). tory infltrate [19, 21, 23, 25]. As diferent bacteria were Thus, the host and geographic range of MDPV appears

1 3 K. A. Sayler et al. primarily restricted to North American managed and wild poxvirus genes [e.g., IMV membrane protein (DPV122)] populations of cervids within the subfamily . underscoring their usefulness in elucidating future cervid- Interestingly, phylogenomic analyses (i.e., comparison poxvirus epidemiological studies [5, 22, 23]. of gene content, size, and nucleotide identity) presented Although route(s) of MDPV transmission in wild or here reveal the MDPV isolate from a mule deer fawn in farmed cervid populations remain unknown, intradermal Wyoming in 1984 is more closely related to MDPV-F iso- injection, scarifcation, intravenous injection, and commin- lated from a Florida white-tailed deer fawn in 2016, as gling led to infection experimentally [24]. Thus, typical pox- compared to an MDPV isolate from a mule deer fawn in virus transmission infection route(s) including percutaneous 1983 only 40 km apart from the Wyoming isolate detected or respiratory transmission may be important [24]. Similar in the following year [5, 19]. Thus, strains of MDPV do to many poxviruses and typifed by capripoxviruses, the not appear restricted to a single cervid species, managed crusts from infected deer have been shown to contain infec- or wild populations, or geographic range within North tious virus and thus may serve to contaminate the environ- America. Nearly identical strains infecting both managed ment leading to persistence of disease in managed popula- and wild cervid species across North America raise impor- tions that commingle within confned enclosures [41]. The tant questions about whether anthropogenic movement of role of arthropod vectors in cervidpoxvirus transmission, cervids has led to the dissemination of MDPV and whether as observed for classifed members of the related genera infected managed populations pose a risk to naïve wild Leporipoxvirus (transmitted by mosquitoes) and Suipoxvirus cervid populations through pathogen pollution. (transmitted by lice), remains a possibility [41]. Future work The genome of the MDPV-F encodes a suite of immu- elucidating the route(s) of transmission would assist in the nomodulatory and host range genes as previously reported development of efective biosecurity measures. Finally, an for mule deer isolates from Wyoming in 1983 (W-848-83) infected managed pudu responded well to antiviral therapy and 1984 (W-1170-84) with the following exceptions: the suggesting chemotherapeutics as an option in cases of severe transforming growth factor beta 1 (DPV163) is only pre- MDPV disease [21]. sent in W-848-83, the CD30-like protein (DPV005) is missing in W-848-83, and the IFN-alpha/beta binding Compliance with ethical standards protein (DPV147) is truncated in W-1170-84 (Supplemen- tal Table 2). MDPV-F encodes an ortholog of a cellular Funding This study was funded by the University of Florida, Institute endothelin (DPV006) with an amino-terminal signal pep- of Food and Agricultural Science Cervidae Health Research Initiative, with funds provided by the State of Florida legislature. tide and a conserved endopeptidase cleavage site (data not shown) as previously reported for W-848-83 and W-1170- Conflict of interest All the authors declare that they have no confict 84 [5] and recently reported in a bat poxvirus [40]. The of interest. role of cervidpoxvirus orthologs of cellular endothelins is unknown; however, it has been hypothesized that their Ethical approval All animal procedures were reviewed and approved by the Institutional Animal Care and Use Committee at University of vasoactive and infammatory properties might contribute Florida (IACUC Protocol #: 201609390). to the proliferative cutaneous lesions observed in cervid- poxvirus infections, or alternatively, they may provide an advantage to the virus by reducing infammation as a com- petitive antagonist to endothelin receptors [5]. It has been postulated that cervidpoxviruses including MDPV-F have References evolved a novel mechanism to thwart IL-1beta-mediated 1. Skinner MA, Buller RM, Damon IK, Lefkowitz EJ, McFadden immune responses by encoding an interleukin 1 receptor G, McInnes CJ, Mercer AA, Moyer RW, Upton C (2012) Pox- antagonist protein (DPV054) [5, 22]. Finally, the MDPV-F viridae. In: King AMQ, Adams MJ, Carstens EB, Lefkowitz EJ encodes other immunomodulatory proteins found in most, if (eds) Virus taxonomy: ninth report of the International Committee not all, cervidpoxiruses, including a c-type lectin-like recep- on taxonomy of viruses. Elsevier Academic Press, New York, pp 291–309 tor protein (DPV144) and a major histocompatibility com- 2. Adams MJ, Carstens EB (2012) Ratifcation vote on taxonomic plex-1 protein (DPV151) that have been argued to modulate proposals to the International Committee on Taxonomy of antiviral natural killer cell immunity [5, 22, 23]. Results Viruses. Arch Virol 157:1411–1422 of BLASTN and BLASTX analyses (data not shown) of 3. Adams MJ, Lefkowitz EJ, King AMQ, Harrach B, Harrison RL, Knowles NJ, Kropinski AM, Krupovic M, Kuhn JH, Mushegian DPV054, DPV144, DPV151 revealed signifcant matches AR, Nibert M, Sabanadzovic S, Sanfaçon H, Siddell SG, Sim- to eukaryotic orthologs (i.e., white-tailed deer) supporting monds P, Varsani A, Zerbini FM, Gorbalenya AE, Davison AJ previous claims that they were captured from and facilitated (2017) Changes to taxonomy and the International Code of Virus adaptation to cervid hosts [4, 21]. These specifc cervidpox- Classifcation and Nomenclature ratifed by the International Committee on Taxonomy of Viruses. Arch Virol 162:2505–2538 virus genes exhibit higher genetic heterogeneity than core

1 3 Deerpox virus in Florida white-tailed deer

4. Upton C, Slack S, Hunter AL, Ehlers A, Roper RL (2003) Poxvi- 24. Bracht AJ, Armién AG, Carrillo C, O’Hearn ES, Fabian AW, rus orthologous clusters: toward defning the minimum essential Moran KE, Lu Z, Ariyakumer DS, Rasmussen JM, Metwally SA poxvirus genome. J Virol 77:590–600 (2013) Isolation and characterization of a cervidpoxvirus from a 5. Afonso CL, Delhon G, Tulman ER, Lu Z, Zsak A, Becerra VM, goitered gazelle (Gazella sugutturosa) from a zoological park in Zsak L, Kutish GF, Rock DL (2005) Genome of deerpox virus. J Minnesota. J Zoo Wildl Med 44:589–595 Virol 79:966–977 25. Bildfell RJ, Thompson KA, Moerdyk-Schauwecker M (2010) 6. Kummuneje K, Krogsrud J (1979) Contagious ecthyma (orf) in Experimental deerpox infection in black-tailed deer (Odocoileus reindeer (Rangifer tarandus tarandus). Vet Rec 105:60–61 hemionus columbianus). J Wildl Dis 46:33–45 7. Horner GW, Robinson AJ, Hunter R, Cox BT, Smith R (1987) 26. Afonso PP, Silva PM, Schnellrath LC, Jesus DM, Hu J, Yang Y, Parapoxvirus infections in New Zealand farmed red deer (Cervus Renne R, Attias M, Condit RC, Moussatché N, Damaso CR (2012) elaphus). N Z Vet J 35:41–45 Biological characterization and next-generation genome sequenc- 8. Scagliarini A, Vaccari F, Turrini F, Bianchi A, Cordioli P, Lavazza ing of the unclassifed Cotia virus SPAn232 (Poxviridae). J Virol A (2011) Parapoxvirus infections of red deer Italy. Emerg Infect 86:5039–5054 Dis 17:684–687 27. Haller SL, Peng C, McFadden G, Rothenburg S (2014) Poxviruses 9. Tikkanen MK, McInnes CJ, Mercer AA, Büttner M, Tuimala and the of host range and virulence. Infect Genet Evol J, Hirvelä-Koski V, Neuvonen E, Huovilainen A (2004) Recent 21:15–40 isolates of parapoxvirus of Finnish reindeer (Rangifer tarandus 28. Inoshima Y, Morooka A, Sentsui H (2000) Detection and diag- tarandus) are closely related to bovine pseudocowpox virus. J Gen nosis of parapoxvirus by the polymerase chain reaction. J Virol Virol 85:1413–1418 Methods 84:201–208 10. Klein J, Tryland M (2005) Characterisation of parapoxviruses 29. VanDevanter DR, Warrener P, Bennett L, Schultz ER, Coulter S, isolated from Norwegian semi-domesticated reindeer (Rangifer Garber RL, Rose TM (1996) Detection and analysis of diverse tarandus tarandus). J Virol 2:79 herpesviral species by consensus primer PCR. J Clin Microbiol 11. Lance WR, Charles HP, DeMartini J (1983) Experimental con- 34:1666–1671 tagious ecthyma in mule deer white-tailed deer and 30. Li Y, Meyer H, Zhao H, Damon IK (2010) GC content-based pan- wapiti. J Wildl Dis 19:165–169 pox universal PCR assays for poxvirus detection. J Clin Microbiol 12. Zarnke RL, Dietrich RA, Neiland KA, Ranglack G (1983) Sero- 48:268–276 logic and experimental investigation of contagious ecthyma in 31. Wernike K, Hofman H, Beer M (2015) Simultaneous detection Alaska. J Wildl Dis 19:170–174 of fve notifable viral diseases of cattle by single-tube multiplex 13. Roess AA, Galan A, Kitces E, Li Y, Zhao H, Paddock CD, Adem real-time RT-PCR. J Virol Methods 217:28–35 P, Goldsmith CS, Miller D, Reynolds MG, Zaki SR, Damon IK 32. Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, (2010) Novel Deer-associated parapoxvirus infection in deer hunt- Kulikov AS, Lesin VM, Nikolenko SI, Pham S, Prjibelski AD, ers. N Engl J Med 363:2621–2627 Pyshkin AV, Sirotkin AV, Vyahhi N, Tesler G, Alekseyev MA, 14. Falk ES (1978) Parapoxvirus infections in reindeer and Pevzner PA (2012) SPAdes: a new genome assembly algorithm associated with unusual human infections. B J Dermatol and its applications to single-cell sequencing. J Comput Biol 99:647–654 19:455–477 15. Smith KJ, Skelton HG, James WD, Lupton GP (1991) Parapoxvi- 33. Langmead B, Salzberg SL (2012) Fast gapped-read alignment rus infections acquired after exposure to wildlife. Arch Dermatol with Bowtie 2. Nat Methods 9:357–359 127:79–82 34. Milne I, Bayer M, Cardle L, Shaw P, Stephen G, Wright F, Mar- 16. Kuhl JT, Huerter CJ, Hashish H (2003) A case of human orf con- shall D (2010) Tablet-next generation sequence assembly visuali- tracted from a deer. Cutis 71:288–290 zation. Bioinformatics 26:401–402 17. Kitchen M, Müller H, Zobl A, Windisch A, Romani N, Huemer 35. Tcherepanov V, Ehlers A, Upton C (2006) Genome Annotation HP (2014) Orf virus infection in a hunter presumably transmitted Transfer Utility (GATU): rapid annotation of viral genomes using by in Western Austria. Acta Dermatovenereol 94:212–214 a closely related reference genome. BMC Genom 7:150 18. Barker IK, Mehren KG, Rapley WA, Gangon AN (1980) Kerato- 36. Katoh K, Standley DM (2013) MAFFT multiple sequence align- conjunctivitis and oral/cutaneous lesion associated with poxvirus ment software version 7: improvements in performance and usa- infection in reindeer. In: Montali RJ, Migaki G (eds) The com- bility. Mol Biol Evol 30:772–780 parative pathology of zoo animals. Smithsonian Institution Press, 37. Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Stur- Washington DC, pp 171–177 rock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, 19. Williams ES, Becerra VM, Graham TJ, Owens MJ, Nunamaker Ashton B, Meintjes P, Drummond A (2012) Geneious basic: CE (1985) Spontaneous poxviral dermatitis and keratoconjunc- an integrated and extendable desktop software platform for tivitis in free-ranging mule deer (Odocoileus hemionus) in Wyo- the organization and analysis of sequence data. Bioinformatics ming. J Wildl Dis 21:430–433 28:1647–1649 20. Patton JF, Nordhausen RW, Woods LW, MacLachlan NJ (1996) 38. Kumar S, Stecher G, Tamura K (2016) MEGA7: molecular evolu- Isolation of a poxvirus from a black-tailed deer (Odocoileus hemi- tionary genetics analysis version 7.0 for bigger datasets. Mol Biol onus columbianus). J Wildl Dis 32:531–533 Evol 33:1870–1874 21. Junge RE, Duncan MC, Miller E, Gregg D, Kombert M (2003) 39. Jin L, McKay A, Green R, Xu L, Bildfell R (2013) Serosurvey for Clinical presentation and antiviral therapy for poxvirus infection antibody to deerpox virus in fve cervid species in Oregon USA. J in pudu (Pudu puda). J Zoo Wildl Med 31:412–418 Wildl Dis 49:186–189 22. Moerdyk-Schauwecker M, Eide K, Bildfell R, Baker RJ, Black 40. Tu SL, Nakazawa Y, Gao J, Wilkins K, Gallardo-Romero N, Li W, Graham D, Thompson K, Crawshaw G, Rohrmann GF, Jin L Y, Emerson GL, Carroll DS, Upton C (2017) Characterization (2009) Characterization of cervidpoxvirus isolates from Oregon, of Eptesipoxvirus a novel poxvirus from a microchiropteran bat. California and eastern Canada. J Vet Diagn Investig 21:487–492 Virus Genes 53:856–867 23. Baughman B, Zhang S, Jin L, Pace LW, Cooley J, Yan L, Zhang 41. Maclachlan NJ, Dubovi EJ, Barthold SW, Swayne DF, Winton JR MZ (2011) Diagnosis of deerpox virus infection in a white- (2017) Fenner’s veterinary virology, 5th edn. Elsevier Academic tailed deer (Odocoileus virginianus) fawn. J Vet Diagn Investig Press, New York 23:965–970

1 3