key: cord-0264487-wswzj7z5
authors: George, Ankita M.; Wille, Michelle; Wang, Jianning; Anderson, Keith; Cohen, Shari; Moselen, Jean; Lee, Leo Yi Yang; Suen, Willy W.; Bingham, John; Dalziel, Antonia E; Hurt, Aeron C.; Williams, David T.; Deng, Yi-Mo; Barr, Ian G.
title: A novel and highly divergent Canine Distemper Virus lineage causing distemper in ferrets in Australia
date: 2021-11-04
journal: bioRxiv
DOI: 10.1101/2021.11.03.467217
sha: a0daaef91895a5bfb1c76cd1b6cd0ac1878e51c6
doc_id: 264487
cord_uid: wswzj7z5

Canine distemper virus (CDV) is a highly contagious systemic viral disease of dogs, that regularly spills-over into other animal species. Despite widespread vaccination, CDV remains endemic in many parts of the world. In this study we report an outbreak of distemper in ferrets in two independent research facilities in Australia. We found that disease severity varied, although most animals had mild to moderate disease signs. Histopathology results of animals with severe disease presented the typical profile of distemper pathology with multi-system virus replication. Through the development of a discriminatory PCR paired with full genome sequencing we revealed that the outbreak at both facilities was caused by a single, novel lineage of CDV. This lineage was highly divergent across the H gene, F signal peptide and full genome and had less than 93% similarity across the H gene to other described lineages, including the vaccine strain. Molecular analysis indicates that this strain belongs to a distinct lineage that diverged from other clades approximately 140 to 400 years ago, and appears to be unique to Australia. Given the differences in key viral proteins of this novel CDV strain, a review of the efficacy of the CDV vaccines currently in use in Australia is warranted to ensure maximum protection of dogs and other vulnerable species. In addition, enhanced surveillance to determine the prevalence of CDV in ferrets, dogs and other at-risk species in Australia would be useful to better understand the diversity of CDV in Australia. Importance Canine distemper virus (CDV) is highly contagious and while dogs are the main reservoir, it may spill over into a number of other animal species. In this study we report an outbreak of distemper in ferrets in two research facilities in Australia. Outcomes of pathology and histopathology suggest ferrets have widespread multi-systemic infection, consistent with previously reported distemper infections in ferrets and dogs. Critically, through sequencing and phylogenetic analysis, we revealed that the outbreak at both facilities was caused by a single, novel and highly divergent lineage of CDV. This virus had less than 93% nucleotide similarity to other described lineages and the vaccine strain. This manuscript adds considerably to the epidemiology, ecology and evolution of this virus, and is one of few reports of distemper in Australia in the literature.

due to the vaccine strain infection have been reported in South Africa and the UK (6, 22, 94 23). In Australia, routine vaccination is available for domesticated dogs, with vaccination 95 normally given to puppies and other domestic animals at risk of contracting CDV, such as 96 domesticated ferrets (11), using a live attenuated CDV. Despite a large-scale ongoing 97 vaccination strategy in Australia, strains of CDV continue to circulate in animal populations 98 (11). For example, in a study conducted from 2006-2014 in dogs and ferrets in Australia, it 99 was reported that there were confirmed or suspected cases of CDV in five states/territories 100 with most cases reported in New South Wales, including two out of three tested ferrets 101 (11). However data pertaining to CDV epidemiology in Australia is limited, and the modes 102 and rates of infection and mortality are not well understood. 103 In this study we report an outbreak of CDV in the ferret population in the Australian state of 104 Victoria in 2019. CDV was detected in ferrets supplied to two independent research 105 facilities from different breeders, indicating that this outbreak likely occurred across wide 106 regions of the state. Herein we describe this CDV outbreak based on clinical signs and 107 qRT-PCR CDV detection and describe the histopathology of severe cases. To better 108 discriminate the dynamics of the outbreak we developed an assay to discriminate between 109 a commercial CDV vaccine and the field strain. By full genome sequencing we also 110 determined that the CDV in this outbreak originated from a novel lineage of CDV. Identification of ferrets with CDV by Two different PCR approaches were used across the two research facilities. At the 159 Doherty Institute a commercial CDV quantitative real-time reverse-transcriptase PCR 160 (qRT-PCR) was initially employed followed by the development of a discriminatory assay.

All ferrets received into the facility were tested. At ACDP, a pan-morbillivirus RT-PCR was 162 applied for detection of CDV from ferrets with clinical symptoms (25).

At the Doherty Institute nasal wash samples were collected from lightly sedated ferrets.

Ferrets were sedated by intramuscular injection of a combination of ketamine (12.5mg/kg, 166 Troy Laboratories) and xylazine (2.5mg/kg, Troy Laboratories). We instilled 1 mL of sterile 167 PBS into one nostril and allowing the liquid to flow out of the other nostril into a collection 168 tube. Nasal wash samples were immediately stored at -80°C until RNA extraction. RNA 169 was extracted from 140µl of nasal wash sample using the QIAamp Viral RNA Mini kit 170 (QIAgen, Australia), according to the manufacturer's instructions. RNA was also extracted 171 from the Protech C3 vaccine (Boehringer Ingelheim, Australia) following reconstitution 172 according to manufacturer's instructions.

Primers were developed for a number of purposes (Table 1) 

To discriminate between the outbreak virus and the vaccine strain a discriminative qRT-189 PCR was developed using a two-probe assay (Table 1) Prevalence of CDV at the Doherty Institute was calculated using the bioconf() function in 217 the Hmisc package (26) and prevalence differences between the field strain and vaccine 218 strain at the Doherty Institute ferret facility were compared using a Chi-squared test. Initial sequencing was done at the Doherty Institute with Sanger sequencing. RNA that 225 was stored at -80°C and had undergone limited freeze thaw cycles was used for RT-PCR 226 using the MyTaq One step RT-PCR kit (Meridian Biosciences) and primers designed for 227 Sanger Sequencing (Table 1) . Product sizes were confirmed using the e-Gel 2% agarose 228 (GP) (Invitrogen, USA) followed by purification using the Exo-SAP-IT PCR product clean-229 up reagent (Thermofisher, Australia) according to manufacturer instructions. The purified 230 template then underwent sequencing using the BigDye Terminator v3.1 Cycle sequencing 231 kit (Thermofisher). The primers used for this step were 10% the concentration of the same 232 primers used to generate PCR products, unless the primer had a M13 tail, in which case a 233 M13 F(5'-TGTAAAACGACGGCCAGT-3') and R (5'-CAGGAAACAGCTATGACC3') was 234 used. Sequencing was performed on a 3500XL Genetic Analyser (Applied Biosystems).

Results were analysed using Lasergene 13 (DNASTAR, Madison, WI, USA).

For subsequent next-generation sequencing (NGS) at the Doherty Institute, RNA was 237 extracted from stored (-80 o C) nasal washes as described previously. Briefly, RT-PCR 238 products were created using the Superscript IV One Step RT-PCR system (Thermofisher) 239 and primers designed based on the results of Sanger Sequencing (Table 1) prior to sampling (approximately 3 weeks prior to nasal wash sampling). Overall, there was 303 a statistically significant difference in the proportion of detections due to the field strain 304 (39%) and vaccine strain (56%) (X 2 =12.921, df=1, p=0.003) (Fig 1) . In 20 ferrets, both the 305 vaccine and field strain of the virus were detected. Despite finding the vaccine strain in all 306 sampling time periods following the routine introduction of CDV vaccination and having a 307 higher overall prevalence, we found that the CDV outbreak caused by the field strain 308 persisted for less than two months (Fig 1) . Following the index cases on 3 April 2019, the 309 field strain was detected in approximately 70% of animals on 7 May, 50% on 21 May, but 310 was no longer detected in ferrets from June 2019. This is in contrast to the vaccine strain, 311 which was consistently detected during the seven months following CDV vaccine 312 implementation on 3 April (Fig 1) . There was no statistically significant difference in 313 disease outcomes (i.e. requiring euthanasia under the ethical guidelines) when comparing 314 ferrets infected with the field strain and the vaccine strain; 3 ferrets positive for the vaccine 315 strain were euthanised and 3 ferrets infected with the field strain were euthanised. An 316 additional ferret that was negative for CDV was euthanised, and three additional ferrets 317 that were not tested for CDV were euthanised (Table 2) . (Fig 2A) inguinal dermatitis, with rough, discoloured patches of skin appearing on 336 the abdomen, (Fig 2B) hyperkeratosis of the footpads, also known as hardpad, (Fig 2C) 337 dermatitis on the chin and mouth, with scaly patches forming, and/or ocular signs including 338 uncontrolled eye twitching with mucopurulent ocular and nasal discharge, resulting in 339 visible crusting around the eyes and nose (Fig 2) . At the Doherty Institute, visible signs Four ferrets with distemper-like clinical signs and one asymptomatic ferret from the ACDP 358 outbreak were submitted for histopathological analysis. Three of the symptomatic ferrets 359 were submitted early in the outbreak, another symptomatic ferret was submitted days after 360 vaccination was introduced, along with an asymptomatic penmate. In these ferrets, 361 necrosis and sloughing of the bronchiolar epithelium were observed in the lung. This was 362 often accompanied by notable diffuse epithelial hyperplasia ( Fig 3A) and occasionally with 363 a lymphohistiocytic infiltrate. In some of the affected areas, frequent round eosinophilic 364 inclusion bodies were detected in the cytoplasm and nucleus of bronchial/bronchiolar 365 epithelial cells (Fig 3A) . Occasional multinucleated syncytial cells were found lining alveoli.

Similar syncytial cells and intracytoplasmic/intranuclear inclusion bodies were identified in 367 hyperplastic parts of the renal pelvic and urinary bladder urothelium (Fig 3B) . The skin 368 over the nasal planum was also hyperplastic with prominent parakeratotic hyperkeratosis 369 emanating most notably from hair follicles. Associated suppurative inflammation was also 370 occasionally observed in the subjacent dermis.

In addition to above lesions, multifocal to coalescing patches of the lymph node cortex 373 were depleted in lymphocytes (Fig 3C) . In the more severely affected lymph nodes, these 374 lymphocyte-depleted areas in the cortex were replaced by oedema, fibrin, histiocytes and 375 fibroplasia ( Fig 3C) . Taken together, the histopathologic features described above were 376 consistent with pathology typically associated with CDV infection in ferrets.

Immunohistochemistry targeting the nucleocapsid protein of CDV showed that viral 379 replication was widespread and intense, affecting many organ systems, even in animals 380 with minimal histopathologic changes. Antigen was detected in the epithelium of the 381 respiratory tract (Fig 3D) , renal pelvis (Fig 3E) , skin of the nasal planum, biliary tract, 382 alimentary tract, female reproductive tract, lacrimal gland and salivary gland. High viral 383 antigen burden was also observed in lymph nodes and spleen. Round cells with 384 morphology consistent with histiocytes and lymphocytes were the main targets in these 385 lymphoid organs (Fig 3F) . In the brain, viral antigen was detected in only few mononuclear Sequence analysis and evolutionary genetics 390 We sequenced 3 complete genomes corresponding to the field strain of CDV, in addition 391 to 8 complete genomes of the CDV vaccine strain from ferrets. Both the field and vaccine 392 strain viral genomes sequenced at the Doherty Institute were recovered from nasal wash 393 samples. The field strain viral genomes from ACDP were sequenced from FFPE brains.

Most vaccine strain viral genomes from ACDP were sequenced from fresh tissue samples 395 (nasal swabs and brain) and only one was from FFPE brain. Analysis of the three field Analysis of the H (Fig 4) and Fsp (Fig 5A) genes of the field strain viral sequences, regions 407 of the CDV genome currently used for lineage discrimination (17), clearly demonstrated 408 the divergent nature of the field strain. Indeed, based upon these genes and the full 409 genome sequences (Fig 5B) , and the current approach for lineage designation (e.g. (17), 410 the field strain should be designated as a novel lineage. Critically, the field strain is highly 411 divergent from both the Protech C3 vaccine itself and from virus sequences from ferrets 412 vaccinated with this attenuated strain. More specifically, the H gene shared less than 93% 413 nucleotide identity with all existing lineages of CDV, including the Protech C3 vaccine 414 sequence [92.5% nucleotide similarity, 91.8% amino acid similarity] which we also 415 sequenced in this study (Fig 4) . Phylogenetically, the outbreak virus belonged to the 416 lineage comprising "Vaccines (both the Protech C3 and the Onderstepoort strain which is 417 used in most attenuated vaccines globally), Asia-3 and North America -1", as defined by 418 (5) (Fig 4) . Furthermore, this sequence is highly divergent from other CDV sequences from 419 mustelids, globally, with the exception of 4 sequences from China which fall into the "Asia-420 3" lineage (Fig 4) . Other mustelid sequences fell into an array of lineages, although the 421 vast majority belonged to the "Asia-1" lineage (Fig 4) . As outlined in previous studies, the 422 H gene phylogeny is dictated largely by geography, and we speculate that the field strain 423 14 reported here is representative of CDV currently circulating in Australia, however, as there 424 are no available sequences in GenBank with which to compare, this cannot be confirmed. 425 We found similar patterns in the Fsp gene and full genome analyses (Fig 4) . Phylogenetic 426 analysis consistently demonstrated the field strain as being highly divergent (Fig 5) is important to take the specific dates merely as a guide. By also running this tree with an 445 uncorrelated relaxed lognormal clock we were able to incorporate the vaccine strain, 446 however this approach moves tMRCA much further back in time. This approach confirms 447 that the divergence of CDV strains causing the outbreak described in this study occurs 448 before the bifurcation of the "vaccine" clade from the "Asia-3" and "North America-1" 449 clades, which is illustrated in all the maximum likelihood trees presented (Fig 4, 5) . constitutes an Australian lineage that has been geographically isolated since the 492 divergence from the Vaccine/Asia-3/North America-1, or constitutes a lineage that was 493 cryptically circulating elsewhere and has only recently been introduced to Australia. It is 494 notable that there are a number of lineages circulating on most continents and in countries 495 where CDV has been investigated (17, 39, 40), so it is not unreasonable to assume long -496 term circulation in Australia, and, whether additional viral diversity will be found in Australia 497 remains to be determined.

In addition to detecting a novel lineage, we found that 51% of ferrets tested with a 499 discriminatory PCR were still positive for the vaccine strain for up to 3 months after 

Ferrets with signs of CDV in this outbreak varied in disease severity, in contrast to the very 523 high mortality rates, with some outbreaks comprising a 100% mortality rate, reported 524 previously in mustelids (11, 46, 47) . This lower mortality, specifically at the Doherty 525 Institute, may have been in part due to intervention measures that were used on 526 suspected CDV infected ferrets, such as injections of vitamin A (36, 48), the brief time 527 period between identification of disease signs and interventions (1-2 days), and the 528 decision to cull ferrets at an early disease stage in order to control the spread of the 529 outbreak. The disease signs observed in animals that required intervention did not display 530 the full extent of morbidity previously reported (12, 13), however, this may similarly be 531 confounded by the decision to intervene early in the disease course in order to contain the 532 outbreak. Therefore, in the current investigation, the true pathogenicity of this new CDV 533 strain remains unclear. A dedicated experimental challenge trial in ferrets will be required 534 in order to determine this. Nevertheless, in ferrets submitted for histopathology, the typical 535 profile of distemper pathology and multi-systemic virus replication were observed which 536 suggests that, at least for a subset of infected ferrets, the highly divergent CDV strain 537 described in our current study is capable of producing typical CDV-induced disease.

Overall CDV continues to pose a large disease burden, globally (49). In addition to dogs, 539 this virus also has an impact on wildlife, with a number of examples of CDV infection in 540 large cats, hyenas and jackals from many parts of Africa (8, 50) . Due to the broad host 541 range of CDV, cross-species infection is known to occur (50, 51), with the same lineages 542 of CDV having been detected in canids, felids, mustelids and even seals. This is of 543 concern, particularly in Australia, where several species of native fauna are currently at 544 risk of extinction due to other factors (52). If this strain of CDV was to potentially infect 545 these at-risk and immunologically naive animals, it could potentially result in widespread 546 disease and high rates of mortality (6). One of the limiting factors in mounting an 547 appropriate response would be the lack of established methods for CDV surveillance 548 across the country (11). Specifically, following vaccination with a live attenuated vaccine, 549 current diagnostic assays may not be able to accurately discriminate between the 550 presence of vaccine or a circulating strain. Due to the geographic isolation of some parts 551 of Australia it is possible that other genetically distinct strains of CDV are present in the 552 country, such that the discriminatory PCR developed here may have limited value in an 553 outbreak with a different lineage. Currently CDV diagnosis is still best achieved through a 554 combination of histopathology and qRT-PCR and should be interpreted in the context of 555 case history and clinical presentation (11). The detection of this novel lineage signals the 556 need, not only for more widespread surveillance of CDV in Australia, but also for 557 sequencing of any viruses that are detected to better understand the diversity of this 558 lineage and to reveal any other lineages that may be circulating on the continent. As the 559 ferrets reported in this study were sourced from regional breeders, the presence of CDV in 560 this population may indicate that the virus is present in both domestic and wild animals in 561 the state of Victoria. Our study describes the endpoint of an outbreak of CDV and 562 highlights the crucial need for CDV surveillance and sequencing to better disentangle CDV 563 ecology and evolution which is a key component for future outbreak response and control. 

ICTV Virus 582 Taxonomy Profile: Paramyxoviridae

Tropism and molecular pathogenesis of Canine Distemper Virus

Diversity of susceptible hosts in canine 588 distemper virus infection: a systematic review and data synthesis

The Canine Morbillivirus strain associated with an epizootic in 592

Caspian Seals provides new insights into the evolutionary history of this virus

Evolution and 595 interspecies transmission of Canine Distemper Virus -an outlook of the diverse 596 evolutionary landscapes of a multi-host virus

Cosby 599 SL. 2019. Canine and Phocine Distemper Viruses: global spread and genetic basis 600 of jumping species barriers

Uncommon acute neurologic presentation of canine distemper in 4 adult 604 dogs

Phylogenetic analysis of canine distemper virus in South African wildlife

Diversity of susceptible hosts in canine 609 distemper virus infection: a systematic review and data synthesis

Canine Distemper Virus infection of giant pandas in China

Epidemiology and clinical presentation of 615 canine distemper disease in dogs and ferrets in Australia

Outbreak of canine 618 distemper in domestic ferrets (Mustela putorius furo)

Small animal models of filovirus 622 disease: recent advances and future directions

Development of spike receptor-binding domain 626 nanoparticle as a vaccine candidate against SARS-CoV-2 infection in ferrets

Infection of eight in 2016-2017

Phylogenetic evidence of the intercontinental circulation of a Canine 634

Distemper Virus lineage in the

Canine distemper: re-636 emergence of an old enemy

Recent increase in canine distemper

Associations between 640 biosecurity and outbreaks of canine distemper on Danish

Reversion to virulence of attenuated Canine Distemper Virus in 643 vivo and in vitro

Canine distemper in African cape hunting dogs (Lycaon 645 pictus): possibly vaccine induced

Lights and shades on an 648 historical vaccine canine distemper virus, the Rockborn strain

Australian code for the care 651 and use of animals for scientific purposes, 8th ed. National Health and Medical 652 Research Council, Canberra

Dolphin and porpoise morbilliviruses are genetically distinct from phocine distemper 655 virus

Hmisc: Harrell Miscellaneous. R package

ggplot2: elegant graphics for data analysis

Fast gapped-read alignment with Bowtie 2. 661

A greedy algorithm for aligning 663 DNA sequences

MAFFT multiple sequence alignment software version 665 7: improvements in performance and usability

New 667 algorithms and methods to estimate maximum-likelihood phylogenies: assessing 668 the performance of PhyML 3.0

Exploring the temporal 670 structure of heterochronous sequences using TempEst (formerly Path-O-Gen)

Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10

Canine 677 morbillivirus (canine distemper virus) with concomitant canine adenovirus, canine 678 parvovirus-2, and Neospora caninum in puppies: a retrospective 679 immunohistochemical study

New aspects of the pathogenesis of canine 683 distemper leukoencephalitis

Management of a Canine Distemper Virus outbreak in a ferret 685 rescue facility

Rabies and 688

Canine Distemper Virus epidemics in the red fox population of northern Italy

Viral diseases of Ferrets

Canine distemper virus in the Serengeti ecosystem: molecular adaptation to 696 different carnivore species

Phylogenetic analysis of the haemagglutinin gene of canine distemper virus 700 strains detected from breeding foxes, raccoon dogs and minks in China

Protective immunity against Canine Distemper Virus in dogs induced 704 by intranasal immunization with a recombinant probiotic expressing the viral H 705 protein

Canine Distemper Viruses expressing a 707 hemagglutinin without N-glycans lose virulence but retain immunosuppression

Canine Distemper Virus (CDV) infection of ferrets as a model for testing 711

Morbillivirus vaccine strategies: NYVAC-and ALVAC-based CDV recombinants 712 protect against symptomatic infection

Antigenic differences in the H proteins of Canine Distemper Viruses

Influence of vaccine strains on the evolution of 718

Canine Distemper Virus

Musculoskeletal and Neurologic Diseases

Vaccination of ferrets for rabies and distemper

Disease manifestations of Canine Distemper Virus infection in ferrets are modulated 725 by vitamin A status

Canine Distemper Virus from diseased 728 large felids: biological properties and phylogenetic relationships

Cross-species

Cross-species transmission of 735 canine distemper virus-an update

Ongoing unraveling of a continental 737 fauna: decline and extinction of Australian mammals since European settlement

Detection of canine distemper 740 virus nucleoprotein RNA by reverse transcription-PCR using serum, whole blood, 741 and cerebrospinal fluid from dogs with distemper

A multiplex quantitative real-time polymerase 743 chain reaction panel for detecting neurologic pathogens in dogs with 744 meningoencephalitis

A fast and simple one-step duplex PCR 746 assay for canine distemper virus (CDV) and canine coronavirus (CCoV) detection

Institute ferret facility. Grey bars indicate when, and the number of samples tested using the 816 discriminatory PCR developed here and tick marks are on the secondary Y axis. Points represent 817 the individual point estimates for the vaccine strain and the field strain. Lines correspond to 818 modelled prevalence using a generalized linear model and shaded areas represent the 95% 819 confidence interval of the model. We were unable to generate a corresponding outbreak figure for 820 ACDP as they did not use a discriminatory PCR.