key: cord-297775-ug4ovsws authors: Hosie, Margaret J; Epifano, Ilaria; Herder, Vanessa; Orton, Richard J; Stevenson, Andrew; Johnson, Natasha; MacDonald, Emma; Dunbar, Dawn; McDonald, Michael; Howie, Fiona; Tennant, Bryn; Herrity, Darcy; Da Silva Filipe, Ana; Streicker, Daniel G; Willett, Brian J; Murcia, Pablo R; Jarrett, Ruth F; Robertson, David L; Weir, William title: Respiratory disease in cats associated with human-to-cat transmission of SARS-CoV-2 in the UK date: 2020-09-23 journal: bioRxiv DOI: 10.1101/2020.09.23.309948 sha: doc_id: 297775 cord_uid: ug4ovsws Two cats from different COVID-19-infected households in the UK were found to be infected with SARS-CoV-2 from humans, demonstrated by immunofluorescence, in situ hybridisation, reverse transcriptase quantitative PCR and viral genome sequencing. Lung tissue collected post-mortem from cat 1 displayed pathological and histological findings consistent with viral pneumonia and tested positive for SARS-CoV-2 antigens and RNA. SARS-CoV-2 RNA was detected in an oropharyngeal swab collected from cat 2 that presented with rhinitis and conjunctivitis. High throughput sequencing of the virus from cat 2 revealed that the feline viral genome contained five single nucleotide polymorphisms (SNPs) compared to the nearest UK human SARS-CoV-2 sequence. An analysis of cat 2’s viral genome together with nine other feline-derived SARS-CoV-2 sequences from around the world revealed no shared catspecific mutations. These findings indicate that human-to-cat transmission of SARS-CoV-2 occurred during the COVID-19 pandemic in the UK, with the infected cats developing mild or severe respiratory disease. Given the versatility of the new coronavirus, it will be important to monitor for human-to-cat, cat-to-cat and cat-to-human transmission. pandemic, naturally occurring SARS-CoV-2 infections linked to transmission from humans have been reported in domestic cats (1, 2) , non-domestic cats (3), dogs (4) and mink (5) . In addition, in vivo experiments have shown that while cats, ferrets and hamsters are susceptible to SARS-CoV-2 infection, ducks, chickens and pigs are apparently not susceptible (6, 7) . Cat-to-cat transmission has been demonstrated experimentally (6) (8) , but the significance of SARS-CoV-2 as a feline pathogen, as well as its reverse zoonotic potential, remains poorly understood. If SARS-CoV-2 were to establish new animal reservoirs, this could have implications for future emergence in humans. At present, there is no evidence of cat-to-human transmission or that cats, dogs or other domestic animals play any appreciable role in the epidemiology of human infections with SARS-CoV-2. However, although the pandemic is currently driven by human-to-human transmission, it is important to address whether domestic animals are susceptible to disease or pose any risk to humans, particularly those individuals who are more vulnerable to severe disease. Domestic animals could also act as a viral reservoir, allowing continued transmission of the virus, even when Ro < 1 in the human population. Recent reports from Dutch mink farms of both mink-to-cat and mink-tohuman transmission of the virus provide support for this scenario (5, 9) We used a range of laboratory techniques to show that two domestic cats from households with suspected cases of COVID-19, and which displayed either mild or severe respiratory disease, were infected with SARS-CoV-2. These findings confirm that human-to-cat transmission of SARS-CoV-2 occurs and can be associated with signs of respiratory disease in cats. Sections of lung tissue were collected post-mortem from cat 1, placed in virus transport medium and stored at -80 °C on 22 April 2020; on 10 June 2020 the virus transport medium (VTM) was removed, and RNAlater ® was added. Lung tissue was also stored in formalin from 22 April until 8 June, when it was processed to wax prior to immunohistochemistry. Infection of cat 2 was identified via a retrospective survey of oropharyngeal and/or conjunctival swabs collected from 387 cats with respiratory signs that had been submitted to the University of Glasgow Veterinary Diagnostic Service (VDS) between March and July 2020 for routine pathogen testing. Ethical approval for this study was granted by the University of Glasgow School of Veterinary Medicine ethics committee (EA27/20). Permission was given for the retrospective analysis of feline swabs submitted to VDS for routine respiratory pathogen testing. Permission was also granted for a public appeal to practising veterinary surgeons via the Veterinary Record, to solicit the submission of samples from suspect SARS-CoV-2 cases (10). This appeal was in line with guidance to veterinarians on the testing of animal samples for SARS-CoV-2 from the Animal and Plant Health Agency (APHA), issued on 13 May (11). This briefing note confirmed that testing of animals for the purpose of clinical research was permitted under appropriate ethical review. Approval to test tissue samples collected post-mortem from cat 1 in the study was obtained from the primary veterinary surgeon. On submitting samples to Scotland's Rural College (SRUC) Veterinary Services, veterinary practices agree that any sample may be used to investigate new and emerging diseases. Samples were received in VTM and screened for feline herpes virus (FHV), feline calicivirus (FCV) and Chlamydia felis (C. felis). DNA extracts from VTM samples were tested for the presence of FHV and C. felis using a multiplex quantitative polymerase chain reaction (qPCR) approach. The assay incorporated published C. felis primers (12) together with primers/probes for FHV and a feline host control gene which were designed in-house. Standard respiratory virus isolation was also attempted using proprietary feline embryonic (FEA) cells. The remnants of these samples were stored at 4°C prior to testing for SARs-CoV-2. TRIzol™ Reagent (ThermoFisher Scientific, Paisley, UK) was added to lyse the sample and ensure inactivation of SARS-CoV-2, followed by organic solvent extraction using chloroform;isoamyl alcohol. Subsequent steps were performed using RNeasy® Mini Kits (Qiagen, Manchester, UK) as per the manufacturer's instructions, with elution of the final RNA sample in 55 µl nuclease-free water. One mock RNA extraction was performed for every seven samples. All samples were tested using two reverse (14) . Negative controls processed in parallel retrieved no viral mapped reads after primer trimming. The created viral genome sequence for cat 2 was uploaded to GISAID with the accession number EPI_ISL_536400. The closest UK human SARS-CoV-2 sequence was initially identified using the COG-UK cluster identification tool civet (https://github.com/COG-UK/civet). A maximumlikelihood phylogenetic tree of all unique human SARS-CoV-2 sequences from the same county as cat 2 (n = 324), along with the cat 2 genome, the closest UK human sequence and the Wuhan-Hu-1 reference, was created using IQ-TREE (15) with the GTR substitution model (selected by IQ-Tree ModelFinder) and 1000 bootstraps. Existing feline (n = 9; Belgium, China, France, Spain, USA) and mink (n = 13; Netherlands) SARS-CoV-2 viral genome sequences were downloaded from the GISAID website (https://www.gisaid.org) on 31 July 2020. suggesting that type I pneumocytes were infected ( Figure 1B ). In contrast, neither viral protein nor RNA was detected in the liver. To characterise cat 2's viral genome, we performed high-throughput sequencing on RNA derived from the clinical specimen. The generated viral genome sequence was 97.2% complete and contained 13 single nucleotide polymorphisms (SNPs) when compared with the original Wuhan_Hu-1 reference sequence. Sequence data from the symptomatic owner were not available and therefore we compared the feline genome with human SARS-CoV-2 sequences, using data from the COVID-19 Genomics UK (COG-UK) consortium. The mutational hamming distance (ignoring Ns and ambiguities) between the cat 2 viral genome and all COG-UK human viral genomes available on 23 August 2020 revealed that the closest human SARS-CoV-2 sequences from the UK differed from the feline sequence by five SNPs (n = 141; Table 1 ); these human sequences were distributed throughout the UK but predominantly (88%) (16) assigned to one lineage. The closest sequences (n = 11) from the same county as cat 2 were an additional SNP away. Phylogenetic analyses of these sequences reinforced the close relationship between the cat 2 viral genome and human-derived UK SARS-CoV-2 genomes ( Figure 3 ). As we do not have the owner's virus sequence, we cannot determine whether the observed mutations in cat 2's viral genome arose in a human prior to transmission. Table 1 details the SNPs observed in the cat 2 viral genome, and their frequency in the existing UK human population and among existing feline SARS-CoV-2 sequences. Six of the 13 SNPs are widespread (>50%) in the UK human population and only three have not been observed previously. It is most likely that the three novel SNPs arose recently as evolutionary bottlenecks during human-to-human transmission and represent an unsampled cluster of human variants. Given that no other feline or mink sequences contained these mutations, there is little indication that these correspond to a host species adaptation of the virus. Next we examined all globally available feline SARS-CoV-2 sequences from the GISAID database for evidence of convergent mutations. Each of the six existing complete feline viral genomes contained 3 SNPs in common with cat 2 resulting in the D614G mutation in Spike, the P323L mutation in nsp12, and a synonymous mutation in nsp3. However, as these mutations are widespread in the human population, it is likely that they evolved in humans and are not associated with feline adaptation. The existing feline viral sequences were mutation distances of 0 (n = 4), 1 (n = 1), and 3 (n = 1) SNPs away from the closest human SARS-CoV-2 sequence in their respective countries. It has been suggested that the D614G mutation in spike (shared by the feline SARS-CoV-2 genomes) confers a fitness advantage to the virus in humans (17, 18), whether the same mutation renders the virus more infectious for cats remains to be established. This is the first report of human-to-cat transmission of SARS-CoV-2 in cats in the UK. Although the ongoing SARS-CoV-2 pandemic is driven by human-to-human These findings have potential implications for the management of cats owned by people who develop SARS-CoV-2 infection. Currently, there is no evidence that domestic cats have played any role in the epidemiology of the COVID-19 pandemic, but a better understanding of how efficiently virus is transmitted from humans to cats will require cats in COVID-19 households to be monitored. The two cases of reverse zoonotic infections that are reported here serve to highlight the importance of a coordinated One Health approach between veterinary and public health organisations. First detection and genome sequencing of SARS-CoV-2 in an infected cat in France First Reported Cases of SARS-CoV-2 Infection in Companion Animals Infection of dogs with SARS-CoV-2 SARS-CoV-2 infection in farmed minks, the Netherlands Susceptibility of ferrets, cats, dogs, and other domesticated animals to SARS-coronavirus 2 Pathogenesis and transmission of SARS-CoV-2 in golden hamsters Transmission of SARS-CoV-2 in Domestic Cats Jumping back and forth: anthropozoonotic and zoonotic transmission of SARS-CoV-2 on mink farms Send cat and dog samples to test for SARS-CoV-2 Factors associated with upper respiratory tract disease caused by feline herpesvirus, feline calicivirus, Chlamydophila felis and Bordetella bronchiseptica in cats: experience from 218 European catteries Fast and accurate short read alignment with Burrows Wheeler Transform An amplicon-based sequencing framework for accurately measuring intrahost virus diversity using PrimalSeq and iVar IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era COG. 2020 The COVID-19 Genomics UK Consortium. Evaluating the effects of SARS-CoV-2 Spike mutation D614G on transmissibility and pathogenicity. medRxiv 2020073120166082 A Critical Needs Assessment for Research in Companion Animals and Livestock Following the Pandemic of COVID-19 in Humans. Vector Borne Zoonotic Dis The risk of SARS-CoV-2 transmission to pets and other wild and domestic animals strongly mandates a onehealth strategy to control the COVID-19 pandemic. One Health This study was supported by an award to MJH, BJW, RFJ, PRM and WW from the The authors have no potential competing interests.