Scottish Wildcat Action (SWA) Specialist Report - Disease Surveillance

We should consider ‘significant’ diseases of wildcats (Felis silvestris) to be those with the capacity to compromise the sustainability of the wild-living population and its ability to withstand change. Interactions amongst and between domestic cats (Felis catus) and wildcats provide opportunities for infectious disease transmission, and prior to the start of this project there was evidence that wild-living wildcats in Scotland were infected by, and exposed to, a number of infectious agents well-known in domestic cats, which can cause disease and mortality in that species. Infectious disease has negatively impacted some endangered populations of wild felids elsewhere in the world and has the potential to impact the wildcat in Scotland, especially in the face of simultaneous threats such as habitat loss, population fragmentation and food scarcity. Therefore, infectious disease threats to wildcats merit further investigation, as does the extent to which wildcats are exposed to environmental toxins, which is largely unknown.

Summary of work

• The aim of this work was to increase our knowledge base relating to disease threats to wildcats in Scotland, most specifically infectious agents and rodenticides (rodent poisons), and to explore relationships between exposure to these agents and potential risk factors such as hybridisation status.

• From 2015-2020 a range of samples was collected from wild-living cats, including feral domestic cats, and cats of wildcat, or domestic-wildcat hybrid, appearance (pelage). The opportunities to access these cats were through the Trap, Neuter, Vaccinate, Release (TNVR) scheme for feral domestic cats and cats with low-scoring hybrid pelage, and through the targeted trapping of live presumed wildcats and cats with high-scoring hybrid pelage. Samples taken included blood samples, and swabs from the mouth, inner eyelids and rectum, to detect evidence of infection with the following feline infectious agents: feline immunodeficiency virus (FIV), feline leukaemia virus (FeLV), feline calicivirus, feline herpesvirus, Bordetella bronchiseptica, Chlamydia felis, Mycoplasma felis, bloodborne Mycoplasma species and Tritrichomonas foetus. In addition, post-mortem examinations were carried out on carcases of presumed wild-living cats, particularly those found on roads, which were submitted by wildlife rangers and members of the public. A proportion of these carcases was radiographed for evidence of gun-shot. Other post-mortem samples were taken, where possible, for infectious agent analyses and detection of rodenticides.

• To date, samples have been collected from 207 live cats trapped through the TNVR scheme and from 17 live cats with wildcat or high-scoring hybrid pelage, trapped through the wildcat trapping scheme. Results are presented for all the trapped wild-living cats as a whole, since through the course of the project it became evident that there was a continuous spectrum of hybridisation within the sampled population.

• All 11 infectious agents we screened for were detected in the sampled population of live wild-living cats. FIV (antigen) was detected in 10 (7%) of 144 live cats tested and was significantly more prevalent in trapped individuals at the ‘domestic cat’ end of the genetic hybrid spectrum.

• Eighty-one presumed wild-living cat carcases were also examined and samples taken where feasible. FIV (antigen) was detected in the blood of two (11%) of 18 cat carcases tested, which were both domestic cat-wildcat hybrids based on genetic testing. There were abnormalities typical of collision with road traffic in at least 58% of the 81 carcases, and lead shot was detected in the carcases of six cats, of which all but one had domestic-cat pelage. However, four were found to be genetic domestic cat-wildcat hybrids. Rodenticides were detected in 30 (61%) of 49 liver samples tested and in 13 (27%) of these 49 cases, the residues were above a threshold level associated with toxic effects in birds and mammals.

• Our findings show that wild-living cats in Scotland are infected with a broad range of infectious agents and have accumulated rodenticides in their tissues. Both infectious agents and rodenticides are potential threats to the health of domestic cat-wildcat hybrids and wildcats. We have also highlighted road-traffic collisions and shooting as causes of mortality in domestic cat-wildcat hybrids.

• We have expanded the range of infectious agents detected in wild-living cats across the hybrid spectrum in Scotland. To our knowledge, this is the first time FIV has been detected in domestic cat-wildcat hybrids and it is a concern, since, like FeLV (which was detected in 3% of cats), in domestic cats the virus typically persists in an individual long-term, predisposing them to other disease conditions and ultimately hastening mortality. An apparent association of FIV with domestic cat genotypes, and its lack of detection in previous wildcat studies in Scotland, suggests that infection is likely to have originated in domestic cats. It is probable that increased interactions between domestic cats, hybrid cats and wildcats in recent decades, as evidenced by increasing levels of introgressive hybridisation, have, in parallel, led to greater exposure of wildcats and hybrid cats to FIV and other pathogens associated with domestic cats. The ability of these infectious agents to cause disease in wildcats and wildcat-domestic cat hybrids, and their potential to impact population dynamics, are still poorly understood and merit further investigation.

• Similarly, the current impacts of road-traffic collisions and shooting on the wild-living cat population are unknown. To better monitor the threat from shooting, we recommend that radiography is included as part of any future post-mortem surveillance scheme. We also recommend further investigations to explore the impacts of rodenticide exposure on cat health.

• Health monitoring will be an important component of continuing conservation management of wildcats, but both the live-sampling protocols and post-mortem schemes will require modifications to maximise their investigative value.

• The presence of a large feral domestic cat population in Scotland poses a challenge to restoration of the wildcat population, not just from the perspective of hybridisation but also with respect to wildcats’ exposure to disease-causing infectious agents. In order to combat this, a high proportion of domestic cats should be neutered and vaccinated in and around areas where any wildcats are reintroduced in the future. We believe tighter national regulation of domestic cat ownership must be introduced if the wildcat’s long-term future in Scotland is to be secured at a broad geographical scale.

• We recommend that a detailed disease risk analysis is conducted to inform plans for any future reintroduction of wildcats, and that continued attention is paid to other key issues such as community engagement and habitat management.

Acknowledgements

The SWA project was supported by the National Lottery Heritage Fund.

It was also funded by the Scottish Government, NatureScot and the Royal Zoological Society of Scotland. Support and resources for associated work was also provided by all the Steering Group members: Cairngorms National Park Authority, Forestry and Land Scotland, National Museums Scotland (NMS), Scottish Land & Estates, NatureScot, Scottish Wildlife Trust, The National Trust for Scotland, The Royal (Dick) School of Veterinary Studies, The Royal Zoological Society of Scotland, The Scottish Gamekeepers Association, The Scottish Government, Wildlife Conservation Research Unit (WildCRU).  In addition, Jenny Bryce and Fiona Strachan helped develop the project while Eileen Stuart, Alan Bantick and Andrew Kitchener chaired the Steering Group.

Additional sponsors, funders and contributors to SWA work included Aigas Field Centre, Ailsa Black, British Association for Shooting and Conservation, Cairngorm Brewery Company, Chester Zoo, Roy Dennis Wildlife Foundation, John Muir Trust, Loch Lomond & The Trossachs National Park, Lukas Ruiter Production, MyPetOnline, NFU Scotland, petDetect, RSPB, Scotland The Big Picture, Swift Ecology and The Highland Council.

We are also very grateful to the many academics, landowners and managers, farmers, foresters, keepers, captive collection managers, veterinary specialists and others who contributed valuable time and effort, members of the public who contributed funds and support, and in particular the many volunteers who got involved in the project.

We are grateful to all our SWA partners, in particular the SWA project officers Keri Langridge, Emma Rawling, Matt Wilson, Hebe Carus, Nicola Tallach and Calum Hislop, for their substantial contributions to this work, and Roo Campbell (SWA Priority Area Manager), Georg Hantke, Andrew Kitchener (NMS) and the WildGenes team (RZSS). We are also indebted to members of the public who reported or submitted carcases; colleagues in veterinary practices undertaking TNVR; Dr Kostas Papasouliotis and the staff at Langford Veterinary Services, the University of Bristol; the Clinical Pathology department at R(D)SVS; IDEXX Laboratories, Ludwigsburg, Germany; Phil Bacon; Lydia Peters and R(D)SVS undergraduate students, including Sima Lionikaite, Callie Neilson, Lydia Peters, Hannah Duncan and Paul Gogerty, for their assistance with post-mortem examinations and related analyses; Tony Sainsbury (Zoological Society of London); and our colleagues Glen Cousquer and Anna-Katharina Schilling. Lastly, we are very grateful to Alex Tomlinson for her detailed and constructive input on a draft version of this report.

1. Introduction

1.1 Background: disease in relation to wildcat conservation

Disease can be described as dysfunction in an individual, more specifically, as ‘any impairment that interferes with or modifies the performance of normal functions, including responses to environmental factors such as nutrition, toxicants and climate; infectious agents; inherent or congenital defects; or combinations of these factors’ (e.g. Wobeser, 2006). Disease threats are omnipresent, and exacerbated by the many current human pressures on ecosystems, such that even robust populations of wildlife are not disease-free (Stephen, 2014). Therefore, we should consider ‘significant’ diseases of wildcats (Felis silvestris) in Scotland to be those with the capacity to compromise the sustainability of the wild-living population and its resilience, for example, to environmental change (Hanisch et al., 2012, Rideout, 2015).

Disease surveillance informs the management of endangered species, such as the wildcat, having the potential to detect existing, new and/or emerging threats, be they infectious or non-infectious, and to improve understanding of their impact on the health of individuals and populations, when analysed alongside population abundance and distribution data. In addition, these insights guide the development of threat-mitigation measures.

Here we investigate the threats posed to wildcats by infectious agents and environmental toxins (anticoagulant rodenticides), and collect opportunistic information about other causes of mortality through post-mortem sampling of presumed wild-living cat carcases. Additional population-level threats to wildcats are addressed in other reports. 

Interactions amongst and between domestic cats (Felis catus) (Gentry et al., 2004) and wildcats provide opportunities for transmission of infectious disease. Prior to the start of this project, studies had already demonstrated that wild-living cats with wildcat pelage, genetic wildcats and genetic domestic cat-wildcat hybrids in Scotland were infected with or exposed to some infectious agents well-recognised in domestic cats (Daniels et al., 1999, McOrist et al., 1991). These included feline leukaemia virus (FeLV) (Table 1), which compromises the immune system and underlies a range of disease conditions in domestic cats, feline herpesvirus (FHV) and feline calicivirus (FCV), causes of ‘cat flu’, feline coronavirus (FeCoV), causal agent of ‘feline infectious peritonitis’ in domestic cats, and Toxoplasma gondii, an intestinal parasite of cats that is zoonotic (transmissible to humans). Feline immunodeficiency virus (FIV) was tested for, but not detected (Daniels et al., 1999, McOrist et al., 1991). The impact of these infections on the health of the sampled cats was unclear. Those authors considered FeLV infection to be of particular concern. Studies by Artois and Remond (1994) and Fromont et al. (2000) provide evidence of an association between FeLV infection and reduced body condition in wild-living wildcats, and the virus has been associated with notable mortality in another endangered, wild-living felid, the Iberian lynx (Lynx pardinus) (Meli et al., 2009).

Based on visual post-mortem findings, McOrist et al. (1991) suspected cat flu to have caused the death of a captive wildcat, and also observed symptoms typical of cat flu in several live captive wildcats. As many authors (e.g. Nowell and Jackson, 1996) have highlighted, infectious disease has particular potential to impact small populations and/or those subject to other stressors. Hence it is a concern for the wildcat population in Scotland, which is ‘alarmingly small’ (Breitenmoser et al., 2019), and impacted by, for example, habitat loss, population fragmentation and periods of food scarcity (Breitenmoser et al., 2019).

A number of other non-infectious conditions also have the potential to adversely impact wildcat health and survival. In parallel to habitat loss, persecution through shooting and trapping is thought to have driven historical declines of wildcat populations in Scotland and elsewhere in Britain. “The wildcat is now legally protected in Great Britain, but there is a risk, because of the phenotypical similarity, that they are still accidentally shot by gamekeepers controlling feral cats on sporting estates… The current impact of persecution and accidental killing on the wildcat population [is] unknown” (Breitenmoser et al., 2019). Similarly, the population-level impact of road-traffic collisions (RTCs) on wildcats in Scotland is poorly understood. The extent to which wildcats are exposed to environmental toxins has also been relatively understudied. McOrist and Kitchener (1994) detected moderate to significant traces of organochlorines in livers from three of 25 wildcat carcases (livers from 18 other cats had slight traces of organochlorines), but wildcats’ exposure to anticoagulant rodenticides (ARs) has not previously been explored and merits investigation.

1.2 Project objectives

The broad aim of this work package was to increase our knowledge base relating to disease threats to wildcats, most specifically infectious agents and ARs. The objectives were to investigate the prevalence of a broad range of infectious agents in wild-living cats in SWA Priority Areas (PAs), and, if possible, to assess the impact of the Trap, Neuter, Vaccinate, Release (TNVR) scheme on infectious agent prevalence, and relationships between infection status and potential risk factors such as hybridisation status (Table 2). Another objective was to establish a protocol for the collection, sampling and archiving of wildcat specimens, in collaboration with the National Museums Scotland (NMS) and other partners (Table 2).

2. Methods

From 2015 to 2020 a selection of samples was taken from live wild-living cats, either captured as part of the TNVR project or through the targeted trapping of wildcats and cats with high-scoring hybrid pelage (seven-point pelage score (7PS) ³ 17, as described by Kitchener et al. (2005).

In addition, carcases of presumed wild-living cats that had been submitted to SWA by wildlife rangers and members of the public were examined and sampled, where possible, in collaboration with NMS.

The sampling and testing approaches used in each scheme are described below.

2.1 Disease surveillance in live wild-living cats

2.1.1 TNVR for feral-domestic cats and cats with low-scoring hybrid pelage

The TNVR project, which had the initial aim of reducing the risk of hybridisation with and disease transmission to wildcats, is described in detail in the separate Trap Neuter Vaccinate Return Programme report. 

Briefly, wild-living cats identified as feral domestic cats or those with low-scoring hybrid pelage (7PS < 16 (Kitchener et al., 2005)) were trapped by project staff and volunteers. Under general anaesthesia, individual cats were examined, sampled, neutered, and vaccinated against FeLV, feline parvovirus (feline panleucopaenia virus, FPV), FCV and FHV (Table 1) by a veterinary surgeon, and then released. The samples collected were blood samples in EDTA; oral (buccal and oropharyngeal), conjunctival and rectal swabs; and, where possible, faecal samples.

Live cats were screened for infection with (and for FIV, exposure to) a range of felid pathogens, which were selected based on their potential to cause significant disease in the domestic cat and/or other felids. These included:

1. Feline retroviruses: FIV and FeLV
2. Feline respiratory pathogens: FCV, FHV, Bordetella bronchiseptica,Chlamydia felis and Mycoplasma felis
3. Bloodborne feline Mycoplasma species: Mycoplasma haemofelis, (cause of feline infectious anaemia), Candidatus (Ca.) Mycoplasma haemominutum and Candidatus (Ca.) Mycoplasma turicensis, and
4. Tritrichomonas foetus, a cause of feline diarrhoea.

The clinical pathological tests we used to screen for these infectious agents are listed in Table 3. A SNAP^® test (IDEXX SNAP^® FIV/FeLV Combo test, Henry Schein) was used in the field to screen blood samples for FeLV antigen (for which a positive result was indicative of active FeLV infection) and FIV antibodies (for which a positive result was indicative of prior or current FIV exposure). The remainder of the blood was submitted to a commercial laboratory (Langford Veterinary Services, University of Bristol) for PCR testing to detect antigens to (i.e. evidence of active infection with) these and the other infectious agents.

2.1.2 Trapping of wildcats and cats with high-scoring hybrid pelage

In a parallel scheme to TNVR, wildcats and cats with high-scoring hybrid pelage (7PS > 17 (Kitchener et al., 2005)) were targeted for trapping to enable collection of morphometric data, disease screening, genetic analysis and GPS-collaring under general anaesthesia. Sample collection for infectious disease surveillance followed the same protocol as for TNVR, as shown in Table 3.

2.1.3 Data analysis

Test results were collated and entered into the Wildcat Action Access database, from which prevalence data were extracted.

As part of a Master’s project (Alves, 2020, see Annex 3), data from a subset of 120 cats was used to investigate risk factors for infection, the spatial distribution of infection and the occurrence of co-infections. These individuals had comprehensive metadata available including Q or genetic score (Senn et al. 2019), sex, SWA PA, social structure, body condition score and age.

2.2 Post-mortem sample collection and ancillary investigations on wild-living cat carcases

2.2.1 Post-mortem scheme and ancillary tests

From 2015-2020, following publicity and encouragement from project staff, carcases of presumed wild-living cats killed on Scottish roads were submitted opportunistically by wildlife rangers, including NatureScot staff, and members of the public. Some carcases had also been received prior to this, through the same channels but on a more ad hoc basis, from 2009-2015. The submissions included domestic ‘tabby’ cats that looked like wildcats, and cats with the appearance of wildcat-domestic hybrids or possible wildcats. The carcases were typically frozen after collection and subsequently submitted to NMS in batches. Also, carcases of captive wildcats were submitted by zoos and wildlife parks, and other dead cats were acquired, for example, when handed in by gamekeepers.

At the National Museums Collection Centre (NMCC) each carcase was skinned by NMS preparators and inspected and sampled by R(D)SVS undergraduate veterinary students trained in a standardised wildcat post-mortem examination and sampling protocol. Since the majority of cats were presumed RTC victims, with obvious traumatic injuries, and owing to other logistical and financial constraints, these were not fully diagnostic post-mortem examinations. The examination and sampling protocols were drawn up at the start of the scheme in collaboration with NMS curators, and are appended (Annexes 1 and 2). An individual cat’s age was assessed based on the state of its teeth and categorised as ‘juvenile’, if estimated to be less than a year old, or ‘adult’. An individual’s sex and body condition score (1-5) was also assessed, and any obvious abnormalities were noted. Tissue samples were collected and stored as per Annexes 1 and 2, and a muscle sample was submitted to the RZSS for genetic analysis, performed as per Senn and Ogden (2015). Also see separate Genetics and Morphology report. Muscle tissue samples were retained in the NMS Biobank.

In the first year of this scheme (2015) carcases were radiographed prior to examination and blood and faecal samples were submitted to commercial laboratories for pathogen screening. Blood was screened for FeLV, FIV, M. haemofelis, Ca. M. haemominutum and Ca. M. turicensis (by PCR at Langford Veterinary Laboratories, University of Bristol); and faecal samples for FPV, FeCoV, T. foetus, Giardia sp., Toxoplasma gondii, Cryptosporidium sp., Salmonella sp., Campylobacter jejuni and C. coli, and other enteric bacteria and parasites, where possible (by PCR, bacterial culture or parasitological analysis, at IDEXX Laboratories, Ludwigsburg, Germany). In subsequent years, no radiographs were taken due to logistical and financial constraints, and tissue samples (bar the ones submitted for genetic analysis and biobanking) were banked by the R(D)SVS for future analysis.

As part of a Masters project (Peters, 2019, see Annex 4) archived liver samples were analysed for ARs at the Chemistry Branch Laboratory, Science & Advice for Scottish Agriculture (SASA).

2.2.2 Data analysis

To date, three R(D)SVS undergraduate student projects have been completed, in which the post-mortem dataset has been summarised and explored (Duncan, 2020, Lionikaite, 2017, Neilson, 2019). Lionikaite (2017) summarised findings from post-mortem investigations conducted from 2015 to early 2017 (the carcases were collected from 2009-2015), including the cases in which radiography and infectious disease screening had been performed.

Peters (2019) determined the level and number of rodenticides present in sampled liver tissue, from a total of six commercially used compounds, and analysed associations with age, sex and hybridisation status (Annex 4).

3. Results

3.1 Disease surveillance in live wild-living cats

To date, 207 cats have been sampled through the TNVR scheme and 17 cats with wildcat or high-scoring hybrid pelage (7PS >17) have been sampled through the wildcat trapping scheme. We have complete results from a total of 146 of these cats. Not all tests were run on each individual, as some samples were insufficient or unsuitable for analysis (for example degraded or contaminated).

Through the course of the project, it became evident that there was no clear morphological or genetic delineation between trapped feral domestic cats, domestic-wildcat hybrids and wildcats. There was a continuous spectrum of hybridisation within our sampled wild-living cat population (Breitenmoser et al., 2019), and no genetically pure wildcats were identified; also see the Genetics and Morphology report. Therefore, we are presenting results for the trapped wild-living cat population as a whole, without attempting to subdivide individuals into these categories. However, associations between infection status and hybridisation status were explored in the Masters thesis of Alves (2020), as discussed below.

All 11 infectious agents tested for were detected in this sampled population (Table 4). Ca. M. haemominutum was detected at the highest prevalence (24% of 144 cats), and FCV was also detected at a relatively high prevalence (21% of 132 cats). Bordetella bronchiseptica was detected (in 12% of 141 cats). Other agents were each detected in <10% of the population. FIV was detected in 7% of 125 cats (a similar proportion was FIV-antibody positive on the field-side SNAP test: see Alves (2020) for further discussion). FeLV was detected in 3% of 125 cats (a similar proportion was FeLV-antigen positive on field-side SNAP test: see Alves (2020) for further discussion). And Chlamydia felis was detected in the lowest proportion of cats (2% of 132). Sample sizes from each PA were not uniform. A small number of faecal samples was retrieved from the live-trapped cats, and these have been archived at R(D)SVS pending future analysis for other pathogens such as FPV and FeCoV (Table 3).

*For these three pathogens, an individual was considered ‘positive’ for infection if at least one of two swabs (taken from different sites – see Table 3) was positive.

Some pathogens were only detected in certain PAs. For example, FIV was detected in cats in northern Strathspey, Strathbogie and Strathavon, and FeLV was detected in northern Strathspey, Strathbogie, Strathpeffer and Morvern. In Morvern, a cluster of cats (n = 5) was found to be co-infected with M. haemofelis and Ca. M. haemominutum.

Table 5 gives a summary of other risk factors found to be associated with a positive infection status for each of the infectious agents analysed, as per Alves (2020). FIV was significantly more prevalent in cats with a high proportion of ‘domestic cat’ DNA, and both FIV and FeLV were significantly more common in colony cats. There was a strong positive correlation between genetic ‘domestic cats’ and colony living. Mycoplasma felis was the only agent found to have a statistically significant association with poor body condition. There was a trend for FIV-infected cats to have poorer body condition scores, but this was not statistically significant. Male cats had a significantly higher prevalence of some infections, including FIV. Several positive correlations between infectious agents were also identified, particularly between FIV and the bloodborne haemoplasmas (M. haemofelis, Ca. M. haemominutum and Ca. M. turicensis). For more comprehensive analyses and discussion of these results, see Alves (2020); an article to be published in a peer-reviewed journal is also currently under preparation.

*These analyses were conducted on a subset of cats for which there was complete metadata (n = 120). For more detailed sub-sample sizes, see Alves (2020)

**See Senn et al. (2019)

3.2 Post-mortem sample collection and ancillary investigations of wild-living cat carcases

3.2.1 Sample population

To date 81 presumed wild-living cat carcases have been examined post-mortem (as well as a number of captive wildcats and a suspected owned domestic cat). The carcases were received from across Scotland, including all PAs. The majority of animals were adult (75%) and male (61% of 61 adults). Of 51 carcases, for which the results of genetic testing were available, using the scoring criteria of Senn et al. (2019) (domestic cat if UBQ ≤ 0.25 and wildcat if LBQ ³ 0.75, see Table 6 legend), 16% (8) were genetically ‘domestic cat’, 82% (42) were ‘hybrid’, and 2% (1) were ‘wildcat’.

3.2.2 Notable findings

The results presented here draw on the student theses of Lionikaite (2017), Neilson (2019), Peters (2019) and Duncan (2020).

Post-mortem blood and faecal samples from 18 individuals were screened for infectious agents. Blood samples from two (11% of these) carcases were positive for FIV infection, and faecal samples from both cases were also positive for T. foetus infection, the only carcases in which either agent was detected. One of these cats was a low-scoring genetic hybrid, which was also positive for Ca. M. haemominutum, and which appeared to have died as a result of ingesting a fish hook, but also had gun-shot present in its tissues (SWA no. 0981, see Table 6). Blood from another case was positive for Ca. M. haemominutum infection (only). Giardia sp. and Cryptosporidium sp. were detected in the faeces of one case each, but were not concurrent with other infections. Cryptosporidium sp. was detected in the only carcase confirmed (to date) as a genetic wildcat, which was an emaciated juvenile male. None of the post-mortem blood samples was positive for FeLV, M. haemofelis or Ca. M. turicensis. Neither were FPV, FeCoV, Toxoplasma gondii, Salmonella sp. or Campylobacter spp. detected in any of the post-mortem faecal samples tested.

On gross visual inspection, gastrointestinal helminths (worms), specifically roundworms and/or tapeworms, were detected in 81% (63) of 78 carcases in which this could be investigated. 15% (12) of these 78 cases had a very heavy helminth burden (> 40 adult tapeworms and/or > 40 adult roundworms visible). Ectoparasites (ticks and/or fleas) were detected in 12% (8) of 69 carcases in which this could be explored.

There were abnormalities typical of collisions with road traffic, such as fractures, bruising, scuffed claws and/or a ruptured diaphragm, in at least 58% (47) of 81 carcases, but not in all cats retrieved from roads.

Lead gun-shot was detected in the carcases of six cats, including 4 (22%) of the 18 carcases in which radiographs were taken. Of these six cats, all but one had domestic cat pelage (Table 6), although four were found to be genetic domestic cat-wildcat hybrids. In four of the cases in which gun-shot was present, which included one carcase found on a roadside, post-mortem observations were consistent with shooting being the cause of death. In another individual found on a roadside, which had domestic cat pelage, but was subsequently found to be a high-scoring genetic hybrid (Table 6), it was not possible to determine the role of shot in the individual’s death. In another case (see above and Table 6) the shot was not considered to have caused fatal injury. Two cats were reported or suspected to have died following a dog attack (Table 6).

Anticoagulant rodenticides (ARs) were detected in the liver of 61% (30) of 49 sampled carcases. In 27% (13) of these 49 cases, the residues were above a recognised toxic threshold (0.2 mg/kg liver) for birds and mammals. The highest concentration recorded (1.62 mg/kg liver) was almost nine times this toxic level. All six ARs screened for were detected in the liver samples. Bromadiolone and difenacoum were the most frequently detected, with a prevalence of 93% and 60% respectively. These compounds appear to be the most commonly used on Scottish arable farms. Samples collected from the Morvern area, which has a small amount of low-intensity farming, had the lowest proportion of AR-positive (40% of five samples), as well as the lowest average AR level. Liver AR residues were not found to be statistically associated with hybridisation status, but were more prevalent and detected at higher concentrations in adult and male cats. See Peters (2019) (Annex 4) for further details.

 “Q = the hybrid score estimate [based on the 35 SNP loci] ranging from 0 (domestic cat) to 1 (wildcat); LBQ = lower boundary of the 90% CI [confidence interval] of the hybrid score Q; and UBQ = upper boundary of the 90% CI hybrid score” (Senn et al., 2019). As per Senn et al. (2019), individuals with LBQ ³ 0.75 are listed as ‘wildcats’, those with UBQ ≤ 0.25 as ‘domestic cats’, and others as ‘hybrids’. Also see the Genetics and Morphology report.

**Using the 7-point pelage score of Kitchener et al. (2005), individuals with scores of < 12 were categorised as domestic cat and those with scores of > 19 as wildcat; other cats were considered domestic cat-wildcat hybrids. Where no score is given, it was not possible to score all characters comprising the 7PS, and identity was estimated from the characters that could be scored and/or the cat’s general appearance (Andrew Kitchener, pers. comm.).

4. Discussion

4.1 Threats to wildcat population health

Our findings demonstrate that wild-living cats in Scotland are infected with a broad range of infectious agents and are also exposed to rodenticides, which has not previously been documented. We have highlighted road-traffic collision and shooting as causes of mortality in domestic cat-wildcat hybrids.

4.1.1 Infectious disease

We have expanded the range of infectious agents detected in wild-living cats across the feral domestic-hybrid-wildcat spectrum in Scotland. This was the first detection of FIV infection in wild-living hybrid cats, to our knowledge, and to date only two studies have detected evidence of FIV exposure (antibodies rather than active infection) in wild-living wildcats in France (Fromont et al., 2000) and the Middle East (Ostrowski et al., 2003). FIV was significantly more prevalent in cats at the ‘domestic cat’ end of the hybrid spectrum (Alves, 2020, see Annex 3 and discussion below). Although not statistically significant, there was a trend for FIV-infected cats to have lower body condition scores (Alves, 2020), and the presence of FIV in wild-living hybrid cats is a cause for concern, because in domestic cats, as with FeLV, the infection typically persists in individuals long-term, compromising the immune system, predisposing them to other disease conditions and ultimately hastening mortality. This recent detection of FIV – previous studies by McOrist et al. (1991) and Daniels et al. (1999) failed to detect FIV in wildcats and hybrid cats in Scotland – and the observed association with domestic cat genotypes, rather than hybrid genotypes, suggests that infection has originated in domestic cats. It is likely that increased interactions between domestic cats, hybrid cats and wildcats in recent decades, as evidenced by increasing levels of introgressive hybridisation (Senn et al., 2019), have, in parallel, led to greater exposure of wildcats and hybrid cats to FIV and other domestic-cat-associated pathogens.

Our detection of FeLV and FCV in the wild-living study population was consistent with the findings of McOrist et al. (1991) and Daniels et al. (1999), as was the detection of FHV (Daniels et al. (1999) found FHV antibodies, i.e. evidence of exposure). We also detected seven other infectious agents that were not investigated in these previous studies. We found B. bronchiseptica, C. felis and M. felis, which, along with FCV and FHV, are pathogens of the respiratory system; M. haemofelis, which causes feline infectious anaemia, and other bloodborne agents Ca. M. haemominutum and Ca. M. turicensis; and T. foetus, Giardia sp. and Cryptosporidium sp., which are protozoan parasites of the gastrointestinal tract that can cause feline diarrhoea. Chlamydia felis, T. foetus and Giardia sp. can also cause zoonotic disease.

That colony living was associated with the domestic cat genotype was to be expected, given that domestic cats can exhibit this behaviour (e.g. Yamaguchi et al., 1996), whereas wildcats are typically solitary (Breitenmoser et al., 2019). FeLV and FCV were, like FIV, found to be more prevalent in colonies, where higher densities of cats create more opportunities for interaction and associated infection transmission (Yamaguchi et al., 1996). Similarly, the higher prevalence of some infectious agents in male cats – which may be more frequently exposed to infection, given they roam and fight more than females – mirrored findings in domestic cats (e.g. Courchamp et al., 1998). The differing patterns of infectious agent prevalence across PAs are noteworthy and provide valuable baseline data for any future wildcat reintroduction into one of those areas.

The positive correlation between FIV and haemoplasma infections reflected the findings from several studies in domestic cats, where FeLV and FIV-positive cats have been found to have a higher risk of haemoplasma infection (summarised in Beugnet and Halos, 2015). Co-infection between these agents may exacerbate the severity of disease in domestic cats (Beugnet and Halos, 2015), giving further potential significance to the simultaneous presence of these infections in a wildcat.

The ability of these agents to cause disease in wildcats and wildcat-domestic cat hybrids has yet to be elucidated. M. felis infection was significantly associated with lower body condition in the sampled population, suggesting this infection negatively impacted cats’ health. However, our relatively small sample size limited inferences regarding this and other risk factors. Considering their known impacts on the health, welfare and population dynamics of domestic cats, many of these infectious agents are likely to have an ability to cause disease in wildcats or their hybrids, and their presence in the extremely fragile and hybridised wildcat population of Scotland may well have the potential to cause local wildcat extinction, particularly in subpopulations subject to other stressors. To further understand the potential effects of infectious disease on wildcat health and mortality, we recommend future disease research and surveillance studies adopt a more comprehensive monitoring and investigative approach (see Section 4.3). For more in-depth discussion of risk factors and co-infections, as well as recommendations for future research and surveillance, see Alves (2020).

4.1.2 Non-infectious conditions

The high proportion (58%) of cat carcases that had injuries consistent with RTCs was to be expected given the emphasis on carcase collection from roadsides. The significance of RTCs as a population-level threat to wild-living cats, and whether they are more common in particular localities, merits exploration. In future, mitigation measures, such as road signs, could be employed as and where appropriate (Duncan, 2020).

The six cases in which lead gun-shot was detected were all found prior to 2016. However, we cannot be certain that gun-shot was absent from carcases submitted in later years, because they were not X-rayed. To better monitor the threat from shooting, we recommend that radiography is included as part of any future post-mortem surveillance scheme (see below). As our results demonstrate, lead gun-shot may be present even in cases where there appears to be another obvious cause of death. Two of the cats, in which lead shot was found, were collected from a roadside and initially presumed to have died from RTCs (one had concurrent RTC-associated injuries), and in another case the shot was an incidental finding (shooting was non-lethal). Details of the work the SWA project conducted to reduce the impacts of shooting on wild-living cats are provided in the Land Management report. As four of the six shot cats in this study demonstrate, unfortunately pelage is not a consistent means of identifying cats with a proportion of wildcat DNA, so that when cats of ‘domestic cat’ appearance are shot, there remains the potential for wildcat genes to be lost from the wild-living cat population. Dog attacks are evidently another potential threat to the wildcat gene pool.

The high proportion of wild-living cats found to be exposed to ARs and the high number with liver levels above a threshold for toxicity in other mammal species (> 25% of sampled carcases) are concerns. As we discuss below, more detailed post-mortem examinations are required to determine the impact of ARs on the health of wild-living cats. For further detailed discussion of these AR results, see Peters (2019 - Annex 4).

4.2 Project objectives: progress and limitations

4.2.1 Objectives 1-3: To assess disease prevalence in wild-living cats in SWA Priority Areas, the impact of TNVR scheme on infectious disease prevalence, and the relationship between hybridisation and disease status

This work has enabled us to determine the prevalence of key feline infectious agents in the Scottish wild-living cat population, which was sampled as part of the TNVR and wildcat trapping schemes, and to start exploring relationships between genetic hybridisation status and infection status. We have also investigated the levels of AR exposure within cats submitted through the opportunistic carcase collection scheme.

However, within the scope of this project, and as discussed above, we have not been able to assess the effect of these agents on the health of individuals (their disease-causing ability), or their impact at a population level, which was also a desired outcome (Table 2; Wildcat Action, 2014). Also, as expected, the timeframe for this project was too short to enable an assessment of whether decreasing numbers of feral cats, or increased vaccination of domestic cats (i.e. the TNVR programme), could lead to reduced infection/disease prevalence in the wild-living cat population. The current sample size would also be insufficient for such an analysis, and such a study would require more-standardised serial monitoring, including estimates of the proportion of the feral cat population neutered or vaccinated (or of neutering or vaccination effort per region), and a more-standardised approach to trapping/sampling for disease screening (see below). The analysis of the relationship between the genetic hybridisation status and infection status was also limited by sample size.

4.2.2 Objective 4. To agree a protocol for the collection and archiving of wildcat specimens/samples for post-mortem and disease screening

Following Breitenmoser et al. (2019), sample collection and storage protocols were agreed with NMS at the start of the project and followed (see Annexes 1 and 2). This has enabled us to collect samples from over 80 wild-living cat carcases from across Scotland, a subset of which have been used to screen for feline infectious agents and ARs.

A number of limitations have prevented us from performing fully diagnostic post-mortem examinations. These include the state of carcases submitted, since many RTC cats suffered extensive trauma. Many carcases were also decomposed to some degree and had been frozen and may have been defrosted more than once prior to inspection. The off-site location of these examinations, in addition to financial constraints, has also presented a challenge, precluding routine radiography of carcases or examination by veterinarians with specialist wildlife pathology expertise. Also, in a proportion of cases, discrepancies in carcase archiving and sample tracking have limited their utility for data recording and analysis.

4.3 Recommendations for future work

Surveillance for both infectious and non-infectious disease will be an important component of a long-term conservation management plan for wildcats, especially because the population is so ‘alarmingly small’ as to now be considered non-viable (Breitenmoser et al., 2019) and given the plans for future reintroduction(s).

With some modifications (see below) there will be benefit in continuing both types of disease surveillance scheme, i.e. targeted surveillance in live cats and opportunistic post-mortem examinations, since they facilitate complementary, but different, outputs. Performed more thoroughly, opportunistic post-mortem examinations would give us the ability to detect new and/or emerging (infectious and non-infectious) threats, as well as monitoring existing ones, and, importantly, to better understand the impact of these threats on the health of individual animals. Post-mortem findings also inform other aspects of wildcat ecology and conservation, for example providing information on diet (as per Lionikaite, 2017). The samples we are able to collect and archive from carcases provide a valuable resource for further research into health threats to wildcats (as demonstrated by Peters’ 2019 study) and other conservation questions. Targeted surveillance for infectious agents in live cats enables us to estimate prevalence and, with the caveats below, to explore the impacts of infections at a population level. In the longer term, it might also enable us to monitor the impact of population management actions, such as TNVR or other future measures, on infection status over time. It will also help to inform future disease risk analyses for reintroduction efforts.

A number of modifications will be needed in future, in order to fulfil the potential of these schemes. Specifically, for the post-mortem scheme:

1. Better preservation of carcases to be submitted for examination, if feasible
2. A more robust system for carcase identification and traceability
3. Submission of carcases to a veterinary institute for full diagnostic post-mortem examination in the first instance, to allow examination by veterinarians with specialist wildlife pathology expertise, including routine radiography of carcases to detect lead shot (or other foreign bodies) and thorough additional diagnostic investigations (e.g. histopathology, cytology, bacterial culture) to accurately diagnose disease conditions.

And for the surveillance scheme in live wild-living cats, to enable population-level analyses:
4. A more systematic approach to trapping and population sampling, i.e. one that is standardised over time and between study areas.
5. Robust longitudinal population data across study areas
6. A longer monitoring timeframe than the current study permitted
7. Collection of clinical data, including a detailed physical examination, to assess health status alongside infection status
8. Estimates of management effort per study area, to explore the impact of management actions on disease/infection prevalence.

Our work has generated a number of research questions, some of which can only be answered with the modifications above to the surveillance schemes. We note that within the SWA project there was relatively little scope to conduct novel research or to implement an experimental sampling design. To some extent this constrained data collection and limited the scope and statistical power of subsequent analyses. Further knowledge of the disease status of wild-living Scottish cats and the associated risks to wildcats could be generated if research objectives received greater emphasis in future conservation programmes for the species. Conservation interventions are a valuable opportunity for research and the utility of this should not be underestimated.

While some research questions cannot be answered at this stage, collaborations are currently being developed with other wildcat projects and researchers in continental Europe that will help our understanding, not only of wildcat ecology and conservation management, but also of wildcat population health and the potential population-level impact of threats such as infectious agents and AR exposure.

The presence of a large feral domestic cat population in Scotland poses a challenge to restoration of the wildcat population, not just from the perspective of hybridisation, but also with respect to wildcats’ exposure to disease-causing infectious agents. The exposure of wildcats and domestic cat-wildcat hybrids to FIV and other agents circulating in domestic cats is only likely to reduce when there are fewer encounters with domestic cats and/or when there is a lower prevalence of infection in the wild-living cat population. In areas inhabited (or inhabitable) by wildcats, this requires a high proportion of domestic and hybrid cats to be neutered and vaccinated. We believe the only way for this to be achieved at a broad geographical scale is through tighter national regulation of domestic cat ownership, and that this will be an essential step in securing a long-term future for the wildcat across Scotland (Meredith et al., 2018).

Bearing in mind these caveats, we recommend that a disease risk analysis is conducted to inform plans for any future reintroduction of wildcats and to generate a rigorous health management protocol for such a scheme (as per International Union for Conservation of Nature (IUCN)/Species Survival Commision (SSC) guidleines (2013)). Alongside the key issues of habitat suitability, community engagement and feral domestic cat management, the presence of infectious agents and ARs in wild-living cats, as well as risks from shooting and road traffic collisions, need to be borne in mind and the relevant threats mitigated effectively in order to facilitate population recovery.

5. Conclusions

Our results show that wild-living cats in Scotland are infected with a broad range of infectious agents and exposed to rodenticides (rodent poisons). Both infectious agents and rodenticides pose a potential threat to the health of wildcats, including any that may be reintroduced in the future. We have confirmed road traffic collisions and shooting as causes of mortality in domestic cat-wildcat hybrids.

This is the first time FIV has been detected in wild-living Scottish domestic cat-wildcat hybrids. The presence of FIV and other feline infectious agents in the wild-living cat population is a concern. Also of concern is the high proportion of these cats exposed to ARs at potentially toxic levels.

From the perspectives of both infectious disease transmission and hybridisation, we argue that tighter national regulation of domestic cat ownership will be an essential step if the long-term future of the wildcat in Scotland is to be secured at a broad geographical scale. Health monitoring should be a component of any long-term conservation management plan for wildcats. However, a number of modifications will be needed to both our live-sampling and post-mortem schemes in future, so that we can assess the actual impact of the potential and confirmed threats we have identified on the health of wildcat individuals and populations, and for us to reliably identify new and/or emerging threats in future.

6. References