Introduction

Tuberculosis (TB) is an infectious disease caused by members of the Mycobacterium tuberculosis (M. tuberculosis) complex (MTBC). In animals, the disease is mainly caused by M. bovis and M. caprae. The microorganism has a wide host range (Nugent 2011) and thrives in complex systems at the wildlife-livestock interface (Thomas and Chambers 2021; Thomas et al. 2021; Buddle et al. 2015). TB in cattle (bovine TB [bTB]) is a major animal welfare and economic problem in many countries (Buddle et al. 2015; Vordermeier et al. 2016; Byrne et al. 2020; Skuce et al. 2012.). The eradication of the disease in livestock is a major goal in most countries, but this is severely hampered due to the transfer from wild animals present in the location of the livestock holdings. In the United Kingdom (UK), badgers (Meles meles) are thought to be the main wildlife reservoir of bTB (Bhuachalla et al. 2015). However, the presence of M. bovis in the red deer (Cervus elaphus) herd of the UK (Delahay et al. 2007; Gowtage-Sequeira et al. 2009) and elsewhere in Europe (Fink et al. 2015; Leth et al. 2019; Santos et al. 2015; Zanella et al. 2008a, b; Reveillaud et al. 2018) has raised the question of whether red deer may be a relevant wild host of bTB. Whether they can act as hosts or not, the presence of bTB in red deer has major implications for the overall welfare of the herd.

Some farmers in this region have recorded bTB reactors during the regular bTB testing of their cattle, and are questioning whether deer might be potential vectors. In order to examine this, up to date data on the incidence of bTB in this region and its relationship to the density of the deer population in specific areas and to the incidence of bTB reactors on local farms are required. This preliminary study will attempt to do this.

Few studies have examined red deer on Exmoor. One study which examined the incidence of TB in various wildlife species in the West of England observed significant infection in fallow deer 4.37% with a lower incidence in roe and red deer (1.02%). The results suggested that deer should be considered potential sources of infection in cattle (Delahay et al. 2007). A report on the health of the Exmoor red deer herd conducted on behalf of the Exmoor National Park Authority (Werrett and Green 2008) showed that bTB existed in the herd, but apart from one hotspot, the incidence was considered not to be above the average for wild deer in the UK.

Bacterial culture is considered to be the ‘gold standard’ method for the diagnosis of TB in all wild animals (Domingos et al. 2019). However, culture is expensive, time-consuming and requires category 3 level laboratory facilities (Thomas and Chambers 2021). In contrast, serological tests are especially useful in wildlife because they are economically attractive, technically easy, enable large scale surveillance and can be applied to both live and dead animals (Thomas et al. 2021). There appears to be no ‘gold standard’ serological assay for bovine TB (Thomas et al. 2021). Among the most useful serological tests for the diagnosis of TB in wild animals are the species-specific lateral flow tests (Chambers 2013), based on the technique of immunochromatography. These are particularly useful in wildlife studies because of the ease with which the tests can be performed and the rapidity of obtaining results. These tests are based on the detection of antibodies to Mycobacterium bovis infection. The most frequently recognised antigens in deer include MPB83, ESAT-6, CFP10 and MPB70 (Harrington et al. 2008).

Materials and methods

Study area

Exmoor is an upland area in the South West of England comprising moorland, farmland and woodland interspersed with rivers and combes. It was designated a National Park in 1954. The Exmoor National Park occupies approximately 692 km2 (267 square miles). Exmoor contains one of the last remaining herds of wild red deer in England.

The red deer herd and the strategies employed to manage it

The National Park Authority working with the Exmoor and District Deer Management Society monitors the health and well-being of the deer population. The two deer management strategies employed on Exmoor are selective culling by hunting with hounds and by stalking. The overall population and general well-being of the wild herd are managed to a large extent by these two procedures. Both strategies despatch the deer by shooting. Although hunting with hounds was banned by the Labour Government in 2004, deer hunting is still permitted under the exempt clause ‘research and observation’. A number of projects were ongoing at the time of this study and we were provided with blood samples by the hunt staff. No deer were specifically culled for this project.

The deer hunting season in the UK revolves around the life cycle of the deer. Autumn stag hunting aims to selectively take out older stags who are past their prime. Autumn stag hunting stops toward the end of October, and before the rut (mating season). Hind hunting begins after the rut and continues until February. The aim of hind hunting is to remove as many of the weaker hinds as possible to leave a strong pool of healthy hinds to deliver their calves in June. Spring stag hunting starts in early March. This selects younger stags and those with misshapen heads (antlers). Again the aim being to preserve a good stock of stags for the future of the herd. All hunting stops at the end of April to allow the pregnant hinds to develop their calves, give birth and bring up their calves relatively undisturbed.

Hunts also offer a service dealing with casualty deer which are more easily located using hounds. The casualty deer are often results of road traffic accidents, entanglement in fences and wounding by careless poachers. The distress caused by such events can be severe (Gentsch 2018) and it is imperative that the injured deer are despatched rapidly to limit their suffering.

The main aim of culling by stalking is to top and tail by culling about 40% of the young deer, i.e. calves and yearlings, and predominantly more females than males, plus a similar number of the older deer. In addition, about 20% of the deer culled by stalking are middle-aged deer, mainly injured or sick deer and deer causing damage to crops or trees. The main aim of culling by stalking reflects that of hunting with hounds and it is to have a good middle-aged herd that are fit and can produce a healthy crop of calves each year.

In this study, the selection of deer to cull was made by the hunts and stalker and not by us. We were supplied with blood on which to conduct this study (see later).

Methods

General

This study was accomplished in two batches, interposed by UK Government imposed COVID-19 restrictions. The first tranche occurred between February 2020 and December 2020. The study resumed in August 2021 and was completed in April 2022.

Blood samples

Blood samples were obtained from deer despatched by the Devon and Somerset Staghounds (DSSH), the Quantock Staghounds and from registered stalkers operating in the area.

Free flowing blood was obtained from neck blood vessels immediately post-mortem. The blood was allowed to clot at room temperature, transferred to the lab and centrifuged at 1000 × g to yield serum. This was stored at − 20 °C for later use.

The total number of red deer samples collected was 106. Male subjects (full grown stags and young male calves) contributed 62 samples (58%), and female subjects (full grown hinds and young female calves) contributed 44 samples (42%).

In addition to blood analysis, a record was made of the sex, approximate age and the site at which the deer were lifted or culled.

Serology

One of the most commonly used lateral flow kits for determining the presence of TB in deer is the cervid STAT-PAK kit (Chembio Diagnostic Systems Inc. Medford NY). The test employs a unique cocktail of MPB83, ESAT-6 and CFP10 antigens in a single-step signal detection system (Harrington et al. 2008). An improved form of the STAT-PAK assay, the cervid DPP Vet TB kit (Chembio Diagnostic Systems Inc.), is available which has a higher specificity than the original test (Buddle et al. 2010). Both assay kits have been evaluated for testing red deer in New Zealand (Buddle et al. 2010) and Great Britain (Gowtage-Sequeira et al. 2009), and in elk and white-tailed deer in the USA (Nelson et al. 2012). This improved test kit was used in this study. The assay was conducted according to the manufacturer’s instructions. The sensitivity of the method has been reported to be as much as 91%, at a specificity approaching 100% (Chambers 2013; Buddle et al. 2010).

Deer densities

The Exmoor and District Deer Management Society conduct annual deer counts as part of the management of the wild herd. In order to conduct the count, the greater Exmoor area is divided into 28 regions of varying acreage. The count is achieved simultaneously across all 28 regions by groups of experienced observers. The count is conducted visually by mobile and stationary observers. The recording includes details of sex and accurate timing of the observation in order to ensure no double counting of the same animal. The deer count does not use distance sampling. Deer density is computed by relating the recorded deer numbers to the acreage of each region. The results of the 2022 deer count were used in this analysis.

Holdings with reactors

The data used in this study was obtained from the UK TB Advisory Service. This is a DEFRA [Department for Environment, Food and Rural Affairs] funded project set up to provide up to date advice to British farmers concerning all aspects of TB in cattle. They publish regularly updated maps of the incidences of TB infections in individual farms (www.lbtb.co.uk). These maps were used to locate holdings with infected cattle on Exmoor.

Statistical analysis

The data regarding deer density and the incidence of bTB in the deer and the number of farms recording bTB reactive cattle showed a non-parametric distribution. Accordingly, the analysis of the relationship between these factors was analysed by the Spearman test.

Results

Serology

The results of the serology are shown in Table 1. Of the total number of 106 samples, there were 30 positive reactors, which equates to an overall infection rate of 28.30% (CI95 19.97–37.1). Table 1 also shows the data derived from samples obtained from the two different modes of deer management. The number of deer showing the presence of bTB was higher in those culled by stalking compared to those culled by hunting. The stalker also reported the presence of visual signs of gross pathology in some of the deer showing positive serology.

Table 1 Raw data from all areas sampled by hunting or stalking

There did not appear to be any major differences in the incidence of bTB between male and female deer. This holds true for both hunted deer (males 21.87%; females 21.05%), and stalked deer (males 34.58%; females 34.62%).

The incidence of bTB infection as a function of age is shown in Table 2. There was no clear pattern of distribution except that there was a higher incidence of bTB in the older [8–9 years] deer (40%) and in the yearlings and under (42.85%). It was not possible to do a full analysis of the distribution of bTB as a function of sex in all groups, but there were similar numbers of male and female bTB positive deer in the yearling and under group, but there were more male (8) than female (1) in the 2–3-year group.

Table 2 The incidence of TB infection as a function of age

As stated above, greater Exmoor is divided into 28 areas for the purpose of the annual deer count. This study sampled deer from 17 areas. Table 3 shows the results for the different areas, in terms of deer density, incidence of bTB and farms with bTB positive cattle. There did not appear to be any particular pattern of distribution of bTB across the 17 regions with the exception of regions 15, 16, 17, 26 and 28. Region 15 (Exmoor Forest) is the largest region by area and naturally provided a high number of samples, of which 19.2% (CI95 11.5–26.9) were found to be bTB positive. Regions 16, 17, 26 and 28 occupy an area in which the observed sick deer and suspected bTB infected deer were targeted. This accounts for the large number of deer culled in a relatively small area. Not surprisingly, the combined figures for the four areas provided figures of 45 samples of which 19 were bTB positive (42.22%; CI95 32.54–51.90). Reports of sick deer were far less frequent in the other areas. Only two casualty deer were reported as sick (as opposed to injured) and neither was positive for bTB.

Table 3 Deer density data for all areas sampled and relationship to farms with bTB reactors

Deer density and farms with reactors

A positive trend was observed in the correlation between deer density and bTB in deer but this did not reach significance (r = 0.413, n = 17). A weak positive trend was seen in the relationship between deer density and the number of farms with bTB reactive cattle (r = 0.181, n = 17). The relationship between the number of farms with bTB reactive cattle and the number of deer with bTB showed a significant positive correlation (r = 0.643, P < 0.02, n = 17).

Discussion

The incidence of bTB in the wild deer herd of Exmoor

Early studies on the incidence of bTB in wild deer populations in the UK indicated a prevalence of below 6% (Clifton-Hadley and Wilesmith 1991). Other more recent studies in Europe, New Zealand and USA have provided figures of 6.94% and 13.22% (Ferreras-Colino et al. 2022), 14% (Vicente et al. 2006), 24% (Zanella et al. 2008b) and as much as 42.5% (Barroso et al. 2020). A meta-analysis and systematic review of all data worldwide provided a pooled prevalence worldwide of 13.71% (Reis et al. 2020) with varying incidences in different countries and regions within countries (see Reis et al. 2020 for details). Overall, the data from this current study of Exmoor deer are within the upper ranges reported in other studies.

In New Zealand, the prevalence of infection was similar in male and female deer (see Nugent et al. 2015) with a higher incidence in younger males (< 2 years) compared to older stags. However, in Spain, the prevalence of infection was found to be lower in female red deer compared to males of the same age (Vincente et al. 2006). Similar to the studies in New Zealand, this study showed a similar incidence of bTB in male and female animals.

The age distribution of the infected deer is interesting. In this study, a high percentage of the yearling deer sampled were infected. There was no difference between male and female yearlings. Vicente et al. (2013) also reported significant infections in young wild boar piglets and red deer calves. In contrast to these and our findings, Lugton et al. (1998) found only 1 deer under 1 year of age out of 18 sampled that had bTB whereas deer over 2 years of age had significantly greater incidences of the disease. Incidences of bTB in young domestic calves have also been observed in some studies (Houlihan et al. 2008; Alameri 2020). The high incidence of bTB in young calves seen in this present study probably reflects the environment in which they live. This is discussed later.

The route of transmission of the disease to these young animals is unclear. Many routes of transmission of bTB have been identified (Skuce et al. 2012; Phillips et al. 2003; Palmer et al. 2004; Allen et al. 2021). However it is generally accepted that M. tuberculosis is primarily transmitted by aerosolised droplets and inhaled into the lungs where it is able to establish infection (Allen et al. 2021; Neill et al. 2005; Lin and Flynn 2015). In contrast, the dose of microbacilli needed to infect via the oral route is roughly 1000 times that needed to infect via the respiratory route (Allen et al. 2021). A recent study reporting zoonotic bTB transmission in humans concluded that the infection was mediated by aerosolised microbacilli encountered when handling cattle in a cattle crush, and not by ingesting infected raw milk (Smith et al. 2004), although a later study implied the transmission of TB to humans and cattle occurred by persistently ingesting infected unpasteurised milk (Doran et al. 2009). The most recent study which examined in detail the development of lesions in a 3-week-old Holstein–Friesian calf concluded that the pattern of lesions, which were exclusively confined to the lungs, suggested the inhaled route of infection (Mekonnen et al. 2021). With this in mind, the most likely explanation of the incidence of bTB in red deer calves seen in this study is that they are being infected as a result of inquisitive nosing around animals or animal carcases infected with bTB. This process would pick up aerosolised microbacilli and lead to the development of the disease.

The apparently high susceptibility of deer calves is likely to be related to the immaturity of the immune system (Basha et al. 2014; Tsafaras et al. 2020). The deficiencies in the neonatal immune system are complex, and beyond the scope of this paper. However, it is known that these deficits render the young animal particularly susceptible to respiratory infections (Basha et al. 2014). The humoral immune system is clearly functional in the young animals investigated in this study because of the presence of antibodies to the CFP10/ESAT6 and MPB83 antigens detected in the serum samples obtained. However, we have no way of knowing to what extent the cellular system is working. In order to mount an effective response to the infection by Mycobacterium bovis, a robust and complex co-ordinated cellular immune response is needed (Pollock et al. 1996, Kaufmann 2002). Consequently, we do not know whether these yearling deer calves will recover as their immune system matures (Basha et al. 2014, Tsafaras et al. 2020), or whether they will succumb to the disease prior to the full development of their immune systems. Studies on wild boar have reported a high incidence of death from bTB in piglets, but that the rate of death declines with age (Barasona et al. 2016).

In our study, the region with the highest incidence of bTB included areas 16, 17, 26 and part of 28 and forms the Holnicote Estate in the North East of the National Park. The relatively high incidence of bTB in this area was partly due to the culling procedure which targeted sick deer and deer suspected of bTB as well as deer selected according to the method outlined in the introduction. The number of sick deer culled in this region was well above the 20% of the total which is normally made up of sick and injured deer, and of a different age range. Interestingly, the smaller number of deer culled away from this region showed no incidence of bTB.

In addition to the high incidence in young deer, deer aged 8–9 years also showed a high level of bTB. It is generally found that the progression of bTB in wild deer increases with age (Vicente et al. 2013). This being so, it would be expected that the incidence in the oldest group (10 + years) would also be high. This was not the case. The number of animals included in these groups is quite low, and until further studies are conducted on a larger group of animals, it is difficult to explain this finding.

The region with the highest number of farms with bTB positive cattle was Exmoor Forest. This was not surprising as this region has the largest area by some distance and had more farms. The Holnicote Estate had the highest incidence of farms with bTB reactive cattle relative to the number of farms in the region with cattle. The major difference in the maintenance of the red deer herd in this estate, compared to elsewhere on Exmoor, is that there is no hunting with hounds on the estate, only stalking. Hunting with hounds was banned by the landowner in 1997. The lack of disturbance by hunting has encouraged the deer to congregate in areas where there is good feeding, such as agricultural pasture and other crops, and not to venture far from this area. Such congregations will encourage the spread of transmissible diseases such as bTB (Carstensen and DonCarlos 2011; Cosgrove et al. 2018). Other studies have shown that areas in which the movement of deer is curtailed or restricted by fencing or other procedures give rise to a higher incidence of bTB (Vicente et al. 2013; Ferreras-Colino et al. 2022).

In a previous study (Werrett and Green 2008), the bTB hot spot was also an area where hunting was not permitted, and in this particular case, deer were also partially enclosed. One of the known effects of hunting is that the combined effects of hounds and hunters encourage deer to leave their home area for periods of time and to limit the congregation in small areas. (Chassagneux et al. 2020). Similar results have been observed in other studies in Europe (Sunde et al. 2009) and USA (Brown et al. 2020). Studies in New Zealand have indicated that year-round hunting has resulted in low-density deer populations that are widely dispersed and less likely to contribute to deer transmission of bTB to members of the herd (Nugent et al. 2015). Considering these points, it is highly likely that the lack of disturbance due to hunting with hounds has had a major impact on the tendency for deer to congregate in high concentrations in regions of the Holnicote Estate and to contribute to the enhanced level of TB seen in this population of deer. This could be easily tested, in a scientifically appropriate manner, by reintroducing hunting with hounds to the estate (under the auspices of research and observation) and seeing how the incidence of bTB in deer and on farms is affected over time. This, in comparison with the findings of this current study, could contribute to a well-controlled study of the benefits, or not, of hunting with hounds on the health and well-being of the wild red deer of Exmoor.

Whatever the reason for the high incidence of bTB on the Holnicote Estate, it represents a welfare cost to the deer that should be addressed.

Deer density and inter-animal transmission

In this study, we failed to observe significant correlations between deer density and the incidence of bTB in deer, and between deer density and the incidence of bTB infection on farms. Previous studies have shown statistically significant correlations between these parameters in Sika deer (Kelly et al. 2021). The lack of significance in this study probably reflects the smaller numbers of animals studied in our study compared to those used in the previous study (Kelly et al. 2021). In agreement with Kelly et al. (2021), we did observe a significant correlation between the number of farms with bTB reactors and the number of bTB positive deer in the regions. Neither our data, nor that of the previous study (Kelly et al. 2021), can be taken as evidence of deer being maintenance hosts of bTB, but it does indicate a strong likelihood of cross species infection and provides support for previous studies which indicate that deer could be vectors for bTB infection in domestic cattle (Fink et al. 2015; Leth et al. 2019).

Deer and cattle share much of the habitat of Exmoor. Deer routinely enter cultivated pasture which is a major source of food for domestic cattle. In addition, cattle also graze areas of the open moor in which the wild deer tend to live. Thus, cattle and deer have a tendency to mix. In addition to deer, the other major wildlife host of bTB on Exmoor is the badger. In an attempt to try to limit the influence of the badger on bTB transmission, the height of water troughs was raised to stop badgers using them. However, deer can still access these water sources. However, the most likely cause of interspecies infection is shared grazing, either on farmland pasture or on the open moor. The route of infection could be oral or respiratory following the re-aerosolisation of deposited sources (Phillips et al. 2003; Palmer et al. 2004). It is also not possible to determine the direction of infection, or to what extent badgers enter the equation. Further studies are needed to clarify this. It also indicates that the eradication of the disease in wild and domesticated animals will require a co-ordinated approach targeting all potential species.

This study examined the situation in a small region of the UK. The results however may be important in the development, or modification, of deer management strategies elsewhere in the UK and beyond.

Future studies

This study suffered to some extent due to the small numbers of animals studied. Continual monitoring will provide greater numbers and improved scientific rigour. The eradication of bTB in wild and domesticated animals must be a long-term aim of future studies. It may not be possible to separate the species in order to limit interspecies infections. Population control is a possible area to investigate. The red deer population of Exmoor is relatively consistent, but there might be a case for increasing the culling of hinds (female deer) in order to limit the overall population. The most likely method of limiting the incidence of bTB in all host species is some form of vaccination. This is a rapidly evolving area. The administration of oral heat-inactivated BCG vaccine has shown some effectiveness in deer challenged with M. bovis infection (Thomas et al. 2017). This requires further study, as it opens a possible avenue to investigate a means of improving the welfare of the wild red deer of Exmoor, and limiting interspecies infection. The small population of deer showing a high incidence of bTB which has been identified in this study would be an ideal group to study as they are a relatively undisturbed colony living in an easily accessible area. Similar studies might be considered in domestic cattle.

Although serological studies such as this are valuable and inexpensive, a combination of serology and pathology, while much more expensive, may be the most appropriate for future studies (Ferreras-Colino et al. 2022).