The variability and seasonality of the environmental reservoir of Mycobacterium bovis shed by wild European badgers

The incidence of Mycobacterium bovis, the causative agent of bovine tuberculosis, has been increasing in UK cattle herds resulting in substantial economic losses. The European badger (Meles meles) is implicated as a wildlife reservoir of infection. One likely route of transmission to cattle is through exposure to infected badger urine and faeces. The relative importance of the environment in transmission remains unknown, in part due to the lack of information on the distribution and magnitude of environmental reservoirs. Here we identify potential infection hotspots in the badger population and quantify the heterogeneity in bacterial load; with infected badgers shedding between 1 × 103 − 4 × 105 M. bovis cells g−1 of faeces, creating a substantial and seasonally variable environmental reservoir. Our findings highlight the potential importance of monitoring environmental reservoirs of M. bovis which may constitute a component of disease spread that is currently overlooked and yet may be responsible for a proportion of transmission amongst badgers and onwards to cattle.

The incidence of Mycobacterium bovis in cattle herds in Great Britain (GB) has increased from 0.01% in 1979 1 to 4.8% in 2012 2 . Control and compensation has cost the taxpayer £500 million over the past decade and this is predicted to increase to £1 billion over the next 10 years if further geographical spread is observed 3 ; making bovine tuberculosis one of the most economically important animal health problems in the UK farming industry 4 .
The European badger is a wildlife reservoir involved in the transmission of M. bovis to cattle in the UK and RoI 5,6 . Once infected, badgers may intermittently shed M. bovis cells in sputum, faeces and urine 7 . One likely route of transmission to cattle is through exposure to infected badger urine and faeces. Although the location and extent of environmental M. bovis has not been previously quantified, indirect contact with infected faeces and urine may be an important exposure pathway 8 . M. bovis genomic DNA can survive outside the host for up to 21 months 9 and cells have been shown to be viable by culture from mice fed soil in which M. bovis had been persisting for months 10 . The survival of shed M. bovis cells is likely to vary in space and time in relation to local environmental conditions and the distribution of infectious badgers. Understanding patterns in environmental contamination (defined as the presence of M. bovis genome equivalents in the environment) could aid in the design of more effective interventions, currently based on culling and vaccination strategies.
The availability of a method to quantify relative levels of environmental contamination with M. bovis could open up possibilities for monitoring spatial and temporal variation in risk and may help direct the implementation of disease control interventions. Currently the only means of measuring levels of infection in badger populations is through trapping and testing with BrockTB Stat Pak ® (Stat Pak) 11 , Interferon gamma (IFNγ ) 12 and culture of clinical samples 13 . Cultivation, particularly from faecal material, has low sensitivity and is qualitative (Fig. 1). A qPCR method for non-invasive environmental monitoring of shedding was developed in our group 14,15 . This qPCR assay quantifies faecal shedding, a measure that correlates strongly with the level of infection within a social group as measured by immunoassay (Spearman's rho = 0.92, p < 0.001) 16 . The only other non-invasive method for monitoring infection in badger populations is culture of faecal material, which is particularly insensitive (Fig. 1). Using this optimised qPCR assay we are able to report on the spatio-temporal reservoir of M. bovis from badger faecal shedding in a natural population over the course of a year. Badgers defecate in latrines within or at the edges of their territories 17 and hence they can be used to identify a defined population of animals.

Methods
Sampling and Trapping. Fresh faecal samples were obtained from latrines associated with 12 badger social groups (Table 1) in Woodchester Park Gloucestershire through 2012 and 2013. Two intensive sampling periods of two weeks each were undertaken during the period of peak badger latrine activity in the spring and autumn of 2012 where up to 10 unique faecal samples were obtained from latrines associated with each social group on alternate days. Faecal samples were taken from latrines in closest proximity to  the main sett of each social group. A second sampling regime was undertaken over a year long period where up to 10 unique fresh, faecal samples were taken from latrines associated with each social group per day over two non-consecutive days in each season, starting two days after trapping operations took place in that location. For the purpose of this study March-May was classified as spring, June-August as summer, September-November as autumn and December-February as winter.
Each of the 12 badger social groups in the study was trapped four times throughout the year, once per season, with variable numbers of animals trapped between groups and seasons (Table S1). Badgers were trapped using baited cage traps placed around the main setts of each social group and identified using a unique tattoo applied at the first capture of that animal. Trapped badgers from each of the 12 social groups were tested by BrockTB Stat-Pak ® , IFNγ and culture of clinical samples. All experimental protocols were approved by the University of Warwick and the Food and Environment Research Agency Ethical Review Committee and carried out in accordance with the approved guidelines and under the license granted by the Home Office under the 1986 Animal (Scientific Procedures) Act.
DNA Extraction and qPCR. Total community DNA was extracted from 0.1 g (+ /− 0.003 g) of faeces using the Fast DNA spin kit for soil (MP Biomedicals) following the manufacturer's instructions. M. bovis was detected and quantified using a qPCR assay which targets the RD4 deletion region unique to the M. bovis genome (Specificity data Table S2). An initial qPCR screen of each sample was performed using an ABI 7500 Fast qPCR machine (ABI) with two technical replicates of each sample. Positive controls (8.5 × 10 5 genome equivalents) and negative controls were also present in duplicate on each plate. PCR reactions were set up using 900 nM of each primer (RD4F 5 'TGTGAATTCATACAAGCCGTAGTCG 3' , RD4R 5 'CCCGTAGCGTTACTGAGAAATTGC 3 '), 250 nM of Taqman probe (6FAM-AGCGCAACACTCTTGGAGTGGCCTAC-TMR), 1 mg ml −1 bovine serum albumen (BSA), 12.5 μ l of Environmental Mastermix 2.0 (ABI), 10 μ l of template and made up to 25 μ l with molecular grade water (Sigma Aldrich). PCR cycling conditions were 50 °C for 2 min followed by 95 °C for 10 min then 40 cycles of 95 °C for 15 sec and 58 °C for 1 min. Samples exhibiting amplification in one or more technical replicates were taken on to full quantification using three technical replicates per sample under the same conditions. If one or more of the technical replicates of the quantification assay exhibited amplification the sample was deemed positive for M. bovis. Serial dilutions of M. bovis BCG Danish 1331 genomic DNA were used as standards for this quantification. An inhibition control assay previously described 14 was used to detect the possibility of false negative results due to inhibition. Where significant inhibition was detected DNA was re-extracted from frozen aliquots and qPCR assays were repeated. The number of M. bovis genome equivalents was quantified independently by qPCR at The University of Warwick and APHA Weybridge (Supplementary Figure 1). Data Analysis. All data analysis was performed using the statistical program R. Logistic regression with social group (Old Oak) as the baseline was used to determine whether the number of positive samples varied amongst social groups throughout the year. Binomial generalised linear models (GLM) were performed to determine differences in M. bovis cells numbers shed between groups and between seasons. For spring two sampling days per social group were chosen to represent cross sectional sampling. Variability within groups was determined by calculating the median, upper and lower quartiles and range for each soil group.
The probability of detecting a false positive rate was 2%, calculated using known negative faecal samples obtained from captive badgers at APHA which were routinely tested for bTB using IFNγ . Negative samples were double blinded and randomly introduced into the experiment at both centres. The probability of detecting x false positive for a given number of samples was calculated using equation 1 where p(x) = the probability of exactly x false positives, f = the false positive rate, n = the number of samples and x = the number of false positives.
The number of confirmatory re-extractions (e) needed to result in the probability (p) of exactly x false positives was calculated using equation 2.

Results
Infection levels within social groups. During the study, 53.6% of trapped badgers were M. bovis positive by Stat-Pak, IFNγ or culture. By qPCR faecal samples from every social group examined were found to be positive (Fig. 2). Although the percentage of infected faecal samples varied considerably ( Table 1, Table S3), the numbers of M. bovis genome equivalents per faecal sample also varied widely ranged from 1 × 10 3 to 4 × 10 5 per gram of faeces (Table 1).
Significant variability in genome equivalents was identified both within and between social groups ( Fig. 3) with social groups Nettle, Top, Septic Tank and West shedding more cells over the year than the other social groups (Table 1). Social groups with a high percentage of positive samples consistently shed amongst the highest cumulative numbers of M. bovis cells during the year (Table 1). Social group Old Oak was exceptional as it has one of the highest cumulative M. bovis genome equivalent values yet had the lowest percentage of positive samples in the study (Table 1). This distribution is consistent with the presence of a relatively small number of animals shedding large amounts of bacteria in some groups. However, as we could not assign faecal samples to individuals we cannot discount within-individual variation in shedding from being responsible for this observation. Hence the need for further research into heterogeneity in transmission risks amongst individual badgers.  (Table 1). This corresponds to immunoassay tests carried out on trapped badgers, which also identified these five groups as the most heavily infected (Table 1). Although there is strong correspondence between immunoassay and qPCR results there are some discrepancies, in particular Nettle and Top are 100% and 90% positive by immunoassay yet there was a large difference in the percentage of positive faecal samples with 42.2% and 10.0% respectively.  The heterogeneities observed in this study between social groups and the consistency with which five groups were identified as highly infected and shedding, suggesting that interventions targeted at particular high risk populations could have a larger impact than random and blanket control strategies 18 . However, we are mindful that any perturbation of badger populations could result in increased rather than decreased transmission 19,20 . The observed discrepancies in the percentage of positive faecal samples for social groups with similar prevalences of infection by immunoassay highlights the need for further work to establish the causes of these differences. Whilst heterogeneity in transmission is a well-known phenomenon, this study is one of the few empirical studies which have attempted to demonstrate the extent of this variability 21 . Although this study does not assess the viability of M. bovis in faeces, previous work has identified the presence of M. bovis 16S rRNA in soil 9 and badger setts and latrines 22 . In addition, studies have had a culture success rate of 2.5% from badger faecal samples 23 and M. bovis has been cultured from cattle faeces several months after excretion 24 . This indicates that at least a proportion of M. bovis cells shed in badger faeces can remain viable in the environment; however, further research is required to determine potential survival and transmissibility of M. bovis in environmental samples.
Whilst the focus in the UK and RoI is on badgers, other wildlife hosts are present 25,26 ; however, little is currently known of their contribution to environmental reservoirs and their relative importance for transmission to cattle 25