Skip to main content
NCBI home page
Immunology logoLink to Immunology
. 2015 Jan 6;144(2):263–270. doi: 10.1111/imm.12369

Association of quantitative interferon-γ responses with the progression of naturally acquired Mycobacterium bovis infection in wild European badgers (Meles meles)

Alexandra J Tomlinson 1,✉, Mark A Chambers 2,3, Robbie A McDonald 4, Richard J Delahay 1,✉
PMCID: PMC4298420  PMID: 25109384

Abstract

Bovine tuberculosis is one of the biggest challenges facing cattle farming in Great Britain. European badgers (Meles meles) are a reservoir host for the causal agent, Mycobacterium bovis. There have been significant recent advances in diagnostic testing for tuberculosis in humans, cattle and badgers, with the development of species-specific assays for interferon-γ (IFN-γ), an important cytokine in tuberculous infections. Using data collected from longitudinal studies of naturally infected wild badgers, we report that the magnitude of the IFN-γ response to M. bovis antigens at the disclosing test event was positively correlated with subsequent progression of disease to a seropositive or excreting state. In addition, we show that the magnitude of the IFN-γ response, despite fluctuation, declined with time after the disclosing event for all badgers, but remained significantly higher in those animals with evidence of disease progression. We discuss how our findings may be related to the immunopathogenesis of natural M. bovis infection in badgers.

Keywords: badger, bovine tuberculosis, interferon-γ

Introduction

The manifestation of disease associated with mycobacterial infections is highly variable both within and between species, and arises as a consequence of complex interactions between host and pathogen.1,2 The Mycobacterium tuberculosis complex consists of several closely related pathogenic mycobacteria including M. tuberculosis, the major cause of tuberculosis in humans, and Mycobacterium bovis, the principal causal agent of bovine tuberculosis. Globally, human tuberculosis is one of the leading causes of adult death due to infectious disease.3 In Great Britain, bovine tuberculosis in cattle is one of the most pressing problems facing British agriculture.4 Endemic M. bovis infection in some European badger (Meles meles) populations has been linked to the local persistence of the disease in cattle in both Great Britain5 and Ireland.6

Tuberculous mycobacteria are intracellular pathogens and, as such, a successful host response is likely to be characterized by a strong cell-mediated component of the adaptive immune response.7 Protection against the establishment and progression of infection with tuberculous mycobacteria is, however, almost certainly more complex than is currently understood, potentially involving different populations of T cells,7 and/or mucosal and serum antibody responses.8,9

Advances in our understanding of the diversity of immunological responses to mycobacterial infections underpin improvements in development of vaccines and diagnostic tests for humans and other animals. The cytokines released from activated T cells, for example, are potentially useful diagnostic and prognostic tools.10 In particular, interferon-γ (IFN-γ) has been recognized as playing a pivotal role in protective immunity, and in humans IFN-γ secretion has been shown to be highly heritable.11 There is, however, no simple correlation between IFN-γ levels and protection from disease in humans,12,13 laboratory animals14,15 or cattle.16 Indeed, several longitudinal case studies in humans have suggested that the magnitude of the early IFN-γ response is positively correlated with a likelihood of disease progression.12,17,18

A diagnostic assay for IFN-γ has been developed specifically for use in badgers,19 and has been deployed since 2006 in field research on bovine tuberculosis in wild, naturally infected badger populations. These studies have some parallels with those conducted on humans following exposure to an M. tuberculosis index case, as both are based on temporal observations subsequent to natural exposure. In the present study, we used data collected from free-living, wild badgers to investigate whether there was any evidence for correlation of the cell-mediated immune response at the incident event with the degree of subsequent disease progression. In particular, we sought to determine whether the magnitude of the IFN-γ response observed at the disclosing event was correlated with later diagnostic test outcomes in individual badgers, and to describe temporal trends in the subsequent cell-mediated immune response.

Materials and methods

Badger life-history data and M. bovis diagnostic test results were collected from two geographically distinct, naturally infected populations of wild badgers in Gloucestershire, southwest England, over similar time periods. The first was a well-studied population at Woodchester Park in Gloucestershire where the IFN-γ assay,19 and the Brock TB Stat-Pak® serological test20 were introduced in July 2006. Test result data collected from then until January 2012 were used in the present study. The second study population was near Cirencester, Gloucestershire (approximately 15 miles from the Woodchester study area), and was trapped and sampled as part of a safety study for the licensing of injectable bacillus Calmette–Guérin (BCG) vaccine for use in badgers (Badger Vaccine Study, BVS).21,22 Data collected from this study area between 2006 and 2009 were combined with those from Woodchester Park.

Both populations were captured and sampled using well-established methods23 resulting in repeated observations of the same individuals over time. At Woodchester Park, each badger social group was trapped four times per year, while in the BVS trapping took place up to twice per year. Following capture, badgers were transported to a nearby sampling facility where they were anaesthetized and individually marked by ventral abdominal tattoo at first capture. Age (cub or adult) and sex were recorded. Blood samples were taken for the IFN-γ and Stat-Pak assays, and a suite of clinical samples (tracheal aspirate, urine, faeces and swabs from bite wounds or abscesses if present) were collected for M. bovis culture.24 Following recovery, badgers were released at the point of capture. Data from all badgers trapped at Woodchester Park were used, but for the BVS study area, we only used data from badgers captured in social groups that were not vaccinated with BCG.

Identifying the IFN-γ incident case

The IFN-γ assay is a quantitative ELISA, measuring the IFN-γ produced by white blood cells in response to incubation with purified protein derivatives from M. bovis (PPD-B), Mycobacterium avium (PPD-A), and positive and negative controls. Data were recorded as optical density (OD) values in response to each of the antigens (two wells, and therefore two values, per antigen). The response to PPD-B was calculated as the average of the two values in response to stimulation with PPD-B, minus the negative control average. The average response to PPD-B minus the average response to PPD-A (hereafter referred to as ‘B–A’) was used as the determinant for classifying a sample as either positive or negative to infection with M. bovis, as this value partially controls for variation in the PPD-B response arising from exposure to or infection by environmental mycobacteria. Values of B–A greater than 0·044 were classified as positive19 for adults. With this threshold value and no age differentiation, the test has a reported sensitivity of 81% and a specificity of 94%.19 For cubs, we used a lower cut-off value of 0·023, on the basis of previous observations of age-related differences in test performance.25 With this lowered threshold, the test has a reported sensitivity of 71% and a specificity of 95% in cubs.25

The Brock TB Stat-Pak® is a lateral flow immunoassay incorporating the antigens MPB83, MPB70, CFP10 and a 16 000 molecular weight antigen, and provides a binary outcome on the basis of serum antibodies binding to one or more of these antigens.20

The periodic nature of the badger trapping events in the present study means that the true incident event (the point at which an individual would have first tested positive in the IFN-γ test) will have been at some point before the actual disclosing test event. Historical analyses of badgers in the Woodchester Park study area have shown that, on average, badgers are trapped twice in each year,26 suggesting that an IFN-γ disclosing event was likely to be no more than 6 months after the true incident event, assuming 100% test sensitivity.

For the purposes of simplicity we have also assumed in our analyses that there was a single exposure event, although in reality multiple exposure events may occur in a population with endemic infection. This limitation is common to similar studies in humans in that there is likely to be ongoing contact between sources of infection, such as family members, and the index case. In the case of human studies, however, this effect may be offset to some degree by treatment of the index case.

Outcome infection status

At each capture event, infection status was assigned to a badger using a one-way, progressive system similar to that described by Delahay et al.26 Individuals were assumed to move from uninfected to IFN-γ positive to Stat-Pak positive and finally to culture positive, based on models of the immunopathogenesis of M. bovis infection in badgers27,28 and humans.29 There was no necessity to pass sequentially through each stage but there was no return to a lesser infection status. Each badger was accordingly assigned an ‘outcome infection status’; either IFN-γ only, Stat-Pak positive or culture positive.

Data analysis

The data set was restricted to ensure that the disclosing test event, specifically the point at which a comparative B–A IFN-γ response greater than the positive cut-off value (0·044 for adults; 0·023 for cubs) was first detected, was as close to the true incident event as possible, and that there was an opportunity to test for subsequent progression of infection. This meant only including badgers that had at least one capture event before the IFN-γ disclosing event at which the results of all three diagnostic tests (IFN-γ, Stat-Pak and culture) were negative. In addition, only badgers that were Stat-Pak and culture negative at the IFN-γ disclosing event were included. Finally, badgers were only included if there was at least one capture, with results from all three diagnostic tests, subsequent to the IFN-γ disclosing event.

The likelihood of detection of a positive Stat-Pak or culture test result may be influenced by the number of times the test is performed and the time elapsed from the IFN-γ disclosing event. Hence, we carried out a multiple logistic regression to determine whether variation in the detection of progression was related to the time to last capture from the IFN-γ disclosing event, and the number of subsequent test events. Badgers were deemed to have progressed from an IFN-γ response if their outcome infection status was either Stat-Pak positive or culture positive, resulting in a binary response variable for this regression analysis.

Two general linear models were constructed to determine if there was significant variation in the magnitude of the IFN-γ response at the disclosing event, in relation to the outcome infection status of an individual badger. The response variables were the response to PPD-B minus negative control in the first model, and the response B–A in the second model. B–A was modelled in addition to the response to PPD-B minus negative control, as this reflects the application of the IFN-γ test in a clinical setting. Both response variables were log-transformed to normalize their distributions. Explanatory variables were: outcome infection status, age at the disclosing event (cub or adult), sex and the source population of the badger (WP or BVS). The predicted means and associated standard errors from the first PPD-B model alone (since outputs were very similar for both models), were back-transformed and plotted to show the predicted mean value of the IFN-γ response, at the disclosing event, for each outcome infection status.

Two generalized linear mixed models were constructed to examine the effects of time on the magnitude of the IFN-γ response shown by each badger at all capture events subsequent to the IFN-γ disclosing event. The data set was restricted to exclude the IFN-γ disclosing events themselves, because we were interested in progression following initial disclosure. As above, the response variables were the response to PPD-B minus negative control, and the response B–A, both log-transformed. Explanatory variables that were included as fixed effects were the time elapsed from the IFN-γ disclosing event (log-transformed), infection status at the capture event (IFN-γ, Stat-Pak positive, culture positive), badger age (cub or adult), sex, and the source population of the badger (WP or BVS). Individual badger was included as a random effect to account for repeat captures of the same individual. An interaction term between infection status and the time elapsed from the disclosing event was included to control for the possibility that the effects of time on the magnitude of the IFN-γ response differed among the levels of infection status.

All analyses were performed in Genstat 15th Edition, (VSN International, Hemel Hempstead, UK). All work was approved by ethical review in the relevant institutions, and carried out under licence granted by the Home Office under the 1986 Animals (Scientific Procedures) Act.

Results

Application of restriction criteria to the data set

Of 751 badgers tested, 313 individuals (128 cubs; 185 adults), were identified as IFN-γ positive at some point in their capture histories. After applying the restriction criteria, the number of eligible badgers was reduced to 56 (13 cubs; 43 adults). The remaining 257 badgers were rejected for the following reasons: the IFN-γ disclosing event was at either the first capture (n = 70), or the last capture (n = 90); the animal was only caught once such that the IFN-γ disclosing event was both the first and last capture event (n = 59); infection status was Stat-Pak positive or culture positive before (n = 13) or at (n = 21) the IFN-γ disclosing event; there was insufficient information to assess infection status before or at the IFN-γ disclosing event (n = 4).

Descriptive statistics

For 38 of the 56 badgers, there was no evidence of infection progression to Stat-Pak-positive or culture-positive status at subsequent captures, resulting in an outcome infection status of IFN-γ only. For the remaining 18 badgers there was evidence of progression to Stat-Pak-positive (n = 13) or culture-positive (n = 5) status. Mycobacterium bovis was cultured from bite wounds in four of the five badgers detected as culture positive. In three of these four badgers the first culture-positive capture event yielded M. bovis only from a bite wound; all other samples were culture negative. For the fifth badger, M. bovis was first cultured from a lymph node abscess.

For the 56 eligible badgers in the restricted data set, there were 184 capture events subsequent to the IFN-γ disclosing event (range = 1–16 captures per badger; mean 3·5 captures per badger, SEM = 0·47). The mean period of time from IFN-γ disclosure to the last capture was 568 days (SEM = 63). Of these 184 capture events, the majority (n = 134, 73%) were IFN-γ negative (based on the comparative B–A value). At an individual level, just under half (n = 27, 48%) of the 56 badgers initially categorised as IFN-γ positive, had no subsequent positive IFN-γ tests.

Statistical analyses

The magnitude of the IFN-γ response at the disclosing event using both measures (response to PPD-B minus negative control, and B–A) differed significantly among the levels of outcome infection status. Positive parameter estimates and test statistics in each model indicated a significantly greater IFN-γ response at disclosure in animals in which there was subsequent evidence of progression (Table1). For both measures of the IFN-γ response, the magnitude at disclosure was greatest if the outcome status of the badger was culture positive (Table1). Similarly, for both test variables, the IFN-γ response was significantly lower (negative parameter estimates) in cubs than adults and lower in male badgers than females. Source population was not a significant factor.

Table 1.

Results from a general linear model constructed to analyse the variation in the magnitude of the IFN-γ response at the disclosing test (recorded as both PPD-B minus negative control and PPD-B minus PPD-A), in relation to the following explanatory variables: outcome infection status, age, sex and source population, for 56 badgers captured in Gloucestershire between 2006 and 2012

Response variable Explanatory variables Level Estimate1 (SE) t(50)2 P value
PPD-B minus negative control Outcome infection status (reference level IFN-γ positive only) Stat-Pak positive 0·53 (0·19) 2·77 0·008
Culture positive 0·96 (0·29) 3·37 0·001
Age (reference level adult) Cub −0·64 (0·19) −3·30 0·002
Sex (reference level female) Male −0·42 (0·16) −2·61 0·012
Source population (reference level BVS) WP 0·13 (0·22) 0·58 0·566
PPD-B minus PPD-A (B–A) Outcome infection status (reference level IFN-γ positive only) Stat-Pak positive 0·54 (0·19) 2·86 0·007
Culture positive 1·01 (0·28) 3·55 < 0·001
Age (reference level adult) Cub −0·59 (0·19) −3·03 0·004
Sex (reference level female) Male −0·38 (0·16) −2·40 0·020
Source population (reference level BVS) WP 0·23 (0·22) 1·06 0·297
Open in a new tab
1

Estimate from the model of the difference in the mean value relative to the reference level, with its associated standard error (SE).

2

The test statistic from the model output with associated degrees of freedom in brackets.

Note: Significant effects are shown in bold, P < 0·05.

Abbreviations: BVS, Badger Vaccine Study; IFN-γ interferon-γ; PPD-A, purified protein derivative from Mycobacterium avium; PPD-B, purified protein derivative from Mycobacterium bovis.

The back-transformed mean values, as predicted by our model, of the response to PPD-B minus negative control at first disclosure, for badgers with subsequent evidence of progression (to either Stat-Pak-positive status or culture-positive status) were each significantly greater than the mean value for badgers with no evidence of subsequent progression (Fig.1).

Figure 1.

Figure 1

Open in a new tab

Predicted mean values of the IFN-γ response to PPD-B minus negative control at the disclosing test, in relation to the outcome infection status of 56 badgers. Error bars represent standard error of the mean. IFN-γ, interferon-γ; PPD-B, purified protein derivatives from Mycobacterium bovis.

There was a significant decline over time (negative parameter estimate) in the magnitude of the IFN-γ response, for both measures (PPD-B minus negative control, and B–A), at capture events subsequent to the IFN-γ disclosing event (Table2). In addition, the magnitude of the IFN-γ response at each capture event, subsequent to the disclosing event, for both measures (PPD-B minus negative control, and B–A) differed significantly among levels of infection status in our models, with culture-positive badgers exhibiting the greater responses (Table2), illustrating that the ordering of response magnitude identified by the IFN-γ disclosing event analysis was maintained as infection progressed. In models of both response variables, the interaction term between time elapsed and infection status was non-significant and was therefore excluded from the final model.

Table 2.

Results from a generalized linear mixed model to analyse the variation in the magnitude of the IFN-γ response, (recorded as both PPD-B minus negative control and PPD-B minus PPD-A), subsequent to the IFN-γ disclosing event, in relation to the following explanatory variables: infection status at the capture event, time from the IFN-γ incident event, age, sex and source population, for 184 test events of 56 badgers captured in Gloucestershire between 2006 and 2012

Response variable Explanatory variables Level Estimate1 (SE) F statistic2 P value
PPD-B minus negative control Infection status (reference level IFN-γ positive) Stat-Pak positive 0·38 (0·26) F2,132·0 = 8·70 < 0·001
Culture positive 1·07 (0·26)
Time from the IFN-γ incident event (log transformed) N/A −0·15 (0·06) F1,176·8 = 6·57 0·011
Age (reference level adult) Cub −0·47 (0·26) F1,173·5 = 3·21 0·075
Sex (reference level female) Male −0·39 (0·15) F1,45·0 = 6·76 0·013
Source population (reference level BVS) WP −0·10 (0·22) F1,81·8 = 0·21 0·650
PPD-B minus PPD-A (B–A) Infection status (reference level IFN-γ positive) Stat-Pak positive 0·37 (0·34) F2,118·1 = 5·09 0·008
Culture positive 1·08 (0·34)
Time from the IFN-γ incident event (log transformed) N/A −0·20 (0·08) F1,175·1 = 6·38 0·012
Age (reference level adult) Cub −0·32 (0·36) F1,176·9 = 0·78 0·379
Sex (reference level female) Male −0·29 (0·17) F1,42·8 = 2·92 0·095
Source population (reference level BVS) WP −0·03 (0·28) F1,105·8 = 0·01 0·914
Open in a new tab
1

Estimate from the model of the difference in the mean value relative to the reference level with its associated standard error (SE).

2

The test statistic from the model output with associated degrees of freedom.

Note: Significant effects are shown in bold, P < 0·05.

Abbreviations: BVS, Badger Vaccine Study; IFN-γ interferon-γ; PPD-A, purified protein derivative from Mycobacterium avium; PPD-B, purified protein derivative from Mycobacterium bovis.

The magnitude of the IFN-γ response at each capture event was lower in males than females, but this was only significant in the PPD-B minus negative control model. Neither badger age nor the source population significantly affected the IFN-γ response at each capture.

There were no significant associations between the likelihood of a higher outcome infection status and either the number of test events subsequent to the IFN-γ disclosing event, or the time from IFN-γ disclosure to last capture (P > 0·05). Hence, there was no evidence for a significant bias in determination of outcome infection status related to the number of tests or the time period.

Discussion

Our longitudinal study provides evidence of temporal variation and progression of cell-mediated immune responses to natural infection in a wildlife population. Our analyses have shown that the magnitude of the early cell-mediated response is significantly related to the likelihood of disease progression. We have quantified these effects using our model, such that the magnitude of the IFN-γ response to PPD-B at its first detection, in badgers that were subsequently culture positive, was more than 2·5 times greater than that of badgers with no evidence of subsequent progression, and 1·5 times greater than that of badgers which subsequently became seropositive but not culture positive (see Fig.1). In addition, the IFN-γ response at the disclosing event of subsequently seropositive badgers was nearly double that of badgers in which there was no evidence of any subsequent progression (see Fig.1).

Subsequent to the disclosing event, IFN-γ responses in badgers with evidence of progression were also higher than those of badgers in which there was no evidence of progression, despite a general decline over time in the magnitude of the IFN-γ response in all badgers.

Our findings rely on the detection of seroconversion and M. bovis excretion. Lack of evidence of progression in individual badgers in our study could be considered a function of the low sensitivities of both the Stat-Pak test (54%),25 and of culture of M. bovis from clinical samples (two independent assessments: 25%30 and 27·5%).31 A greater number of test events combined with a longer time period for progression to occur could increase the likelihood of detecting disease progression. We demonstrated, however, that there was no significant bias in our categorization of the outcome infection status for each badger related to either the duration of the period from the first detection of infection, or the number of times it was tested. Hence, we can be confident that our analytical approach was a valid means of examining the relationship between IFN-γ responses and subsequent disease progression. Furthermore, although the diagnostic approaches have their limitations,32 previous studies have consistently demonstrated their value in distinguishing among epidemiologically meaningful categories of infection status.33–36

Similar correlations between early IFN-γ responses and subsequent progressive disease have been reported from experimental infection models in cattle37–40 and captive badgers.41 This body of evidence is consistent with an immunopathological contribution to observed subsequent disease. In humans, the value of quantifying the magnitude of the early IFN-γ response relates to stratification of the progression of risk in an individual, to target prophylactic treatment more effectively.18,29,42 In wild animals, the magnitude of the IFN-γ response at the disclosing event may therefore hold value as a predictor for disease progression in wild animals. However, its practical value as a field tool for the selective identification of individual badgers for management intervention is likely to be limited, not least because the current format of the test requires overnight incubation of blood samples.19 As further data become available it may be feasible to derive a probabilistic assessment of the risk of a badger becoming Stat-Pak positive or excreting M. bovis on the basis of its IFN-γ test outcomes. In addition, analysing quantitative data from the IFN-γ release assay, may give us further insight into the weight of infection at a social group level and, therefore, the distribution of infection risk in the population. The application of risk-based approaches that incorporate epidemiological information is not a new idea in the human context,18,29,43–45 but applying it in a veterinary setting is more novel.

The positive correlation between initial IFN-γ responses and subsequent disease progression in our study may also relate to the infective dose. There is some evidence for association with the infective dose from both experimental and natural infection models. In an experimental study in captive badgers, the magnitude of the cell-mediated response to PPD-B, as measured by a lymphocyte transformation assay, was positively correlated with the challenge dose of M. bovis.41 Dose-related responses to BCG have also been reported in badgers using the ELISpot assay, which measures the number of IFN-γ-producing T cells rather than the magnitude of the IFN-γ response.46 In humans, the level of contact with an index case has also been positively correlated with initial IFN-γ responses.47

It is also possible that the route of infection may influence the magnitude of IFN-γ responses and subsequent pathology. For example, there is evidence to suggest that infection via biting may be associated with more aggressive pathology in badgers,48,49 and experimental intra-dermal injection of M. bovis has been linked to progressive systemic infection.50 Unfortunately there were no measures of the IFN-γ response in these studies. In the present study, there were five badgers for which there was evidence of progression to M. bovis excretion, four of which involved isolation of M. bovis from a bite wound. In addition, in three of these badgers M. bovis was cultured only from bite wounds at the culture-positive incident event. Unfortunately, the very small sample sizes preclude drawing any conclusions, but we suggest that our findings are not inconsistent with the route and/or the dose affecting the magnitude of the initial IFN-γ response and the likelihood of future progression.

We identified a reduction in the magnitude of the IFN-γ response with time, subsequent to the disclosing event, with no subsequent positive test results (based on the comparative B–A response) in just under half of the badgers. Although this may not initially appear to be consistent with a reported sensitivity value of 81%19 for the comparative test, our observations are of a restricted group of individuals subsequent to the detection of infection, and it is therefore difficult to draw any firm conclusions. In badgers with evidence of progressive infection, the magnitude of the IFN-γ response declined over time following initial detection, despite remaining higher than in badgers with no evidence of progression. A similar decline in the IFN-γ response has been observed in experimentally infected badgers51 and wild boar (Sus scrofa);52 in both cases associated with seroconversion and progressive clinical disease. Possible explanations for these observed reductions include a shift from a T-helper type 1-biased response to a T helper type 0-biased response, as observed in experimentally infected cattle,1 and/or progressive anergy in response to continued antigenic stimulation; anergy to tuberculin skin tests (indirect measures of the cell-mediated response) has been reported in cattle,53,54 and in humans.55

In badgers with no evidence of subsequent progression, the magnitude of the IFN-γ response was initially lower than for badgers with subsequent progression but also faded with time, which could be consistent with latent infection.56 Where the response faded completely, this could potentially have been related to full resolution and clearance of infection.

The magnitude of the IFN-γ response as measured by B–A, and in response to PPD-B minus negative control, was also consistently lower in males than females. This is consistent with evidence of enhanced cell-mediated immune responses reported in female humans, in association with the differences in sex hormone profiles between males and females.57

In summary, we have shown that the magnitude of the early IFN-γ responses of badgers to M. bovis arising from naturally acquired infection, is positively correlated with subsequent progressive disease. In addition, we have shown that IFN-γ responses in all badgers reduce over time, for which we offer several hypotheses. Progress in understanding the immunobiology of naturally acquired M. bovis infection in wild badgers should assist with the future development of diagnostic tests, and vaccination and disease management strategies, as mirrored by the study of tuberculosis in humans.

Acknowledgments

The authors wish to express their thanks to the field workers and all other staff at FERA and AHVLA who contributed to data collection in both field studies. Thanks to AHVLA staff at the National Wildlife Management Centre, Langford and Weybridge, and to FERA staff, for diagnostics and assistance with data management. We are also grateful to all the farmers and landowners in both study areas for their co-operation. This work was funded by Defra.

Disclosures

None.

References

  1. Welsh MD, Cunningham RT, Corbett DM, Girvin RM, McNair J, Skuce RA, Bryson DG, Pollock JM. Influence of pathological progression on the balance between cellular and humoral immune responses in bovine tuberculosis. Immunology. 2005;114:101–11. doi: 10.1111/j.1365-2567.2004.02003.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Dharmadhikari AS, Nardell EA. What animal models teach humans about tuberculosis. Am J Respir Cell Mol Biol. 2008;39:503–8. doi: 10.1165/rcmb.2008-0154TR. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. World Health Organisation. Global Tuberculosis Report 2012. WHO/HTM/TB/2013.11. Geneva: World Health Organisation; 2013. [Google Scholar]
  4. Godfray HCJ, Donnelly CA, Kao RR. A restatement of the natural science evidence base relevant to the control of bovine tuberculosis in Great Britain. Proc R Soc Lond B Biol Sci. 2013;280:20131634. doi: 10.1098/rspb.2013.1634. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bourne FJ, Donnelly C, Cox D, Gettinby G, McInerney J, Morrison I, Woodroffe R, Bovine TB. The Scientific Evidence, A Science Base for a Sustainable Policy to Control TB in Cattle, An Epidemiological Investigation into Bovine Tuberculosis . Final Report of the Independent Scientific Group on Cattle TB: Department for Environment, Food and Rural Affairs, London, 2007: 83–5.
  6. Griffin JM, Williams DH, Kelly GE, Clegg TA, O'Boyle I, Collins JD, More SJ. The impact of badger removal on the control of tuberculosis in cattle herds in Ireland. Prev Vet Med. 2005;67:237–66. doi: 10.1016/j.prevetmed.2004.10.009. [DOI] [PubMed] [Google Scholar]
  7. Boom WH. The role of T-cell sub-sets in Mycobacterium tuberculosis infection. Infect Agents Dis. 1996;5:73–81. [PubMed] [Google Scholar]
  8. Glatman-Freedman A. The role of antibody-mediated immunity in defence against Mycobacterium tuberculosis: advances towards a novel vaccine strategy. Tuberculosis (Edinb) 2006;86:91–197. doi: 10.1016/j.tube.2006.01.008. [DOI] [PubMed] [Google Scholar]
  9. Hoft D. Tuberculosis vaccine development: goals, immunological design and evaluation. Lancet. 2008;372:164–75. doi: 10.1016/S0140-6736(08)61036-3. [DOI] [PubMed] [Google Scholar]
  10. Lalvani A, Millington KA. T cells and tuberculosis: beyond interferon-γ. J Infect Dis. 2008;197:941–3. doi: 10.1086/529049. [DOI] [PubMed] [Google Scholar]
  11. Cobat A, Gallant CJ, Simkin L. High heritability of antimycobacterial immunity in an area of hyperendemicity for tuberculosis disease. J Infect Dis. 2010;201:15–9. doi: 10.1086/648611. [DOI] [PubMed] [Google Scholar]
  12. Doherty TM, Demissie A, Olobo J, Wolday D, Britton S, Eguale T, Ravn P, Andersen P. Immune responses to the Mycobacterium tuberculosis-specific antigen ESAT-6 signal and subclinical infection among contacts of tuberculosis patients. J Clin Microbiol. 2002;40:704–6. doi: 10.1128/JCM.40.2.704-706.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fletcher HA. Correlates of immune protection from tuberculosis. Curr Mol Med. 2007;7:319–25. doi: 10.2174/156652407780598520. [DOI] [PubMed] [Google Scholar]
  14. Mittrücker H, Steinhoff U, Köhler A, Krause M, Lazar D, Mex P, Miekley D, Kaufmann SHE. Poor correlation between BCG vaccination-induced T cell responses and protection against tuberculosis. Proc Natl Acad Sci USA. 2007;104:12434–9. doi: 10.1073/pnas.0703510104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Badell E, Nicolle F, Clark S. Protection against tuberculosis induced by oral prime with Mycobacterium bovis BCG and intranasal subunit boost based on the vaccine candidate Ag85B-ESAT-6 does not correlate with circulating IFN-γ producing T-cells. Vaccine. 2009;27:28–37. doi: 10.1016/j.vaccine.2008.10.034. [DOI] [PubMed] [Google Scholar]
  16. Buddle BM, Wedlock DN, Denis M, Skinner MA. Identification of immune response correlates for protection against bovine tuberculosis. Vet Immunol Immunopathol. 2005;108:45–51. doi: 10.1016/j.vetimm.2005.08.002. [DOI] [PubMed] [Google Scholar]
  17. Adetifa IMO, Brookes R, Lugos MD, de Jong BC, Antonio M, Adegbola RA, Hill PC. Rising ELISPOT count prior to the onset of symptoms of full-blown tuberculosis disease. Int J Tuberc Lung Dis. 2007;11:350–2. [PubMed] [Google Scholar]
  18. Metcalfe JZ, Cattamanchi A, Vittinghof E, Ho C, Grinsdale J, Hopewell PC, Kawamura M, Nahid P. Evaluation of a quantitative IFN-γ response for risk stratification of active tuberculosis suspects. Am J Respir Crit Care Med. 2010;181:87–93. doi: 10.1164/rccm.200906-0981OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Dalley DJ, Davé D, Lesellier S, Palmer S, Crawshaw T, Hewinson RG, Chambers MA. Development and evaluation of a gamma-interferon assay for tuberculosis in badgers (Meles meles. Tuberculosis (Edinb) 2008;88:235–43. doi: 10.1016/j.tube.2007.11.001. [DOI] [PubMed] [Google Scholar]
  20. Chambers MA, Crawshaw T, Waterhouse S, Delahay R, Hewinson RG, Lyashchenko KP. Validation of the BrockTB Stat-Pak assay for detection of tuberculosis in Eurasian badgers (Meles meles) and influence of disease severity on diagnostic accuracy. J Clin Microbiol. 2008;46:1498–500. doi: 10.1128/JCM.02117-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Chambers MA, Rogers F, Delahay RJ. Bacille Calmette–Guérin vaccination reduces the severity of tuberculosis in badgers. Proc R Soc Lond B Biol Sci. 2011;278:1913–20. doi: 10.1098/rspb.2010.1953. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Carter SP, Chambers MA, Rushton SP. BCG vaccination reduces risk of tuberculosis infection in vaccinated badgers and unvaccinated badger cubs. PLoS ONE. 2012;7:e40933. doi: 10.1371/journal.pone.0049833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Vicente J, Delahay RJ, Walker N, Cheeseman CL. Social organisation and movement influence the incidence of bovine tuberculosis in an undisturbed high density badger (Meles meles) population. J Anim Ecol. 2007;76:348–60. doi: 10.1111/j.1365-2656.2006.01199.x. [DOI] [PubMed] [Google Scholar]
  24. Gallagher J, Horwill DM. A selective oleic acid albumin agar medium for the cultivation of Mycobacterium bovis. J Hyg (Lond) 1977;79:155–60. doi: 10.1017/s0022172400052943. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Chambers MA, Waterhouse S, Lyashchenko K, Delahay R, Sayers R, Hewinson RG. Performance of TB immunodiagnostic tests in Eurasian badgers (Meles meles) of different ages and the influence of duration of infection on serological sensitivity. BMC Vet Res. 2009;5:42–9. doi: 10.1186/1746-6148-5-42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Delahay RJ, Langton S, Smith GC, Clifton-Hadley RS, Cheeseman CL. The spatio-temporal distribution of Mycobacterium bovis (bovine tuberculosis) infection in a high-density badger population. J Anim Ecol. 2000;69:428–41. [Google Scholar]
  27. Mahmood KH, Stanford JL, Rook GAW, Stuart FA, Pritchard DG, Brewer JI. The immune responses in two populations of wild badgers naturally infected with bovine tubercle bacilli. Tubercle. 1987;68:119–25. doi: 10.1016/0041-3879(87)90027-4. [DOI] [PubMed] [Google Scholar]
  28. Lesellier S, Corner L, Costello E. Antigen specific immunological responses of badgers (Meles meles) experimentally infected with Mycobacterium bovis. Vet Immunol Immunopathol. 2008;122:35–45. doi: 10.1016/j.vetimm.2007.11.005. [DOI] [PubMed] [Google Scholar]
  29. Andersen P, Doherty M, Pai M, Weldnigh K. The prognosis of latent tuberculosis: can disease be predicted? Trends Mol Med. 2007;13:175–82. doi: 10.1016/j.molmed.2007.03.004. [DOI] [PubMed] [Google Scholar]
  30. Pritchard DG, Stuart FA, Wilesmith JW, Cheeseman CL, Brewer JI, Bode R, Sayers PE. Tuberculosis in East Sussex III. Comparison of post-mortem and clinical methods for the diagnosis of tuberculosis in badgers. J Hyg (Lond) 1986;97:27–36. doi: 10.1017/s0022172400064329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Chambers MA, Pressling WA, Cheeseman CL, Clifton-Hadley RS, Hewinson RG. Value of existing serological tests for identifying badgers that shed Mycobacterium bovis. Vet Microbiol. 2002;86:183–9. doi: 10.1016/s0378-1135(02)00012-3. [DOI] [PubMed] [Google Scholar]
  32. Drewe J, Tomlinson AJ, Walker NJ, Delahay RJ. Diagnostic accuracy and optimal use of three tests for tuberculosis in live badgers. PLoS ONE. 2010;5:e11196. doi: 10.1371/journal.pone.0011196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Wilkinson D, Smith GC, Delahay RJ, Rogers LM, Cheeseman CL, Clifton-Hadley RS. The effects of bovine tuberculosis (Mycobacterium bovis) on mortality in a badger (Meles meles) population in England. J Zool. 2000;250:389–95. [Google Scholar]
  34. Delahay RJ, Walker N, Smith GC, Wilkinson D, Clifton-Hadley RS, Cheeseman CL, Tomlinson AJ, Chambers MA. Long-term temporal trends and estimated transmission rates for Mycobacterium bovis infection in an undisturbed high-density badger (Meles meles) population. Epidemiol Infect. 2013;141:1445–56. doi: 10.1017/S0950268813000721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Graham J, Smith GC, Delahay RJ, Bailey T, McDonald RA, Hodgson D. Multi-state modelling reveals sex-dependent transmission and severity of tuberculosis in wild badgers. Epidemiol Infect. 2013;141:1429–36. doi: 10.1017/S0950268812003019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Tomlinson AJ, Chambers MA, Carter SP, Wilson GJ, Smith GC, McDonald RA, Delahay RJ. Heterogeneity in the risk of Mycobacterium bovis infection in European badgers (Meles meles) cubs. Epidemiol Infect. 2013;141:1458–66. doi: 10.1017/S095026881300040X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Vordermeier HM, Chambers MA, Cockle PJ, Whelan AO, Simmons J, Hewinson RG. Correlation of ESAT-6-specific gamma interferon production with pathology in cattle following Mycobacterium bovis BCG vaccination against experimental bovine tuberculosis. Infect Immun. 2002;70:3026–32. doi: 10.1128/IAI.70.6.3026-3032.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Villarreal-Ramos B, McAulay M, Chance V, Martin M, Morgan J, Howard CJ. Investigation of the role of CD8+ T cells in bovine tuberculosis in vivo. Infect Immun. 2003;71:4297–303. doi: 10.1128/IAI.71.8.4297-4303.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Thacker TC, Palmer MV, Waters WR. Associations between cytokine gene expression and pathology in Mycobacterium bovis infected cattle. Vet Immunol Immunopathol. 2007;119:204–13. doi: 10.1016/j.vetimm.2007.05.009. [DOI] [PubMed] [Google Scholar]
  40. Hope JC, Thom ML, McAulay M, Mead E, Vordermeier HM, Clifford D, Hewinson RG, Villarreal-Ramos B. Identification of surrogates and correlates of protection in protective immunity against Mycobacterium bovis infection induced in neonatal calves by vaccination with M. bovis BCG Pasteur and M. bovis BCG Danish. Clin Vaccine Immunol. 2011;18:373–9. doi: 10.1128/CVI.00543-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Lesellier S, Corner L, Costello E. Immunological responses following experimental endobronchial infection of badgers (Meles meles) with different doses of Mycobacterium bovis. Vet Immunol Immunopathol. 2009;127:174–80. doi: 10.1016/j.vetimm.2008.09.012. [DOI] [PubMed] [Google Scholar]
  42. Pai M. Spectrum of latent tuberculosis – existing tests cannot resolve the underlying phenotypes. Nat Rev Microbiol. 2010;8:242. doi: 10.1038/nrmicro2236-c1. [DOI] [PubMed] [Google Scholar]
  43. Del Corral H, París SC, Marín ND. IFNγ response to Mycobacterium tuberculosis, risk of infection and disease in household contacts of tuberculosis patients in Colombia. PLoS ONE. 2009;4:e8257. doi: 10.1371/journal.pone.0008257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Ringshausen FC, Nienhaus A, Schablon A, Schlösser S, Schultze-Werninghaus G, Rohde G. Predictors of persistently positive Mycobacterium-tuberculosis-specific interferon-γ responses in the serial testing of health care workers. BMC Infect Dis. 2010;10:220. doi: 10.1186/1471-2334-10-220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Corbière V, Pottier G, Bonkain F. Risk stratification of latent tuberculosis defined by combined interferon-γ release assays. PLoS ONE. 2012;7:e43285. doi: 10.1371/journal.pone.0043285. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Lesellier S, Palmer S, Gowtage-Sequiera S. Protection of Eurasian badgers (Meles meles) from tuberculosis after intra-muscular vaccination with different doses of BCG. Vaccine. 2011;29:3782–90. doi: 10.1016/j.vaccine.2011.03.028. [DOI] [PubMed] [Google Scholar]
  47. Hill PC, Fox A, Jeffries DJ. Quantitative T cell assay reflects infectious load of Mycobacterium tuberculosis in an endemic case contact model. Clin Infect Dis. 2005;40:273–8. doi: 10.1086/427030. [DOI] [PubMed] [Google Scholar]
  48. Clifton-Hadley RS, Wilesmith JW, Stuart FA. Mycobacterium bovis in the European badger (Meles meles): epidemiological findings in tuberculous badgers from a naturally infected population. Epidemiol Infect. 1993;111:9–19. doi: 10.1017/s0950268800056624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Jenkins HE, Morrison WI, Cox DR. The prevalence, distribution and severity of detectable pathological lesions in badgers naturally infected with Mycobacterium bovis. Epidemiol Infect. 2008;136:1350–61. doi: 10.1017/S0950268807009909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Pritchard DG, Stuart FA, Brewer JI. Experimental infection of badgers (Meles meles) with Mycobacterium bovis. Epidemiol Infect. 1987;98:145–54. doi: 10.1017/s0950268800061859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Lesellier S, Corner L, Costello E. Immunological responses and protective immunity in BCG vaccinated badgers following endobronchial infection with Mycobacterium bovis. Vaccine. 2009;27:402–9. doi: 10.1016/j.vaccine.2008.10.068. [DOI] [PubMed] [Google Scholar]
  52. Garrido J, Sevilla IA, Beltrán-Beck B. Protection against tuberculosis in Eurasian wild boar vaccinated with heat-inactivated Mycobacterium bovis. PLoS ONE. 2011;6:e24905. doi: 10.1371/journal.pone.0024905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Lepper AWD, Pearson CW, Corner LAL. Anergy to tuberculin in beef cattle. Aust Vet J. 1977;53:214–6. doi: 10.1111/j.1751-0813.1977.tb00188.x. [DOI] [PubMed] [Google Scholar]
  54. Houlihan MG, Dixon FW, Page NA. Outbreak of bovine tuberculosis featuring anergy to the skin test, udder lesions and milkborne disease in young calves. Vet Rec. 2008;163:357–61. doi: 10.1136/vr.163.12.357. [DOI] [PubMed] [Google Scholar]
  55. Maher J, Kelly P, Hughes P, Clancy L. Skin anergy and tuberculosis. Respir Med. 1992;86:481–4. doi: 10.1016/s0954-6111(96)80006-0. [DOI] [PubMed] [Google Scholar]
  56. Ewer K, Millington KA, Deeks JJ, Alvarez L, Bryant G, Lalvani A. Dynamic antigen-specific T-cell responses after point-source exposure to Mycobacterium tuberculosis. Am J Respir Crit Care Med. 2006;174:831–9. doi: 10.1164/rccm.200511-1783OC. [DOI] [PubMed] [Google Scholar]
  57. Boumann A, Heinemann MJ, Faas MM. Sex hormones and the immune response in humans. Hum Reprod Update. 2005;11:411–23. doi: 10.1093/humupd/dmi008. [DOI] [PubMed] [Google Scholar]

Articles from Immunology are provided here courtesy of British Society for Immunology

RESOURCES