Low oral BCG doses fail to protect cattle against an experimental challenge with Mycobacterium bovis

Tuberculosis 91 (2011) 400e405

Contents lists available at ScienceDirect

Tuberculosis journal homepage: http://intl.elsevierhealth.com/journals/tube

BOVINE TUBERCULOSIS

Low oral BCG doses fail to protect cattle against an experimental challenge with Mycobacterium bovis Bryce M. Buddle a, *, Frank E. Aldwell b, Geoffrey W. de Lisle c, H. Martin Vordermeier d, R. Glyn Hewinson d, D. Neil Wedlock a a

AgResearch, Hopkirk Research Institute, Private Bag 11008, Palmerston North 4442, New Zealand Centre for Innovation, University of Otago, P.O. Box 56, Dunedin, New Zealand AgResearch, Wallaceville, P.O. Box 40063, Upper Hutt, New Zealand d Veterinary Laboratory Agency, Weybridge, Surrey, UK b c

a r t i c l e i n f o

s u m m a r y

Article history: Received 23 May 2011 Received in revised form 17 June 2011 Accepted 7 July 2011

Studies were undertaken to determine whether a dose of oral Mycobacterium bovis bacillus Calmette eGuérin (BCG) which did not induce skin test reactivity could protect cattle against bovine tuberculosis (TB). Groups of calves (n ¼ 9) were vaccinated by administering 108, 107 or 106 colony forming units (CFU) of BCG orally or 106 CFU subcutaneous (s.c.) BCG. A control group (n ¼ 10) was not vaccinated. All animals were challenged with M. bovis 18 weeks after vaccination and euthanized and necropsied at 16 weeks following challenge. Positive responses in the single cervical tuberculin skin test (severe interpretation) at 15 weeks post-vaccination were only observed in the s.c. BCG and 108 CFU oral BCG groups (four of nine animals/group). Following experimental challenge with M. bovis, both these BCG-vaccinated groups had significant reductions in lesion scores and bacterial counts whereas there was no protection in calves vaccinated with oral doses of 106 or 107 CFU of BCG. In conclusion, low oral doses of BCG did not induce skin test responses, IFN-g responses or protection against TB, however, in the BCG vaccine groups where protection was observed, there was no correlation between protection and skin test responses or IFN-g responses. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Mycobacterium bovis Oral vaccination BCG Tuberculin skin test Bovine tuberculosis

1. Introduction Bovine tuberculosis (TB), caused by Mycobacterium bovis, is a major cause of economic loss in many countries where it is endemic, and has been estimated that 50 million cattle worldwide are infected with M. bovis causing economic losses of about $3 billion per year.1 Test and slaughter of infected cattle can eradicate the disease, although these control measures are less effective where wildlife reservoirs of bovine TB exist or in many developing countries where test and slaughter programs are not economically or socially acceptable.2 Vaccination of cattle against M. bovis has the potential to be an attractive option for control of bovine TB. M. bovis bacillus CalmetteeGuérin (BCG) has been widely used for vaccination against human TB despite its variable efficacy. In cattle, BCG has been shown to induce a significant level of

* Corresponding author. AgResearch, Hopkirk Research Institute, Grasslands Research Centre, Private Bag 11008, Palmerston North 4442, New Zealand. Tel.: þ64 6 351 8679; fax: þ64 6 353 7853. E-mail address: [email protected] (B.M. Buddle). 1472-9792/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.tube.2011.07.001

protection against M. bovis infection when cattle are experimentally challenged3e5 or more recently in the field.6,7 BCG has a number of favourable attributes as a vaccine: it is safe and inexpensive to produce, however, one of the major problems with the use of BCG in cattle is that vaccination induces reactivity to the tuberculin skin test and thus compromises diagnosis of TB in cattle.8,9 When vaccination against bovine TB is not totally effective, it is important in most situations that the use of tests for the diagnosis of the disease are retained and vaccination does not compromise the use of these tests. BCG vaccine for protection against human TB was originally developed in France for delivery via the oral route and in Brazil, oral delivery of this vaccine was retained until the mid-1970s when it was replaced by the intradermal route.10 This change in route of delivery was mainly instigated by pressure from medical practitioners based on the poor responses of oral immunized subjects to purified protein derivative (PPD) skin tests.11 This finding, together with evidence there is dissociation between cells that passively transfer protective immunity and those that transfer delayed-type hypersensitivity to tuberculin12 suggest that it could be possible to identify a TB vaccine that protects, but does not induce skin test

B.M. Buddle et al. / Tuberculosis 91 (2011) 400e405

reactivity. In fact, delivery of BCG to humans by mucosal routes such as the oral route may inhibit delayed-type hypersensitivity (DTH) to purified protein derivative of Mycobacterium tuberculosis.13 Previous studies in cattle have shown that administration of formulated BCG by an oral route can induce significant protection against tuberculosis.9,14 Comparable levels of protection to subcutaneous (s.c.) injection of 106 colony forming units (CFU) of BCG were observed in groups of cattle vaccinated orally with a dose of 109 or 108 CFU of BCG. Importantly, there was a reduction in the numbers of animals responding positively in the caudal fold skin test for the oral vaccinated groups.9 It would be advantageous to have a vaccine which does not induce a DTH response to tuberculin. The objective of the studies was to determine whether low doses of BCG formulated in a lipid matrix and delivered via the oral route could protect cattle against experimental challenge with M. bovis without sensitizing animals to the tuberculin skin test. 2. Materials and methods 2.1. Animals Forty-six Friesian-cross female calves, approximately 6 months old were obtained from a TB-free accredited herd from an area of New Zealand where both farmed and feral animals were free of bovine TB. Prior to the studies, all calves tested negative in the IFNg test using purified protein derivatives (PPDs) prepared from M. bovis (bovine PPD) and Mycobacterium avium (avian PPD). The cattle were grazed on pasture in an isolation unit. 2.2. Bacterial strains M. bovis BCG strain, Danish 1331, was used to vaccinate the calves and M. bovis WAg202, originally isolated from a tuberculous possum (Trichosurus vulpecula) in New Zealand was used as the challenge strain. WAg202 has been used in previous vaccination/ challenge studies in cattle.3,4,9 Both BCG and the M. bovis challenge strains were grown to mid-log phase in Tween albumin broth (Dubos broth base; Difco Laboratories, Detroit, Mich.) supplemented with 0.006% (vol/vol) alkalinized oleic acid, 0.5% (wt/vol) albumin fraction V and 0.25% (wt/vol) glucose. Dilutions were made in Tween albumin broth to obtain the appropriate doses for inoculation. The number of CFU inoculated was determined retrospectively by plating ten-fold dilutions on Middlebrook 7H11 (Difco) supplemented with 0.5% (wt/vol) albumin, 0.2% (wt/vol) glucose and 1% (wt/vol) sodium pyruvate. 2.3. Oral vaccine formulation BCG was formulated for oral vaccination in a lipid matrix, as previously described9; this medium is liquid at 37  C, but solidifies below 30  C. Dilutions of BCG were prepared in Tween albumin broth and mixed in the lipid medium with all media kept at 37  C during the mixing procedure. Each vaccination dose was contained in a 10 ml volume which was kept in a 10 ml syringe and was stored at 4  C prior to use. Viability of BCG was determined by warming the vaccine to 37  C, extracting the BCG from the lipids, culturing on 7H11 agar plates and counting CFU. 2.4. Antigens The preservative-free PPDs prepared from M. avium (avian PPD) and M. bovis (bovine PPD) for the IFN-g testing were supplied by Prionics (Schlieren, Switzerland). Avian and bovine

401

PPD for skin testing was obtained from AsureQuality (Upper Hutt, New Zealand).

2.5. Vaccination The calves were divided into five groups of 9e10 calves using a randomised stratified sampling system so that all groups contained animals with a similar distribution of IFN-g responses to avian PPD in the weeks prior to the start of the study. Nine calves were each dosed orally with a 10 ml of vaccine containing approximately 2  108 CFU of Danish 1331 BCG, and a second group of nine calves were dosed in a similar manner with a 1  107 CFU of BCG dose while a third group of nine calves were dosed orally with 1  106 CFU of BCG. For administration of the oral vaccine, the 10 ml syringe containing the vaccine was warmed to ambient temperature (15e20  C), inserted into the side of the mouth of the calf and squirted over the back of the tongue of the animal. A fourth group of nine calves (positive controls) were each injected subcutaneously in the neck with 1  106 CFU of BCG Danish 1331, obtained from Statens Serum Institute (Copenhagen, Denmark) and reconstituted from lyophilised state. An additional group of 10 calves served as nonvaccinated controls (negative control). 2.6. M. bovis challenge, and necropsy procedure The calves were challenged intratracheally with 5  103 CFU of virulent M. bovis as previously described3 at 18 weeks after vaccination. All cattle were killed 16 weeks after challenge. Procedures for identifying macroscopic tuberculous lesions and processing for histopathology have been previously described.3 Samples from four thoracic lymph nodes (left and right bronchial and anterior and posterior mediastinal) were collected from all of the animals for bacterial culture and histology. Additional samples were collected from any tuberculous lesions observed in lungs, other lymph nodes or organs. For bacterial culture, tissue samples were homogenized in a Tenbroeck grinder (Wheaton, Millville, N.J.), decontaminated in 0.75% cetylpyridinium chloride for 40 min, centrifuged at 3500g for 20 min and processed for isolation of mycobacteria as previously described.3 2.7. IFN-g assay Heparinised blood samples were collected from the calves at regular intervals to analyse cellular immune responses. To follow the kinetics of the IFN-g response, blood cultures were set up within 6 h of collection. Blood samples (1.5 ml) were dispersed into wells of a 24-well plate and bovine or avian PPD (20 mg/ml final concentration; Prionics), ESAT-6/CFP10 (4 mg/ml final concentration; Statens Serum Institute, Copenhagen, Denmark) or phosphate-buffered saline (PBS; Nil) was added. After incubation at 37  C for 20 h, the plasma supernatants were harvested and their IFN-g levels measured using a sandwich ELISA kit (Prionics). Results were represented as absorbance values by expressing the optical density 450 nm (OD) value for bovine PPD minus OD value for PBS (Nil). For assessment of IFN-g responses at 3 weeks after tuberculin skin testing (18 weeks after vaccination), the official New Zealand standard15 was used with blood cultures set up the following day (20e28 h post-collection) and processed by AgResearch, Upper Hutt. A positive result in the standard IFN-g test was defined as bovine PPD OD minus avian PPD OD  0.100 OD, while a positive result in the ESAT-6/CFP10 test was ESAT-6/CFP10 OD minus Nil OD  0.040 OD.

402

B.M. Buddle et al. / Tuberculosis 91 (2011) 400e405

2.8. Tuberculin skin test Comparative cervical tuberculin skin test was undertaken at 15 weeks after vaccination and at 2 weeks prior to slaughter. For this test, cattle were inoculated intradermally with 0.1 ml volumes containing either 0.05 mg avian PPD or 0.1 mg bovine PPD (AsureQuality, Upper Hutt, New Zealand) at separate sites on the right side of the neck. The skin-fold thickness was measured with callipers prior to injection and 72 h after injection for both bovine and avian PPDs. Operational test cut-offs were those used in UK surveillance operations (www.defra.gov.uk/foodfarm/farmanimal/ diseases/atoz/tb/control/tuberculin.htm) and positive responses were defined as an increase in skin thickness for bovine PPD minus the increase for avian PPD of >2 mm for severe interpretation and >4 mm for standard interpretation. Results for a single cervical skin test reported at 15 weeks post-vaccination were determined by using data from the bovine PPD site in the comparative cervical skin test, with positive cut-offs of >2 mm and >4 mm. 2.9. Statistical analysis Statistical analysis was performed using Minitab (edition 15), with statistical significance fixed at P < 0.05. An analysis of IFN-g levels after vaccination and challenge was undertaken on log10transformed data using analysis of variance with pair-wise comparisons. For analysis of tuberculin skin test responses, pathological scores and bacterial counts from the pulmonary lymph nodes, the ManneWhitney U test with pair-wise comparisons was used. The proportions of animals with lesions and those responding in immunological tests were analysed using Fisher’s exact test with pair-wise comparisons. Within group correlations between the lesion scores and positive responses in the single cervical skin test or IFN-g at 15 weeks after vaccination were calculated using Pearson’s correlation. 3. Results 3.1. Skin test responses The proportions of animals in each vaccinated group responding in the cervical skin tests undertaken at 15 weeks after vaccination are shown in Table 1. For the severe interpretation of the single cervical skin test, there were significantly greater proportions of animals responding positively in the s.c. BCG and 108 CFU oral BCG groups compared to that for the non-vaccinated group (P < 0.05). One animal in each of the s.c. BCG and 108 CFU oral BCG groups responded positively in the comparative skin test (severe interpretation). No animals vaccinated with 106 or 107 CFU oral BCG

responded positively in either skin test following vaccination. The median responses for bovine PPD in the s.c. BCG and 108 CFU oral BCG groups were greater than that for non-vaccinated group and the median responses for avian PPD in the BCG s.c. group were also significantly greater than that for the non-vaccinated group (P < 0.05, Figure 1). At 15 weeks post-challenge there were no significant differences between the skin test responses for bovine PPD between the groups (data not shown). 3.2. IFN-g responses after vaccination and challenge The kinetics of T cell responses to M. bovis antigens were determined by measuring the release of IFN-g from whole blood stimulated with bovine PPD with blood cultures set up within 6 h of collection (Figure 2). Vaccination with BCG by either the s.c. route or an oral dose of 108 CFU induced a significant increase in the IFN-g responses to bovine PPD compared to the non-vaccinates, while no increases were detected for the 106 and 107 CFU oral BCG groups. Compared to the non-vaccinated group, the mean IFN-g responses for the s.c. BCG group were significantly greater at all time points between 3 to 18 weeks post-vaccination and for the 108 CFU oral BCG group at 8, 15 and 18 weeks post-vaccination (P < 0.05). The 108 CFU oral BCG group had significantly greater mean IFN-g responses than those for the 106 and 107 CFU oral BCG groups at 8, 11, 15 and 18 weeks post-vaccination (P < 0.05). There was a boost in IFN-g responses at 3 weeks after the skin test, particularly for the s.c. BCG and 108 CFU oral BCG groups. After challenge the only significant differences in IFN-g responses between groups were a lower mean IFN-g response for the non-vaccinated BCG group at 2 weeks post-challenge than those for all the BCG groups and a lower mean response for the s.c. BCG group than that for the 106 CFU oral BCG group at 5 weeks post-challenge (P < 0.05). None of the vaccinated animals were positive in the standard blood IFN-g assay where blood cultures were set up 20e28 h after blood collection (Table 1). A total of four animals from three different BCG-vaccinated groups gave a positive response to ESAT6/CFP10 with the median response for these animals of 0.193 (range 0.068, 0.204) OD units. These four animals had strong IFN-g responses to avian PPD and for three of these animals, the avian PPD response was >2 OD units. 3.3. Pathological and microbiological findings Compared to the non-vaccinated group, the s.c. BCG group had a lower median lymph node lesion score following M. bovis challenge and both the s.c. BCG and 108 CFU oral BCG groups had a lower median number of M. bovis from pulmonary lymph nodes (P < 0.05, Table 2). Compared to the 106 CFU oral BCG group, both

Table 1 Immunological responses following vaccination with BCG. Vaccine group

Proportion of animals positive Comparative cervical skin test*

Non-vaccinated Subcut BCG (106 CFU) Oral BCG (108 CFU) Oral BCG (107 CFU) Oral BCG (106 CFU)

Single cervical skin test*

Standard IFN-g testy

ESAT-6/CFP10 IFN-g testy

B  A, >2 mm

B  A, >4 mm

B, >2 mm

B, >4 mm

B  A, 0.100 OD

Ag  Nil, 0.040 OD

0/10 1/9 1/9 0/9 0/9

0/10 0/9 0/9 0/9 0/9

0/10 4/9* 4/9* 0/9 0/9

0/10 2/9 1/9 0/9 0/9

0/10 0/9 0/9 0/9 0/9

0/10 1/9 2/9 1/9 0/9

B  A response for bovine PPD minus response for avian PPD; B response for bovine PPD, Ag  Nil response for ESAT-6/CFP10 minus response for PBS (Nil). *Significantly greater than that for the non-vaccinated group (P < 0.05). * The cervical skin tests were undertaken at 15 weeks after vaccination of the animals. Positive responses of >2 mm and >4 mm in the comparative cervical test represent severe and standard interpretations, respectively for TB diagnosis as defined by UK surveillance operations.22 y The standard and ESAT-6/CFP10 IFN-g tests were undertaken 18 weeks after vaccination of the animals.

B.M. Buddle et al. / Tuberculosis 91 (2011) 400e405

403

and their combined lesion score. The relationship between the combined lesion score of the animals and whether they produced a positive result in the single cervical skin test (severe interpretation) is shown in Figure 3. In addition there was no correlation within these groups between the IFN-g responses for bovine PPD at 15 weeks after vaccination and their combined lesion scores or with animals which produced a positive result in the single cervical skin test (data not shown). 4. Discussion

Figure 1. Box and whisker plots of skin test responses to avian (,) and bovine PPD (-) 15 weeks after vaccination showing the median, 25 & 75 percentiles and range. The data are expressed as the increases in skin thickness (mm) between the time of inoculation and 72 h later. *Denotes significantly higher than the median for the nonvaccinated group (P < 0.05).

the s.c. BCG and 108 CFU oral BCG groups had a lower median lymph node lesion score and lower median number of M. bovis from pulmonary lymph nodes (P < 0.05). In addition, the s.c. BCG group had a lower median lung lobe lesion score than that for the 106 CFU oral BCG groups and a lower median number of M. bovis from lymph nodes than that for the 107 CFU oral BCG group (P < 0.05). The combined lung lobe and pulmonary lymph node lesions scores are shown in Figure 3. The BCG s.c. group had a significantly lower median score than those for the non-vaccinated, 106 and 107 CFU oral BCG groups, while the median score for the 108 CFU oral BCG group was significantly lower than that for the 106 CFU oral group (P < 0.05). There was no evidence of protection against the M. bovis challenge in the 106 and 107 CFU oral BCG groups assessed on the basis of reduced lesion scores or bacterial counts. Despite the lower lesion scores in the s.c. BCG and 108 CFU oral BCG groups, M. bovis was isolated from the pulmonary lymph nodes of all of the vaccinated animals. Within the vaccinated groups with positive skin test responses, there was no correlation between animals which produced a positive skin test response to bovine PPD in the single cervical skin test

Skin test Challenge

3.50

Optical Density

3.00 2.50 2.00 1.50 1.00 0.50 0.00 0

5

10

15

20

25

30

Weeks after vaccination Figure 2. IFN-g responses after vaccination challenge with M. bovis. Mean IFN-g released from bovine PPD-stimulated whole-blood cultures from cattle vaccinated with (-) S/C BCG, (:) oral BCG (108 CFU), () oral BCG (107 CFU), (C) oral BCG (106 CFU), (>) non-vaccinated. The data is expressed as mean optical density (OB) values for bovine PPD minus the Nil OD value. Error bars indicate standard error of the means.

Findings from the current study have established that vaccination of cattle with doses of 106 and 107 CFU oral BCG induced no positive skin test responses at 15 weeks after vaccination or release of IFN-g from bovine PPD-stimulated blood cultures. In contrast, vaccination with 106 CFU s.c. BCG or 108 CFU oral BCG induced positive responses in four of nine animals in each group in the single cervical skin test and one animal from each group in the comparative test using the severe interpretation of these tests. In addition, animals from these groups produced an IFN-g response to bovine PPD. There were few positives in the comparative cervical skin test or standard IFN-g test as the responses for avian and bovine PPD were similar. Results from an earlier study reported a reduced proportion of cattle reacting positively in the caudal fold test when vaccinated with 108 CFU oral BCG compared to those vaccinated with 106 CFU s.c. BCG.9 However, these results are not directly comparable to those in single cervical skin test from the current study. The caudal fold skin test from the earlier study reported a positive as an increase in skin thickness of 1 mm for bovine PPD (New Zealand standard interpretation15) and was undertaken 8 weeks after vaccination when DTH responses may not have been at their maximal level following oral vaccination with BCG. In previous studies, oral vaccination with 108 or 109 CFU of BCG induced peak bovine PPD-specific IFN-g responses from blood cultures up to 10 weeks after those observed following s.c. injection of BCG.9,14 A number of different factors may influence whether BCGvaccinated animals react in a skin test other than the dose and route of administration of the BCG vaccine. The results from the current study emphasize that the type (comparative or single cervical) and interpretation (standard or severe) of the skin test may influence whether BCG-vaccinated cattle react positively in the skin test and these regulations are set by the specific country undertaking the tests. The duration of time between vaccination and skin testing can also have an effect and Moodie8 reported that from more than 100 BCG-vaccinated cattle, 86% were positive in the comparative skin test at 5 weeks post-vaccination, but none were positive at 76 weeks post-vaccination. The strain of BCG used for vaccination has been shown to have an effect on the size of the skin test response and calves vaccinated with the Pasteur strain of BCG have produced greater skin test responses than those vaccinated with the Danish strain, although the same dose of vaccine and similar cohorts of animals were used.16 Calves vaccinated with the Pasteur strain of BCG as neonates or at 6 months of age have produced similar skin test responses for bovine PPD,16,17 but their responses to avian PPD may differ due to different exposures to environmental mycobacteria. A surprising result was that a total of four animals from three of the BCG-vaccinated groups produced positive responses in the ESAT-6/CFP10 IFN-g test. The mycobacterial proteins, ESAT-6 and CFP10 are expressed from genes in RD1 region of the M. bovis genome which is absent from the genome of BCG.18 These proteins have been incorporated in the whole blood IFN-g test as a means of differentiating between BCG-vaccinated and M. bovis-infected cattle.18,19 While the genes for ESAT-6 and CFP10 are absent from

404

B.M. Buddle et al. / Tuberculosis 91 (2011) 400e405

Table 2 Pathological and bacteriological findings following Mycobacterium bovis challenge. Vaccine group

Non-vaccinated Subcut BCG (106 CFU) Oral BCG (108 CFU) Oral BCG (107 CFU) Oral BCG (106 CFU)

Proportion with: Lung lesions

LN lesions

6/10 2/9 5/9 5/9 6/9

7/10 5/9 6/9 6/9 9/9

Median lung lobe lesion score* 2.5 0 1 3 3

(0, (0, (0, (0, (0,

11){ 5)y 7) 7) 12)

Median LN lesion scorey 3.5 1 1 4 4

(0, (0, (0, (0, (1,

12){ 4)*y 9)y 12) 12)

Mean no. of lesioned LNs/animal 2.1 0.7 1.1 1.9 2.6

    

0.5** 0.2*y 0.4y 0.5 0.4

Median no. of M. bovis/g of pulmonary LN 2.18 0.77 0.81 1.94 2.53

(0.10, (0.15, (0.15, (0.14, (0.19,

4.48){ 3.24)*y 3.32)*y 4.24) 4.19)

*Significantly less than that for the non-vaccinated group (P < 0.05). ySignificantly less than that for the oral BCG 106 CFU group (P < 0.05). * Lung lesion score: 0, no lesions; 1, 1e9 lesions; 2, 10e29 lesions; 3, 30e99 lesions; 4, 100e199 lesions; 5, 200 lesions; scores for the individual lobes are pooled. y Total pulmonary lymph node (LN) lesion score per animal; score for individual node: 0, no lesions; 1, 1e19 small lesions (1e2 mm diameter); 2, 20 small lesions or medium size lesion (3e5 mm diameter); 3, large lesion (5 mm diameter). Scores pooled for the four pulmonary lymph nodes. { Median (range). ** Mean  standard error.

most non-tuberculous mycobacteria, they are present in Mycobacterium kansasii and it has been reported that cattle infected with M. kansasii have responded to ESAT-6/CFP10 in the IFN-g test.20 All of the cattle which responded positively to ESAT-6/CFP10 in the IFN-g test in the current trial had strong responses to avian PPD at the same time, with three of these avian PPD responses >2 OD units, compared to responses of 0.2 OD for ESAT-6/CFP10. This suggested that the response was a consequence of co-infection with environmental mycobacteria, possibly M. kansasii and not as a result of vaccination with BCG. Based on lesion scores and pulmonary lymph node bacterial counts, oral doses of 106 and 107 CFU of BCG failed to protect the cattle against a subsequent experimental M. bovis challenge compared those seen following vaccination with s.c. BCG or an oral dose of 108 CFU of BCG. The median combined lung and lymph node lesion score and bacterial count for the s.c. BCG group was significantly lower than those for the non-vaccinated, 106 and 107 CFU oral BCG groups and the median score and bacterial count for the 108 CFU oral BCG group was significantly lower for that for the 106 CFU oral BCG group. Previous studies with an oral dose of 109 CFU BCG in cattle have demonstrated significant protection against challenge with M. bovis compared to non-vaccinated animals and the level of protection between groups vaccinated with 108 and 109 CFU oral BCG was similar.9,14 Results from the current trial indicated that doses of oral BCG which induced no or

Figure 3. Combined lung lobe and lymph node lesion scores in cattle challenged with M. bovis. Median for group is denoted by a line. Closed symbols represent animals which did not produce a positive result in the single cervical skin test (severe interpretation) at 15 weeks after vaccination and open symbols represent animals which produced a positive result in the single cervical skin test. *Denotes significantly lower than the median of the non-vaccinated group (P < 0.05).

minimal skin test reactivity and peripheral blood IFN-g responses did not protect against a M. bovis challenge. However, in the BCG groups where protection was observed, there was no correlation between lesion scores and the animals which produced a positive skin test response in the single cervical skin test after vaccination or IFN-g responses for bovine PPD measured at the same time. These findings are comparable with other recent data produced in neonatally-vaccinated cattle that also demonstrated a lack of correlation of tuberculin skin test reactivity post-BCG vaccination and protection (A. Whelan and H.M. Vordermeier, unpublished observations). In humans, there was controversy for many years whether postBCG DTH correlated with protection against TB, although the few trials which have investigated this issue did not find a positive correlation. In the British MRC trial, BCG vaccine gave 80% protection against TB in individuals who were negative to tuberculin at the time of vaccination, but there was no evidence of a correlation between this protection and results of tuberculin tests undertaken one year after vaccination.21 Similarly, an analysis of 10 BCG trials found no association between skin test conversion rates and protective efficacy against TB.22 Finally, a study in Malawi found that persistent vaccine-derived hypersensitivity to mycobacterial antigens was not a correlate of vaccine-derived protection against mycobacterial diseases.23 These findings concur with those from the current study in cattle that post-BCG tuberculin reactivity does not correlate with protection. The challenge remains to develop a live mycobacterial vaccine for cattle that does not induce tuberculin reactivity in a population, but still protects against TB. A problem is that a dose of BCG needed to generate sufficient levels of effector cells also induces a subset that enters the circulation capable of migrating into the skin in response to inflammation leading to the DTH reaction. Different T cell populations are associated with DTH to PPD and protection against TB and immunization with killed M. tuberculosis has resulted in DTH reactivity to PPD, without simultaneous expression of protective immunity to TB.24 Although the T cells recruited after challenge with live M. tuberculosis and cells recruited during a DTH reaction have a similar phenotype and cytokine profile, predominantly CD4þ T cells expressing IFN-g and TNF-a, the two cell populations recognise different mycobacterial proteins.25 This suggests that using specific mycobacterial antigens in a TB vaccine or skin test reagent may allow differentiation between the DTH response and protection. In conclusion, administration of oral BCG to cattle at doses which did not induce a skin test response did not induce protection against an experimental challenge with M. bovis. It remains to be determined if this is also correct against a natural M. bovis challenge. It is possible that a further titration of oral BCG may reveal

B.M. Buddle et al. / Tuberculosis 91 (2011) 400e405

a dose which protects against TB and has a reduced ability to induce skin test reactivity. However, use of a very specific defined concentration of BCG for oral application in the field would be problematic due to the difficulty in accurately counting mycobacteria due to clumping of the bacteria and possible loss of viability with storage. A more promising approach may be to use defined diagnostic proteins which are not expressed in BCG in skin tests26 or blood-based tests27 or alternatively, use sub-unit vaccines which do not induce responses to standard skin test reagents.

10.

11. 12.

13.

Acknowledgements We thank Allison McCarthy, Tania Wilson, Keith Hamel, Richard Green and Gary Yates for excellent technical assistance, John Napier and Robert Madsen for care of the animals, Donwhen Luo for statistical analyses.

14.

15.

16.

Funding: This work was supported financially by the Foundation for Research, Science and Technology (New Zealand) and the Department for Environment, Food, and Rural Affairs (UK).

17.

Competing interests: None declared. Ethical approval: Animal procedures were approved by an independent animal ethics committee.

18.

References

19.

1. Steele JH. Regional and country status report. In: Thoen CO, Steele JH, editors. Mycobacterium bovis infection in animals and humans. Ames: Iowa Press; 1995. p. 169e72. 2. Cousins DV. Mycobacterium bovis infection and control in domestic livestock. Res Sci Tech 2001;20:71e85. 3. Buddle BM, de Lisle GW, Pfeffer A, Aldwell FE. Immunological responses and protection against Mycobacterium bovis in calves vaccinated with a low dose of BCG. Vaccine 1995;13:1123e30. 4. Wedlock DN, Denis M, Skinner MA, Koach J, de Lisle GW, Vordermeier HM, et al. Vaccination of cattle with a CpG ODN-formulated mycobacterial protein vaccine and BCG induces levels of protection against bovine tuberculosis superior to vaccination with BCG alone. Infect Immun 2005;73:3540e6. 5. Hope JC, Thom ML, Villarreal-Ramos B, Vordermeier HM, Hewinson RG, Howard CJ. Vaccination of neonatal calves with Mycobacterium bovis BCG induces protection against intranasal challenge with virulent M. bovis. Clin Exp Immunol 2005;139:48e56. 6. Ameni G, Vordermeier M, Aseffa A, Young DB, Hewinson RG. Field evaluation of the efficacy of Mycobacterium bovis Bacillus CalmetteeGuérin against bovine tuberculosis in neonatal calves in Ethiopia. Clin Vaccine Immunol 2010;17:1533e8. 7. Lopez-Valencia G, Renteria-Evangelista T, de Jesús Williams J, Licea-Navarro A, De la Mora-Valle A, Medina-Basulto G. Field evaluation of the protective efficacy of Mycobacterium bovis BCG vaccine against bovine tuberculosis. Res Vet Sci 2010;88:44e9. 8. Moodie PA. Tuberculin reactions in BCG. Br Vet J 1977;133:642e5. 9. Buddle BM, Aldwell FE, Skinner MA, de Lisle GW, Denis M, Vordermeier HM, et al. Effect of oral vaccination of cattle with lipid-formulated BCG on

20.

21.

22. 23. 24.

25.

26.

27.

405

immune responses and protection against bovine tuberculosis. Vaccine 2005;23:3581e9. Benévolo-de-Andrade TC, Monteiro-Maia R, Cosgrove C, Castello-Branco LR. BCG Moreau Rio de Janeiro: an oral vaccine against tuberculosis e review. Mem Inst Oswaldo Cruz 2005;100:459e65. Gheorghiu M, Lagranderie M, Balazuc AM. Stabilisation of BCG vaccines. Dev Biol Stand 1996;87:251e61. Orme IM, Collins FM. Adoptive protection of the Mycobacterium tuberculosisinfected lung. Dissociation between cells that passively transfer protective immunity and those that transfer delayed-type hypersensitivity to tuberculin. Cell Immunol 1984;84:113e20. Holt DF, Brown RM, Belshe RB. Mucosal bacilli CalmetteeGuérin vaccination of humans inhibits delayed-type hypersensitivity to purified protein derivative but induces specific interferon-g responses. Clin Infect Dis 2000;30(Suppl. 3):S217e22. Buddle BM, Denis M, Aldwell FE, Vordermeier HM, Hewinson RG, Wedlock DN. Vaccination of cattle with Mycobacterium bovis BCG by a combination of systemic and oral routes. Tuberculosis 2008;88:595e600. Anonymous. National bovine tuberculosis pest management strategy e national operational plan: 1 July 2005 to 30 June 2013. Wellington, NZ: Animal Health Board; 2005. Wedlock DN, Denis M, Vordermeier HM, Hewinson RG, Buddle BM. Vaccination of cattle with Danish and Pasteur strains of Mycobacterium bovis BCG induce different levels of IFN-g post-vaccination, but induce similar levels of protection against bovine tuberculosis. Vet Immunol Immunopathol 2007;118:50e8. Buddle BM, Wedlock DN, Parlane NA, Corner LAL, de Lisle GW, Skinner MA. Revaccination of neonatal calves with Mycobacterium bovis BCG reduced the level of protection against tuberculosis induced by a single vaccination. Infect Immun 2003;71:6411e9. Vordermeier HM, Whelan A, Cockle PJ, Farrant L, Palmer N, Hewinson RG. Use of synthetic peptides derived from the antigens ESAT-6 and CFP-10 for differential diagnosis of bovine tuberculosis in cattle. Clin Diagn Lab Immunol 2001;8:571e8. Buddle BM, Parlane NA, Keen DL, Aldwell FE, Pollock JM, Lightbody K, et al. Differentiation between Mycobacterium bovis BCG-vaccinated and M. bovisinfected cattle by using recombinant mycobacterial antigens. Clin Diagn Lab Immunol 1999;6:1e5. Waters WR, Palmer MV, Thacker TC, Payeur JB, Harris NB, Minion FC, et al. Immune responses to defined antigens of Mycobacterium bovis in cattle experimentally infected with Mycobacterium kansasii. Clin Vaccine Immunol 2006;13:611e9. Hart PDA, Sutherland I, Thomas J. The immunity conferred by effective BCG and vole bacillus vaccines, in relation to individual variations in tuberculin sensitivity and to technical variations in the vaccines. Tubercle 1967;48:201e10. Comstock GW. Identification of an effective vaccine against tuberculosis. Am Rev Respir Dis 1988;138:479e80. Fine PE, Sterne JA, Ponnighaus JM, Rees RJ. Delayed-type hypersensitivity, mycobacterial vaccines and protective immunity. Lancet 1994;344:1245e9. Orme IM. Induction of non-specific acquired resistance and delayed-type hypersensitivity, but non specific acquired resistance, in mice inoculated with killed mycobacterial vaccines. Infect Immun 1988;56:3310e2. Pais TF, Silva RA, Smedgaard B, Appelberg R, Andersen P. Analysis of T cells recruited during delayed-type hypersensitivity to purified protein derivative (PPD) versus challenge with tuberculosis infection. Immunology 1998;95:69e75. Whelan AO, Clifford D, Upadhyay B, Breadon EL, McNair J, Hewinson RG, et al. Development of a skin-test for bovine tuberculosis for differentiating infected from vaccinated animals. J Clin Microbiol 2010;48:3176e81. Sidders B, Pirson C, Hogarth PJ, Hewinson RG, Stoker NG, Vordermeier HM, et al. Screening of highly expressed mycobacterial genes identifies Rv3615c as a useful differential antigen for the Mycobacterium tuberculosis complex. Infect Immun 2008;76:3932e9.