Minimum Infective Dose of Mycobacterium bovis in Cattle
http://www.100md.com
感染与免疫杂志 2005年第10期
TB Research Group, Veterinary Laboratories Agency, Weybridge, United Kingdom KT15 3NB
ABSTRACT
The aim of this work was to determine the minimum infective dose of Mycobacterium bovis necessary to stimulate specific immune responses and generate pathology in cattle. Four groups of calves (20 animals) were infected by the intratracheal route with 1,000, 100, 10, or 1 CFU of M. bovis. Specific immune responses (gamma interferon [IFN-] and interleukin-4 [IL-4] responses) to mycobacterial antigens were monitored throughout the study, and the responses to the tuberculin skin test were assessed at two times. Rigorous post mortem examinations were performed to determine the presence of pathology, and samples were taken for microbiological and histopathological confirmation of M. bovis infection. One-half of the animals infected with 1 CFU of M. bovis developed pulmonary pathology typical of bovine tuberculosis. No differences in the severity of pathology were observed for the different M. bovis doses. All animals that developed pathology were skin test positive and produced specific IFN- and IL-4 responses. No differences in the sizes of the skin test reactions, the times taken to achieve a positive IFN- result, or the levels of the IFN- and IL-4 responses were observed for the different M. bovis doses, suggesting that diagnostic assays (tuberculin skin test and IFN- test) can detect cattle soon after M. bovis infection regardless of the dose. This information should be useful in modeling the dynamics of bovine tuberculosis in cattle and in assessing the risk of transmission.
INTRODUCTION
Bovine tuberculosis caused by Mycobacterium bovis is a worldwide animal health problem and remains a major threat to public health in the countries in which people live in close contact with their cattle and milk is not pasteurized (7, 11). Early experimental studies (5, 15) suggested that the principal route of M. bovis transmission is most likely to be aerogenous (rather than oral). These experiments indicated that lower doses of M. bovis could be used to infect cattle intranasally compared with the larger doses required when an oral route of delivery was used. Furthermore, tuberculous lesions in the gut, a relatively rare event in naturally infected animals, were common only in cattle that were experimentally infected via the oral route (10, 15). Observations from more recent experimental infections confirmed that infection of cattle via the intranasal route results in pathology that is largely confined to the upper respiratory tract, while intratracheal infection tends to cause lesions in the lower respiratory tract. Naturally infected field reactor cattle most commonly have lesions in the lower respiratory tract and pulmonary lymph nodes, while involvement of the upper respiratory tract occurs more rarely (2, 6, 17, 19, 30).
Bovine tuberculosis may spread by cattle-to-cattle transmission and also through the involvement of wildlife reservoirs (16). However, experimental intranasal and intratracheal M. bovis infections in cattle have shown that bacterial shedding (the presence of viable M. bovis in the nasal mucus) is, at best, transient and involves extremely low numbers (approximately 70 CFU) of bacilli (14). Such a low dose of M. bovis has not previously been considered relevant in the bovine model of tuberculosis, even though a low infection dose for Mycobacterium tuberculosis in humans has been accepted for many decades (21, 24, 25, 27). These historical studies showed that the numbers of primary calcified M. tuberculosis lesions in otherwise healthy people were low (between one and three lesions per individual) and also established that the infectious dose of M. tuberculosis could be as low as 1 to 10 bacilli.
Previous experimental infections of cattle with M. bovis have suggested that the infective dose can have a profound influence on the severity of the disease that follows. For example, in the intranasal model, 5 x 105 to 106 CFU resulted in multiple respiratory lesions, while 5 x 102 to 104 CFU resulted in a more variable pathology (some animals had multiple lesions, and some animals had no lesions) and 102 CFU resulted in no visible lesions at all. Although the latter group remained skin test negative, M. bovis was isolated from the nasal mucus of one animal 100 days after infection (17). We and other workers have shown that in the intratracheal model low doses of M. bovis (800 to 6 x 103 CFU) can result in animals that are skin test negative, have no visible lesions at post mortem, and are M. bovis culture negative (2, 4, 20, 23). Interestingly, Rhodes et al. (23) also measured specific cytokine responses and showed that the skin test-negative animals in their study that were gamma interferon (IFN-) and interleukin-2 (IL-2) negative gave positive specific IL-4 responses, indicating that these animals had been successfully exposed to the M. bovis inoculum but had dealt with the infection differently.
To date, the doses of M. bovis used in the cattle model have been relatively high, and natural infections with such doses are unlikely. The current study was therefore undertaken to determine the lowest infective dose of M. bovis in cattle that resulted in pathology. To this end, we infected calves with logarithmically decreasing doses from 1,000 CFU to 1 CFU per animal. Immunological monitoring (specific IFN- and IL-4 responses) throughout the study and tuberculin skin testing at two separate time points were carried out in order to obtain information on the effect of the M. bovis dose on these parameters.
MATERIALS AND METHODS
Culture medium and antigens. The culture medium used for the whole blood cultures and peripheral blood mononuclear cell stimulation was RPMI 1640 with Glutamax (Gibco) supplemented with 5% CSPR3 (control serum protein replacement; Gibco), nonessential amino acids (Gibco), 100 U/ml penicillin, 100 μg/ml streptomycin, and 5 x 10–5 M 2-mercaptoethanol (Gibco); 10% fetal calf serum (Gibco) was used instead of CSPR3 in the B-cell cultures. Tuberculin preparations purified protein derivative A and B (PPD-B and PPD-A) were produced at the Veterinary Laboratories Agency (Weybridge, United Kingdom). Recombinant ESAT6 and CFP10 proteins (29) were obtained from M. Singh (GBF, Braunschweig, Germany).
Bacterial strain. The infection inoculum was prepared from a mid-log-phase frozen seed stock of M. bovis strain 2122/97, whose concentration had been determined previously. The seed stock was thawed and then diluted to the required concentration in Middlebrook 7H9 medium containing 0.05% Tween 80. The infective dose was confirmed retrospectively by plating the inoculum in triplicate on a modified formulation of Middlebrook 7H11 agar and enumerating colonies following 4 weeks of incubation at 37°C. To determine the extent of clumping, the inoculum was sonicated for 10 s at 21% amplitude using a Sonics Vibra Cell VCX500 sonicator fitted with a 4-mm-diameter tip. Bacterial enumeration of the sonicated sample was performed on modified Middlebrook 7H11 agar. One CFU was confirmed to contain between 6 and 10 viable bacilli.
Cattle. Twenty Friesian Holstein heifers and bullocks that were a minimum of 6 months old from a tuberculosis-free herd were randomly divided into four groups. The disease-free status of individual animals was further confirmed by an IFN- test prior to recruitment. Calves were inoculated intratracheally with M. bovis field strain GB (AF 2122/97) by using the protocol of Buddle et al. (2). Briefly, an 80-cm endotracheal tube containing a fine cannula was inserted per os into the trachea of an anesthetized animal. The appropriate dose of inoculum in 1.5 ml prepared as described above was injected through the cannula and flushed out with 2 ml of saline. This route of infection was selected as the route most similar to natural infection, since previous experiments indicated that intratracheal infection replicates more closely the lesion distribution seen in the majority of naturally infected cattle. One group of six calves (calves 2805, 2858, 2865, 2871, 2877, and 2878) received 1 CFU, and one group of six calves (calves 2806, 2861, 2863, 2866, 2869, and 2870) received 10 CFU. In addition, one group of four animals (calves 2802, 2859, 2927, and 2928) received 100 CFU, and one group of four animals received 1,000 CFU (calves 2923, 2924, 2925, and 2926). Each of the four groups of animals was housed separately in a high-security self-contained isolation unit under negative pressure with 14 air changes per h for the duration of the experiment. (In previous experiments conducted in the facility, in which uninfected calves were kept in contact with animals infected with a much higher dose of M. bovis than was used in this study, no natural cattle-to cattle-transmission occurred; it is therefore unlikely that any transmission occurs between animals.) Blood samples were taken by venepuncture of the jugular vein for immunological studies before infection and then weekly until 6 weeks postinfection (p.i.). Subsequently, samples were taken every 2 weeks until the end of the experiment.
Skin test. The single comparative intradermal tuberculin skin test was performed as described previously (1) in accordance with the standard protocol used in the United Kingdom. All the animals were skin tested at week 12 postinfection and again shortly before the end of the experiment (weeks 23 to 25 postinfection). An increase in the skin induration response to PPD-B that was greater than the response to PPD-A by at least 10 mm was considered a positive response.
Post mortem. Calves were euthanized by intravenous injection of sodium pentobarbitone, and a detailed post mortem examination was carried out. Lymph nodes were removed aseptically (right and left submandibular, right and left medial retropharyngeal, right and left lateral retropharyngeal, right and left parotideal, cranial mediastinal, caudal mediastinal, right and left bronchial, cranial tracheobronchial, right and left tonsil). Lymph nodes were serially sliced (2-mm slices) and examined for the presence of lesions. Samples of lesion material and also apparently unaffected tissue were used for histopathological examination and for culture of M. bovis. Other lymph nodes were macroscopically examined in situ. The lungs were serially sliced (5-mm slices) following the bronchial tree, and the slices were palpated and inspected. Small pieces of lung tissue were removed for culture and histopathology. The nasal passages were also opened and inspected. Individual tissues were assigned a pathology score depending upon the number, size, and character of the lesions observed in accordance with the standard methodology used in this laboratory (28).
Whole-blood culture and IFN- enzyme-linked immunosorbent assay. Duplicate cultures of peripheral whole blood were diluted 1:1 with culture medium (100 μl per well in 96-well flat-bottom microtiter plates) and cultured in the presence or absence of antigen (final concentration of PPD-B and PPD-A, 10 μg/ml; final concentration of ESAT6 and CFP10, 5 μg/ml; or mitogen control, 2 μg/ml [staphylococcus enterotoxin B]) for 24 h (9). Supernatants were then assessed for gamma interferon content using a commercially available enzyme-linked immunosorbent assay kit (BOVIGAM; Biocore, United States) according to the manufacturer's instructions. The results were expressed as the increase in the mean optical density at 450 nm (OD450) in the presence of antigen compared to the OD450 in the medium or unstimulated supernatants. A change in the OD450 of more than 0.1 was considered a positive response.
IL-4 bioassay. The IL-4 bioassay was carried out as previously described (12, 23). Briefly, peripheral blood mononuclear cells from naive skin test-negative cattle kept at Veterinary Laboratories Agency were positively sorted using monoclonal antibody IL-A58 (obtained courtesy of the Institute for Animal Health, Compton, Berkshire, United Kingdom), goat anti-mouse immunoglobulin G-coated microbeads, and the MACS column separation system (Miltenyi Biotech, Surrey, United Kingdom) to obtain a highly enriched B-cell population. Whole blood was incubated 1:1 with antigen, mitogen, and medium or a control as described above for the IFN- analysis, but the incubation time was 6 days. Culture supernatants were then harvested, and 50 μl undiluted supernatant was added to 100 μl of a B-cell suspension at a concentration of 106 cells/ml in 96-well round-bottom microtiter plates. The plates were incubated for 24 h before they were pulsed with tritiated thymidine (Amersham UK) and then harvested after a further 24 h. The results are expressed as stimulation indices (SI) (i.e., mean counts per minute for B-cell proliferation in the presence of antigen-stimulated supernatant divided by the mean counts per minute for B-cell proliferation in the presence of medium or control unstimulated supernatant). A stimulation index greater than 3.0 was considered a positive response. Previous work (22) using a monoclonal antibody specific for bovine IL-4 that had been shown to inhibit specifically the biological activity of recombinant bovine IL-4 (CC303; 10 μg/ml) showed that this antibody was able to block B-cell stimulatory activity in the assay.
RESULTS
Skin test and pathological findings for M. bovis-infected cattle. Groups of calves were infected with 1, 10, 100, and 1,000 CFU M. bovis at the beginning of the experiment.
The skin test results at week 12 showed that 14 of the 20 animals were skin test positive (Table 1). The six remaining negative animals were distributed among the four groups as follows: three animals that received 1 CFU (calves 2085, 2871, and 2877,) one animal that received 10 CFU (calf 2866), one animal that received 100 CFU (calf 2859), and one animal that received 1,000 CFU (calf 2924). These results were confirmed at the second skin test at weeks 23 to 25 p.i. (data not shown). There was no significant difference as determined by a chi-square test (P = 0.6267) in the distribution of skin test-positive and -negative animals between the different groups. All skin test-positive calves were found to have visible lesions in the respiratory lymph nodes, and most also contained lesions in the upper lung lobes (all except calf 2861). Mycobacterial culture on modified Middlebrook 7H11 agar and acid-fast staining confirmed the presence of M. bovis in tissues with lesions. In accordance with our previous findings, no lesions were detected in the head nodes, and no mycobacteria were cultured from these areas. The degrees of pathology as determined by the pathology scores (Table 2) of all the animals in the groups were comparable, and there were no significant differences as determined by the Kruskal-Wallis test (P = 0.3896) between groups that received different infective doses. All skin test-negative animals presented without gross pathology (no visible lesions). M. bovis could not be cultured from tissue samples from these animals, and no histopathological signs of tuberculosis were detected (data not shown). Thus, we were able to infect cattle with 1 CFU of M. bovis.
Immunological findings for skin test-positive cattle. All skin test-positive calves developed strong positive specific IFN- responses. The kinetics of the IFN- response of the three skin test-positive calves infected with 1 CFU are shown in Fig. 1. In all 14 skin test-positive calves a positive IFN- response was observed by 3 to 5 weeks p.i., after which the response fluctuated (but remained positive) throughout the course of infection. The time to positivity (number of weeks postinfection until a positive response was observed) (Table 3) and the intensity of the response (OD450) (data not shown) did not vary significantly between groups that received different infective doses. The responses to the recombinant antigens ESAT6 and CFP10 were variable. Of the 14 calves that produced IFN-, 2 (calves 2802 and 2806) responded to CFP10 but not to ESAT6, while 5 (calves 2863, 2865, 2869, 2925, and 2928) responded to ESAT6 but not to CFP10. The remaining seven animals (calves 2858, 2861, 2876, 2878, 2923, 2926, and 2927) responded equally to both proteins (data not shown). There was no correlation between the infective dose of M. bovis and the responses to these two proteins (data not shown).
Six-day whole-blood supernatants from 13 of 14 skin test-positive calves were shown to stimulate the proliferation of B cells in the IL-4 bioassay. Positive IL-4 responses were transient and were observed soon after infection, between 5 and 7 weeks p.i. for most animals. The IL-4 responses of the three skin test-positive calves that received 1 CFU M. bovis are shown in Fig. 2. There was no apparent difference in the strength of the IL-4 response (SI) between groups that received different M. bovis doses. A summary of the IFN- and IL-4 data for all skin test-positive animals is shown in Table 3.
The presence of another anti-inflammatory cytokine, IL-10, was also investigated in both skin test-positive and -negative animals in our study, since specific IL-10 responses have recently been reported to occur during experimental bovine paratuberculosis infection (S. Marche, K. Walravens, C. J. Howard, J. C. Hope, V. Rosseels, K. Huygen, and J. Godfroid, 7th Int. Vet. Immunol. Symp., Quebec, Canada, 2004, abstr. 370). However no significant IL-10 responses were detected in any of our cattle (data not shown).
DISCUSSION
This study clearly demonstrated that 1 CFU of M. bovis is able to cause bovine tuberculosis. Interestingly, this finding is in accord with a mathematical analysis of experimental bovine tuberculosis performed by Neill et al. (18), who postulated that a single M. bovis organism should be capable of causing disease in cattle. In our experiment 1 CFU was found to contain between 6 and 10 viable bacilli. Infection with just 1 CFU resulted in pathology whose severity was equivalent to that seen in animals which received far higher doses (up to 1,000 CFU). The presence of pathology correlated with a positive tuberculin skin test and a strong sustained specific IFN- response, which supports previous findings by us and other workers that antigen (ESAT6)-specific IFN- responses correlate with the severity of pathology in cattle (13, 28). No M. bovis dose-related effect was observed in the pathology score, the strength of the tuberculin skin test, the strength of the IFN- response, or the time taken to obtain a positive IFN- response. Our data therefore do not support a dose-related pathological outcome in the animals in which disease becomes established, but they seem to suggest that comparable levels of pathology may develop regardless of the number of CFU inoculated.
Whether the skin test-negative animals in our study, if left for a much longer time, would have converted to a positive phenotype is not known. Previous data from experimental intratracheal infections of cattle in which M. bovis doses of 800 CFU and 1,000 CFU were used (2, 4) showed that 40% of the cattle remained negative for the tuberculin skin test, specific IFN- test, and M. bovis culture. However, infection with a slightly higher dose, 2,000 CFU, gave more variable results; that is, there were fewer negative animals, and a proportion of these animals showed late-onset IFN- responses (3, 20). This again raises the question of latency in cattle and its potential importance in disease transmission. It has been suggested that further immunological analyses of low-dose infections could be employed to investigate latency in cattle (16); however, such studies would by necessity be long term, and their outcomes would be unreliable.
Our previous study (23) suggested that skin test-, IFN--, and IL-2-negative M. bovis-infected cattle could produce specific IL-4 responses. This provided some hope that IL-4 might be a detectable marker in this otherwise negative phenotype. However, these results were not repeated in the current study. Significant positive IL-4 responses were found in only 13 of 14 of the skin test-positive animals. These responses were transient, occurring in most animals between weeks 5 and 7 p.i., although individual spikes of IL-4 activity were also seen at other times in some animals. We also measured IL-4 activity in peripheral blood mononuclear cell supernatants of skin test-negative cattle, but the results were negative (data not shown). However, the lowest dose used in the previous study was 6 x 103 CFU, which is six times greater than the highest dose (1,000 CFU) used in the current experiments. Whether this difference in M. bovis dose is relevant for IL-4 responses is not known.
The IL-4 bioassay is currently the only reliable assay for measuring bovine IL-4 activity. The fact that clinical studies identify a role for IL-4 in the pathological process (8, 26) clearly indicates the need for more specific assays to assess the role of this cytokine and its splice variants in bovine tuberculosis.
In summary, we found that 1 CFU of M. bovis is sufficient to cause established tuberculous pathology in cattle. This pathology is identical to that resulting from significantly higher experimental doses (up to 1,000 CFU in this study) and reflects the pathology seen in naturally infected field reactor cattle. Cattle infected with 1 CFU that developed pathology exhibited strong positive responses to the diagnostic tuberculin skin test. Furthermore, the infectious dose of M. bovis had no bearing on the time taken to obtain a positive IFN- response in the animals that went on to develop pathology.
Our data are in accord with very low numbers of bacilli transmitted aerogenously between cattle, potentially by nasal shedding. Comfortingly, the animals that do go on to develop pathology and therefore become a likely source of contamination within a herd can be detected at an early stage with the IFN- test and also provide a positive tuberculin skin test response.
ACKNOWLEDGMENTS
We thank Jayne Hope and Elinor Mead (Institute for Animal Health, Compton, Berkshire, United Kingdom) for assessing our supernatants for IL-10.
This work was funded by the Department of Environment and Rural Affairs (DEFRA) U.K.
REFERENCES
1. Anonymous. 2002. European Commission Regulation (EC) NO 1226/2002 of 8 July 2002 amending annex B to Council Directive 64/432/EEC. Off. J. Eur. Union (English) L 179:13.
2. Buddle, B. M., G. W. DeLisle, A. Pfeffer, and F. E. Aldwell. 1995. Immunological responses and protection against Mycobacterium bovis in calves vaccinated with a low dose of BCG. Vaccine 13:1123-1130.
3. Buddle, B. M., D. Keen, A. Thomson, G. Jowett, A. R. McCarthy, J. Heslop, G. W. DeLisle, J. L. Stanford, and F. E. Aldwell. 1995. Protection of cattle from bovine tuberculosis by vaccination with BCG by the respiratory or subcutaneous route, but not by vaccination with killed Mycobacterium vaccae. Res. Vet. Sci. 59:10-16.
4. Buddle, B. M., N. A. Parlane, D. L. Keen, F. E. Aldwell, J. M. Pollock, K. Lightbody, and P. Andersen. 1999. Differentiation between Mycobacterium bovis BCG-vaccinated and M. bovis-infected cattle by using recombinant mycobacterial antigens. Clin. Diag. Lab. Immunol. 6:1-5.
5. Chausse, P. 1913. Des methodes a employer pour realiser la tuberculose experimentale par inhalation. Bull. Soc. Med. Vet. 31:267-274.
6. Corner, L. A. 1994. Post mortem diagnosis of Mycobacterium bovis infection in cattle. Vet. Microbiol. 40:53-63.
7. Daborn, C. J., J. M. Grange, and R. R. Kazwala. 1996. The bovine tuberculosis cycle: an African perspective. Soc. Appl. Bacteriol. Symp. Ser. 25:27S-32S.
8. Demissie, A., M. Abebe, A. Aseffa, G. Rook, H. Fletcher, A. Zumla, K. Weldingh, I. Brock, P. Andersen, T. M. Doherty, and the VACSEL Study Group. 2004. Healthy individuals that control a latent infection with Mycobacterium tuberculosis express high levels of Th1 cytokines and the IL-4 antagonist IL-42. J. Immunol. 172:6938-6943.
9. Emery, D. L., F. H. Duffy, and P. R. Wood. 1988. An analysis of cellular proliferation and synthesis of lymphokines and specific antibody in vitro by leucocytes from immunized cattle. Vet. Immunol. Immunopathol. 18:67-80.
10. Francis J. 1947. Bovine tuberculosis, including a contrast with human tuberculosis. Staples Press Ltd., London, United Kingdom.
11. Hardie, R. M., and J. M. Watson. 1992. Mycobacterium bovis in England and Wales: past, present and future. Epidemiol. Infect. 109:23-33.
12. Kuhnle, G., R. A. Collins, J. E. Scott, and G. M. Keil. 1996. Bovine interleukin-2 and interleukin-4 expressed in recombinant bovine herpesvirus-1 are biologically active secreted glycoproteins. J. Gen. Virol. 77:2231-2240.
13. Lyashchenko, K., A. O. Whelan, R. Greenwald, J. M. Pollock, P. Andersen, R. G. Hewinson, and H. M. Vordermeier. 2004. Association of tuberculin-boosted antibody response with pathology and cell-mediated immunity in cattle vaccinated with Mycobacterium bovis BCG and infected with M. bovis. Infect. Immun. 72:2462-2467.
14. McCorry, T. A., A. O. Whelan, M. D. Welsh, J. McNair, E. Walton, D. G. Bryson, R. G. Hewinson, and H. M. Vordermeier. Shedding of Mycobacterium bovis in the nasal mucus of cattle experimentally infected with tuberculosis by the intranasal and intratracheal routes. Vet. Rec., in press.
15. McFadyean, J. 1910. What is the common method of infection in tuberculosis J. Comp. Pathol. 23:239-250 and 289-298.
16. Morrison, W. I., F. J. Bourne, D. R. Cox, C. A. Donnelly, G. Gettinby, J. P. McInerney, and R. Woodroffe. 2000. Pathogenesis and diagnosis of infections with Mycobacterium bovis in cattle. Vet. Rec. 26:236-242.
17. Neill, S. D., J. Hanna, J. J. O'Brien, and R. M. McCracken. 1988. Excretion of Mycobacterium bovis by experimentally infected cattle. Vet. Rec. 123:340-343.
18. Neill, S. D., J. J. O'Brien, and J. Hanna. 1991. A mathematical model for Mycobacterium bovis excretion from tuberculous cattle. Vet. Microbiol. 28:103-109.
19. Neill, S. D., J. M. Pollock, D. B. Bryson, and J. Hanna. 1994. Pathogenesis of Mycobacterium bovis infection in cattle. Vet. Microbiol. 40:41-52.
20. Ng, K. H., F. E. Aldwell, D. N. Wedlock, J. D. Watson, and B. M. Buddle. 1997. Antigen-induced interferon-gamma and interleukin-2 responses of cattle inoculated with Mycobacterium bovis. Vet. Immunol. Immunopathol. 57:59-68.
21. O'Grady, F., and R. L. Riley. 1963. Experimental airborne tuberculosis. Adv. Tuberc. Res. 12:150-190.
22. Rhodes, S. G., J. M. Cocksedge, R. A. Collins, and W. I. Morrison. 1999. Differential cytokine responses of CD4+ and CD8+ T cells in response to bovine viral diarrhoea virus in cattle. J. Gen. Virol. 80:1673-1679.
23. Rhodes, S. G., N. Palmer, S. P. Graham, A. E. Bianco, R. G. Hewinson, and H. M. Vordermeier. 2000. Distinct response kinetics of gamma interferon and interleukin-4 in bovine tuberculosis. Infect. Immun. 68:5393-5400.
24. Rich, A. R. 1946. Pathogenesis of tuberculosis. Thomas, Springfield, Ill.
25. Riley, R. L., C. C. Mills, W. Nyka, N. Weinstock, P. B. Storey, L. U. Sultan, M. C. Riley, and W. F. Wells. 1995. Aerial dissemination of pulmonary tuberculosis. A two year study of contagion in a tuberculosis ward, 1959. Am. J. Epidemiol. 142:3-14.
26. Rook, G. A. W., R. Hernandez-Pando, K. Dheda, and G. T. Seah. 2004. IL-4 in tuberculosis: implications for vaccine design. Trends Immunol. 25:483-488.
27. Sonin, L. S. 1951. The role of particle size in experimental airborne infection. Am. J. Hyg. 53:337-354.
28. Vordermeier, H. M., M. A. Chambers, P. J. Cockle, A. O. Whelan, J. Simmons, and R. G. Hewinson. 2002. Correlation of ESAT-6-specific gamma interferon production with pathology in cattle following Mycobacterium bovis BCG vaccination against experimental bovine tuberculosis. Infect. Immun. 70:3026-3032.
29. Vordermeier, H. M., A. Whelan, P. J. Cockle, L. Farrant, N. Palmer, and R. G. Hewinson. 2001. The use of synthetic peptides derived from the antigens ESAT-6 and CFP-10 for differential diagnosis of bovine tuberculosis in cattle. Clin. Diag. Lab. Immunol. 8:571-578.
30. Whipple, D. L., C. A. Bolin, and J. M. Miller. 1996. Distribution of lesions in cattle infected with Mycobacterium bovis. J. Vet. Diagn. Investig. 8:351-354.(Gillian S. Dean, Shelley )
ABSTRACT
The aim of this work was to determine the minimum infective dose of Mycobacterium bovis necessary to stimulate specific immune responses and generate pathology in cattle. Four groups of calves (20 animals) were infected by the intratracheal route with 1,000, 100, 10, or 1 CFU of M. bovis. Specific immune responses (gamma interferon [IFN-] and interleukin-4 [IL-4] responses) to mycobacterial antigens were monitored throughout the study, and the responses to the tuberculin skin test were assessed at two times. Rigorous post mortem examinations were performed to determine the presence of pathology, and samples were taken for microbiological and histopathological confirmation of M. bovis infection. One-half of the animals infected with 1 CFU of M. bovis developed pulmonary pathology typical of bovine tuberculosis. No differences in the severity of pathology were observed for the different M. bovis doses. All animals that developed pathology were skin test positive and produced specific IFN- and IL-4 responses. No differences in the sizes of the skin test reactions, the times taken to achieve a positive IFN- result, or the levels of the IFN- and IL-4 responses were observed for the different M. bovis doses, suggesting that diagnostic assays (tuberculin skin test and IFN- test) can detect cattle soon after M. bovis infection regardless of the dose. This information should be useful in modeling the dynamics of bovine tuberculosis in cattle and in assessing the risk of transmission.
INTRODUCTION
Bovine tuberculosis caused by Mycobacterium bovis is a worldwide animal health problem and remains a major threat to public health in the countries in which people live in close contact with their cattle and milk is not pasteurized (7, 11). Early experimental studies (5, 15) suggested that the principal route of M. bovis transmission is most likely to be aerogenous (rather than oral). These experiments indicated that lower doses of M. bovis could be used to infect cattle intranasally compared with the larger doses required when an oral route of delivery was used. Furthermore, tuberculous lesions in the gut, a relatively rare event in naturally infected animals, were common only in cattle that were experimentally infected via the oral route (10, 15). Observations from more recent experimental infections confirmed that infection of cattle via the intranasal route results in pathology that is largely confined to the upper respiratory tract, while intratracheal infection tends to cause lesions in the lower respiratory tract. Naturally infected field reactor cattle most commonly have lesions in the lower respiratory tract and pulmonary lymph nodes, while involvement of the upper respiratory tract occurs more rarely (2, 6, 17, 19, 30).
Bovine tuberculosis may spread by cattle-to-cattle transmission and also through the involvement of wildlife reservoirs (16). However, experimental intranasal and intratracheal M. bovis infections in cattle have shown that bacterial shedding (the presence of viable M. bovis in the nasal mucus) is, at best, transient and involves extremely low numbers (approximately 70 CFU) of bacilli (14). Such a low dose of M. bovis has not previously been considered relevant in the bovine model of tuberculosis, even though a low infection dose for Mycobacterium tuberculosis in humans has been accepted for many decades (21, 24, 25, 27). These historical studies showed that the numbers of primary calcified M. tuberculosis lesions in otherwise healthy people were low (between one and three lesions per individual) and also established that the infectious dose of M. tuberculosis could be as low as 1 to 10 bacilli.
Previous experimental infections of cattle with M. bovis have suggested that the infective dose can have a profound influence on the severity of the disease that follows. For example, in the intranasal model, 5 x 105 to 106 CFU resulted in multiple respiratory lesions, while 5 x 102 to 104 CFU resulted in a more variable pathology (some animals had multiple lesions, and some animals had no lesions) and 102 CFU resulted in no visible lesions at all. Although the latter group remained skin test negative, M. bovis was isolated from the nasal mucus of one animal 100 days after infection (17). We and other workers have shown that in the intratracheal model low doses of M. bovis (800 to 6 x 103 CFU) can result in animals that are skin test negative, have no visible lesions at post mortem, and are M. bovis culture negative (2, 4, 20, 23). Interestingly, Rhodes et al. (23) also measured specific cytokine responses and showed that the skin test-negative animals in their study that were gamma interferon (IFN-) and interleukin-2 (IL-2) negative gave positive specific IL-4 responses, indicating that these animals had been successfully exposed to the M. bovis inoculum but had dealt with the infection differently.
To date, the doses of M. bovis used in the cattle model have been relatively high, and natural infections with such doses are unlikely. The current study was therefore undertaken to determine the lowest infective dose of M. bovis in cattle that resulted in pathology. To this end, we infected calves with logarithmically decreasing doses from 1,000 CFU to 1 CFU per animal. Immunological monitoring (specific IFN- and IL-4 responses) throughout the study and tuberculin skin testing at two separate time points were carried out in order to obtain information on the effect of the M. bovis dose on these parameters.
MATERIALS AND METHODS
Culture medium and antigens. The culture medium used for the whole blood cultures and peripheral blood mononuclear cell stimulation was RPMI 1640 with Glutamax (Gibco) supplemented with 5% CSPR3 (control serum protein replacement; Gibco), nonessential amino acids (Gibco), 100 U/ml penicillin, 100 μg/ml streptomycin, and 5 x 10–5 M 2-mercaptoethanol (Gibco); 10% fetal calf serum (Gibco) was used instead of CSPR3 in the B-cell cultures. Tuberculin preparations purified protein derivative A and B (PPD-B and PPD-A) were produced at the Veterinary Laboratories Agency (Weybridge, United Kingdom). Recombinant ESAT6 and CFP10 proteins (29) were obtained from M. Singh (GBF, Braunschweig, Germany).
Bacterial strain. The infection inoculum was prepared from a mid-log-phase frozen seed stock of M. bovis strain 2122/97, whose concentration had been determined previously. The seed stock was thawed and then diluted to the required concentration in Middlebrook 7H9 medium containing 0.05% Tween 80. The infective dose was confirmed retrospectively by plating the inoculum in triplicate on a modified formulation of Middlebrook 7H11 agar and enumerating colonies following 4 weeks of incubation at 37°C. To determine the extent of clumping, the inoculum was sonicated for 10 s at 21% amplitude using a Sonics Vibra Cell VCX500 sonicator fitted with a 4-mm-diameter tip. Bacterial enumeration of the sonicated sample was performed on modified Middlebrook 7H11 agar. One CFU was confirmed to contain between 6 and 10 viable bacilli.
Cattle. Twenty Friesian Holstein heifers and bullocks that were a minimum of 6 months old from a tuberculosis-free herd were randomly divided into four groups. The disease-free status of individual animals was further confirmed by an IFN- test prior to recruitment. Calves were inoculated intratracheally with M. bovis field strain GB (AF 2122/97) by using the protocol of Buddle et al. (2). Briefly, an 80-cm endotracheal tube containing a fine cannula was inserted per os into the trachea of an anesthetized animal. The appropriate dose of inoculum in 1.5 ml prepared as described above was injected through the cannula and flushed out with 2 ml of saline. This route of infection was selected as the route most similar to natural infection, since previous experiments indicated that intratracheal infection replicates more closely the lesion distribution seen in the majority of naturally infected cattle. One group of six calves (calves 2805, 2858, 2865, 2871, 2877, and 2878) received 1 CFU, and one group of six calves (calves 2806, 2861, 2863, 2866, 2869, and 2870) received 10 CFU. In addition, one group of four animals (calves 2802, 2859, 2927, and 2928) received 100 CFU, and one group of four animals received 1,000 CFU (calves 2923, 2924, 2925, and 2926). Each of the four groups of animals was housed separately in a high-security self-contained isolation unit under negative pressure with 14 air changes per h for the duration of the experiment. (In previous experiments conducted in the facility, in which uninfected calves were kept in contact with animals infected with a much higher dose of M. bovis than was used in this study, no natural cattle-to cattle-transmission occurred; it is therefore unlikely that any transmission occurs between animals.) Blood samples were taken by venepuncture of the jugular vein for immunological studies before infection and then weekly until 6 weeks postinfection (p.i.). Subsequently, samples were taken every 2 weeks until the end of the experiment.
Skin test. The single comparative intradermal tuberculin skin test was performed as described previously (1) in accordance with the standard protocol used in the United Kingdom. All the animals were skin tested at week 12 postinfection and again shortly before the end of the experiment (weeks 23 to 25 postinfection). An increase in the skin induration response to PPD-B that was greater than the response to PPD-A by at least 10 mm was considered a positive response.
Post mortem. Calves were euthanized by intravenous injection of sodium pentobarbitone, and a detailed post mortem examination was carried out. Lymph nodes were removed aseptically (right and left submandibular, right and left medial retropharyngeal, right and left lateral retropharyngeal, right and left parotideal, cranial mediastinal, caudal mediastinal, right and left bronchial, cranial tracheobronchial, right and left tonsil). Lymph nodes were serially sliced (2-mm slices) and examined for the presence of lesions. Samples of lesion material and also apparently unaffected tissue were used for histopathological examination and for culture of M. bovis. Other lymph nodes were macroscopically examined in situ. The lungs were serially sliced (5-mm slices) following the bronchial tree, and the slices were palpated and inspected. Small pieces of lung tissue were removed for culture and histopathology. The nasal passages were also opened and inspected. Individual tissues were assigned a pathology score depending upon the number, size, and character of the lesions observed in accordance with the standard methodology used in this laboratory (28).
Whole-blood culture and IFN- enzyme-linked immunosorbent assay. Duplicate cultures of peripheral whole blood were diluted 1:1 with culture medium (100 μl per well in 96-well flat-bottom microtiter plates) and cultured in the presence or absence of antigen (final concentration of PPD-B and PPD-A, 10 μg/ml; final concentration of ESAT6 and CFP10, 5 μg/ml; or mitogen control, 2 μg/ml [staphylococcus enterotoxin B]) for 24 h (9). Supernatants were then assessed for gamma interferon content using a commercially available enzyme-linked immunosorbent assay kit (BOVIGAM; Biocore, United States) according to the manufacturer's instructions. The results were expressed as the increase in the mean optical density at 450 nm (OD450) in the presence of antigen compared to the OD450 in the medium or unstimulated supernatants. A change in the OD450 of more than 0.1 was considered a positive response.
IL-4 bioassay. The IL-4 bioassay was carried out as previously described (12, 23). Briefly, peripheral blood mononuclear cells from naive skin test-negative cattle kept at Veterinary Laboratories Agency were positively sorted using monoclonal antibody IL-A58 (obtained courtesy of the Institute for Animal Health, Compton, Berkshire, United Kingdom), goat anti-mouse immunoglobulin G-coated microbeads, and the MACS column separation system (Miltenyi Biotech, Surrey, United Kingdom) to obtain a highly enriched B-cell population. Whole blood was incubated 1:1 with antigen, mitogen, and medium or a control as described above for the IFN- analysis, but the incubation time was 6 days. Culture supernatants were then harvested, and 50 μl undiluted supernatant was added to 100 μl of a B-cell suspension at a concentration of 106 cells/ml in 96-well round-bottom microtiter plates. The plates were incubated for 24 h before they were pulsed with tritiated thymidine (Amersham UK) and then harvested after a further 24 h. The results are expressed as stimulation indices (SI) (i.e., mean counts per minute for B-cell proliferation in the presence of antigen-stimulated supernatant divided by the mean counts per minute for B-cell proliferation in the presence of medium or control unstimulated supernatant). A stimulation index greater than 3.0 was considered a positive response. Previous work (22) using a monoclonal antibody specific for bovine IL-4 that had been shown to inhibit specifically the biological activity of recombinant bovine IL-4 (CC303; 10 μg/ml) showed that this antibody was able to block B-cell stimulatory activity in the assay.
RESULTS
Skin test and pathological findings for M. bovis-infected cattle. Groups of calves were infected with 1, 10, 100, and 1,000 CFU M. bovis at the beginning of the experiment.
The skin test results at week 12 showed that 14 of the 20 animals were skin test positive (Table 1). The six remaining negative animals were distributed among the four groups as follows: three animals that received 1 CFU (calves 2085, 2871, and 2877,) one animal that received 10 CFU (calf 2866), one animal that received 100 CFU (calf 2859), and one animal that received 1,000 CFU (calf 2924). These results were confirmed at the second skin test at weeks 23 to 25 p.i. (data not shown). There was no significant difference as determined by a chi-square test (P = 0.6267) in the distribution of skin test-positive and -negative animals between the different groups. All skin test-positive calves were found to have visible lesions in the respiratory lymph nodes, and most also contained lesions in the upper lung lobes (all except calf 2861). Mycobacterial culture on modified Middlebrook 7H11 agar and acid-fast staining confirmed the presence of M. bovis in tissues with lesions. In accordance with our previous findings, no lesions were detected in the head nodes, and no mycobacteria were cultured from these areas. The degrees of pathology as determined by the pathology scores (Table 2) of all the animals in the groups were comparable, and there were no significant differences as determined by the Kruskal-Wallis test (P = 0.3896) between groups that received different infective doses. All skin test-negative animals presented without gross pathology (no visible lesions). M. bovis could not be cultured from tissue samples from these animals, and no histopathological signs of tuberculosis were detected (data not shown). Thus, we were able to infect cattle with 1 CFU of M. bovis.
Immunological findings for skin test-positive cattle. All skin test-positive calves developed strong positive specific IFN- responses. The kinetics of the IFN- response of the three skin test-positive calves infected with 1 CFU are shown in Fig. 1. In all 14 skin test-positive calves a positive IFN- response was observed by 3 to 5 weeks p.i., after which the response fluctuated (but remained positive) throughout the course of infection. The time to positivity (number of weeks postinfection until a positive response was observed) (Table 3) and the intensity of the response (OD450) (data not shown) did not vary significantly between groups that received different infective doses. The responses to the recombinant antigens ESAT6 and CFP10 were variable. Of the 14 calves that produced IFN-, 2 (calves 2802 and 2806) responded to CFP10 but not to ESAT6, while 5 (calves 2863, 2865, 2869, 2925, and 2928) responded to ESAT6 but not to CFP10. The remaining seven animals (calves 2858, 2861, 2876, 2878, 2923, 2926, and 2927) responded equally to both proteins (data not shown). There was no correlation between the infective dose of M. bovis and the responses to these two proteins (data not shown).
Six-day whole-blood supernatants from 13 of 14 skin test-positive calves were shown to stimulate the proliferation of B cells in the IL-4 bioassay. Positive IL-4 responses were transient and were observed soon after infection, between 5 and 7 weeks p.i. for most animals. The IL-4 responses of the three skin test-positive calves that received 1 CFU M. bovis are shown in Fig. 2. There was no apparent difference in the strength of the IL-4 response (SI) between groups that received different M. bovis doses. A summary of the IFN- and IL-4 data for all skin test-positive animals is shown in Table 3.
The presence of another anti-inflammatory cytokine, IL-10, was also investigated in both skin test-positive and -negative animals in our study, since specific IL-10 responses have recently been reported to occur during experimental bovine paratuberculosis infection (S. Marche, K. Walravens, C. J. Howard, J. C. Hope, V. Rosseels, K. Huygen, and J. Godfroid, 7th Int. Vet. Immunol. Symp., Quebec, Canada, 2004, abstr. 370). However no significant IL-10 responses were detected in any of our cattle (data not shown).
DISCUSSION
This study clearly demonstrated that 1 CFU of M. bovis is able to cause bovine tuberculosis. Interestingly, this finding is in accord with a mathematical analysis of experimental bovine tuberculosis performed by Neill et al. (18), who postulated that a single M. bovis organism should be capable of causing disease in cattle. In our experiment 1 CFU was found to contain between 6 and 10 viable bacilli. Infection with just 1 CFU resulted in pathology whose severity was equivalent to that seen in animals which received far higher doses (up to 1,000 CFU). The presence of pathology correlated with a positive tuberculin skin test and a strong sustained specific IFN- response, which supports previous findings by us and other workers that antigen (ESAT6)-specific IFN- responses correlate with the severity of pathology in cattle (13, 28). No M. bovis dose-related effect was observed in the pathology score, the strength of the tuberculin skin test, the strength of the IFN- response, or the time taken to obtain a positive IFN- response. Our data therefore do not support a dose-related pathological outcome in the animals in which disease becomes established, but they seem to suggest that comparable levels of pathology may develop regardless of the number of CFU inoculated.
Whether the skin test-negative animals in our study, if left for a much longer time, would have converted to a positive phenotype is not known. Previous data from experimental intratracheal infections of cattle in which M. bovis doses of 800 CFU and 1,000 CFU were used (2, 4) showed that 40% of the cattle remained negative for the tuberculin skin test, specific IFN- test, and M. bovis culture. However, infection with a slightly higher dose, 2,000 CFU, gave more variable results; that is, there were fewer negative animals, and a proportion of these animals showed late-onset IFN- responses (3, 20). This again raises the question of latency in cattle and its potential importance in disease transmission. It has been suggested that further immunological analyses of low-dose infections could be employed to investigate latency in cattle (16); however, such studies would by necessity be long term, and their outcomes would be unreliable.
Our previous study (23) suggested that skin test-, IFN--, and IL-2-negative M. bovis-infected cattle could produce specific IL-4 responses. This provided some hope that IL-4 might be a detectable marker in this otherwise negative phenotype. However, these results were not repeated in the current study. Significant positive IL-4 responses were found in only 13 of 14 of the skin test-positive animals. These responses were transient, occurring in most animals between weeks 5 and 7 p.i., although individual spikes of IL-4 activity were also seen at other times in some animals. We also measured IL-4 activity in peripheral blood mononuclear cell supernatants of skin test-negative cattle, but the results were negative (data not shown). However, the lowest dose used in the previous study was 6 x 103 CFU, which is six times greater than the highest dose (1,000 CFU) used in the current experiments. Whether this difference in M. bovis dose is relevant for IL-4 responses is not known.
The IL-4 bioassay is currently the only reliable assay for measuring bovine IL-4 activity. The fact that clinical studies identify a role for IL-4 in the pathological process (8, 26) clearly indicates the need for more specific assays to assess the role of this cytokine and its splice variants in bovine tuberculosis.
In summary, we found that 1 CFU of M. bovis is sufficient to cause established tuberculous pathology in cattle. This pathology is identical to that resulting from significantly higher experimental doses (up to 1,000 CFU in this study) and reflects the pathology seen in naturally infected field reactor cattle. Cattle infected with 1 CFU that developed pathology exhibited strong positive responses to the diagnostic tuberculin skin test. Furthermore, the infectious dose of M. bovis had no bearing on the time taken to obtain a positive IFN- response in the animals that went on to develop pathology.
Our data are in accord with very low numbers of bacilli transmitted aerogenously between cattle, potentially by nasal shedding. Comfortingly, the animals that do go on to develop pathology and therefore become a likely source of contamination within a herd can be detected at an early stage with the IFN- test and also provide a positive tuberculin skin test response.
ACKNOWLEDGMENTS
We thank Jayne Hope and Elinor Mead (Institute for Animal Health, Compton, Berkshire, United Kingdom) for assessing our supernatants for IL-10.
This work was funded by the Department of Environment and Rural Affairs (DEFRA) U.K.
REFERENCES
1. Anonymous. 2002. European Commission Regulation (EC) NO 1226/2002 of 8 July 2002 amending annex B to Council Directive 64/432/EEC. Off. J. Eur. Union (English) L 179:13.
2. Buddle, B. M., G. W. DeLisle, A. Pfeffer, and F. E. Aldwell. 1995. Immunological responses and protection against Mycobacterium bovis in calves vaccinated with a low dose of BCG. Vaccine 13:1123-1130.
3. Buddle, B. M., D. Keen, A. Thomson, G. Jowett, A. R. McCarthy, J. Heslop, G. W. DeLisle, J. L. Stanford, and F. E. Aldwell. 1995. Protection of cattle from bovine tuberculosis by vaccination with BCG by the respiratory or subcutaneous route, but not by vaccination with killed Mycobacterium vaccae. Res. Vet. Sci. 59:10-16.
4. Buddle, B. M., N. A. Parlane, D. L. Keen, F. E. Aldwell, J. M. Pollock, K. Lightbody, and P. Andersen. 1999. Differentiation between Mycobacterium bovis BCG-vaccinated and M. bovis-infected cattle by using recombinant mycobacterial antigens. Clin. Diag. Lab. Immunol. 6:1-5.
5. Chausse, P. 1913. Des methodes a employer pour realiser la tuberculose experimentale par inhalation. Bull. Soc. Med. Vet. 31:267-274.
6. Corner, L. A. 1994. Post mortem diagnosis of Mycobacterium bovis infection in cattle. Vet. Microbiol. 40:53-63.
7. Daborn, C. J., J. M. Grange, and R. R. Kazwala. 1996. The bovine tuberculosis cycle: an African perspective. Soc. Appl. Bacteriol. Symp. Ser. 25:27S-32S.
8. Demissie, A., M. Abebe, A. Aseffa, G. Rook, H. Fletcher, A. Zumla, K. Weldingh, I. Brock, P. Andersen, T. M. Doherty, and the VACSEL Study Group. 2004. Healthy individuals that control a latent infection with Mycobacterium tuberculosis express high levels of Th1 cytokines and the IL-4 antagonist IL-42. J. Immunol. 172:6938-6943.
9. Emery, D. L., F. H. Duffy, and P. R. Wood. 1988. An analysis of cellular proliferation and synthesis of lymphokines and specific antibody in vitro by leucocytes from immunized cattle. Vet. Immunol. Immunopathol. 18:67-80.
10. Francis J. 1947. Bovine tuberculosis, including a contrast with human tuberculosis. Staples Press Ltd., London, United Kingdom.
11. Hardie, R. M., and J. M. Watson. 1992. Mycobacterium bovis in England and Wales: past, present and future. Epidemiol. Infect. 109:23-33.
12. Kuhnle, G., R. A. Collins, J. E. Scott, and G. M. Keil. 1996. Bovine interleukin-2 and interleukin-4 expressed in recombinant bovine herpesvirus-1 are biologically active secreted glycoproteins. J. Gen. Virol. 77:2231-2240.
13. Lyashchenko, K., A. O. Whelan, R. Greenwald, J. M. Pollock, P. Andersen, R. G. Hewinson, and H. M. Vordermeier. 2004. Association of tuberculin-boosted antibody response with pathology and cell-mediated immunity in cattle vaccinated with Mycobacterium bovis BCG and infected with M. bovis. Infect. Immun. 72:2462-2467.
14. McCorry, T. A., A. O. Whelan, M. D. Welsh, J. McNair, E. Walton, D. G. Bryson, R. G. Hewinson, and H. M. Vordermeier. Shedding of Mycobacterium bovis in the nasal mucus of cattle experimentally infected with tuberculosis by the intranasal and intratracheal routes. Vet. Rec., in press.
15. McFadyean, J. 1910. What is the common method of infection in tuberculosis J. Comp. Pathol. 23:239-250 and 289-298.
16. Morrison, W. I., F. J. Bourne, D. R. Cox, C. A. Donnelly, G. Gettinby, J. P. McInerney, and R. Woodroffe. 2000. Pathogenesis and diagnosis of infections with Mycobacterium bovis in cattle. Vet. Rec. 26:236-242.
17. Neill, S. D., J. Hanna, J. J. O'Brien, and R. M. McCracken. 1988. Excretion of Mycobacterium bovis by experimentally infected cattle. Vet. Rec. 123:340-343.
18. Neill, S. D., J. J. O'Brien, and J. Hanna. 1991. A mathematical model for Mycobacterium bovis excretion from tuberculous cattle. Vet. Microbiol. 28:103-109.
19. Neill, S. D., J. M. Pollock, D. B. Bryson, and J. Hanna. 1994. Pathogenesis of Mycobacterium bovis infection in cattle. Vet. Microbiol. 40:41-52.
20. Ng, K. H., F. E. Aldwell, D. N. Wedlock, J. D. Watson, and B. M. Buddle. 1997. Antigen-induced interferon-gamma and interleukin-2 responses of cattle inoculated with Mycobacterium bovis. Vet. Immunol. Immunopathol. 57:59-68.
21. O'Grady, F., and R. L. Riley. 1963. Experimental airborne tuberculosis. Adv. Tuberc. Res. 12:150-190.
22. Rhodes, S. G., J. M. Cocksedge, R. A. Collins, and W. I. Morrison. 1999. Differential cytokine responses of CD4+ and CD8+ T cells in response to bovine viral diarrhoea virus in cattle. J. Gen. Virol. 80:1673-1679.
23. Rhodes, S. G., N. Palmer, S. P. Graham, A. E. Bianco, R. G. Hewinson, and H. M. Vordermeier. 2000. Distinct response kinetics of gamma interferon and interleukin-4 in bovine tuberculosis. Infect. Immun. 68:5393-5400.
24. Rich, A. R. 1946. Pathogenesis of tuberculosis. Thomas, Springfield, Ill.
25. Riley, R. L., C. C. Mills, W. Nyka, N. Weinstock, P. B. Storey, L. U. Sultan, M. C. Riley, and W. F. Wells. 1995. Aerial dissemination of pulmonary tuberculosis. A two year study of contagion in a tuberculosis ward, 1959. Am. J. Epidemiol. 142:3-14.
26. Rook, G. A. W., R. Hernandez-Pando, K. Dheda, and G. T. Seah. 2004. IL-4 in tuberculosis: implications for vaccine design. Trends Immunol. 25:483-488.
27. Sonin, L. S. 1951. The role of particle size in experimental airborne infection. Am. J. Hyg. 53:337-354.
28. Vordermeier, H. M., M. A. Chambers, P. J. Cockle, A. O. Whelan, J. Simmons, and R. G. Hewinson. 2002. Correlation of ESAT-6-specific gamma interferon production with pathology in cattle following Mycobacterium bovis BCG vaccination against experimental bovine tuberculosis. Infect. Immun. 70:3026-3032.
29. Vordermeier, H. M., A. Whelan, P. J. Cockle, L. Farrant, N. Palmer, and R. G. Hewinson. 2001. The use of synthetic peptides derived from the antigens ESAT-6 and CFP-10 for differential diagnosis of bovine tuberculosis in cattle. Clin. Diag. Lab. Immunol. 8:571-578.
30. Whipple, D. L., C. A. Bolin, and J. M. Miller. 1996. Distribution of lesions in cattle infected with Mycobacterium bovis. J. Vet. Diagn. Investig. 8:351-354.(Gillian S. Dean, Shelley )