* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Research Project Final Report
Marburg virus disease wikipedia , lookup
Bovine spongiform encephalopathy wikipedia , lookup
Middle East respiratory syndrome wikipedia , lookup
Trichinosis wikipedia , lookup
Brucellosis wikipedia , lookup
Hepatitis C wikipedia , lookup
African trypanosomiasis wikipedia , lookup
Human cytomegalovirus wikipedia , lookup
Tuberculosis wikipedia , lookup
Onchocerciasis wikipedia , lookup
Dirofilaria immitis wikipedia , lookup
Sarcocystis wikipedia , lookup
Schistosomiasis wikipedia , lookup
Leptospirosis wikipedia , lookup
Neonatal infection wikipedia , lookup
Hepatitis B wikipedia , lookup
Hospital-acquired infection wikipedia , lookup
Coccidioidomycosis wikipedia , lookup
Oesophagostomum wikipedia , lookup
General enquiries on this form should be made to: Defra, Science Directorate, Management Support and Finance Team, Telephone No. 020 7238 1612 E-mail: [email protected] SID 5 Research Project Final Report Note In line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects. 1. Defra Project code 2. Project title This form is in Word format and the boxes may be expanded or reduced, as appropriate. 3. ACCESS TO INFORMATION The information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000. Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors. SID 5 (Rev. 3/06) Project identification SE3024 Low dose infection in cattle: disease dynamics and diagnostic strategies Contractor organisation(s) Veterinary Laboratories Agency Woodham Lane New Haw, ADDLESTONE Surrey KT15 3NB 54. Total Defra project costs (agreed fixed price) 5. Project: Page 1 of 27 £ 466,621 start date ................ 01/10/2002 end date ................. 30/09/2006 6. It is Defra’s intention to publish this form. Please confirm your agreement to do so. ................................................................................... YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow. Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer. In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000. (b) If you have answered NO, please explain why the Final report should not be released into public domain The balance of these data are being written up for publication in scientific journals at the moment. Publication of this form by Defra could be viewed as disclosure of the data, and this could exclude these data from publications in such peer-reviewed journals. However, we intend to submit these papers in the next 3-6 months, and I am therefore aksing Defra not to publish the report before March 2007. Executive Summary 7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work. Bovine TB may spread via cattle-to-cattle transmission and also through the involvement of environmental reservoirs. In the majority of cattle (ca. 60 %) presenting with bovine tuberculosis in GB, pathology is restricted to the lower respiratory tract (lung and/or lymph nodes draining the lung) and the most appropriate experimental model best reproducing this pathology presentation is challenge causing disease in the lower respiratory tract, ie the intratracheal challenge model, which is being used in these studies. An additional model of experimental infection via the aerosol challenge route was also established, which also targets the lower respiratory tract. Applying both models side-by-side synergistically allowed us to define factors relevant to transmission, pathogenesis, and the performance of immuno-diagnostic tests, with particular emphasis on studying the disease in animals that presented with low or very low bacillary loads as well as latently infected animals. The specific aims of this project were as follows: To determine the minimum infective dose of M. bovis in cattle (objs. 1, 2). To generate a “memory cow” model of protective immunity, re-infection/re-exposure and potential latency of M. bovis in cattle (Obj. 3). To establish an aerosol challenge model for M. bovis infection in cattle (Obj. 4). In objecties 1 and 2, we determined the minimum infective dose of Mycobacterium bovis necessary to stimulate specific immune responses and generate pathology in cattle. Calves were infected by the intratracheal route with 1,000, 100, 10 and 1 colony-forming units (CFU) of M. bovis. Results showed that half of the animals infected with 1 CFU of M. bovis developed pulmonary pathology typical of bovine tuberculosis, whereas the other half of the animals infected with 1 CFU presented without signs of disease. No signficiant difference in the severity of pathology was observed after intratracheal challenge after infection with the 1-1000 CFU dose range, although higher infective doses (>10,000 CFU) resulted in more severe disease and pathology (data not shown), in line with the more severe disease observed after aerosol infection with a similar high dose. All animals that developed pathology were skin test positive and produced specific IFN- and IL-4. There was no difference in the size of the skin test reaction, the time taken to achieve a positive IFN- result, or in the levels of IFN- and IL-4 between animals infected with the different doses of M. bovis, suggesting that current diagnostic tests (skin test and IFN-test) can detect cattle soon after M .bovis infection regardless of dose. In objective 4, an aerosol challenge model for cattle was established by titrating infective M. bovis doses, and the results demonstrated that even high doses delivered by aerosols (ca. >104 CFU) resulted in visible pathology confined to the lower respiratory tract although bacilli could also be detected in some animals in the upper respiratory tract. Therefore, the following conclusions can be made from the results of objs. 1,2 and 4 (see also detailed SID 5 (Rev. 3/06) Page 2 of 27 discussion of objective 4 below) can be summarised as follows: Intratracheal and aerosol routes of experimental challenge routes are the most appropriate routes because they best reflect the lesion most commonly associated with natural infection (in GB). The lower respiratory tract is highly susceptible, infective doses of 1 CFU will result in pathology and disease in the lung and/or associated lymph nodes. Predominant disease phenotype seen in GB is likely to be caused by aerosol infection of small numbers of bacilli delivered by small aerosol particles (3-5 organisms) to the lung. At these low infection doses, it is possible to produce, apart from VL/Cu+ animals, also animals that are NVL/cu+ or NVL/cu- yet positive in skin test and IFN- test: Such animals are therefore not ‘falsepositives’ but infected sub-clinically and could pose future infection risks to herds. Objective 3. In order to study immune responses in cattle containing the infection with low, or undetectable bacillary loads, we generated a “memory cow” model that we adapted from a drug-assisted model of protective immunity developed originally for mice. We aimed to address the following specific questions: How do diagnostic tests work in animals with very low bacillary loads, or in a state of latency? Does a primary infection protect against cattle against M.bovis re-infection? Identify possible correlates of protection/markers of disease severity. Identify potential protective antigens. We could show that cattle infected with M. bovis (spoligotype 9) and then treated with isoniazid (INH) harbour minimal or no pathology compared to untreated animals, yet still presented with strong cellular immune responses (IFN-, DTH). In addition, INH-treated cattle were protected against rechallenge with M. bovis, and we used these animals to explore immunological correlates of protection as well as to define potential protective antigens. Therefore, this objective had cross-cutting relevance also for Defra’s cattle TB vaccine development programme. For example, we could show that IL-4 splice variants have been shown to antagonise/down-regulate IL4-mediated cellular responses, furthermore, that one splice variant, IL43, was elevated in this model of protective immunity against M. bovis. In relation to the expected outputs from Obj. 3, the following conclusions can be made: BOVIGAM responses contract with treatment/reduction in bacillary load, but animals present with positive results at most time points, i.e. would still be detected by BOVIGAM assay even at low or undetectable bacillary loads (see also comments below to objective 4). Primary infection confers significant degree of protection against re-infection. This protection was against heterologous M. bovis strain (cross-protection), which is encouraging for vaccine development using BCG or one particular attenuated M. bovis strain. Successful chemotherapy requires active immunity. IFN- has been confirmed as a marker of bacterial load/pathology. The IL43 splice variant constitutes a surrogate of protection. Objective 4. Development of an aerosol delivery model. A model was developed to accurately deliver M. bovis to calves via the aerosol route. This model was then used to deliver decreasing infective doses (104, 102, 101 CFU). The results indicated a decrease in pathology with aerosol delivery resulting between the high dose and the two lower doses. However, even with the highest dose almost exclusively in pathology restricted to the lower respiratory tract (lung and lung associated lymph nodes). These data, together with results from obj. 1, support the conclusions discussed above in the context of obj. 1. The value of the aerosol infection system lies with the combination of the natural host, the natural route of infection and the use of a low M. bovis dose to infect calves. This approach has given further insight into the relationship between infectious dose, the immunology of infection and the type of pathological picture that follows. The lower infectious doses in particular, induced an immune state more akin to that seen often in the field, where the relationship between disease status and immune response is complex. In particular, animals infected with the lowest infection dose not only presented as NVL/culture-negative but were also skin test negative yet still detectable using the BOVGIAM test. Results of the aerosol challenge model will be discussed in relation to the intratracheal challenge model. Project Report to Defra 8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: SID 5 (Rev. 3/06) Page 3 of 27 the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer). Objectives 1: To define the relationship between low infectious doses of M. bovis and immunological parameters, diagnostic tests and severity of pathology in cattle; and objective 2: End-point titration of infective dose in the intratracheal challenge model. Abstract. The aim of this work was to determine the minimum infectious dose of M. bovis necessary to stimulate specific immune responses and generate pathology in cattle. Four groups of calves (20 animals) were infected by the intra-tracheal route with 1000, 100, 10 or 1 CFU of M. bovis. Specific immune responses (IFN- and IL-4) to mycobacterial antigens were monitored throughout the study, and responses to the tuberculin skin test assessed [1] at two time points. Detailed post mortem examinations [2] were performed to determine the presence of pathology, and samples were taken for microbiological and histopathological confirmation of M. bovis infection. In the group infected with the lowest dose (1 CFU), one half of the animals developed pulmonary typical of bovine tuberculosis: The minimal infective dose using the intratracheal infection route to infect cattle with M.bovis is therefore 1 CFU. The large majority of the animals infected with the higher doses also developed pathology typical of bovine tuberculosis, and M. bovis could be cultured from tissues. No difference in the severity of pathology were observed for the different M. bovis doses. All animals that developed pathology were skin testpositive and produced specific IFN- and IL-4 responses. No differences in the size of the skin test reactions, the times taken to achieve a positive IFN- result, or the levels of IFN-and IL-4 responses were observed for the different M. bovis doses, suggesting that current 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 modelling the dynamics of bovine TB in cattle and in assessing the risk of transmission. Results. 1. Skin test and pathological findings. Table 1 shows the skin test results at 12 weeks post infection with M. bovis (Table 1). Of the 20 animals infected, 14 became skin test-positive (applying the standard interpretation of the skin test). The 6 skin test-negative animals were distributed among the 3 groups as follows; 3 calves receiving 1CFU (animal no.s 2805, 2871, 2877), 1 calf receiving 10 CFU (animal no. 2866), 1 calf receiving 100 CFU (animal no. 2859), 1 calf receiving 1000 CFU (animal no. 2924). These results were confirmed at the second skin test at weeks 23-25 post-infection (data not shown). There was no significant difference (Chi-squared p=0.6267) in the distribution of skin testpositive and negative animals between the different groups. All skin test-positive animals had visible pathology in the respiratory lymph nodes and most also contained lesions in the upper lung. Mycobacterial culture on modified Middlebrook 7H11 agar and acid fast staining confirmed the presence of M. bovis in lesioned tissues. The degree of pathology (see Table 2. Pathology Scores) of all animals was comparable, there was no significant difference (Kruskal-Wallis p=0.3896) between groups that had received infective doses between 1 and 100 CFU M. bovis. All skin test-negative animals presented without gross pathology and were M. bovis culture-negative. SID 5 (Rev. 3/06) Page 4 of 27 2. Immunological responses. All skin test-positive calves (14 animals) developed positive specific IFN- responses (24-hour whole blood assay and BOVIGAM ELISA) between 3-5 weeks post-infection (p.i.) (see Table 3. for Summary of IFN- and IL-4 responses). The time to positivity (number of weeks p.i. until a positive response was observed) and the intensity of the response did not vary significantly between groups that received different doses of M. bovis. There was no significant difference in the cytokine response to the immunodominant proteins ESAT6 and CFP10 relative to M. bovis dose (data not shown). 6-day whole blood supernatants from 13/14 calves produced specific IL-4 (B-cell bioassay). IL-4 responses were transient and initially observed between 5-7 weeks p.i. for most animals. There was no difference in the strength of the IL-4 response relative to M. bovis dose. SID 5 (Rev. 3/06) Page 5 of 27 Concluding remarks. In summary, data from Objectives 1 and 2 showed that 1CFU 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 1000 CFU in this study) and reflects the pathology seen in naturally infected field reactor cattle. Cattle infected with 1CFU that developed pathology exhibited strong positive responses to the diagnostic tuberculin skin test. Furthermore, the infectious dose of M. bovis had no bearing upon the time taken to achieve a positive IFN- response in those animals that developed pathology. These data are in accord with very low numbers of bacilli transmitted aerogenously between cattle. 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- and tuberculin skin tests. Furthermore, the Intratracheal route of experimental challenge is appropriate because it best reflect the lesion most commonly associated with natural infection (in GB), t he lower respiratory tract is highly susceptible, infective doses of 1 CFU will result in pathology and disease in the lung and/or associated lymph nodes, and the predominant disease phenotype seen in GB is caused by aerosol infection of small numbers of bacilli delivered by small aerosol particles (3-5 organisms) to the lung. Orignial Objective 2. To define the degree of protection conferred to cattle by low doses of M. bovis against sub-sequent high-dose M. bovis re-challenge and what impact this outcome has on the sensitivity of diagnostic tests. This was removed after approval by Defra (17 November 2003) to allow lower titration of M. bovis (Obj. 1) to proceed down to 1CFU/animal. Objective 3. (VLA) To define how diagnostically useful immune parameters (e.g. skin test and IFNresponses) develop over time in animals that recover from primary M. bovis infection and to test whether such animals are protected against sub-sequent re-infection (cattle memory model). Experiment 3A (VLA). Isoniazid-attenuated primary infection with virulent M. bovis: assessment of drug effectiveness and safety. Abstract. The aim of this preliminary experiment was to assess the potential usefulness of a drug-attenuated acute M. bovis infection of cattle as a means of generating cattle presenting with low or very low bacillary load and to assess the dynamics of the immune response in such animals during the course of treatment. 16 calves were infected with a dose of M. bovis known to result in pathology (350 CFU). At 4 weeks p.i. (when a stable specific immune response was observed) 8 of these were treated with the TB drug isoniazid (INH; 25mg/kg/day) for a period of 10 weeks. INH treatment was stopped 5 days prior to necropsy. Skin tests were performed 1-2 weeks prior to necropsy. INH-treated cattle were compared with the untreated M. bovis challenge controls in terms of their specific IFN- responses and pathology. Serum samples were taken to measure liver function markers (known side effects of INH include liver damage), and plasma INH monitored to ensure drug clearance. Our results demonstrated the effectiveness of INH-treatment to reduce disease burden (ie bacillary loads and SID 5 (Rev. 3/06) Page 6 of 27 severity of pathology), and that this system is appropriate for use in objective 3B. In addition, we could demonstrate that no adverse hepato-toxcicity was induced by INH treatment. Results. 1. Skin test and pathological findings. All animals in both groups were skin test positive at 13 weeks post infection. Mean pathology scores at necropsy were significantly lower in the INH treated animals than in the control group (table 4). Histopathology (FIG.1) confirmed that lesions in the INH treated group were less severe than in the control group FIG.1. Histopathology after INH treatment CW3196. Infected Isoniazid treated CW3199. Infected and untreated Left tracheobronchial lymph node. Granuloma with central mineralization. Numerous associated giant cells surround mineralized debris (arrows). Note lack of necrosis and scant peripheral fibrosis. H&E 100x Caudal mediastinal lymph node. Central mineralization and necrosis (arrowhead) with scattered lymphocytes and macrophages at the periphery. Giant cells present rimming central mineralization (arrow). H&E 100x TABLE 4. Pathology scores Animal M. bovis + INH CW 3195 CW 3196 CW 3197 CW 3198 101721 201708 601726 601740 Mean M. bovis CW 3199 CW 3200 CW 3201 CW 3202 101714 401724 501718 701734 Mean *, p-values (Mann-W hitney) Lymph Node Lung Sub total Culture Score 0 3 0 2 0 0 2 0 0.9 2 0 0 3 0 0 3 2 1.2 2 3 0 5 0 0 5 2 2.1 1 2 0 2 0 2 7 4 2 9 2 3 3 3 6 15 7 6.0 7 3 4 5 4 3 4 3 4.1 16 5 7 8 7 9 19 10 10.1 16 4 11 7 3 17 18 22 12.3 0.0032 0.0027 0.0014 0.0039 Semi-quantitative scoring of pathology taken from Vordermeier et al (2002), Infect. Immun. 70(6):3026-3032 Semi-quantitative scoring of CFU from cultured tissues courtesy of A.O.W helan. 2. IFN- responses There was no difference in mean antigen specific interferon- production between the INH treated group and the control group (FIG.2). However, individual results indicated that there was a tendency in some INH treated animals towards lower levels of specific IFN-production SID 5 (Rev. 3/06) Page 7 of 27 FIG. 2. Mean IFN-gamma responses of INH-treated and untreated animals. Data expressed as the mean OD450 responses (+ SEM), with OD values of medium control wells subtracted from OD values from wells stimulated with the indicated antigens. Mean IFN responses INH treated animals 1.0 OD 450nm Mean IFN responses untreated animals CFP10/ESAT6 PPDA PPDB 0.5 0.0 1 -0.5 2 3 4 5 6 7 8 9 1.5 CFP10/ESAT6 PPDA PPDB 1.0 OD 450nm 1.5 0.5 0.0 10 1 Weeks post infection 2 3 4 5 6 7 8 9 10 Weeks post infection -0.5 3. Liver function markers. The key liver function markers aspartate aminotransferase, gamma glutamyl transferase and total bilirubin were measured using kits available from Olympus UK Ltd (London, UK). It was found that in all the animals, levels of these markers remained constant and within normal limits, showing that none of the animals were experiencing hepatotoxic side effects of INH treatment (data not shown). 4. INH clearance. HPLC measurement of plasma INH levels during treatment and 3 days prior to post mortem showed that the daily dosing regime maintained INH in the circulation during treatment, and that all INH was cleared from the system within 1 week of treatment cessation (FIG.3). Consequently, a 4-week resting period between treatment cessation and re-infection was sufficient to ensure that no residual INH affected the bacilli used to re-infect the calves in Obj. 3B (see below). FIG. 3. Pharmacokinetic of INH-treatment of cows. Serum INH levels were measured at the beginning of treatment (week 5) during treatment (week 8) and one after cessation of therapy. Data from representative animal is shown. sample data from one animal circulating INH g per ml 5 4 3 2 1 14 w ee k 8 ee k w w ee k 5 0 Experiment 3B (VLA/IAH). INH-assisted protective immunity against re-infection with M. bovis. Abstract. The aim of this experiment was to assess whether INH-treatment of acute M. bovis infection as a regime for generating protective immunity to re-infection, and to investigate possible immunological correlates of protection as well as determining the performance of the BOVIGAM assay in animals presenting with no or minor gross pathology and low bacillary loads. 24 calves (3 groups of 8) were used in this experiment. 2 groups of 8 calves were infected with 350 CFU M. bovis (spoligotype 9, sp. 9) at week 0. From 3 weeks p.i. all 16 M. bovis (sp.9)-infected calves received treatment with INH (25mg/kg/day) which lasted for 14 weeks (i.e. until week 17 of the experiment). Animals were then taken off INH treatment and rested for 4 weeks. At week 22 of the experiment, one group of M. bovis (sp.9)-infected/INH-treated calves, and the third group of 8 naïve calves were challenged with 1000 CFU M. bovis (sp.35). All 24 animals were skin tested 1 week prior to necropsy at week 34. Immunological responses (IFN-, IL-4, IL-10, TNF-) were monitored. In addition PBMC were collected for mRNA extraction to investigate IL4 splice variants, recently described in cattle, and correlated with immune protection in humans infected with M. tuberculosis [3-5]. Serum and plasma samples were also collected as above. Detailed post mortem examinations, microbiological culture of tissues and histopathological analyses were carried out on all animals. We could show that BOVIGAM responses contract with treatment/reduction in bacillary load, but animals present with positive results at most time points, i.e. would still be detected by BOVIGAM assay even at low or undetectable bacillary loads, that primary infection confers significant degree of protection against reinfection, that FN- is a marker of bacterial load/pathology, and that the IL43 splice variant constitutes a surrogate of protection. SID 5 (Rev. 3/06) Page 8 of 27 Results. 1. Skin test, pathological findings and M. bovis recovery from PM tissues All 16 INH treated animals were skin test positive at week 32, immediately prior to post mortem, as were the 8 type 35 challenge controls. Animals treated with INH and rested (Group A) showed significantly less pathology than the challenge controls (Group C, table 5). Group B animals (INH-treated animals re-challenged with sp. 35) also presented with substantially reduced gross pathology compared to the sp. 35 infected control group C (table 5). Thus INH-treatment of M.bovis infected cattle resulted in the development of protective immunity. M. bovis was cultured from tissue collected and the spoligotypes of the recovered organism established: Visibly lesioned animals in group B were either positive for type 9 M.bovis only (ie were fully protected against sp 35 rechallenge), for type 35 alone (ie chemotherapy had cleared the sp. 9 primary inoculum but animals were not fully protected against sp. 35 re-challenge), or harboured both type 9 and type 35 M. bovis . Table 5. Gross pathology Pathology Scores Treatment Calf Lymph Nodes Lung Subtotal Gp.A. M.bovis type 9 + INH alone 3375 3376 3377 3378 101233 201946 600672 400770 Total 1 1 0 1 1 0 0 2 6 0 0 0 0 1 0 0 5 6 1 1 0 1 2 0 0 7 12 Gp.B. M.bovis type 9 + INH type 35 challenge 3371 3372 3373 3374 500177 701232 700946 701633 Total 1 0 0 (culture +ve) 0 1 0 1 3 4 6 0 0 2 0 3 0 12 9 p value (Mann-Whitney) Gp.A-Gp.B = 0.597, nsd 1 0 0 4 10 0 2 3 20 Gp.C. M.bovis challenge control, type 35 alone 3421 3422 3423 3424 101938 401139 601625 SID 5 (Rev. 3/06) 5 0 0 6 6 3 0 3 1 0 9 5 0 3 Page 9 of 27 8 1 0 15 11 3 3 701612 5 12 17 Total 25 33 58 p value (Mann-Whitney) Gp.A-Gp.C = 0.04 statistically significantly different p value (Mann-W hitney) Gp.B-Gp.C = 0.126 nsd ( ) indicates the numbers of areas of the lung affected 2. INH clearance HPLC analysis for both INH and its metabolite N-acetyl INH (revealed through treatment of the plasma sample with HCl at 800 C prior to derivatization) showed INH to have been completely cleared from the circulation prior to infectious rechallenge at week 22 (FIG.4). FIG.4. Pharmacokinetics of INH in serum. Results from all 8 calves that were subsequently re-challenged with M. bovis artweek 22 are shown. total circulating INH + N acetyl INH g per ml INH g per ml 20 3371 3372 3373 3374 3375 3376 3377 3378 15 10 5 0 0 5 10 15 20 weeks post infection 3. Cytokines a) IFNCytokine assays for IFNshowed a similar pattern of fluctuation with time to that seen in previous experiments. In all four groups, specific responses developed within three weeks of infection (FIG.5.). These responses gradually cionytracted during INH treatment until week 17 when treatment was stopped. Subsequent to treatment cessation, a sharp increase was observed in some animals in all groups, suggesting that cessation of treatment was allowing mycobacteria to resume growth (homologous rechallenge). Interestingly, INH-treated animals that were not re-challenged but presented without gross pathology (NVL) exhibited a peak of IFN- responses both to PPD-B and ESAT-6/CFP-10 shortly after treatment cessation (Fig. 5, upper right panel) that was not as evident in the corresponding animals that showed gross pathology (VL, Fig. 5, upper left panel). This suggests that an active memory immune response post-treatment cessation - generated during the INH-treatment - is required for effective chemotherapy. INH-treated calves that were re-challenged with the heterologous M.bovis strain (sp. 35) at week 22 could also be grouped into animals with and without gross pathology (VL and NVL animals, respectively). In both groups of animals we could observe the brief response peak following treatment cessation (ie prior to re-challenge), although the peak was less pronounced in the VL group (Fig. 5 lower left panel) than for the NVL animals (Fig. 5, lower right panel). Interestingly, following sp. 35 re-challenge only the NVL animals developed a further IFN- response peak, mainly to ESAT-6 and CFP-10 stimulation followed by a generally downward trend over the postrechallenge period (Fig. 5, lower right). In contrast, the ESAT-6/CFP-10 peak following re-challenge in the VL animals was less pronounced and closely followed the PPD-B responses. Further, both responses did not decrease over time but remained stable or followed an upward trend (Fig. 5, lower left). As ESAT-6 and CFP-10induced IFN- responses correlate with bacterial load and the severity of pathology we propose that these response peaks after INH-treatment cessation, and following re-challenge in the ‘fully protected’ (ie NVL) animals represent the reduction of bacterial loads as consequence of anamnestic and protective immune responses [3]. FIG.5. Kinetics of in vitro IFN- responses as measured by BOVIGAM assay. Results are presented as mean OD450 (+ SEM) after stimulation with PPD-B or ESAT-6/CFP-10 with medium background levels subtracted. Upper panels: Group A animals (INH-treatment, but no re-challenge). VL animals, left panel; NVL animals, right panel. Lower panels: Group B animals (INHtreated, then sp. 35 re-challenged). Left panel, VL animals; right panel, NVL animals. SID 5 (Rev. 3/06) Page 10 of 27 Optical Density 450nm 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 CFP10/ESAT6 PPDB 10 20 30 40 weeks PI 20 30 Optical Density 450nm Optical Density 450nm 10 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 10 20 30 40 weeks PI INH mean Bovigam data type 35 challenge NVL (n=2) INH mean Bovigam data type 35 challenge VL (n=6) 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 optical density 450nm Mean Bovigam data NVL no challenge (n=3) Mean Bovigam data VL no challenge (n=5) 40 weeks after first infection 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 10 20 30 40 weeks after first infection INH INH Challenge week 22 Challenge week 22 b) IL10 cytokine assays using five day whole blood culture supernatants showed that a peak of specific IL10 activity was always detected within two weeks of commencement of INH treatment (FIG.6). Interestingly, following M. bovis sp. 35 challenge of un-treated animals we did not observe a corresponding IL-10 response (data not shown), nor was an IL-10 response induced in the INH-treated animals following M. bovis sp. 35 re-challenge (data not shown). This observation can be interpreted in several ways. Firstly, the presence of a specific IL-10 response in protected animals before re-challenge could suggest that IL-10 constitutes a potential marker (surrogate/correlate) of protection, for example T regulatory cell produced IL-10 could be involved in the formation of solid T cell memory. Equally feasible is the interpretation that T regulatory cell-produced IL-10 is acting a antiinflammatory agent counteracting the effects of IFN- that would otherwise contribute to immunopathology [7]. Only further experiments will allow us to assign a function to these IL-10 responses. FIG.6. Mean specific IL10 (n=16) responses. IL-10 in culture supernatants was determined by ELISA, IL-10 production from cultures stimulated with either PPD-B or a peptide cocktail of ESAT-6 and CFP-10, with medium background values subtracted, is expressed as mean units (+ SEM). Shown are results of all 16 INH-treated animals up to the time of rechallenge (week 22). No further IL-10 production was observed post-rechallenge (not shown). 25 PPDB ESAT6/CFP10 Units of IL10 20 15 10 5 0 2 4 -5 6 8 10 12 14 16 18 20 22 weeks post infection start INH treatment INH stopped c)TNFand IL-4cytokine ELISA assays did not yield any data that could usefully be correlated with either INH treatment, pathology or culture data. d) IL4 splice variants The presence of 2 IL4 splice variants IL42 and IL43 were assessed in bovine ex vivo un-stimulated PBMC. Results from cattle from SE3024 Objective 3B (INH-assisted protection model) were compared with results from BCG-vaccinated cattle, naïve controls and naturally infected field reactors. mRNA was quantified using paired primers and taqman probes specific for IL42 and IL43. GAPDH was used as the house-keeping gene to which cytokines were normalised in order to remove differences in the amounts of mRNA within each sample (see Technical Annexe for complete methods). Firstly, we investigated field reactors versus naïve controls for the presence of the splice variants in naturally infected populations. Field reactors were obtained from multi-reactor breakdowns with culture-confirmed bovine tuberculosis. Dividing the IL43-positive field reactors (which were also ESAT-6 and/or CFP-10 as well as PPD-B IFN- positive) into animals with and without gross visible pathology (VL, visible lesion and NVL, non-visible lesion, respectively) (Fig.7) revealed that the highest expression of IL43 was found in the NVL group. IL43 was significantly higher in this group of compared to the VL group (Mann SID 5 (Rev. 3/06) Page 11 of 27 Whitney p=0.0091). Thus a high IL43 mRNA expression appeared to correlate with the absence of pathology in cattle naturally infected with M. bovis. FIG. 7. IL43 expression in NVL and VL field rectors. Results are expressed as mean IL43 expression/unit GAPDH (+ SEM). IL43-positive field reactors * IL43/unit GAPDH 20 15 10 5 0 NVL VL pathology status We next investigated IL4 splice variants in cattle protected from virulent M. bovis challenge by the INH-treatment described above. Increased IL43 was observed following M. bovis (sp.9) infection and INH treatment from week 13, and this reached statistical significance on weeks 21 and 25 (Mann Whitney p=0.0159 and p=0.0317 respectively) (Fig. 8). As expected, increased IL43 was also observed in the group B animals before rechallenge (Fig. 9), compared to group A animals (that were INH-treated but not re-challenged). After re-challenge, IL43 responses remained higher in group B comparEd to group A animals, although this difference in responses was not significant (Fig. 9). Similar increases in IL43 expression were also observed in BCG vaccinated animals (not shown, as not part of SE3024). In conclusion, therefore, ex vivo expression of IL43 constitutes a potential surrogate for protection. FIG.8. IL43 expression during INH-attenuated M. bovis infection in group A animals. Results of individual animals are presented with horizontal bars indication median responses. IL43/unit GAPDH 100 * * 10 1 0.1 0.01 0.001 INH 0 0 3 7 13 21 25 30 34 Mb. sp.9 Fig. 9. IL43 expression during INH-attenuated M. bovis infection in animals from all three groups. For clarity and ease of making comparisons between groups, only median values are shown. Group A, SP9/INH (blue); group B, SP9/NIH/Sp35 (red), group C, Sp35 infection only (green). SID 5 (Rev. 3/06) Page 12 of 27 Interestingly as discussed above, it was IL43, and not IL42 that appeared to be the dominant splice variant in these investigations with un-stimulated PBMC. We did note however that stimulation of PBMC with antigen (PPDB) resulted in different patterns of expression, often with strong IL4 and IL42 responses but weak IL43. Figure 10 below shows the mean PPDB-specific responses in M. bovis-infected, BCG-vaccinated and naïve cattle. FIG. 10. PPDB-specific cytokine induction in cultured PBMC after in vitro stimulation with bovine tuberculin( PPD-B). Compared are responses of M. bovis sp. 9 infected animals, BCG vaccinated calves as well as uninfected/unvaccinated controls (naïve). fold increase mRNA 10 IL4 IL42 IL43 8 6 4 2 0 INFECTED BCG NAIVE Differences in the kinetics of bovine splice variants have been noted previously [8], and so taken together the data suggest differential activation and possibly also different functions for IL42 and IL43 in bovine tuberculosis. We are now able to pursue this hypothesis of differential splice variant activity in cattle. 4. Potential protective antigens. INH-treatment conferred a significant degree of protection against re-challenge, which provided us with an opportunity to compare antigen profiles between treated and sp. 35 re-challenged cows (protected animals) and animals that were only infected with sp. 35 (unprotected controls). In previous experiments we had shown that BCG vaccinated animals that were subsequently challenged with M. bovis developed rapid, anamnestic responses about 2 weeks post-challenge, at time points where responses in unvaccinated control animals had not developed [6]. We hypothesised that antigens recognised at this early timepoint post-infection would therefore constitute protective antigens. As discussed above, we observed similar response peaks after re-challenge. Therefore, we evaluated a panel of 20 antigens for their capacity to induce IFN- responses in INH-treated and re-challenged cattle and compared their responses to untreated control animals only infected with sp. 35. These experiments were undertaken 2, 3, and 4 weeks post-challenge with sp. 35 (ie at weeks 24, 25, 26). The results are presented in Fig. 11, and demonstrate that 2 weeks post-infection very few antigens are recognised in the control animals, whereas at least 3 antigens induced IFN- responses in the treated animals (antigens indicated by red boxes in Fig. 11), whilst responses against these and other antigens could be demonstrated in both groups at the later time points. These three antigens are now being evaluated for their ability to act as subunit vaccines in protecting cattle against bovine TB (as part of project SE3224). Fig. 11. IFN- response after stimulation with a panel of antigens. Responses are expressed as mean IFN- production in pg + SEM. Potential subunit vaccine candidates are indicated by red boxes. Definition of a potential protective antigen: responses 2 weeks post-challenge in protected cattle (M. bovis type 9 infectino, INH-treatment, re-challenge with sp. 9, bottom panel), no responses as 2 weeks post-challenge in control animals (M.bovis sp. 35 infection only, top panel). Please note that the identity of the tested antigens is being withheld to protect IP. SID 5 (Rev. 3/06) Page 13 of 27 Objective 4. To establish and validate an M. bovis aerosol challenge model for cattle (VSD component, report prepared by Dr Jim McNair and edited into the body of the complete report by the project co-ordinator, Dr Martin Vordermeier). Bovine tuberculosis is primarily a disease of the respiratory tract and the main route of transmission, is aerogenous. The typical patterns of disease, as seen in field cases, are predominantly within the lower respiratory tract, and to a lesser extent in the upper tract, with a smaller number of cases observed through disclosure of lesions in the mesenteric lymph nodes. The dose of M. bovis considered to be infectious is thought to be very small although this has not been defined experimentally. The prime objective of this experimental series was to combine the aerogenous route of infection with a defined number of colony forming units to clarify the infectious doses required to initiate infection. Objective 4 comprised three distinct phases. The first phase was based on the preparation of unicellular cultures of M. bovis with defined colony forming units. The second phase was based on developing expertise with the modified Madison aerosol chamber and to define particle size in aerosol suspensions and M. bovis survival curves. The third phase was the application of the chamber for the delivery of defined aerosolised doses to naïve cattle followed by measurement of immune responses and definition of the pathological picture. As the procedures used in this objective are direct outputs of this project, and integral to the success of these experiments, we will describe them in the context of the individual experiments in detail rather than listing them in the methodological annexe. 1. Development of the aerosol model to deliver M. bovis to cattle, and determination of particle sizes. a) Preparation of unicellular M. bovis stocks. Our approach is based on the growth of M. bovis with disruption of multi cellular clumps and cell enumeration. M. bovis strain AF2122/97 was used throughout all experiments. 100ml volumes of 7H9 broth (0.22um filter sterilised) were seeded with 1ml stock culture and incubated for 7 days at 37oC. 10ml of the 7H9 broth were transferred to a 7H11 agar flat and allowed to rest for one hour before the addition of 100 ml 7H9 broth. At 14 days post inoculation, cultures were rocked gently to remove the surface bacterial growth after which supernatants were pooled. 40ml aliquots of live cultures were transferred to 100ml vaccine bottles complete with a sealed septum and passed through a 23-guage needle a total of ten times. The cell suspension was then filtered through a 5µm PVDF filter and stored (frozen at -70oC) in 2ml aliquots. Examination of these cultures by Ziel-Nielsen staining revealed a predominance of single cells, with only a few, small clumps present. b) Quantificaton of aerosol delivery. M. bovis aerosolised by nebulisation and induced into the airstream of the Madison chamber was sampled using an all-glass-impinger (AGI) system to trap bacteria and to allow enumeration. This system also allowed definition of the M. bovis curve generated by the aerosol stream. With the Madison chamber operating normally and delivering M. bovis to the airstream, air samples were taken at various SID 5 (Rev. 3/06) Page 14 of 27 time points for culture to estimate CFU delivered. Titration studies based on M. bovis cultures with defined CFU indicated a strict relationship between the numbers of M. bovis added to the nebuliser, the generation of a stream containing aerosolised bacteria and cell survival over a defined period. In addition, the decrease in cell viability pre and post nebulisation was minimal. Furthermore, our studies indicated that the particle size generated by the nebuliser was in the range 0.8 to 15µm, a range which is compatible with M. bovis in single cell suspensions. c) Delivery of non-infectious aerosols, estimation of particle sizes. In order to gain confidence in the operation of the Madison chamber while delivering an aerosol stream to calves, a number of experiments were carried out using the chamber, to deliver a non-infectious aerosol to calves. Originally, it was envisaged that calves would require sedation using Rompum in order to apply the mask over the muzzle for the length of time required to deliver an infectious dose by aerosol. However, in a series of preliminary experiments, it was evident that calves, when treated quietly and gently, would accept the mask for up to five minutes. Recordings taken from a pneumotachometer also indicated that there was no respiratory distress associated with the mask during this time period. Pneumotachometer data gave extremely accurate and consistent volumes of inspired and expired air over a given time period. Coupled with the data generated by M. bovis nebulisation experiments, we were confident that we were able to accurately deliver defined infective doses of M. bovis to individual calves. Objective 4, milestones 04/04, 04/05, 04/06: Aerosol delivery of M. bovis to calves, dose titrations Exp. 1. Aerosol model tested with low dose (10 4 CFU) Exp. 2. Aerosol model tested with low dose (10 2 CFU) Exp. 3. Aerosol model tested with low dose (101 CFU) These experiments were designed to assess the effect of delivering aerosolised M. bovis at various doses to the respiratory tract of cattle, and to assess the effect of dose on immunology and pathogenesis while using the natural route of infection. This experimental series was based on groups of five calves each experiment, all of which were purchased from farms with a TB-free history over the previous five years. All calves were Friesian cross bred, castrated males aged between 4 and 6 months. Each infection was carried out according to the same protocol. Calves were restrained individually in a cattle crush and a member of staff placed and held the mask in position. After a short period of familiarisation, the aerosol stream carrying M. bovis was directed to the mask for a period during which 100 litres of air containing the infectious dose was inhaled. The mask was kept in position for a further period of two minutes to allow M. bovis exhaled from the lungs to be expelled and trapped within the HEPA filters linked to the aerosol chamber. All calves were treated similarly within a thirty minute period, following which, the chamber was removed from the animal accommodation for fumigation and decontamination. Over a period of nine months, animals were kept and were blood sampled on a regular basis. Immunological tests were carried out either immediately or, for antibody analysis, samples were stored at -20oC until assayed. One week before post mortem examination, calves were skin tested using conventional PPDA and PPDB reagents and interpretation of results. A detailed post mortem examination was carried out when all the tissues and lymph nodes associated with the upper and lower respiratory tract were examined and sampled for mycobacterial culture. a) Infectious dose delivered by aerosol chamber. Prior to each experimental infection, the infectious dose was calculated from stock cultures of known CFU and titration curves derived from aerosol experiments. Table 6 summarises the mean values calculated for each experiment. When airstream samples taken during the 10 1 CFU infection and cultured, no bacterial growth was recorded on any plates. This failure to support bacterial growth meant that the infectious dose given to this group of calves could not be verified. Table 6: Mean values of infectious dose and inhalation time, calculated for each group of calves (5 calves per group) infected by aerosolised M. bovis using the modified Madison chamber. Experiment Intended dose Inhalation time Actual dose Range (Secs) 1 104 CFU 170 7.68 (0.29) x 104 7.2– 7.9x104 2 2 10 CFU 147 226 (21.9) 101 - 341 3 101 CFU 169 Unknown* NA * Failure to culture M. bovis on this batch of agar plates meant that this infectious dose could not be calculated There was a high degree of uniformity, within each experimental infection, of the dose delivered to the respiratory tract of calves. This uniformity was due to a combination of accurate colony forming units defined in stock M. bovis cultures and the use of a pneumotachometer to measure the volume of air inspired. b) In vitro cell mediated responses following infection with M. bovis. Before and after experimental infection, all calves were blood sampled and tested for IFN- release using whole blood cultures stimulated with various complex and recombinant antigens. Supernatant fluids were harvested and tested for the presence of IFN- using the Bovigam test kit. Pre-infection, all cattle were found to have negligible IFN- responses. When tested post SID 5 (Rev. 3/06) Page 15 of 27 infection using PPDB, there was no difference in the onset of the mean IFN- response in each of the three groups. However, in general terms, the larger the dose of M. bovis delivered, the greater the IFN- response seen post infection (Figs. 12-14). Within the 104 and 102 CFU groups all animals remained positive after three weeks post infection. However, in the group with the lowest infectious dose (10 1 CFU) there were inconsistent responses throughout the post infection period. All animals were positive at various time points but only two animals (2042-1 and 2275-5) were consistently positive post infection. IFN- responses to ESAT-6 were also greater with the larger infectious doses, when judged by the mean group response. However, in both the 102 CFU and 101 CFU groups there was a large degree of variation observed. In the 102 CFU infection group, two animals (2237-2 and 2242-7) were consistently negative to ESAT-6 and in the 101 CFU group most animals on most occasions remained unresponsive to ESAT-6. Immune responses to CFP10 were very similar to ESAT-6 but not identical. In the 104 CFU group, the mean response post infection, was positive throughout the post infection period with the exception of four individual time points. On three of these occasions, if an animal was negative to CFP10, it was positive to ESAT-6. The IFN- response to CFP10 in the 102 CFU group was also consistent, with four calves responding to this antigen post infection while one animal (2239-4), did not respond at any time post infection. Generally, all animals were unresponsive to CFP10 in the 101 CFU group with the exception of three animals on only one occasion. This M. bovis titration series has shown that as the infectious dose is lowered, the more inconsistent the immune response to the two recombinant antigens becomes. c) Skin testing with conventional PPDA and PPDB. Two animals infected with 104 CFU became ill during the experiment and were euthanized on animal welfare grounds during weeks 13 and 19 post infection and were not skin tested. All of the other cattle were skin tested using PPDA and PPDB by intra dermal injection, with skin thickness measured at 0 and 72 hours post inoculation (table 7). All calves in the 10 4 CFU and 102 CFU groups that were skin tested were positive using test interpretation criteria as applied under field conditions. The mean increase in skin thickness for each group was very similar, 17.2 and 13.8 mm respectively. In contrast, from the five cattle tested in the 101 CFU group, only one (2275-5) was skin test positive. Coincidently, this animal was also M. bovis culture positive (cranial mediastinal lymph node) and had the greatest responses to PPDB by IFN- release. This animal had also intermittent but inconsistent IFN- responses to ESAT-6 and CFP10. Table 7. Summary of the skin test results from each experimental infection (10 4, 102 and 101 CFU), based on the increase in skin thickness (PPDB – PPDA) at 72 hours post intradermal injection with PPDA and PPDB. Experiment 1 2 3 Infectious dose 7.68 x 104 226 (21.9) Unknown* Number tested 3 5 5 Number positive 3 5 1 Mean increase (mm) 17.2 13.8 1.3 Range (mm) 9-26.5 7-27 -1 to 5 d) Post mortem examinations. All calves were given a detailed examination post mortem, to identify and record the presence of tuberculous lesions typical of M. bovis infection. The presence and size of lesions were recorded and used to calculate a post mortem score for each tissue or lymph node examined and which was used to compare individual animals or groups (table 8). Samples of tissue, irrespective of the presence of lesions, were taken for bacteriological culture. For each infectious dose given, there was a very different pathological picture as defined by the distribution of lesions. All calves given 10 4 CFU were found to have tuberculous lesions that were confined to the lower respiratory tract with the exception of calf 1952-2 where a single lesion was found in the left palatine tonsil. All lung lobes were found to have lesions with the exception of calf 2246-2, were the middle and accessory lobes were free from lesions. Lesions were found in the cranial and caudal mediastinal lymph nodes of all calves, however, only two calves (1952-2 and 1953-3) were found to have lesions in the cervical lymph node. The post mortem score in this group ranged between 16 and 48, with the mean value calculated at 33.8 (SD = 12.2). Table 8. Comparison of the post mortem scores calculated after each experimental infection (10 4, 102 and 101 CFU). Experiment Infectious Number Number Mean PM score Range dose examined lesion (mm) positive 1 7.68 x 104 5 5 33.8 (12.2) 16-48 2 226 (21.9) 5 3 3.6 (3.9) 0-9 3 Unknown* 5 0 0 0 By reducing the infectious dose to 10 2 CFU, a much more restricted pattern of lesion development was observed. No lesions were found in two calves (2022-2 and 2247-4) at gross post mortem examination, while three calves were found to have lesions in the left and right bronchial, cranial and caudal mediastinal lymph nodes. Only two SID 5 (Rev. 3/06) Page 16 of 27 lung lobes were found to be lesioned, the left cranial lung and the right caudal lung. All other lung lobes were found to be lesion free. The mean post mortem score for this group was 3.6 (SD = 3.9) At the lowest infectious dose (101 CFU) no lesions were disclosed in any of the calves exposed, in any of the upper or lower respiratory tract tissues examined. All tissues examined were of a normal appearance and size. The only exception was the palatine tonsils which were seen to have discreet, pus filled areas which were culture negative for M. bovis. The post mortem score for these calves, individually, and for the group was 0. At the time of writing, one tissue sample (the cranial mediastinal lymph node from calf 2275-5) was found to be culture positive for acid fact bacilli and later confirmed as M. bovis AF2122/97 by spoligotyping and VNTR analysis. The presence of this isolate, albeit from only one animal, validates the infectious dose given to this group and helps to explain the immune responses generated following exposure. There is a close association between the higher infectious dose (104 CFU), the pattern of lesion development and the immune responses measured. All five caves were lesioned, all were skin test positive and all had strong, persistent in vitro IFN- responses to PPDB, ESAT-6 and CFP10. In contrast, infections with either 102 or 101 CFU were characterised by a reduction or absence of lesions, and, inconsistent in vitro immune responses. When 102 CFU was given, the two calves without lesions were skin test positive, had strong in vitro IFN- responses to PPDB and although seen at a much lower level, IFN- responses to ESAT-6 and CFP10 were also present. In contrast, the IFN- response to CFP10 was absent in an animal that was lesioned. The greatest degree of variation was seen in the group given 10 1 CFU were only one animal was skin test positive. This calf was also culture positive for M. bovis. All animals in this group were positive to PPDB at a number of time points post infection, however, most animals were negative to both ESAT-6 and CFP10 for most of the post infection period and were only positive on an intermittent basis. There is an implication that at this infectious dose, animals are more likely to be positive by PPDB than by recombinant antigens. This pattern was also seen in the animal which was culture positive, when PPDB induced responses to a greater extent than responses provoked by either ESAT-6 or CFP10. e) Bacteriological culture. There was evidence of spread from the lower to the upper respiratory tract following culture of tissue samples taken from calves infected with 104 CFU. In two calves, M. bovis was isolated from the upper trachea. In a third animal, samples from the lower trachea and the left medial retropharyngeal were culture positive. In a fourth animal the right and left palatine tonsil, the left pharyngeal tonsil and the nasopharynx were culture positive. There was no evidence of disease spread from the lower tract in calves infected with 10 2 CFU. In this group, M. bovis was confirmed in lesioned tissue samples in the three calves where lesions were recorded. However, M. bovis was cultured from one animal (2022-2, caudal mediastinal lymph node) where no lesions were recorded) but not isolated from any tissue samples taken from animal 2247-4, which was not lesioned. At the time of writing, only one M. bovis culture positive tissue (caudal mediastinal lymph node taken from animal 2275-5) has been recorded in the group infected with 101 CFU. All cultured samples taken from this group will be kept for a total of 86 days before recording the final result. Conclusions This work highlights the value of using the natural host for M. bovis infection models and the importance of using a natural route of infection to interrogate the effects of dose on infection establishment, pathogenesis, host immunity and diagnosis. This approach, intended to mimic tield type infections, will give invaluable insight into naturally acquired M. bovis infection by using a natural route of infection in cattle. In particular, the lower infectious doses induced an immune state more akin to frequently observed in the field, where the relationship between disease status and immune response can be complex. This model can be exploited to explore these complexities and to improve our diagnostic capabilities. Discussion of data in objective 4 in relation to results from objective 1. Interestingly, and seemingly in contrast to the intratracheal infection model, a reduction in pathology was observed with decreasing infective doses (i.e. 10,000 CFU compared to the two lower doses). However, not the same dose titrations were performed in both infection models and we could show in previous studies with the intratracheal challenge model that more severe disease and pathology resulted after infection with of > 10,000 CFU than with the lower doses applied here (< 1000 CFU). Therefore, the results between high and low doses (>10,000 CFU and below 1000 CFU) are comparable between the intratracheal and the aerosol challenge routes. The dose range between 100 and 1000 CFU need to be evaluated in the aerosol model to allow full comparison with the intratracheal model. Further, aerosol infection with 10 CFU resulted in no visible pathology, although the exact infective dose could not be determined in this experiment. However, at face value, these findings seem to be in contrast to the intratracheal model where doses between 1 and 10 CFU resulted in visible pathology. However, even intratracheal infection with 1 and 10 CFU did not result in visible pathology in all animals (for example only 3/6 animals infected with 1 CFU presented as VL animals, or 5/6 infected with 10 CFU), furthermore the severity of pathology induced after intratracheal challengewith 10 or 100 CFU compared to that induced after aerosol infection with 100 CFU was of similar range and not significantly different (1-12 with intratracheal infection at these doses compared to 0-9 with 102 CFU delivered by aerosol). Thus, and taking into accout the relatively small numbers of animals that we were able to use in these experiments, it is difficult to conclude unambiguously that fundamental differences exist between these two models at these low and limited infection doses. Further studies will be required to determine this without ambiguity. One obvious theoretical difference between the two SID 5 (Rev. 3/06) Page 17 of 27 models is that the intratracheal route delivers the infective inoculum concentrated to the same site in the lung (right cranial lobe), which from previous studies we deem to be highly susceptible. Therefore such aninfection results in a more uniform disease pattern which is a signficiant advantage when performing, for example vaccination studies, where financial constraints necessitate that all animals develop disease. The aerosol route will deliver the dose dispersed over the whole lung, i.e. it is possible that areas of the lungs are targeted that are not equally susceptible to infection, therefore resulting potentially resulting in less uniform disease development. However, more studies will be needed to address this hypothesis. SID 5 (Rev. 3/06) Page 18 of 27 Figure 12. Mean IFN- responses to PPDB, ESAT-6 and CFP10 following infection with 104 CFU M. bovis. SID 5 (Rev. 3/06) Page 19 of 27 Figure 13. Mean IFN- responses to PPDB, ESAT-6 and CFP10 following infection with 102 CFU M. bovis. SID 5 (Rev. 3/06) Page 20 of 27 Figure 14. Mean IFN- responses to PPDB, ESAT-6 and CFP10 following infection with 101 CFU M. bovis. SID 5 (Rev. 3/06) Page 21 of 27 Project SE3024: Summary of main findings and suggestions for further research A. Main results: Intratracheal and aerosol routes of experimental challenge routes are the most appropriate routes because they best reflect the lesions most commonly associated with natural infection (in GB). The lower respiratory tract is highly susceptible, infective doses of 1 CFU delivered directly to the lung by the intratracheal route will result in pathology and disease in the lung and/or associated lymph nodes. Taking results of objectives 1 and 4 together, one conclude that the predominant disease phenotype seen in GB is caused by aerosol infection of small numbers of bacilli delivered by small aerosol particles (3-5 organisms) to the lung. The observations above therefore will allow a more informed and precise analysis of transmission risks, and of the virulence of M. bovis infection of cattle. BOVIGAM responses contract with reduction in bacillary load, but animals present with positive results at most time points, although results using the lowest infective dose delivered by aerosol showed a more variable response with the defined antigens (ESAT-6/CFP-10) not performing consistently as well as tuberculin. Animals infected with this dose and route also at certain time points would have escaped detection by the BOVIGAM test. Using low infective doses or chemotherapy, it is possible to produce, apart from VL/Cu+ animals, also animals that are NVL/cu+ or NVL/cu- yet positive in skin test and IFN- test: Such animals are therefore not ‘false-positives’ but infected sub-clinically and could pose future infection risks to herds. This is particularly relevant in the case of new outbreaks in otherwise non-endemic regions of GB, where the rapid removal of any infected animal is a priority. Furthermore, it emphasises also the relevance of using alternative diagnostic tests like the BOVGIAM IFN-test in this scenario to maximise the identification and removal of the number of infected cattle. This conclusion is also supported by results from objective 4 where animals infected with the lowest infection dose not only presented as NVL/culture-negative (ie could be considered as latently infected) but were also skin test negative yet detectable using the BOVGIAM test. Primary infection confers significant degree of protection against re-infection. This protection was against a heterologous M. bovis strain (cross-protection), which is encouraging for vaccine development using BCG or one particular attenuated M. bovis strain. Successful chemotherapy requires active immunity. IFN- has been confirmed as a marker of bacterial load/pathology. The IL43 splice variant constitutes a surrogate of protection. Memory cow model can also be used to prioritise potential protective antigen for subunit vaccine development, and 3 candidates have been identified in this experiment. Aerosol infection was intended to mimic field type infections, and gave invaluable insight into naturally acquired M. bovis infection by using a natural route of infection. In particular, the lower infectious doses induced an immune state more akin to frequently observed in the field, where the relationship between disease status and immune response can be complex. This model can be exploited to explore these complexities and to improve our diagnostic capabilities. Results from project SE3024 have also cross-cutting relevance for Defra’s cattle TB vaccine development programme. These cross-cutting research results are indicated in this summary by italics. B. Possible future work: Investigations into the nature and degree of latency of TB in cattle, both by field and experimental studies. To investigate the functions of IL-4 splice variants in the regulation of immunity in cattle. To determine the protective potential of subunit candidates identified. SID 5 (Rev. 3/06) Page 22 of 27 Technical Annexe: materials and methods used throughout this project (objectives in brackets, methods specific to objective 4 are discussed in the context of the results section above). Mycobacterial antigens. Tuberculin PPD preparations from M. bovis (PPDB) and from M. avium (PPDA) were produced at the Veterinary Laboratories Agency (VLA) of Weybridge as described previously. Recombinant ESAT6 and CFP10 were obtained from Dr. Singh, GBF, Braunschweig, Germany, peptide cocktails representing ESAT-6 and CFP-10 (Vordermeier et al., 2001) were prepared from individual peptides prepared by Pepscan Ltd, Lelystad, the Netherlands. Experimental M. bovis infection (objectives 1 and 3). Friesian-cross females or castrated males were obtained from a herd free of M. bovis infection, as defined by a history of negative skin test results. All animals were also negative in the IFN- diagnostic test for bovine tuberculosis. Calves were infected intra-tracheally using the GB field isolate AF2122/97 (spoligotype 9) in all experiments unless otherwise stated. GB field isolate AF 61-1307-01 (spoligotype 35) was used for re-infection of calves in OBJ.3B. During these studies animals were housed in an high-security isolation unit under negative pressure, and expelled air was filtered. The single comparative tuberculin skin test (SCITT) was performed on all animals at the time points stated in the text for each objective above. Euthanasia was carried out by intravenous injection of sodium pentobarbitone, and a detailed post mortem analysis performed as described below. For the lymph nodes, each of the following types of lymph node (or lymph node chain) was removed aseptically: right and left lateral retropharyngeal, right and left medial retropharyngeal, right and left submandibular, right and left cervical superficial, right and left bronchial, crainial mediastinal chain, caudal mediastinal chain, and mesenteric. These were serially sliced (approx. 2mm slices) with a scalpel and inspected. Samples from individual nodes and lesions were obtained for mycobacterial culture and histopathological examination. The remaining superficial and visceral lymph nodes were inspected in situ. For the lungs, each lobe was serially sliced with a sharp knife. All slices were palpated and inspected. The respiratory airways were opened as far into the lung parenchyma as possible and inspected. Isoniazid (INH) treatment of cattle (obj. 3). Isoniazid tablets (isonta/28/100) were obtained from UCB Pharma Limited (Bedfordshire, U.K.). Tablets were ground up using a conventional coffee bean grinder and the correct dose/animal (25mg/kg/day, as determined by individual animal weight [9]) suspended in water and fed orally using a Phillips Cattle Drencher (Cox Surgical, U.K.). Measurement of Plasma INH using High Performance Liquid Chromatography (HPLC, obj. 3). Plasma INH was measured following the methodology of Smith et al [10]. Plasma samples were deproteinated by precipitation with 10% trichloracetic acid prior to derivatization with cinnemaldehyde and neutralization with ammonium acetate. INH in plasma was then measured. In addition, to determine levels of circulating N acetyl INH (a metabolite), a second sample was treated with hydrochloric acid prior to derivatization in order to convert N acetyl INH to INH in the sample. The amounts of INH in each sample was determined by HPLC. Each sample, and a set of standards treated in the same way, at the same time as the samples, were chromatographed on a Jupiter C18 5 column (250 x 4.6 mm) using an isocratic solvent system: 0.1 M ammonium acetate, pH 5.6/acetonitrile/methanol (47/23/30 v/v/v) at 1ml/min. The amount in each standard was determined by reading against a line of peak area against amount of INH plotted for the corresponding set of standards. N.B. Standard for INH assay kindly provided by Celltech Manufacturing Services Ltd., U.K.). Liver function assays (ob. 3). Filter-sterilised (0.2m) serum samples for individual calves were analysed at VLA-Shrewsbury for 3 key liver function markers; aspartate aminotransferase, gamma glutamyl transferase and total bilirubin, using kits available from Olympus UK Ltd (London, UK) Whole blood culture for cytokines (objs. 1-3). Whole blood cultures were performed in 96-well plates in 0.25ml/well aliquots in duplicate by mixing 0.225ml whole blood/well with 0.025ml of antigen-containing solution (at 10X final concentration required). Supernatants were harvested after 24 hours (for IFN- and TNF-) and 5 days (for IL-4 and IL-10) [2]. IFN- assay (all objectives). 24-hour whole blood supernatants were determined using the BOVIGAM enzyme immunoassay kit and following the manufacturers instructions (Biocore, USA). TNF-assay (objective 3B). 24-hour whole blood supernatants were determined using the bovine TNFscreening kit (Endogen). IL-4 ELISA assay (objective 3B). 3-day whole blood supernatants were determined using the bovine IL-4 screening kit (Endogen). IL4/B-cell Bioassay (obj. 1-3). 5-day whole blood supernatants were tested in the B cell bioassay as previously described [11]. Bovine B cells from a naïve skin-test-negative animal were positively sorted from PBMC using the SID 5 (Rev. 3/06) Page 23 of 27 monoclonal antibody ILA58 (IAH, Compton, Berkshire) specific for bovine immunoglobulin (Ig) light chains, which both labels and preactivates B cells by cross-linking the surface Ig. PBMC were then incubated with rat-antimouse IgG2(a+b)-coated microbeads before positive sorting using the MACS column separation system (Miltenyi Biotech, Surrey, UK). B cells were eluted, washed and resuspended to 10 6/ml in culture medium (RPMI 1640 with Glutamax [Gibco] supplemented with 10% foetal bovine serum [Gibco], nonessential amino acids [Gibco], 100U/ml penicillin, 100g/ml streptomycin and 5x10-5M 2-mercaptoethanol), and 0.1ml added/well to 96-well round-bottomed culture trays together with 50l/well of supernatant to be tested, in duplicate. Plates were incubated overnight at 37oC and 5%CO2, pulsed with tritiated thymidine, and incubated for a further 24 hours before harvesting and counting using -scintillation. IL-10 assay (objective 3B). A bovine IL10 antibody pair obtained from Serotec was used for this assay. 96-well plates were coated overnight with bovine anti-IL10 prior to adding supernatant samples. Binding of the biotinylated detection antibody was visualised using horseradish-peroxidase conjugated streptavidin (Sigma) and Pierce Super Signal ELISA femto maximum sensitivity substrate, the resulting light emissions being read on a luminometer. IL10 standard was kindly provided by J. Hope, IAH, Compton. IFN-ELISPOT [12] (objectives 1 and 3). ELISPOT plates (Immobilon polyvinylidene difluoride membranes, Millipore, France) were coated overnight at 4oC with the bovine IFN- -specific monoclonal antibody clone 5D10 (Biosource International, California, USA). Unbound antibody was removed by washing, and the wells were blocked with 10% foetal calf serum in RPMI 1640 medium (Life Technologies, Paisley, Scotland, United Kingdom). PBMC (2 x 105/well suspended in tissue culture medium [RPMI 1640 supplemented with 5% controlled process serum replacement type 1; Sigma Aldrich; nonessential amino acids; Sigma Aldrich; 5 x 10-5 M 2mercaptoethanol, 100 U of penicillin/ml, and 100 µg of streptomycin sulfate/ml; Sigma Aldrich]) were then added and cultured at 37°C and 5% CO2 in a humidified incubator for 24 hours. Spots were developed with rabbit serum specific for IFN- followed by incubation with an alkaline phosphatase-conjugated monoclonal antibody specific for rabbit immunoglobulin G (Sigma Aldrich). The spots were visualized with 5-bromo-4-chloro-3-indolylphosphatenitroblue tetrazolium substrate (Sigma Aldrich). RNA extraction, reverse transcription and real-time quantitative PCR (polymerase chain reaction, obj. 3B). Previous clinical studies have shown that the IL42 splice variant is to be found in un-stimulated PBMC [3-5]. Therefore we used freshly isolated PBMC for all cattle in this report. RNA was prepared from 10 6 PBMC/sample using the RNeasy mini kit system and following the manufacturers instructions (Qiagen Ltd., West Sussex, UK). PBMC were isolated by layering over Histopaque 1077 (Sigma) and centrifuging at 800g for 40 minutes at room temperature, washed twice in Hanks Balanced Salt Solution (HBSS, Life Technologies) and counted. 106 PBMC were aliquoted into sterile screw top vials, the cells pelleted by centrifugation and the supernatant discarded. Lysates were prepared by resuspending the pelleted cells in 350l of RLT buffer according to manufacturers instructions (Qiagen Ltd), and lysates stored at minus 80 oC. Cell lysates were centrifuged through DNA shredder columns (Qiagen Ltd) prior to ethanol extraction and RNA isolation using the RNeasy column system. RNA preparations were finally treated with Turbo DNase (Ambion (Europe) Ltd., Huntingdon, UK) for 30 minutes at 37oC. Reverse transcription (RT) was carried out using the Qiagen Reverse iT TM T-primed First Strand Synthesis Kit and following the manufacturers instructions. RT controls (no RTase enzyme) were included for every sample. The PCR assay was optimized using plasmids containing the constructs for IL4, IL42 and IL43, as well as using bovine PBMC (ex vivo, antigen-stimulated and mitogen-stimulated). PCR was carried out in triplicates for each sample using a Universal PCR Mastermix (Applied Biosystems, USA) plus forward and reverse primers and Taqman probes as described previously [8]. For a positive signal, at least 2 out of the 3 PCR reactions in any triplicate had to give detectable Ct values. Cytokines were normalized to the housekeeping gene GAPDH [13]. All primers and probes were manufactured by MWG-Biotech, Ebersberg, Germany. Standard curves for each cytokine and GAPDH were prepared from the PBMC of two field reactor animals using serial dilutions of cDNA, extracted and processed as above. The same set of standard curves was used to extract measurements of RNA from all experiments contained in this report. Variation in GAPDH per samples ranged from 3.5 to 885 arbitrary units, with 50% samples falling between 17.9 and 304 units GAPDH. A further DNA contamination control (DNase-treated RNA sample prior to reverse transcription) was included for each sample on each run. Reactions were carried out using a Rotor-Gene RG300 (Corbett Research, Cambridge, UK). Cycling conditions of an initial activation step of 10 minutes at 95oC followed by 45 cycles of 95oC for 15 seconds and 60oC for 60 seconds were used for all primers and probes. A calibration control plasmid was included on each PCR run in duplicate such that and any minor differences (virtually none) between runs could be normalized and therefore all PCR runs comparable SID 5 (Rev. 3/06) Page 24 of 27 References cited in text: 1. Anonymous. 2002. European Commission Regulation (EC) NO 1226/2002 of 8 July 2002 amending annexe B to Council Directive 64/432/EEC. Off. J. Eur. Union (English) L 179: 13. 2. Dean, G.S., S.G. Rhodes, M. Coad, A.O. Whelan, P.J. Cockle, D.J. Clifford, R.G. Hewinson and H.M. Vordermeier. 2005. Minimum infective dose of Mycobacterium bovis in cattle. Infect. Immun., 73(10): 6467-6471. 3. Dheda, K., J. S. Chang, R. A. M. Breen, A. Jamanda, J. A. Haddock, M. Lipman, L. U. Kim, J. F. Huggett, M. A. Johnson, G. A. W. Rook and A Zumla. 2005. Expression of a novel cytokine, IL4delta2, in HIV and HIVtuberculosis co-infection. AIDS, 19: 1601-1606. 4. Fletcher, H., P. Owiafe, D. Jeffries, P. Hill, G. A. W. Rook, A. Zumla, T. M. Doherty, R. H. Brookes and the VACSEL Study Group. 2004. Increased expression of mRNA encoding interleukin (IL)-4 and its splice variant IL42 in cells from contacts of Mycobacterium tuberculosis, in the absence of in vitro stimulation. Immunology, 112:669-673. 5. Demissie, A., L. Wassie, M. Abebe, A. Aseffa, G. Rook, A. Zumla, P. Andersen, T. M. Doherty and the VACSEL Study Group. 2006. The 6-kilodalton early secreted antigenic target-responsive asymptomatic contacts of tuberculosis patients express elevated levels of interleukin-4 and reduced levels of gamma interferon. Infect. Immun., 74(5): 2817-2822. 6. Vordermeier, H.M., M.A. Chambers, B.M. Buddle, J.M. Pollock and R.G. Hewinson. 2006. Progress in the development of vaccines and diagnostic reagents to control tuberculosis in cattle. Vet. J., 171(2): 229-244. 7. O’Garra, A. and P. Vieira. 2004. Regulatory T cells and mechanisms of immune system control. Nat. Med., 10(8): 801-805. 8. Waldvogel, A. S., M-F. Lepage, A. Zakher, M. P. Reichel, R. Eicher and V. T. Heussler. 2004. Expression of interleukin-4, interleukin-4 splice variants and interferon-gamma mRNA in calves experimentally infected with Fasciola hepatica. Vet. Immunol. Immunopathol., 97: 53-63. 9. Leite, R.M.H., R.C. Leite, J.A. Lima, C.B. Foscolo, P.M.P.C. Mota, F.C.F. Lobato and A.P. Lage. 2000. HPLC identification of isoniazid residues in bovine milk. Arq. Bras. Med. Vet. Zootec., 52(6): 662-668. 10. Smith, P., J. van Dyk and A. Fredericks. 1999. Determination of rifampicin, isoniazid and pyrazinamide by high performance liquid chromatography after their simultaneous extraction from plasma. Int. J. Tuberc. Lung Dis., 3(suppl.): 325-328. 11. Rhodes, S.G., N. Palmer, S.P. Graham, A.E. Bianco, R.G. Hewinson and H.M. Vordermeier. 2000. Distinct responses of gamma interferon and interleukin-4 in bovine tuberculosis. Infect. Immun., 68(9): 5393-5400. 12. Vordermeier, H.M., M. A. Chambers, P.J. Cockle, A.O. Whelan, J. Simmons and R.G. Hewinson. 2002. Correlation of ESAT6-specific gamma-interferon production with pathology in cattle following Mycobacterium bovis BCG vaccination against experimental bovine tuberculosis. Infect. Immun., 70(6): 3026-3032. 13. Leutenegger, C. M., A. M. Alluwaimi, W. L. Smith, L. Perani and J. S. Cullor. 2000. Quantitaion of bovine cytokine mRNA in milk cells of healthy cattle by real-time Taqman polymerase chain reaction. Vet. Immunol. Immunopathol., 77(3-4): 275-287. SID 5 (Rev. 3/06) Page 25 of 27 References to published material 9. This section should be used to record links (hypertext links where possible) or references to other published material generated by, or relating to this project. SID 5 (Rev. 3/06) Page 26 of 27 Scientific journals and books: Dean, G.S., S.G. Rhodes, M. Coad, A.O. Whelan, P.J. Cockle, D.J. Clifford, R.G. Hewinson and H.M. Vordermeier. Minimum infective dose of Mycobacterium bovis in cattle. 2005. Infect. Immun., 73(10): 6467-6471. Rhodes, S.G., J. Sawyer, A.O. Whelan, G.S. Dean, M. Coad, K.J. Ewer, A.S. Waldvogel, A. Zakher, D.J. Clifford, R.G. Hewinson*, H.M. Vordermeier*. IL-4delta3 splice variant expression in bovine tuberculosis: a marker of protective immunity? J. Immunol., September 2006, submitted manuscript. Rhodes, S.G. The cattle model of tuberculosis: studies in the natural host. In; Focus on Tuberculosis Research. P187-201. Ed. Lucy T. Smithe, Nova Science Publishers Inc., 2005. ISBN I-59454-137-X. J.D. Rodgers, N.L. Connery, J. McNair, M.D. Welsh, R.A. Skuce, D.G. Bryson, D.N. McMurray & J.M. Pollock. Experimental exposure of cattle to a precise aerosolised challenge of Mycobacterium bovis: a novel model for study of tuberculosis. Manuscript in preparation. J. McNair, J.D. Rodgers, N.L. Connery, , M.D. Welsh, R.A. Skuce, D.G. Bryson, & J.M. Pollock. Manuscript in preparation.The impact of Mycobacterium bovis aerosol challenge, at different concentrations, on the immunology and pathology of disease. Scientific meetings with published abstracts: Dean, G.S., S.G. Rhodes, M. Coad, A.O. Whelan, P.J. Cockle, D.J. Clifford, R.G. Hewinson and H.M. Vordermeier. Minimum infective dose of Mycobacterium bovis in cattle. Presentation for the Annual Congress of the British Society for Immunology, Harrogate, UK, 7-10th Dec. 2004. Dean, G.S., S.G. Rhodes, M. Coad, A.O. Whelan,B. Villarreal-Ramos, E. Mead, D.J. Clifford, R.G. Hewinson and H.M. Vordermeier. Isoniazid treatment of bovine tuberculosis: development of a memory model. Presentation for the 7th International Veterinary Immunology Symposium, Quebec City, Canada, 25-30th July, 2004. Dean, G.S., S.G. Rhodes, M. Coad, A.O. Whelan,B. Villarreal-Ramos, E. Mead, D.J. Clifford, R.G. Hewinson and H.M. Vordermeier. Isoniazid treatment of bovine tuberculosis: development of a memory model. Presentation for the 1st Joint Meeting of European National Societies of Immunology, Paris 6-9th September, 2006. Rhodes, S.G., J. Sawyer, A.O. Whelan, G.S. Dean, M. Coad, K.J. Ewer, A.S. Waldvogel, A. Zakher, D.J. Clifford, R.G. Hewinson*, H.M. Vordermeier*. IL-4delta3 splice variant expression in bovine tuberculosis: a marker of protective immunity? Presentation for the 1st Joint Meeting of European National Societies of Immunology, Paris 6-9th September, 2006. Pollock, J.M., Rodgers, J.D., Welsh, M.D. and McNair, J. (2005). Pathogenesis of bovine tuberculosis: the role of experimental models of infection. The 4th International Conference on Mycobacterium bovios, 2226 August 2005, Dublin Castle, Dublin, Ireland. Pollock, J.M., Welsh, M.D. and McNair, J. (2005). Development of new strategies for the control of bovine tuberculosis in Northern Ireland. Conference Proceedings Symposium Tropical Animal Health, Utrecht, The Netherlands Rodgers, J.D., Connery, N., McNair, J., Welsh, M.D., Bryson, T.D.G. and Pollock, J.M. (2005). Aerosol exposure of cattle to Mycobacterium bovis: A novel method for the study of Tuberculosis. The 6th International Conference of Mycobacterial pathogenesis, 30 June-3 July 2005, Saltsjobaden, Sweden, June 2005. Rodgers, J.D., Connery, N., Mahaffy, H., Breadon, E.L., Milligan, C.R., Colhoun, L., Welsh, M.D., McNair, J., Bryson, T.D.G. and Pollock, J.M. (2006). Immune responses in cattle using an aerosol infection model of tuberculosis when challenged with a high dose of Mycobacterium bovis. Poster Presentation at Cellular mechanisms in host-pathogen interactions, 1-6 June 2006, Elsinore, Denmark. SID 5 (Rev. 3/06) Page 27 of 27