Introduction

Cell-mediated, but not humoral- or antibody-mediated, immune responses protect humans and animals against tuberculosis (Tb). Effective vaccination against Tb must produce an 'immunological imprint' so that a cell-mediated immune response is induced in all vaccinated individuals. Protective immunity to these facultative intracellular bacteria has been shown to correlate to the acquisition of a population of antigen-specific T cells, which are capable of activating macrophages to become mycobacteriocidal. Both CD₄ ⁺ and CD₈ ⁺ T cells are thought to be involved in eliciting protection.^{1, 2} Therefore, any prospective vaccine against Tb must be able to elicit and activate these particular cell populations.

It is the type 1 population of T cells that is considered to be essential to the development of a protective, cell-mediated immune response.^{3, 4} The principal mechanism by which this protection is achieved is through the production of IFN-, which is capable of enhancing antimicrobial activity in macrophages.⁵ The importance of this cytokine is suggested in studies that show that IFN- knockout mice are highly susceptible to mycobacterial infection.^{6, 7, 8} Conversely, a type 2 immune response (characterized by the production of IL-4, IL-5 and IL-10) is not protective against Tb.^{3, 4, 9} The conventional vaccine used against Tb is bacillus Calmette Guerin (BCG). This vaccine was first developed over 70 years ago and has not been improved since. At best, BCG provides 80% protection; however, under certain conditions it has also been shown to induce no protection from disease.¹⁰ This variation in efficacy has resulted in the search for improved vaccines against Tb. In 1994, O'Donnell et al. produced a recombinant version of BCG (rBCG), which constitutively expressed IL-2 via an extra-chromosomal shuttle plasmid,¹¹ and elicited high levels of IFN-. This system has been used to express a number of different cytokines from BCG, including IL-18. It has been established clearly that IL-18 is a powerful costimulatory factor, which, in concert with IL-12, upregulates IFN- production.¹² The purpose of the present study was to determine whether two rBCG, one secreting IL-2 and the other IL-18, were capable of eliciting an improved protective, type 1 immune response as compared to normal, parental BCG.

Methods

Animals

Healthy BALB/c mice were obtained from the Department of Animal Laboratory Sciences, University of Otago, Dunedin, New Zealand. Male mice were used at 6-8 weeks of age. All animals were housed in cages containing five animals and kept on a daily 12 h cycle of light and dark. The mice were fed ad libidium on rat pellets.

Ethical approval for these experiments was granted by the Otago University Animal Ethics Committee (grant number 39/98).

Immunizations

Groups of five mice were vaccinated s.c. with 10⁶ recombinant BCG (rBCG) or normal BCG (nBCG). nBCG and rBCG (Pasteur 1173P2) were cultured in 7H9 broth (Difco, Detroit, USA) and organisms were allowed to grow to mid-log phase before vaccination. Mice were subsequently left for either 6 or 16 weeks before being killed. Each experiment was repeated three times.

Cell proliferation assay

Spleens were removed aseptically and teased apart to extract the lymphocytes. Splenocytes were washed in media three times and cultured in DMEM (Gibco BRL, Life Technologies, Gaithersberg, USA) containing 5% FCS (Gibco BRL), 25 mmol/L HEPES (Sigma, St Louis, MO, USA) and 40 g/mL Gentamicin Sulphate (David Bull Laboratories, Melbourne, Australia). Cells were adjusted to a concentration of 3 10⁶/mL and then 100 L was plated along with either media alone, 50 g/mL PPD-B [CSL, Parkville, Australia], or 12 g/mL Concanavalin A (Con A; Sigma) in 96-well plates (Roskilde, Denmark). The final volume was 150 L in each well. Each sample was plated in triplicate and cultured for 3 days in a humidified atmosphere containing 5% CO₂. Wells were then pulsed with 0.5 Ci of [³H]TdR (Amersham Life Science, Buckinghamshire, England) and left for a further 18 h before being harvested onto microfibre filters (Whatman International, Maidstone, England). Proliferation was determined by measuring [³H]TdR incorporation, using a microbeta counter (Wallac, Turku, Finland), and was expressed as counts per minute. Background levels of proliferation were subtracted from results before graphing. Actual values are shown in figure legends.

Cytokine analysis

Splenocytes were isolated, washed and adjusted to 3 10⁶ cells/mL and plated in 24-well plates (Nunc). Samples were cultured along with either DMEM (Gibco BRL), 50 g/mL PPD-B (CSL), or 12 g/mL ConA (Sigma) for a period of 72 h. Supernatants were removed, pooled and analysed for the presence of IL-2, IL-4 and IFN- using a sandwich ELISA. The ELISA were performed according to the PharMingen manufacturer instructions. In brief, 96-well, flat-bottomed plates were coated overnight with 50 L of a 2 g/mL concentration of primary anticytokine capture antibody (PharMingen). Plates were washed and blocked with 200 L/well of PBS + 10% FCS (Gibco BRL) and incubated at room temperature for 2 h. Plates were washed again, and doubling dilutions of standards and supernatants were added. The samples were then incubated at 4°C overnight. The plates were washed and 100 L of biotin-conjugated anticytokine-detecting monoclonal antibody (PharMingen) at a concentration of 1 g/mL was added to each well. After a 1 h incubation at room temperature, the plates were washed and 100 L of a 1/500 dilution of streptavidin horseradish peroxidase was added (Amersham Life Science) per well. The plates were incubated for a further 45 min at room temperature before being rewashed and having 100 L/well of TMB substrate (Bio-Rad, Hercules, USA) added. The reaction was stopped by the addition of 100 L/well of 0.5 mol/L H₂SO₄ and the OD was determined at 450 nm, using a spectrophotometer (Bio-Rad). The amount of each cytokine in the supernatant was extrapolated from the standard curve. The standards were recombinant cytokines (PharMingen) diluted in 1 : 2 dilutions from 1000 to 15 pg/mL for IL-4, 500-0.3 ng/mL for IFN-, and 10-0.1 ng/mL for IL-2.

Reverse transcription PCR

Total cellular RNA was isolated from activated splenocytes by acid guanidium isothiocyanate/phenol chloroform extraction as previously described.¹³ Levels of cytokine mRNA were assessed by a semiquantitative reverse transcription (RT)-PCR. RNA was reverse transcribed, using reverse transcriptase and oligo(dT) primers (Boeringer Mannheim, Germany) according to the manufacturer's protocol. The cDNA was diluted in trisethylenediaminetetraacetic acid (TE) to give a final volume of 15 L. cDNA from each animal was then amplified using primers to murine IFN-, and to -actin. PCR was performed in a reaction mixture containing 2 L cDNA, 25 mmol/L dNTP, 0.4 mol/L of each primer and 2.5 U of Taq DNA polymerase. Samples were amplified with a thermocycler (Hybaid) using the following protocol: initial denaturing of 95°C for 1 min, then at 95°C for 30 s, at 54°C for 45 s, and at 72°C for 45 s (32 cycles for each cytokine) with a final extension at 72°C for 5 min. Samples were electrophoresed in a 2% agarose gel and the results were determined by OD readings and arbitrarily expressed as mRNA units. All data were standardized to -actin.

Serum antibody detection

Serum was collected and pooled from the various groups of mice. ELISA plates were coated overnight with 50 g/mL of PPD-B (CSL). Plates were washed and blocked, and serial dilutions of serum samples and standards were added. Plates were then incubated overnight at 4°C. Anti-IgG1 and IgG2a antibodies (Sigma) were added at recommended dilutions, and the plates were incubated for 1 h at room temperature. Anti-goat, peroxidase labelled, monoclonal antibody was added to each well and the plates were incubated for a further 45 min, before the addition of TMB substrate (Bio-Rad). The reaction was stopped by addition of 0.5 mol/L H₂SO₄ and the OD was read at 450 nm.

Intranasal infection with live bacillus Calmette Guerin

Sixteen weeks post-vaccination, mice were infected with live M. bovis BCG. Mice were anaesthetized and given 5 10⁴ BCG in a 50 L volume. This was administered intranasally by placing the BCG dropwise onto the nares and allowing the mice to inhale it. The mice were left for a further 4 weeks before being killed. Spleens were then removed aseptically and homogenized. Tenfold dilutions of the homogenate were plated on 7H9 agar plates and incubated at 37°C in a humidified atmosphere containing 5% CO₂. Colonies were counted weekly from 4 to 8 weeks after plating of samples.

Aerosol challenge with Mycobacterium bovis

Mice were challenged with a virulent M. bovis strain WAg201, a New Zealand bovine isolate, 16 weeks after vaccination. Single-cell suspensions of the challenge isolate were prepared using a modification of a method described by Grover et al. ¹⁴ and stored at -70°C. To prepare these suspensions, the bacterial cells were dispersed by sonication for 30 s and filtered through an 8 m filter. The mice were infected via the respiratory route using an aerosol chamber that produces droplet nuclei of the size appropriate for entry into alveolar spaces.^{15, 16} The concentration of viable M. bovis in the nebulizer fluid was empirically adjusted to result in the inhalation and retention of approximately 10 viable organisms per mouse. This challenge dose had previously been estimated from the culture of mouse lungs within 20 h of infection. The mice were killed 5 weeks after infection and serial dilutions of individual lung and spleen homogenates were plated out on modified 7H11 agar.¹⁷ Bacterial colonies were counted 28 days later, following incubation at 37°C in humidified air. The aerosol infection and subsequent maintenance and manipulation of infected mice were performed under strict isolation conditions in a BL3 biohazard facility.

Statistical analysis

Data for each experiment were expressed as a mean (SEM) and the results were analysed for statistical significance using unpaired Student's t-tests from the software package INSTAT 2.01 (GraphPad Software, San Diego, USA). Analysis of bacterial counts from the M. bovis-challenged mice was carried out by ANOVA on log₁₀-transformed data and compared by Fisher's pairwise comparison. The level of significance was set at P < 0.05.

Results

Longevity of the splenocyte response to vaccination with rBCG/IL-2 and nBCG

It has been reported that the primary immune response to BCG lasts for approximately 4-6 weeks.^{4, 18} After this time point, an anamnestic or recall response is seen in vaccinates. Titration of the BCG vaccines showed that 10⁶ organisms was the optimal dose required to produce optimal levels of lymphocyte transformation and antigen-specific IFN- production (data not shown). The lymphocytes of BALB/c mice were assayed for their ability to proliferate in response to specific antigen (PPD) both 6 and 16 weeks after vaccination with nBCG or rBCG/IL-2. Six weeks after vaccination, both rBCG/IL-2 and nBCG induced high levels of lymphocyte transformation in vitro. There was no significant difference between these responses. However, after 16 weeks, lymphocyte transformation was greatly reduced (P < 0.01) in the nBCG-vaccinated group compared to mice examined at 6 weeks (Figure 1). There was no significant change in the rBCG/IL-2 group. To ensure that these results were not due to persistence of the vaccine, half of the spleen of each mouse was homogenized and plated onto 7H11 agar plates, 16 weeks after vaccination. Viable bacteria were assessed weekly for 2-3 months. No viable bacteria were found in any samples taken 16 weeks after vaccination.

Figure 1.

The ability of bacillus Calmette Guerin (BCG) and recombinant BCG-secreting IL-2 (rBCG/IL-2) to activate splenic lymphocytes in BALB/c mice. In vitro proliferation of splenocytes to restimulation with 50 g/mL PPD were measured 6 and 16 weeks after vaccination with either 10⁶ c.f.u. of BCG or recombinant BCG-secreting IL-2. Mean background levels were less than 500 c.p.m., and were subtracted from the final values. Lymphocyte transformation was significantly reduced in mice 16 weeks after vaccination with nBCG as compared to 6 weeks after vaccination (P < 0.01). There was no significant reduction in lymphocyte transformation over time in mice vaccinated with rBCG/IL-2. , 6 weeks; , 16 weeks.

Full figure and legend (13K)

Cytokine production in response to nBCG and rBCG/IL-2

The role that type 1, IFN--producing, cells play in immunity to intracellular bacteria is central to protection. In particular, IFN- is considered to be essential.⁵ IFN- levels were assayed by both ELISA (protein) and RT-PCR (mRNA). The ELISA results showed that early in the immune response (6 weeks) BALB/c mice produced higher levels of IFN- in response to rBCG/IL-2 vaccination than in response to nBCG vaccination (P < 0.05; Figure 2a). Levels had waned in both groups by 16 weeks; however, production of IFN- continued to be significantly higher in mice vaccinated with rBCG/IL-2, compared to those vaccinated with nBCG (P = 0.001). In order to confirm the data derived from the ELISA, we analysed the levels of IFN--specific mRNA in response to the two vaccines (Figure 2b). Sixteen weeks after vaccination with the rBCG/IL-2, splenocytes from BALB/c mice continued to produce high levels of IFN- mRNA, compared to those vaccinated with nBCG. There were threefold higher levels of IFN- transcription with rBCG/IL-2 compared to nBCG, indicating a trend similar to that found for IFN- protein expression.

Figure 2.

Production of IFN- by BALB/c splenocytes after in vitro restimulation with 50 g/mL PPD. Protein was measured by (a) ELISA (6 and 16 weeks after vaccination with either 10⁶ c.f.u. of normal bacillus Calmette Guerin [nBCG] or rBCG/IL-2) and (b) mRNA by semiquantitative PCR (6 and 16 weeks after vaccination with either 10⁶ c.f.u. of nBCG or rBCG/IL-2). No IFN- was detected in background (unstimulated) splenocytes. Mice vaccinated with rBCG/IL-2 produced significantly more IFN- at both 6 (P < 0.05) and 16 weeks (P = 0.001) post-vaccination, compared to those vaccinated with nBCG. (c) Levels of IL-2, measured by ELISA, produced by splenocytes after in vitro restimulation with 50 g/mL PPD (6 and 16 weeks after vaccination with either 10⁶ c.f.u. of nBCG or rBCG/IL-2). At 6 weeks post-vaccination, the mean value of background levels of IL-2 was less than 100 pg/mL. At 16 weeks post-vaccination, no IL-2 was detected in control samples. While levels of IL-2 were not significantly reduced at 16 weeks, compared to 6 weeks in animals vaccinated with rBCG/IL-2, levels were significantly reduced by 16 weeks in mice vaccinated with nBCG (P < 0.001). j, 6 weeks; , 16 weeks.

Full figure and legend (32K)

The production of IL-2 in response to vaccination with rBCG/IL-2 was also maintained over 16 weeks, showing little decrease (Figure 2c). By contrast, in mice vaccinated with nBCG, IL-2 levels were almost undetectable 16 weeks after vaccination (P < 0.001). No IL-4 was detected by ELISA in any of the samples, whether taken at 6 or 16 weeks (data not shown).

Serum antibody isotypes induced by nBCG and rBCG/IL-2

The levels of IgG1 and IgG2a were measured in the serum of mice 16 weeks after vaccination. Figure 3 shows that vaccination with rBCG/IL-2 induced a serum IgG response dominated by IgG2a, whereas the nBCG induced a response dominated by IgG1.

Figure 3.

Levels of IgG1 and IgG2a antibody found in the serum of BALB/c mice 16 weeks after vaccination with 10⁶ c.f.u. of recombinant (rBCG/IL-2) or normal bacillus Calmette Guerin (nBCG). Comparison of serum antibodies show that mice vaccinated with rBCG/IL-2 had significantly higher levels of IgG2a and significantly lower levels of IgG1 than mice vaccinated with nBCG (P < 001). No detectable levels of PPD-specific antibody were found in unvaccinated controls. , IgG1; , IgG2a.

Full figure and legend (16K)

Clearance of live BCG by animals vaccinated with nBCG and rBCG/IL-2

We compared rBCG/IL-2 and nBCG as to their ability to increase antimycobacterial activity in mice and limit systemic spread. Sixteen weeks after subcutaneous vaccination with either nBCG or rBCG/IL-2, mice were exposed to live BCG via the nose. Four weeks later, the dissemination of live BCG to the spleen, or the growth of the organisms within the spleen, was assayed by assessing the numbers of culturable organisms present. Figure 4 shows that vaccination with nBCG significantly reduces the number of BCG present in the spleen (P < 0.01), but that the rBCG/IL-2 gave significantly greater protection than the nBCG (P < 0.05).

Figure 4.

In vivo clearance of live bacillus Calmette Guerin (BCG) (5 10⁴) delivered intranasally to BALB/c mice vaccinated, subcutaneously, 16 weeks previously with normal BCG (nBCG) or rBCG/IL-2. Control animals were not vaccinated. Culturable BCG present in the spleens were determined 4 weeks after intranasal infection. Animals vaccinated with rBCG/IL-2 contained significantly less BCG in their spleens than mice vaccinated with BCG (P < 0.05).

Full figure and legend (11K)

Protective efficacy of rBCG/IL-2 against virulent Mycobacterium bovis

The ability of the cytokine-secreting BCG vaccines to protect BALB/c mice from aerosoled infection with virulent M. bovis was compared to nBCG (Table 1). The results show that despite the positive in vitro indicators, rBCG/IL-2 showed no significantly increased ability to protect animals against M. bovis as compared to nBCG. Both vaccines significantly reduced the numbers of c.f.u. in the lung and spleen when compared to unvaccinated controls (P < 0.05).

Table 1 - The ability of the cytokine-secreting BCG vaccines to protect BALB/c mice from aerosoled infection with virulent Mycobacterium bovis compared to nBCG.

Full table

Immune responses to vaccination with rBCG/IL-18 and its protective efficacy against Mycobacterium bovis

The failure of the rBCG/IL-2 to improve protection against virulent challenge leads us to consider whether the synergistic affect of IL-18 on IFN- production might have a more pronounced effect. Therefore, a recombinant BCG vaccine that secretes IL-18 was tested for its effects on immune responses to BCG antigens. Splenocytes of BALB/c mice were assayed for antigen-specific proliferation and IFN- production 6 weeks after vaccination with an IL-18-secreting recombinant BCG. The data show that while the lymphocyte transformation responses were similar (Figure 5a), the levels of IFN- in cultures from mice vaccinated with rBCG/IL-18 were significantly reduced (P < 0.01).

Figure 5.

In vitro immune responses of splenocytes from BALB/c mice to restimulation with PPD 6 weeks after vaccination with 10⁶ c.f.u. of normal bacillus Calmette Guerin (nBCG) or rBCG/IL-18. (a) Lymphocyte transformation; and (b) production of IFN- were measured by tritiated thymidine incorporation and ELISA, respectively. There was no significant difference in lymphocyte transformation responses, but there was a significant reduction in IFN- production by splenocytes from mice vaccinated with rBCG/IL-18 (P < 0.01).

Full figure and legend (24K)

When compared for its protective efficacy against nBCG, it was found that rBCG/IL-18 gave significantly less protection against infection than nBCG (P < 0.05; Table 1).

Discussion

The most efficient and cost-effective method of protecting against and eradicating infectious disease is by vaccination. Unfortunately, the vaccine used widely against Tb, BCG, has been the subject of considerable criticism due to its failure in several vaccine trials.¹⁰ However, evidence from laboratory experiments suggest that BCG is highly immunogenic and contains most, if not all, the antigens thought to be important in stimulating protective immunity.¹⁹ So, if BCG is strongly antigenic, why does it sometimes fail to induce protective immunity? A possible reason is that BCG is unable to reshape immune responses or immune capabilities. Most evidence favours a dominant role for the type 1 cytokines, IFN- and IL-2, as opposed to the type 2 cytokines, such as IL-4, in a protective immune response to Tb infection. If BCG cannot reprogramme an existing response, but merely reinforces it, such as in hosts genetically predisposed to generating a type 2 immune response or where immune suppression has compromised the host's ability to generate a type 1 immune response,^{10, 20} then the failure of BCG under some field conditions is understandable. Vaccination with BCG would not guarantee an effective type 1 response in all hosts under all conditions, and its efficacy would vary accordingly. Here we have referred to type 1 and type 2 immune responses rather than Th1 and Th2 responses, as it is now recognized that a number of cell types, including cells of the innate immune response, can participate in the development of these responses.²¹ Our data do not directly implicate any particular cell type, although the longevity of the responses are indicative of T-cell involvement.

In our study, BALB/c mice respond poorly to BCG vaccination, failing to produce the in vitro immune correlates of a strong and sustained type 1 immune response. Therefore, we chose this strain of mice to determine if cytokine-secreting BCG vaccines could alter these responses. This laboratory and others have shown that IL-2 can be used to improve lymphocyte responsiveness to antigens from BCG and M. bovis. ^{22, 23, 24} Further, it has been shown that IL-2 secreted from BCG can induce upregulation of antigen-specific IFN- production by splenocytes restimulated in vitro. ^{11, 25} We have extended these findings to show that cytokines secreted by BCG can modulate immune responses to vaccination in BALB/c mice. This confirms our earlier findings that IL-2 is able to modulate the immune responses of a naturally susceptible host species (deer) to BCG vaccination.²⁴ Here we have shown that the IL-2 produced by the recombinant BCG is able to strongly upregulate lymphocyte responses and, in particular, IFN- production over a prolonged period. It also induces an antibody isotype shift, characteristic of a type 1 immune response. We have also shown that IL-2 production by restimulated splenocytes can be detected in rBCG/IL-2-vaccinated animals long after it has waned in nBCG-vaccinated mice. IL-2 is a recognized growth factor for both NK and T cells,²⁶ which are known to produce IFN- early in immune responses.²⁷ IL-2 is also required for the activation and proliferation of cytotoxic (CD8+) T cells,^{28, 29} which develop later in the acquired immune response and are also a major source of IFN-. Therefore, the IL-2 secreted from the rBCG may be acting to upregulate IFN- production both early and late in the response to BCG. This would create a microenvironment that favours development of a type 1 immune response. In our experience and that of others,^{24, 25} rBCG/IL-2 is eliminated at the same rate, if not slightly more rapidly, than normal BCG in vivo and is unlikely to contribute to the results observed 16 weeks later in the vaccine response. This is supported by our inability to culture rBCG/IL-2 from spleen samples taken 16 weeks after subcutaneous vaccination (unpubl. data). Therefore, we have concluded that IL-2 production by rBCG/IL-2 is unlikely to be the reason for the observed results. An alternative explanation is that the IL-2 secreted by the recombinant BCG acts initially in a paracrine manner, initiating the endogenous production of IL-2 by host T cells, which then maintain IL-2 production after the elimination of the rBCG/IL-2. This, in turn, may result in prolonged IFN- production. Despite the generation of a favourable immune profile in vitro, the vaccine was unable to confer increased levels of protection against virulent M. bovis challenge when compared to nBCG.

Therefore, we investigated a second cytokine known to increase IFN- production, IL-18, which has been shown to be a potent effector of immunity to Listeria monocytogenes, even in the absence of IFN-.³⁰ IL-18 is also known to synergize with IL-12 to increase IFN- production.¹² Therefore, we tested a rBCG that secreted IL-18 for its ability to improve immune responses and protective efficacy against virulent infection. We showed that, rather than improving the immune response to BCG, the rBCG/IL-18 inhibited IFN- production and compromised the induction of protective immunity against virulent M. bovis in BALB/c mice. Hoshino et al.³¹ reported that aberrant expression of IL-18, in a transgenic mouse model, resulted in the upregulation of both type 1 and type 2 cytokines. In our study, BALB/c mice did respond in a type 2 manner to vaccination with BCG; therefore, the IL-18 may be reinforcing this effect. We have no direct evidence to support the concept that IL-18 acts to upregulate the type 2 response; for example, we were not able to show increased IL-5 production in mice vaccinated with rBCG/IL-18 (data not shown). However, the reduction in IFN- levels suggests that IL-18 is inhibiting the development of protective type 1 responses in BALB/c mice. The administration of IL-18 to BALB/c mice has been shown to exacerbate infections by Leishmania major,³² and it has been suggested that the underlying mechanism is that IL-18, in the absence of IL-12, could upregulate type 2 responses.

The present study confirms that in vitro markers of immunity, such as IFN-, are poor predictors of protective immunity against virulent M. bovis challenge. Indeed, studies of Tb in a number of animal models have shown that lymphocytes from diseased animals secrete large amounts of the type 1 cytokine IFN-, which can be used for immunodiagnosis of disease.³³ While IFN- is essential, it is obviously not sufficient to confer protection against Tb and additional immune factors are required.³⁴ The inclusion of immunopotentiating factors in future BCG vaccines will improve the chances that protective immunity can be induced predictably in a genetically diverse population. This will require the identification of the factor(s) that, in conjunction with IFN-, are required for protection against Tb. Our data suggest that care will need to be taken in selecting these factors.

References

1. Ravn P, Boesen H, Pedersen B, Andersen P. Human T cell responses induced by vaccination with Mycobacterium bovis BCG. J. Immunology 1997; 158: 1949–55. | ChemPort |
2. Flynn J, Chan J. Immunology of tuberculosis. Ann. Rev. Immunol. 2001; 19: 93–129. | Article | ISI | ChemPort |
3. Kaufmann S. Immunity to intracellular bacteria. Ann. Rev. Immunol. 1993; 11: 129–63. | Article | ChemPort |
4. Orme I. Immunity to mycobacteria. Curr. Opin. Immunol. 1993; 5: 497–502. | Article | PubMed | ChemPort |
5. Yang J, Mitsuyama M. An essential role for endogenous interferon- in the generation of protective T cells against Mycobacterium bovis BCG in mice. Immunology 1997; 91: 529–35. | Article | PubMed | ChemPort |
6. Cooper A, Dalton D, Stewart T, Griffin J, Russel D, Orme I. Disseminated tuberculosis in interferon- gene-disrupted mice. J. Exp. Med. 1993; 178: 2243–9. | Article | PubMed | ISI | ChemPort |
7. Jouanguy E, Altare F, Lamhamedi S et al. Interferon receptor deficiency in an infant with fatal BCG infection. N. Engl. J. Med. 1996; 335: 1956–61. | Article | PubMed | ISI | ChemPort |
8. Murray P, Young RA, Daley G. Haematopoietic remodeling in interferon-gamma-deficient mice infected with mycobacteria. Blood 1998; 91: 5743–50.
9. Hook S, Griffin F, Mackintosh C, Buchan G. Activation of an interleukin-4 mRNA-producing population of peripheral blood mononuclear cells after infection with Mycobacterium bovis or vaccination with killed, but not live, BCG. Immunology 1996; 88: 269–74. | PubMed | ChemPort |
10. Fine P. Variation in protection by BCG. Implications of and for heterologous immunity. Lancet 1995; 346: 1339–45. | Article | PubMed | ISI | ChemPort |
11. O'Donnell MA, Aldovini A, Duba R et al. Recombinant Mycobacterium bovis BCG secreting functional interleukin-2 enhances gamma interferon production by splenocytes. Infect. Immun. 1994; 62: 2508–14. | PubMed | ChemPort |
12. Okamura H, Tsutsui H, Komatsu T et al. Cloning of a new cytokine that induces IFN- production by T cells. Nature 1995; 378: 88–91. | Article | PubMed | ISI | ChemPort |
13. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 1987; 162: 156–9. | Article | PubMed | ISI | ChemPort |
14. Grover A, Kim H, Wiegeshaus E, Smith D. Host-parasite relationships in experimental airborne tuberculosis II. Reproducible infection by means of an inoculum preserved at -70C. J. Bacteriol. 1967; 94: 832–5. | PubMed | ChemPort |
15. Wiegeshaus E, McMurray D, Grover A, Harding G, Smith D. Host-parasite relationships in experimental airborne tuberculosis III. Relevance of microbial enumeration to acquired resistance in guinea pigs. Am. Rev. Resp. Dis. 1970; 102: 422–9. | PubMed | ChemPort |
16. McMurray D, Carlomagno M, Mintzer C, Tetzlaff C. Mycobacterium bovis BCG vaccine fails to protect protein-deficient guinea pigs against respiratory challenge with virulent Mycobacterium tuberculosis. Infect. Immun. 1985; 50: 555–9. | PubMed | ChemPort |
17. Wedlock D, Aldwell F, Collins D, de Lisle G, Wilson T, Buddle B. Immune responses induced in cattle by virulent and attenuated Mycobacterium bovis strains: correlation of delayed-type hypersensitivity with ability of strains to grow in macrophages. Infect. Immun. 1999; 67: 2172–7. | PubMed | ChemPort |
18. Cooper A, Flynn J. The protective immune response to. Mycobacterium tuberculosis. Curr. Opin. Immunol. 1995; 7: 512–6. | Article | PubMed | ChemPort |
19. Collins FM. Antituberculous immunity: New solutions to an old problem. Rev. Infect. Dis. 1991; 13: 940–50. | PubMed | ChemPort |
20. Griffin J, MacIntosh C, Slobbe L, Thomson A, Buchan G. Vaccine protocols to optimise the protective efficacy of BCG. Tuber. Lung Dis. 1999; 79: 135–43. | Article | PubMed | ChemPort |
21. Jankovic D, Lui Z, Gause W. Th1- and Th2-cell commitment during infectious disease: asymmetry in divergent pathways. Trends Immunol. 2001; 22: 450–7. | Article | PubMed | ISI | ChemPort |
22. Baldwin S, D'Souza C, Roberts A et al. Evaluation of new vaccines in the mouse and guinea pig model of tuberculosis. Infect. Immun. 1998; 66: 2951–9. | PubMed | ISI | ChemPort |
23. Colizzi V, Malkovsky M. Augmentation of interleukin-2 production and delayed hypersensitivity in mice infected with Mycobacterium bovis and fed a diet supplemented with vitamin A acetate. Infect. Immun. 1985; 48: 581–3. | PubMed | ChemPort |
24. Slobbe L, Lockhart E, O'Donnell M, Mackintosh C, de Lisle G, Buchan G. An in vitro comparison of bacillus Calmette-Guerin (BCG) and cytokine-secreting BCG vaccines. Immunology 1999; 96: 517–23. | Article | PubMed | ChemPort |
25. Kong D, Kunimoto D. Secretion of human interleukin 2 by recombinant Mycobacterium bovis BCG Infect. Immun. 1995; 63: 799–803. | PubMed | ChemPort |
26. Kjeldsen K, Quale A, Skalhegg B, Sioud M, Forre O. Selective activation if resting human T lymphocytes by interleukin 2. Eur. J. Immunol. 1993; 23: 2092–9. | PubMed | ChemPort |
27. Henny C, Kuribayashi K, Kern D, Gillis S. Interleukin-2 augments natural killer cell activity. Nature 1981; 291: 335–8. | Article |
28. Smith KA. Interleukin-2, inception, impact and implications. Science 1988; 240: 1169–76. | Article | PubMed | ISI | ChemPort |
29. Yang X, Hayglass K. IFN-, but not IL-4, synthesis by antigen-primed murine T cells is IL-2 dependent. J. Immunol. 1993; 150: 4354–63. | PubMed | ChemPort |
30. Neighbors M, Xu X, Barrat F et al. A critical role for interleukin 18 in primary and memory effector responses to Listeria monocytogenes that extends beyond its effects on interferon production. J. Exp. Med. 2001; 194: 343–54. | Article | PubMed | ISI | ChemPort |
31. Hoshino T, Kawase Y, Okamoto M et al. IL-18-Transgenic mice: in vivo evidence of a broad role for IL-18 in modulating immune function. J. Immunol. 2001; 166: 7014–8. | PubMed | ISI | ChemPort |
32. Xu D, Trajkovic V, Hunter D et al. IL-18 induces the differentiation of Th1 or Th2 cells depending upon the cytokine milieu and genetic background. Eur. J. Immunol. 2000; 30: 3147–56. | Article | PubMed | ISI | ChemPort |
33. Wood P, Corner L, Plackett P. Development of a simple, rapid in vitro cellular assay for bovine tuberculosis based on the production of gamma-interferon. Res. Vet. Sci. 1990; 49: 46–9. | PubMed | ChemPort |
34. Leal I, Smedegard B, Andersen P, Appelberg R. Failure to induce enhanced protection against tuberculosis by increasing T-cell-dependent interferon- generation Immunology 2001; 104: 157–61 | Article | PubMed | ISI | ChemPort |

Acknowledgements

Funding for this project was provided by the Foundation for Research, Science and Technology, New Zealand.