Skip to main content
NCBI home page
Clinical and Experimental Immunology logoLink to Clinical and Experimental Immunology
. 2000 Jan;119(1):92–98. doi: 10.1046/j.1365-2249.2000.01100.x

Bacille Calmette–Guérin (BCG)-associated inflammation and fibrosis: modulation by recombinant BCG expressing interferon-gamma (IFN-γ)

A Wangoo *, I N Brown *, B G Marshall *, H T Cook †, D B Young *, R J Shaw *
PMCID: PMC1905541  PMID: 10606969

Abstract

Immunization with existing BCG vaccines has failed to confer consistent protection against tuberculosis. One of the ways to improve the efficacy of BCG is by enhancing its ability to induce a type-1 T cell response. However, this approach carries the risk that enhanced immunoreactivity may exacerbate tissue pathology associated with vaccination. The aim of the present study was to determine whether use of a recombinant BCG expressing IFN-γ (BCG-IFN) would result in an alteration in the pattern of inflammation and local tissue fibrosis. A murine intravenous BCG infection model was used in which there was a time- and dose-dependent increase in the weight and number of granulomas in the liver. Infection was associated with increased inflammatory activity in the liver, as shown by the increase in expression of inducible nitric oxide synthase (iNOS) assessed by immunochemistry and by measurement of specific mRNA, and in fibrosis measured by hydroxyproline content of the liver and percentage of granuloma cells staining positively for type 1 procollagen. Infection with BCG-IFN resulted in a reduction in organ weight and bacterial load on day 21 compared with infection with control BCG transformed with vector alone (BCG-plasmid). By day 21, there was also a reduction in iNOS mRNA and iNOS+ cells in granulomas in mice infected with BCG-IFN compared with infection with BCG-plasmid, and a similar reduction in both total number of granulomas and liver hydroxyproline content. These results demonstrate that the granulomas in the areas of mycobacterial infection are active sites of both inflammation and fibrosis, and that the local expression of IFN-γ by the recombinant BCG results in more efficient bacterial clearance which is accompanied by a reduction in tissue pathology.

Keywords: bacille Calmette–Guérin, fibrosis, interferon-gamma

INTRODUCTION

Tuberculosis (TB) is a common and serious infectious disease with a global mortality estimated at three million annually [1]. The World Health Organization estimates that the number of cases of TB will rise from 7.5 million in 1990 to over 10 million by the year 2000 [2]. BCG remains the most controversial of all the currently used vaccines, as its protective efficacy has varied widely in different parts of the world [3]. At a time when TB is increasing in both developing and industrialized countries, there is a renewed interest and urgency in improving the efficacy of BCG. Protection against mycobacterial infection requires activation of macrophages mediated by production of IFN-γ by type 1 (Th1) CD4+ T cells [4–6]. The importance of this pathway in murine models is evident from experiments involving T cell depletion, or disruption of the IFN-γ gene [7], and in man by the enhanced susceptibility to mycobacterial infection in individuals with mutations affecting the IFN-γ receptor [8]. IFN-γ production may act in part by inducing production of inducible nitric oxide (iNOS) [9]. Neutralization of IFN-γ has been shown to inhibit nitrite production in spleen cells stimulated with viable BCG. Thus, enhancement of the ability to induce a type 1 T cell response represents an attractive strategy for improving BCG vaccination.

However, conventional BCG vaccination causes tissue pathology, including induration and ulceration of the vaccination site [10], and intradermal BCG in man is associated with a dose-dependent inflammatory response which can result in scarring [11]. Previous work has also identified evidence of active fibrosis in the local dermal immune response to mycobacterial antigens [12,13]. The cellular and molecular mechanisms involved in this pathology are poorly understood, and there is a danger that modifications designed to enhance type 1 responses associated with protection may also exacerbate pathological sequelae associated with vaccination.

Recently, BCG strains secreting cytokines have been developed and have been shown to modify and potentiate the immune response to subsequent challenge with mycobacterial antigens [14]. Among these strains, the most profound cell-mediated immune responses were induced by BCG secreting IFN-γ (BCG-IFN). To explore further the potential use of this approach for development of improved BCG vaccines, we used a murine fibrosis model to measure BCG-induced pathology, and report on responses generated by infection with BCG-IFN compared with those induced by control BCG.

MATERIALS AND METHODS

Mice

Female BALB/c mice raised under specific pathogen-free conditions were purchased from Harlan Olac (Bicester, UK). The mice were maintained for at least 1 week and used for experiments at 7–8 weeks old.

Mycobacterium bovis BCG

Mycobacterium bovis BCG (Pasteur strain) was grown at 37°C in Middlebrook 7H9 broth (Difco Labs, Detroit, MI) containing 0.2% (v/v) glycerol, 0.05% (v/v) Tween-80 and supplemented with 10% (v/v) Middlebrook ADC enrichment (Difco Labs), to an optical density (OD) of 1.0 at 600 nm. BCG grown in this medium has a smooth appearance, does not grow in large clumps and is more easily dispersed [15]. This stock culture was stored frozen as 1 ml aliquots at −80°C. Vials were thawed and the concentration of BCG, expressed as colony-forming units (CFU) per ml, was determined by plating diluted cultures on Middlebrook 7H11 agar containing 0.5% (v/v) glycerol and supplemented with 10% (v/v) Middlebrook OADC enrichment (Difco Labs).

A recombinant BCG strain that secretes functional murine IFN-γ (BCG-IFN) and a BCG strain with plasmid alone (BCG-plasmid) were a kind gift from P. J Murray (Whitehead Institute, Cambridge, MA). To construct BCG-IFN, the gene encoding murine IFN-γ was cloned in a mycobacterial shuttle plasmid, pRBD4, with a kanamycin resistance gene [14]. Recombinant BCG strains were cultured as described above except for the inclusion of 20 mg/ml of kanamycin in growth media. Culture supernatants from recombinant BCG were used to measure secreted IFN-γ by ELISA using paired antibodies from Pharmingen (San Diego, CA).

Injection of BCG into mice

BCG cultures were thoroughly dispersed before injection by repeated gentle passage through a 27 G needle. Mice were injected with 2 or 4 × 106 CFU intravenously in the lateral tail vein in a 200-ml volume in saline. After injection, a sample of the BCG was diluted and plated on to 7H11 agar with or without kanamycin to determine the number of CFU injected into each mouse. Mice were weighed and killed by cervical dislocation on different days post-infection (as indicated in Figures). The spleen, liver and lungs were removed, weighed and the pathology monitored using the parameters described below.

Preparation of histological specimens and measurement of bacterial infection

The progress of bacterial growth was monitored at various time points in spleen, liver and lungs from infected mice. Organs were weighed before a small portion of each tissue was removed and fixed by immersion in 10% formal saline solution. Samples were embedded in paraffin wax, cut into sections and mounted onto slides. Sections were then stained with haematoxylin–eosin or Ziehl–Neelsen stain to assess the extent of granuloma formation and number of acid-fast bacilli (AFB) in granulomas. An accumulation of at least 10 mononuclear inflammatory cells including epithelioid macrophages was considered as a granuloma. The number of granulomas was counted in 50 microscopic fields (× 400 magnification) for each slide. A portion of liver and lungs was weighed and homogenized in 3 ml saline in a stomacher. These homogenates were then plated on Middlebrook 7H11 agar plates in serial 10-fold dilutions in duplicate, and were incubated at 37°C for 2 weeks. Colonies were counted and the number of CFU per organ was calculated.

Immunohistochemistry

Immunostaining was carried out using the avidin–biotin complex (ABC) method as described previously [13]. This method incorporated the use of biotinylated antibody as a link antibody. The primary antibodies used in this study were directed to type I procollagen [13] and to iNOS (Transduction Labs, Lexington, KY). All the secondary antibodies and the ABC kit were from Dako Ltd (High Wycombe, UK). After immunohistochemical staining, each slide was inspected by light microscopy and the total number of cells and the number of cells positive for iNOS or type 1 procollagen were counted in each granuloma. The scoring system for the histological changes have been validated by A.W., B.G.M., and H.T.C. [12,13]. The mean percentage of stained cells was calculated for 20 granulomas. In all cases the observer was blind to the code of the experiment.

Hydroxyproline assay

Hydroxyproline content was determined by the methods described by Stegemann & Stalder [16]. Briefly, a portion of liver or lungs was hydrolysed in 6 n HCl at 100°C overnight. The sample was mixed with chloramine T solution and incubated for 20 min at room temperature. After incubation, aldehyde/perchloric acid solution was added and incubated at 60°C. After 15 min, the samples were cooled under tap water and absorbance was measured at 550 nm. The hydroxyproline content was expressed for whole liver or lungs or as μg/100 mg tissue.

Isolation and reverse transcriptase-polymerase chain reaction amplification of mRNA

Small pieces of liver and lung tissues were snap-frozen in liquid nitrogen and stored at −70°C. Total RNA from the frozen tissue was isolated by homogenizing the organs in 0.5 ml of guanidine isothiocyanate. RNA was extracted by a modification of the acid-GTC phenol chloroform method as described previously [17]. First strand cDNA synthesis from 1 μg total RNA was carried out as described by Wangoo et al. [18]. The gene-specific primers were: IFN-γ 5′-AACGCTACACACTGCATCT-3′ (sense), 5′-AGCTCATTGAATGCTTGG-3′ (anti-sense); tumour necrosis factor-alpha (TNF-α) 5′-GCGACGTGGAACTGGCAGAAG-3′ (sense), 5′-GGTACAACCCATCGGCTGGCA-3′ (anti-sense); iNOS 5′-TCACTGGGACAGCACAGAAT-3′ (sense), 5′-TGTGTCTGCAGATGTGCTGA-3′ (anti-sense); and β-actin 5′-ATGGATGAC-GATATCGCT-3′ (sense), 5′-ATGAGGTAGTCTGTCAGGT-3′ (anti-sense). The predicted sizes of IFN-γ, TNF-α, iNOS and β-actin DNA products were 398, 382, 499 and 570 bp, respectively.

To permit the same number of cycles in the polymerase chain reaction (PCR) to be used for measurement of the cDNA of IFN-γ, TNF-α, iNOS and that of β-actin, and thus for the assays to be performed concurrently, cDNA was diluted 1:10 for IFN-γ, TNF-α and iNOS, and 1:50 for β-actin. A five-fold correction factor was subsequently used to calculate β-actin ratios. The PCR amplification mixture consisted of 10× PCR buffer, 1.25 mm MgCl2, 1 U of Taq Polymerase (Promega, Southampton, UK), 200 μm deoxyribonucleosides (dATP, dGTP, dCTP, dTTP), 0.3 μm of each primer, and appropriate dilution of cDNA, made up to 50 μl with sterile distilled water. Amplification was carried out for 33 cycles in a DNA thermal cycler (Perkin Elmer Ltd, Beaconsfield, UK) under the following reaction conditions: 94°C/1 min, 55°C/2 min, 72°C/2 min. PCR products (15 μl) were separated by electrophoresis on a 2% agarose gel containing ethidium bromide. The gel was photographed and the bands for IFN-γ, TNF-α, iNOS and β-actin were quantified using ‘Quantiscan’ software (Biosoft, Cambridge, UK). Results were first calculated as a ratio of IFN-γ, TNF-α, iNOS to that of β-actin for each sample, and finally expressed as mean ± s.e.m.

Statistical analysis

A paired t-test was used to indicate a statistically significant difference between control and experimental groups. P < 0.05 was taken to be statistically significant.

RESULTS

BCG fibrosis model

Intravenous infection of mice with BCG resulted in a time-dependent increase in liver weight and in the number of granulomas in the liver. Organ weights and the number of granulomas were maximal at day 21 and then started to decline. The weight of livers increased from 1002 ± 66.4 mg on day 7 to a peak of 1276 ± 102 mg on day 21 post-infection. Similarly the number of granulomas per 50 microscopic fields (× 400) increased from 14.5 ± 3.1 on day 7 to 180.8 ± 25.5 on day 21 post-infection, after which there was a slight decline. There was a dose-dependent relationship between the number of BCG administered (1–4 × 106) and mean number of granulomas in liver and granulomas positive for AFB. The numbers of granulomas and granulomas positive for AFB were greater in mice injected with 4 × 106 CFU compared with 1 × 106 CFU on day 21 post-infection. However, on day 28 post-infection there was a decline in both the number of granulomas and the liver granulomas positive for AFB compared with day 21 post-infection.

The importance of granulomas as a focus of inflammatory activity was shown by an increase in the percentage of granuloma cells which stained positively for iNOS by immunochemistry in the liver (Fig. 1a). The number of iNOS+ cells increased from day 7 post-infection until day 21, after which there was a slight decline. No iNOS activity was observed outside the granulomas. Similarly, the abundance of total liver iNOS mRNA associated with the granulomas also increased from week 1 post-infection (Fig. 1b), demonstrating increased iNOS expression at the transcriptional level as well as at the level of protein. The iNOS:β-actin ratio increased from 0.06% at day 7 to 0.13% at day 21.

Fig. 1.

Fig. 1

Open in a new tab

(a) Percentage of cells positive for inducible nitric oxide synthase (iNOS) by immunochemistry in granulomas of liver on different days post-infection in BALB/c mice infected with BCG. (Results represent mean of six mice ± s.e.m.) (b) Abundance of mRNA expression for iNOS by reverse transcriptase-polymerase chain reaction in liver on different days post-infection in BALB/c mice infected with BCG. Lanes 1, 2 and 3 represent three representative samples from each group.

Within the liver, there was an increase in hydroxyproline content (Fig. 2a), which served as a measure of total collagen. This was matched by an increase in the percentage of cells within the granuloma staining for type 1 procollagen (Fig. 2b), a marker of new collagen synthesis [19].

Fig. 2.

Fig. 2

Open in a new tab

Amount of hydroxyproline in liver (a) and percentage of cells positive for type I procollagen by immunochemistry in granulomas of liver (b) on different days post-infection in BALB/c mice infected with BCG. (Results represent mean of six mice ± s.e.m.)

Infection with recombinant BCG expressing IFN-γ

To evaluate the effect of increased local production of IFN-γ, BCG-IFN was compared with BCG-plasmid in the fibrosis model. Initially, the two recombinant strains were compared with respect to their growth rate in vitro over a period of 3 weeks in the presence or absence of kanamycin. There was no significant difference in growth curves of the two strains in the presence or absence of kanamycin over a 3-week period. Mean doubling time during the first 2 weeks of culture was 21.34 ± 2.45 h for BCG-plasmid and 19.95 ± 3.22 h for BCG-IFN. Secretion of IFN-γ was measured by ELISA using supernatants from in vitro cultures: the mean amount of IFN-γ observed from 4–14-day cultures of BCG-IFN was 793 ± 225 pg/ml. Biological activity of this BCG-IFN was not measured in the present study. However, in previous studies the antigen-specific proliferation and cytokine release by murine splenocytes was found to be substantially greater in cells from animals injected with BCG-IFN compared with BCG containing a plasmid which did not express cytokine [14].

Using the fibrosis model described above, mice were infected with BCG-IFN or BCG-plasmid and were killed on day 21. Infection with recombinant BCG expressing IFN-γ resulted in a statistically significant reduction in spleen and liver weights at day 21 when compared with infection with plasmid-control BCG (Fig. 3a) (P < 0.05). A similar significant reduction in bacterial load in spleen and liver was observed 21 days after infection with BCG-IFN (Fig. 3b) (P < 0.05).

Fig. 3.

Fig. 3

Open in a new tab

(a) Weight of spleen and liver as percentage of body weight in mice infected with 2 × 106 BCG transformed with empty plasmid (BCG-plasmid; ▪) or BCG expressing IFN-γ (BCG-IFN; □) on different days post-infection. Results represent mean of 6–12 mice ± s.e.m. in two separate experiments at each day post-infection; *P < 0.05 on day 21 in spleen and liver. (b) BCG colony-forming units (CFU) in mice infected with 2 × 106 BCG carrying only plasmid (BCG-plasmid; ▪) or BCG expressing IFN-γ (BCG-IFN; □) on different days post-infection. Results represent mean of 6–12 mice ± s.e.m. in two separate experiments at each day post-infection; *P < 0.05 on day 21 in spleen and liver. (c) Number of granulomas (i), amount of hydroxyproline in liver (ii) and percentage of cells positive for procollagen 1 in liver granulomas (iii) in mice infected with 2 × 106 BCG transformed with empty plasmid (BCG-plasmid; ▪) or BCG expressing IFN-γ (BCG-IFN; □) on day 21 post-infection. Results represent mean of 12 mice; *P < 0.05.

Effect of recombinant BCG expressing IFN-γ on local fibrosis and inflammation

Mice infected with BCG-IFN had fewer hepatic granulomas per 50 microscopic fields than mice infected with BCG-plasmid (Fig. 3c(i)). Infection with BCG-IFN resulted in a reduction in day 21 liver hydroxyproline content in comparison with BCG-plasmid (P < 0.05) (Fig. 3c(ii)). The reduced hydroxyproline levels were not simply a reflection of the total organ weight. The increase in liver hydroxyproline following BCG-plasmid infection was sustained at day 35 (545.1 ± 62.3 μg/liver) at a time when the liver weight had started to return to normal (6.64% body wt). A more modest increase in hydroxyproline (510.3 ± 68.8 μg/liver) was also observed at day 35 following IFN-BCG infection when liver size was similar to that following control infection (6.66% body wt). Reduced fibrosis in mice infected with BCG-IFN was further demonstrated by a lower percentage of cells within the granulomas staining for type 1 procollagen (Fig. 3c(iii)).

In the mice infected with BCG-IFN there was a decrease in iNOS/β-actin mRNA ratio in the liver compared with mice infected with BCG-plasmid (Fig. 4a), and a similar decrease in the total number of cells staining for iNOS in the liver granulomas on day 21 (Fig. 4b). However, there was no significant difference in IFN-γ/β-actin mRNA ratio or TNF-α/β-actin mRNA ratio on day 21 (Fig. 4a).

Fig. 4.

Fig. 4

Open in a new tab

(a) Comparison of abundance of mRNA expression for inducible nitric oxide synthase (iNOS), IFN-γ and tumour necrosis factor-alpha (TNF-α) as ratio of β-actin in liver of mice infected with BCG transformed with empty plasmid (BCG-plasmid; ▪) or BCG expressing IFN-γ (BCG-IFN; □) on day 21 post-infection. *P < 0.05. (b) Comparison of immunoscoring for iNOS+ cells in granulomas of liver in mice injected with BCG transformed with empty plasmid (BCG-plasmid; ▪) or BCG expressing IFN-γ (BCG-IFN; □) on day 21 post-infection. *P < 0.05.

DISCUSSION

Infection with mycobacteria induces a vigorous cell-mediated immune response which can contribute both to protection of the host and to disease-associated pathology. Similarly, vaccination with BCG is accompanied by a dose-dependent inflammatory reaction in children [11]. The precise mechanisms involved in protection and pathology are incompletely understood, and there is a risk that attempts to enhance vaccine efficacy could inadvertently result in an exacerbation of immunopathology. The purpose of the present study was to assess this possibility in the case of a recombinant BCG vaccine engineered to express murine IFN-γ. A previous study of this vaccine had demonstrated an enhanced ability to induce a type 1 T cell response to mycobacterial antigens [14]; we wished to determine if the augmented type 1 response resulted in an increase in vaccine-associated pathology.

A murine fibrosis model was established in which infection with BCG induced formation of granulomas in the liver. The formation of granulomas is a characteristic feature of mycobacterial infection. Depending on the infection resistance/susceptibility status of the host, lesions may either regress with destruction of bacteria within them, or the disease may progress with surviving lesions going on to caseate and liquefy [20]. Although widespread caseous necrosis can be seen in gene disrupted mice (IFN-γ, TNF, etc.), this event does not seem to spread in normal mice, perhaps because of fibrosis and collagen deposition in these areas, which acts to wall off these pockets [21]. In the present study, a corresponding increase in liver hydroxyproline content and a high percentage of granuloma cells with positive staining for type 1 procollagen demonstrated that granuloma formation contributed to the process of fibrosis and possible consequent tissue damage.

BCG-induced granulomas were characterized by the induction of iNOS expression. Nitric oxide synthase is induced in macrophages by exogenous microorganisms and/or their products, for example lipopolysaccharide (LPS), and by endogenous type 1 cytokines, notably IFN-γ [22]. Nitric oxide (NO) can exert both beneficial and deleterious effects on tissues in the setting of infection or other disease processes. In murine models, NO has been demonstrated to be important in the defence against several microbial species, including M. tuberculosis [23]. Treatment of human alveolar macrophages with a potent inhibitor of iNOS, N(G)-monomethyl-l-arginine monoacetate (L-NMMA), has been shown to reduce killing of BCG in human alveolar macrophages [24]. NO may also play a regulatory role in the granulomatous response. Inhibition of iNOS activity during formation of granulomas induced by mycobacterial antigens resulted in enlarged granulomatous lesions containing increased numbers of neutrophils and eosinophils [25]. This regulation may be related to vascular events that permit leucocyte infiltration, or elevated levels of nitric oxide may directly alter the type 1 cytokine profile [25]. iNOS has been also shown to enhance wound repair using iNOS knockout mice [26]. In the present study iNOS expression may have a dual role. First, IFN-γ from BCG-IFN-infected mice might up-regulate iNOS in earlier stages of infection than BCG-plasmid-infected mice and therefore induce more rapid BCG killing. Second, iNOS production at the sites of granulomas might serve to limit the extent of immunopathology by down-regulating production of proinflammatory cytokines. However, other studies have also associated NO with tissue injury [27] and cytotoxic effects [28]. Increased expression of iNOS and NO production have been shown in a number of chronic inflammatory diseases of upper airways and lungs, for example [29,30], and macrophages in the lungs of patients with clinically active TB often express increased levels of iNOS [31]. Recently, iNOS has been shown also to play a paradoxical role, both in the effector mechanisms of resistance and in the down-regulation of the type 1 cytokine response, which is ultimately required for NO production [32]. This raises the possibility that in our study, iNOS might also play a role in down-regulation of IFN-γ and hence have a negative feedback on iNOS production before any tissue damage is caused.

Production of IFN-γ is a functional marker of murine T cells that confer adoptive immunity against M. tuberculosis [33]. Therefore, immunological resistance to mycobacterial infection in mice is probably mediated by Th1 cells producing IFN-γ [33], whereas mice susceptible to certain mycobacterial infections have been shown to produce higher levels of Th2 cytokines, such as IL-4, and relatively little IFN-γ [34], or less IFN-γ-inducing cytokines [35]. Other work has shown that an experimental mouse model of TB consists of an early protective T cell response in which IFN-γ is elaborated at sites of bacterial implantation to contain and destroy the bacterial infection, and an antibody response occurs in later stages of infection mediated by type 2, IL-4-secreting cells to process dead bacteria by macrophages [36]. Alternatively, Rook and colleagues suggest that when a Th2 response is superimposed upon a pre-existing Th1 response, the resulting cell-mediated inflammatory site may become exquisitely sensitive to cytokine-mediated damage [37], thus exacerbating local tissue damage [38]. It is apparent from all these studies that Th2 cytokines, such as IL-4, do not contribute directly to protection in models of mycobacterial infection. However, our data are consistent with previous evidence suggesting an enhancement of the type 1 immune response by cytokine-secreting BCG constructs [14]. We observed a more rapid clearance of bacteria in mice infected with the recombinant BCG strain expressing IFN-γ. Importantly, this effect was not associated with enhancement in any of the markers of tissue pathology. Granuloma formation, iNOS expression and fibrosis were all less prominent in mice infected with BCG-IFN than in mice infected with BCG-plasmid. A possible explanation for these findings is that the local expression of recombinant IFN-γ triggered early activation of a protective type 1 response, promoting bacterial killing in the absence of fibrosis and reducing the number of organisms capable of inducing subsequent pathology. Alternatively, an increase in the level of IFN-γ may be responsible directly for preventing fibrosis. IFN-γ has been shown to reduce collagen formation in a bleomycin-induced lung fibrosis model, e.g. in bleomycin-mouse model [39], and in murine schistosomiasis [40].

The important conclusion from this study is that, as judged by the tissue response, it may be possible to enhance the potentially protective type 1 immune response to BCG vaccination without necessarily compromising its safety, or exacerbating local inflammatory sequelae.

Acknowledgments

We thank Peter Murray for the generous gift of recombinant BCG strains. This work was supported by grants from the Wellcome Trust, UK.

REFERENCES

  • 1.Snider DE, Jr, Montague JR. The neglected global tuberculosis problem: a report of the 1992 world congress on tuberculosis. J Infect Dis. 1994;169:1189–96. doi: 10.1093/infdis/169.6.1189. [DOI] [PubMed] [Google Scholar]
  • 2.Raviglione MC, Snider DE, Jr, Kochi A. Global epidemiology of tuberculosis. Morbidity and mortality of a worldwide epidemic. JAMA. 1995;273:220–6. [PubMed] [Google Scholar]
  • 3.Colditz GA, Brewer TF, Berkey CS, et al. Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature. JAMA. 1994;271:698–702. [PubMed] [Google Scholar]
  • 4.Kaufmann SHE, Flesch I. Function and antigen recognition pattern of L3T4+ T cell clones from Mycobacterium tuberculosis-immune mice. Infect Immun. 1986;54:291–6. doi: 10.1128/iai.54.2.291-296.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Ottenhoff THM, Kal AB, Van Embden JDA, et al. The recombinant 65 kD heat shock protein of Mycobacterium bovis BCG/M. tuberculosis is a target molecule for CD4+ cytotoxic T lymphocytes that lyse human monocytes. J Exp Med. 1988;168:1947–52. doi: 10.1084/jem.168.5.1947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Barnes PF, Mistry SD, Cooper CL, et al. Compartmentalization of a CD+ T lymphocyte subpopulation in tuberculous pleuritis. J Immunol. 1989;142:1114–9. [PubMed] [Google Scholar]
  • 7.Flynn JL, Chan J, Triebold KJ, et al. An essential role for interferon gamma in resistance to Mycobacterium tuberculosis infection. J Exp Med. 1993;178:2249–54. doi: 10.1084/jem.178.6.2249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Newport MJ, Huxley CM, Huston S, et al. Mutation in the interferon-gamma-receptor gene and susceptibility to mycobacterial infection. N Engl J Med. 1996;335:1941–9. doi: 10.1056/NEJM199612263352602. [DOI] [PubMed] [Google Scholar]
  • 9.Chan J, Xing Y, Magliozzo RS, Bloom BR. Killing of virulent Mycobacterium tuberculosis by reactive nitrogen intermediates produced by activated murine macrophages. J Exp Med. 1992;175:1111–22. doi: 10.1084/jem.175.4.1111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Bloom BR, Fine PEM. Tuberculosis: pathogenesis, protection and control. Washington: American Society for Microbiology; 1994. The BCG experience: implications for future vaccines against tuberculosis; p. 531. [Google Scholar]
  • 11.Miles MM, Shaw RJ. Effect of inadvertent intradermal administration of high dose percutaneous BCG vaccine. Br Med J. 1996;312:1205. doi: 10.1136/bmj.312.7040.1205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Marshall B, Wangoo A, Cook HT, Shaw RJ. Increased inflammatory cytokines and new collagen formation in cutaneous tuberculosis and sarcoidosis. Thorax. 1996;51:1253–61. doi: 10.1136/thx.51.12.1253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Wangoo A, Cook HT, Taylor GM, Shaw RJ. Enhanced expression of type 1 procollagen and transforming growth factor beta in tuberculin induced delayed type hypersensitivity. J Clin Pathol. 1995;48:339–45. doi: 10.1136/jcp.48.4.339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Murray PJ, Aldovini A, Young RA. Manipulation and potentiation of antimycobacterial immunity using recombinant bacille Calmette–Guérin strains that secrete cytokines. Proc Natl Acad Sci USA. 1996;93:934–9. doi: 10.1073/pnas.93.2.934. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Brown CA, Brown IN. Mycobacterium bovis, BCG, modulation of murine antibody responses: influence of dose and degree of aggregation of live or dead organisms. Br J Exp Pathol. 1982;63:133–43. [PMC free article] [PubMed] [Google Scholar]
  • 16.Stegemann H, Stadler K. Determination of hydroxyproline. Clin Chim Acta. 1967;18:267. doi: 10.1016/0009-8981(67)90167-2. [DOI] [PubMed] [Google Scholar]
  • 17.Wangoo A, Taylor IK, Haynes AR, Shaw RJ. Upregulation of alveolar macrophage platelet derived growth factor B (PDGF B) mRNA by interferon gamma from Mycobacterium tuberculosis antigen (PPD) stimulated lymphocytes. Clin Exp Immunol. 1993;94:43–50. doi: 10.1111/j.1365-2249.1993.tb05975.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Wangoo A, Laban C, Cook HT, et al. Interleukin-10 and corticosteroid induced reduction in type I procollagen in a human ex vivo scar culture. Int J Exp Pathol. 1997;78:33–41. doi: 10.1046/j.1365-2613.1997.d01-241.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.McDonald JA, Broekelmann TJ, Matheke ML, et al. A monoclonal antibody to the carboxyterminal domain of procollagen type I visualizes collagen-synthesizing fibroblasts. Detection of an altered fibroblast phenotype in lungs of patients with pulmonary fibrosis. J Clin Invest. 1986;78:1237–44. doi: 10.1172/JCI112707. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Doenhoff MJ. Granulomatous inflammation and the transmission of infection: schistosomiasis—and TB too? Immunol Today. 1998;19:462–7. doi: 10.1016/s0167-5699(98)01310-3. [DOI] [PubMed] [Google Scholar]
  • 21.Orme IM. The immunopathogenesis of tuberculosis: a new working hypothesis. Trends Microbiol. 1998;6:94–7. doi: 10.1016/s0966-842x(98)01209-8. [DOI] [PubMed] [Google Scholar]
  • 22.Modolell M, Corraliza IM, Link F, et al. Reciprocal regulation of the nitric oxide synthase/arginase balance in mouse bone marrow-derived macrophages by TH1 and TH2 cytokines. Eur J Immunol. 1995;25:1101–4. doi: 10.1002/eji.1830250436. [DOI] [PubMed] [Google Scholar]
  • 23.Green SJ, Scheller LF, Marletta MA, et al. Nitric oxide: cytokine-regulation of nitric oxide in host resistance to intracellular pathogens. Immunol Letters. 1994;43:87–94. doi: 10.1016/0165-2478(94)00158-8. [DOI] [PubMed] [Google Scholar]
  • 24.Nozaki Y, Hasegawa Y, Ichiyama S, et al. Mechanism of nitric oxide-dependent killing of Mycobacterium bovis BCG in human alveolar macrophages. Infect Immun. 1997;65:3644–7. doi: 10.1128/iai.65.9.3644-3647.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Hogaboam CM, Chensue SW, Steinhauser ML, et al. Alteration of the cytokine phenotype in an experimental lung granuloma model by inhibiting nitric oxide. J Immunol. 1997;159:5585–93. [PubMed] [Google Scholar]
  • 26.Yamasaki K, Edington HD, McClosky C, et al. Reversal of impaired wound repair in iNOS-deficient mice by topical adenoviral-mediated iNOS gene transfer. J Clin Invest. 1998;101:967–71. doi: 10.1172/JCI2067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Mulligan MS, Hevel JM, Marletta MA, Ward PA. Tissue injury caused by deposition of immune complexes is L-arginine dependent. Proc Natl Acad Sci USA. 1991;88:6338–42. doi: 10.1073/pnas.88.14.6338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Fukuo K, Inoue T, Morimoto S, et al. Nitric oxide mediates cytotoxicity and basic fibroblast growth factor release in cultured vascular smooth muscle cells. A possible mechanism of neovascularization in atherosclerotic plaques. J Clin Invest. 1995;95:669–76. doi: 10.1172/JCI117712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Furukawa K, Harrison DG, Saleh D, et al. Expression of nitric oxide synthase in the human nasal mucosa. Am J Respir Crit Care Med. 1996;153:847–50. doi: 10.1164/ajrccm.153.2.8564142. [DOI] [PubMed] [Google Scholar]
  • 30.Hamid Q, Springall DR, Riveros-Moreno V, et al. Induction of nitric oxide synthase in asthma. Lancet. 1993;342:1510–3. doi: 10.1016/s0140-6736(05)80083-2. [DOI] [PubMed] [Google Scholar]
  • 31.Nicholson S, Bonecini-Almeida MG, Lapa e, Silva JR, et al. Inducible nitric oxide synthase in pulmonary alveolar macrophages from patients with tuberculosis. J Exp Med. 1996;183:2293–302. doi: 10.1084/jem.183.5.2293. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.James SL, Cheever AW, Caspar P, Wynn TA. Inducible nitric oxide synthase-deficient mice develop enhanced type 1 cytokine-associated cellular and humoral immune responses after vaccination with attenuated Schistosoma mansoni cercariae but display partially reduced resistance. Infect Immun. 1998;66:3510–8. doi: 10.1128/iai.66.8.3510-3518.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Orme IM, Miller ES, Roberts AD, et al. T lymphocytes mediating protection and cellular cytolysis during the course of Mycobacterium tuberculosis infection. J Immunol. 1992;148:189–96. [PubMed] [Google Scholar]
  • 34.Huygen K, Abramowicz D, Vandenbussche P, et al. Spleen cell cytokine secretion in Mycobacterium bovis BCG-infected mice. Infect Immun. 1992;148:2887–93. doi: 10.1128/iai.60.7.2880-2886.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Kobayashi K, Kai M, Gidoh M, et al. The possible role of interleukin (IL)-12 and interferon-gamma-inducing factor/IL-18 in protection against experimental Mycobacterium leprae infection in mice. Clin Immunol Immunopathol. 1998;88:226–31. doi: 10.1006/clin.1998.4533. [DOI] [PubMed] [Google Scholar]
  • 36.Orme IM, Roberts AD, Griffin JP, Abrams JS. Cytokine secretion by CD4 T lymphocytes acquired in response to Mycobacterium tuberculosis infection. J Immunol. 1993;151:518–25. [PubMed] [Google Scholar]
  • 37.Rook GA, Hernandez-Pando R. T cell helper types and endocrines in the regulation of tissue-damaging mechanisms in tuberculosis. Immunobiology. 1994;191:478–92. doi: 10.1016/S0171-2985(11)80454-7. [DOI] [PubMed] [Google Scholar]
  • 38.Hernandez-Pando R, Orozcoe H, Sampieri A, et al. Correlation between the kinetics of Th1, Th2 cells and pathology in a murine model of experimental pulmonary tuberculosis. Immunology. 1996;89:26–33. [PMC free article] [PubMed] [Google Scholar]
  • 39.Gurujeyalakshmi G, Giri SN. Molecular mechanisms of antifibrotic effect of interferon gamma in bleomycin-mouse model of lung fibrosis: down regulation of TGF-beta and procollagen I and III gene expression. Exp Lung Res. 1995;21:791–808. doi: 10.3109/01902149509050842. [DOI] [PubMed] [Google Scholar]
  • 40.Czaja MJ, Weiner FR, Takahashi S, et al. Gamma-interferon treatment inhibits collagen deposition in murine schistosomiasis. Hepatology. 1989;10:795–800. doi: 10.1002/hep.1840100508. [DOI] [PubMed] [Google Scholar]

Articles from Clinical and Experimental Immunology are provided here courtesy of British Society for Immunology

RESOURCES