Search

Tamm-Horsfall protein (THP) is the major glycoprotein component of urine, yet its biological role remains obscure. Recent reports have suggested that a concanavalin A (Con A)-binding fraction of THP from pregnancy urine can bind the cytokines tumour necrosis factor-alpha (TNF-alpha) and interleukin-1 (IL-1). In order to investigate this claim in relation to THP from normal adult urine we raised monoclonal antibodies to THP and sought THP/TNF-alpha interactions in three separate assay systems. We found no evidence that THP binds to TNF-alpha under physiological conditions, but we observed that it exerts a weak, probably not physiologically relevant, but reproducible inhibitory effect on the toxicity of TNF-alpha for monolayers of L929 cells, even when the cells are pretreated with the THP, and washed before addition of the cytokine. Since our preparations of THP do not interact directly with TNF-alpha we postulated an interaction with the cells themselves, or with their extracellular matrix. The THP was found by ELISA, immunoblotting and immunohistology, to bind to as yet unidentified components of the extracellular matrix in a manner dependent on cations, pH and carbohydrates. These data, considered in the light of the published amino acid sequence and biochemical properties, suggest that THP is a member of a structural glycoprotein family known to modulate cell adhesion.

In this brief review, inspired partly by a symposium at the autumn meeting of the British Society for Immunology, 1992, varying hypotheses concerning the etiopathogenesis of rheumatoid arthritis (RA) are explored and tested against current evidence. Immunogenetic considerations, whilst of interest, have not aided our understanding of the development of this disease. The association with restricted HLA-DR beta chain hypervariable sequences does not hold true with all cases of RA (but may be related to disease severity) and studies of T cell receptor (TCR) beta chain usage fail to show consistent oligoclonality of infiltrating T cells in the synovial compartment. Etiologies based on triggering by bacteria are also considered: homologies between the 'shared epitope' sequences of HLA-DR1 and DR4 beta chains, Escherichia coli dnaJ and Proteus haemolysin do not indicate any feasible mechanisms for the development of RA, and cannot explain the many cases in which such DR sequences do not occur, though new data from man and animals enhance interest in the role of bowel flora. Finally, the striking parallels between slow bacterial infections and RA, in terms of immunogenetics, pathology, IgG glycosylation abnormalities and autoimmune manifestations, are put forward as circumstantial evidence that such bacterial infections may underly, or trigger, this serious disease.

Objective: To investigate the prevalence of Mycobacterium tuberculosis DNA in granulomatous tissues from patients with sarcoidosis and from controls matched for age, sex, and tissue by using the polymerase chain reaction., Design: Single blind control trial., Subjects: 16 patients with sarcoidosis who had undergone diagnostic biopsy of lung, skin, or lymph node and 16 patients with squamous cell carcinoma or Hodgkin's disease to act as controls. In addition, four lung specimens infected with M tuberculosis were included as positive controls., Results: M tuberculosis DNA was present in sarcoid tissues containing granulomas from seven of the 16 patients and one of the 16 matched controls. Two of the four specimens known to be infected with M tuberculosis were positive in the controlled experiment., Conclusion: These figures suggest that M tuberculosis DNA is detected as readily in patients with sarcoidosis as in patients with frankly tuberculous tissues and imply that M tuberculosis may be linked to the cause of sarcoidosis.

Objective: To assess the claim that the human 60-kd heat-shock protein (HSP) is highly expressed in the joints of patients with rheumatoid arthritis (RA), but is not readily detected in normal tissues., Methods: Monoclonal antibodies were raised against the human 60-kd mitochondrial heat-shock protein (P1 protein; hsp60), and their specificity was established. They were then applied to synovial tissue., Results: HSP was expressed similarly in normal, osteoarthritic, and RA synovium. Low levels of hsp60 were detected in synovial fluid by immunoprecipitation., Conclusion: Minor differences in the distribution of hsp60 in synovium from RA joints were attributable to increased cellularity and to the disorganization of the tissue architecture.

It has previously been shown that the inherently tumour necrosis factor-alpha (TNF-alpha)-sensitive L929 murine fibroblast cell line becomes much more sensitive to the cytotoxic effect of this cytokine after exposure to Mycobacterium tuberculosis in culture. In this study it is now shown that normal human cells of types likely to be involved in tuberculous lesions are affected in a similar way. Growth of normal human fibroblasts is usually stimulated by TNF-alpha in vitro, but after exposure to M. tuberculosis or to extracts of this organism, these cells are killed rather than stimulated by subsequent exposure to TNF-alpha. Similarly, human endothelial cells become susceptible to doses to TNF-alpha which do not normally affect viability. Moreover this enhancement of sensitivity to TNF-alpha is not confined to its toxicity. Endothelial cells and HeLa cells exposed to M. tuberculosis express increased levels of ICAM-1 after subsequent exposure to TNF-alpha, implying synergy between the two stimuli. It is suggested that these effects contribute to the ability of M. tuberculosis to distort the normal protective role of TNF-alpha so that the cytokine becomes detrimental to the host.

Serum immunoglobulin (Ig) G responses to a variety of mycobacterial antigens were measured in children from the UK, in children with tuberculosis from Hyderabad, India and Dhaka, Bangladesh, classified according to whether the disease was disseminated or localized, and in non-tuberculous controls. Anti-lipoarabinomannan (LAM) IgG responses in UK children showed a marked trough between 6 months and 3 years coincident with the reported peak incidence of disseminated tuberculosis. Geometric mean IgG responses to sonicates of slow-growing mycobacteria (rich in LAM) in 36 children with disseminated tuberculosis were markedly lower than in 99 children with localized tuberculous lesions (for Mycobacterium scrofulaceum P < 0.01, for M. tuberculosis P < 0.01, and for M. vaccae P < 0.01). Responses to purified LAM were also lower in the disseminated tuberculosis group (P < 0.05) but there was no difference between the groups in their response to mycobacterial 65 kDa protein. Multiple regression analysis showed that the reduced response to sonicated mycobacterial antigens and to LAM in children with disseminated disease was independent of age, nutritional status, skin test reactivity, duration of previous symptoms, and city of origin. There was no evidence for sequestration of antibody to immune complexes. These findings are compatible with the hypothesis that children with low levels of antibody to sonicated mycobacterial antigen and to LAM, or those who cannot mount an antibody response, are predisposed to dissemination. A role for antibody in preventing disseminated forms of tuberculosis in childhood has implications for the development of improved vaccines and for the optimum timing of vaccination with bacille Calmette-Guérin.

The effect of pregnancy on the incidence and severity of pristane-induced arthritis was examined along with the glycosylation status of IgG during the ante-natal and post-partum periods. It was found that pristane-induced arthritis is prevented by pregnancy. In addition, the levels of agalactosyl IgG fall during pregnancy but rise to greater than normal within a few days of parturition, before resetting towards the norm shortly afterwards. Interestingly, the level of agalactosyl IgG correlates with the severity of arthritis. As previously reported IL-6 may be an important factor, not necessarily the only one, in the production of agalactosyl IgG. Here it is clearly demonstrated that the kinetics of IL-6 activity post-pristane injection parallels the kinetics of agalactosyl IgG production. In addition, the overshoot in agalactosyl IgG levels immediately post-partum coincides with a burst in IL-6 activity. It is considered that these changes in IgG glycoform levels, or the factors which control them, may be related to the mechanisms underlying prevention/remission of arthritis during pregnancy.

Tissue sites involved in certain types of inflammation become sensitive to the destructive effect of a subsequent injection of tumour necrosis factor alpha (TNF alpha). To try to further delineate the cascade of effector and regulatory events controlling this activity, a new model is described and its main properties characterized. C57BL/GrFa mice received mycobacterial products subcutaneously in the footpads. Recombinant TNF alpha was injected 24 h later into the same sites. To assess the tissue-destructive effect of TNF alpha in these "primed" footpads, swelling and haemoglobin content of injected footpads were measured, 16 h and 20 h respectively after the injection of TNF alpha. When loaded with either Escherichia coli LPS (10 micrograms) or Mycobacterium vaccae soluble sonicate (17 micrograms), footpads were reactive to the subsequent injection of 1 microgram recombinant TNF alpha, as assessed by both swelling and haemoglobin content. When C57BL/GrFa mice received 10(9) autoclaved M. vaccae subcutaneously in the back 10 days before the footpad was "primed" with soluble M. vaccae sonicate, the destructive effect of TNF alpha was significantly enhanced, becoming 5-10-fold greater than that seen in sites "primed" with an optimal dose of LPS. This higher reactivity was abrogated by a single dose of anti-CD4 given just before the injection of TNF alpha. This local reactivity to TNF alpha of skin sites loaded with mycobacterial products is compared to the local LPS-dependent Shwartzman reaction, and the relevance of this assay as a model with which to delineate the mechanisms of tissue damage in tuberculosis is discussed.

Mycobacterial lesions and skin sites challenged with soluble mycobacterial antigen are very sensitive to the necrotizing effect of tumour necrosis factor (TNF). We have used a model that permits separate quantitative assessment of swelling and haemorrhage to show that when these reactions are elicited in mice that have not been deliberately immunized, pretreatment of the mice with lipopolysaccharide (LPS), or with a MoAb to CR3 which blocks emigration of myeloid cells into the tissues, will block both the swelling and the haemorrhage. On the other hand, treatment with an inhibitor of platelet-activating factor (PAF), or with misoprostol (a synthetic prostaglandin E1 analogue), or with cobra venom factor (CVF) which depletes complement, preferentially blocks the haemorrhagic component, while leaving the swelling relatively unaltered. As swelling occurs before the haemorrhage is seen, it is possible that these factors act at a late stage in the cascade of events leading to the tissue damage. However, LPS and CVF were able to inhibit swelling and haemorrhage in the massive reactions elicited in pre-immunized animals, whereas the PAF inhibitor had no detectable effect.

In this article, Graham Rook and John Stanford propose that a group of idiopathic diseases that are often associated with a degree of autoimmunity and arthritis, including rheumatoid arthritis, inflammatory bowel disease, sarcoidosis and psoriasis, are caused by extremely slow-growing bacteria. They suggest that these diseases are one end of a continuous spectrum caused by related slow-growing genera, which ranges from rheumatoid arthritis, through Takayasu's arteritis and Whipple's disease, to reach the conventional mycobacterioses such as tuberculosis and leprosy.

Tumour Necrosis Factor alpha (TNF/Cachectin) is a cytokine produced mainly by macrophages, which has been shown to cause endothelial cell damage, pyrexia and weight loss, clinical features of tuberculosis, but not of sarcoidosis which is in many other respects a similar disease. 1,25 di-hydroxy Vitamin D and gamma interferon, factors which are present in vivo in both tuberculosis and sarcoidosis, enhance the ability of macrophages to release TNF in vitro. We have studied the ability of pulmonary alveolar macrophages (PAM) harvested by broncho-alveolar lavage (BAL) to produce TNF in response to stimulation with E. coli endotoxin lipopolysaccharide (LPS). 25 patients undergoing bronchoscopy and BAL were studied: 9 with sarcoidosis, 7 with tuberculosis (TB) and 9 (non-neoplastic) disease controls. TNF was assayed by Enzyme Linked Immunosorbent Assay (ELISA) in lavage fluid and cell culture supernatants. No TNF was detected in lavage fluid from any of the groups. PAMs from control patients released no detectable TNF spontaneously, but released 59 +/- 31 units after LPS stimulation. Cells from patients with sarcoidosis and tuberculosis released TNF spontaneously in vitro (TB 226 +/- 106 units; Sarcoidosis 293 +/- 176). TNF release by these cells was not increased further by addition of an optimal concentration of LPS. Thus, the pulmonary macrophages of patients with sarcoidosis and tuberculosis released significantly more TNF than those of controls.

In addition to its well known effects on reproductive organs and lactation the pituitary hormone prolactin (PRL) also influences immune functions. The present investigation was performed in order to clarify the regulatory role of prolactin in autoimmune disease by using the collagen II arthritis model. Groups of virgin female DBA/1 mice were subjected to different short-term treatment protocols (5-10 days) with rat PRL and the drugs bromocriptine (inhibits prolactin secretion) and haloperidol (enhances prolactin secretion). Treatments were performed at different stages of disease development, and effects on clinical scores, anti-collagen II antibody titres as well as agalactosyl IgG levels were recorded. The effects of the treatment protocols on serum PRL levels were assessed by using a radioimmunoassay (RIA) system. Although the accumulated results of the present study indicate that PRL does not fulfil a major role in the regulation of collagen II arthritis, some interesting observations were made. High levels of PRL (PRL injections) made the arthritis worse only if treatment was performed during the induction stage of the disease. Bromocriptine treatment during the immunisation period did not significantly affect the course of arthritis, but treatment at later stages tended to cause exacerbation (significant at the onset period only). These results indicate that PRL has different effects during early and late stages of the development of collagen II-induced arthritis.

IL-6 titres in sera and peritoneal exudate fluids (PEF) derived from pristane injected DBA/1 and CBA/Igb mice were measured. Arthritic DBA/1 mice had significantly higher serum IL-6 titres than nonarthritic or normal mice at 160 days post pristane injection. By contrast, although both arthritic and non-arthritic CBA/Igb mice had higher serum IL-6 titres than normal mice, there was no significant difference in serum IL-6 titre between these two groups at day 200-230. In both strains of mice, the IL-6 titres in PEF were more than 10 times serum levels regardless of arthritis. As previously reported for CBA/Igb mice, agalactosyl IgG levels are raised in pristane injected DBA/1 mice and the percentage is higher in arthritic animals than that in non-arthritic mice. An association between serum agalactosyl IgG levels and PEF IL-6 in pristane injected DBA/1 was demonstrated. Moreover, the injection of recombinant IL-6 into normal mice increased their serum agalactosyl IgG levels. However, it is considered that IL-6 is not the only factor involved in the production of agalactosyl IgG.

It was found that the percentage of IgG-associated agalactosyl N-linked oligosaccharides (G0) falls during normal human pregnancy and rises to values higher than before conception following delivery (n = 10, 39-55 days after delivery). Serial bleeds from a normal pregnant woman showed a fall in the percentage G0 during gestation and a rapid rise post-partum. A similar study on a pregnant arthritic woman with a pathologically elevated percentage G0 also showed a fall in percentage G0 during pregnancy and a rapid rise post-partum. The changes in IgG glycosylation in the pregnant arthritic woman occurred simultaneously with the pregnancy-induced remission and post-partum recurrence of disease. A further seven pregnant women with rheumatoid arthritis were studied and analysis of their G0 values pre- and post-partum confirmed the result. In a further series of experiments using an animal model of rheumatoid arthritis, DBA/1 mice with collagen-induced arthritis were found to have elevated G0 levels compared with control mice. The percentage G0 was found to fall simultaneously with pregnancy-induced remission to the same value as non-arthritic pregnant mice. Post-partum recurrence of arthritis in these mice was also accompanied by a simultaneous and rapid rise in percentage G0. Pseudopregnancy did not result in a change in the percentage G0, confirming the effect of true pregnancy. Since the proportion of agalactosyl IgG is abnormally high in the serum of patients with rheumatoid arthritis these changes in IgG glycoform levels, or the factors which control them, may be related to the mechanisms underlying remission of arthritis in humans during pregnancy.

Unlike Mycobacterium leprae, Mycobacterium tuberculosis is not found inside cells other than macrophages and polymorphonuclear cells in vivo, yet previous work has revealed that in vitro it readily enters all cell lines tested. Moreover, these cells are not killed by the intracellular mycobacteria. We report here that when fibroblasts take up live (but not killed) M. tuberculosis H37Rv, they develop greatly increased sensitivity to the toxic effects of tumor necrosis factor (TNF) whether the cell line is inherently sensitive to TNF or not. Ultrasonically disrupted M. tuberculosis also has this property. The increased sensitivity is seen in the absence of metabolic inhibitors, although addition of emetine, an inhibitor of protein synthesis, causes the effect to manifest itself earlier and at a lower concentration of TNF. In contrast, infection with Mycobacterium bovis bacillus Calmette-Guérin induces little or no increased sensitivity to TNF, whereas Mycobacterium avium and M. tuberculosis H37Ra have intermediate sensitivities. We discuss the possibility that virulent tuberculosis strains produce a factor which distorts the normal protective function of TNF, rendering it toxic to host tissues and leading to the classical immunopathology of tuberculous lesions.

Tumor necrosis factor-alpha (TNF) is a cytokine involved in the pathogenesis of shock and in granuloma formation, tissue necrosis, and fibrosis, in many organ systems, including the lung. It has been suggested that cells from patients infected by the human immunodeficiency virus (HIV + ve) are primed for TNF release. We postulated that TNF release from the alveolar macrophages (AM) of such patients with lung disease might lead to their observed pulmonary dysfunction. We present data confirming that peripheral blood monocytes (PBM) and demonstrating that AM from HIV + ve patients with pulmonary manifestations show significantly greater TNF production than those from HIV-negative (HIV - ve) subjects. In addition, we found sequentially significant increases in TNF production from AM and PBM of HIV + ve patients with no pathogens detected at bronchoscopy (NB), bacterial pneumonia (BP), and those with Pneumocystis carinii pneumonia (PCP). The overall TNF levels were greater from AM than PBM in all groups other than spontaneous production from HIV - ve subjects. Adherent populations of PBM and AM were incubated for 4 h with lipopolysaccharide (10 micrograms/ml) or control medium alone. Cell-free supernatants were examined for the presence of TNF using an immunoassay. The TNF levels (mean +/- SD) in IU/ml from stimulated PBM of the PCP, BP, NB, and control groups, respectively, were 186 +/- 36, 140 +/- 30, 95 +/- 18, and 55 +/- 10 and the spontaneous levels were 123 +/- 25, 100 +/- 22, 75 +/- 24, and 11 +/- 5.(ABSTRACT TRUNCATED AT 250 WORDS)

Recombinant phage clones, TB1 and TB2, were selected from a Mycobacterium tuberculosis lambda gt11 DNA expression library by screening with a polyclonal antiserum raised against the antigen 85 complex of Mycobacterium bovis BCG. Analysis of recombinant DNA inserts and expressed fusion proteins showed that two new genes had been isolated. The product of clone TB2 was identified as a member of the 30/31-kDa antigen 85 complex. Restriction enzyme analysis showed that this gene differs from previously cloned members of this antigen complex, with detailed serological analysis indicating that it may encode the 85C component. Antisera raised against the expressed product of clone TB1 recognized a 55-kDa protein in M. tuberculosis extracts. The 55-kDa protein also has fibronectin-binding activity and, like the 30/31-kDa family, is a prominent target of the antibody response in patients with mycobacterial disease. Although the clones were selected by using the same antiserum, detailed analysis by serology and by DNA hybridization showed that they represent two quite distinct types of fibronectin-binding activities expressed by M. tuberculosis. Further analysis of the fibronectin-binding antigens of M. tuberculosis may provide important insights into their role in mediating the interaction with the host immune system.

Pregnancy is known to influence the course of rheumatoid arthritis (RA) in women, as well as type II collagen-induced arthritis (CIA) in DBA/1 mice. A characteristic feature is the remission during gestation and the exacerbation of the diseases during the post-partum period. In the case of CIA in DBA 1 mice, two hormonal changes have been assumed to be critical for the induction of the post-partum flare: (i) the fall in steroid hormone levels from those present during pregnancy; and (ii) surges of prolactin (PRL) release at and after delivery. Our results show that treatment with oestradiol during a short period immediately after parturition protects the mouse from a post-partum flare of the disease, and that treatment with bromocriptine, a drug known to inhibit the endogenous PRL release, has a significant though less marked effect. Studies of lactating (i.e. animals with physiological stimulation of endogenous PRL release) and non-lactating arthritic mice revealed no clear-cut differences, indicating that PRL is of minor importance for the induction of the post-partum flare. Some steroids other than oestradiol, which may be implicated in the exacerbation of arthritis, namely progesterone and hydrocortisone, had no clinical effect. Analyses of agalactosyl IgG levels in mice with CIA, and anti-collagen II antibodies in sera collected at the end of the experiments revealed no significant differences between the oestradiol and the control groups. The successful oestradiol treatment of the mice indicates that the drop in endogenous oestradiol levels prior to delivery ends the oestrogen-mediated protection against arthritis during pregnancy.

We outline the mechanisms contributing to the human form of the Koch phenomenon, which we define as necrosis occurring within 24-48 h of injection of mycobacterial antigen into the skin of past or present tuberculosis patients. It is probable that tissue damage mediated in the same way occurs in the lesions themselves. We suggest that the necrosis is mediated in part by cytokines, particularly Tumour Necrosis Factor (TNF), and that this occurs for three reasons. First, Mycobacterium tuberculosis evokes an immunoregulatory abnormality characterised by raised agalactosyl IgG. This abnormality, also found in rheumatoid arthritis, Crohn's disease, and Erythema Nodosum Leprosum, seems to be associated with dysregulation of cytokine release. Secondly, M. tuberculosis itself triggers further cytokine release. Thirdly, the normally protective role of TNF is distorted by several interacting properties of components of M. tuberculosis, which render the cytokine toxic to the host tissues. The immunoregulatory abnormality may be susceptible to correction by immunotherapy.

Several prominent mycobacterial protein antigens involved in antibody and T cell responses have been identified as members of highly conserved heat shock protein families. In particular, immune responses to the mycobacterial 65 kD heat shock protein (hsp65) have been implicated in the pathogenesis of autoimmune diseases both in experimental animal models and in man. Additionally, hsp65 has been shown to modulate the course of autoimmune disease in such experimental animal systems. In this report, we have examined the synthesis of heat shock proteins by a fast growing mycobacterial strain, M. vaccae, in heat stressed cultures and used the pristane induced arthritis model to investigate the immunoprophylactic and immunotherapeutic potential of heat killed M. vaccae. Heat shock of M. vaccae cultures at 48 degrees C demonstrated a 43-fold increase in hsp65 over that expressed at 37 degrees C. It is therefore suggested that heat killed M. vaccae contains sufficient hsp that can be presented in the context of appropriate adjuvant properties for use as an effective immunomodulatory agent. Immunisation experiments with M. vaccae revealed that protection or exacerbation of pristane induced arthritis was dependent on the dose (given in an oil or aqueous suspension), route and time of immunisation. In addition, it was demonstrated that the development of arthritis correlated with high levels of agalactosyl IgG and that "protected" animals had significantly depressed levels.

Tuberculosis is characterised by fever, weight loss and necrosis in both lesions and tuberculin skin test sites (Koch phenomenon), although the antigens of Mycobacterium tuberculosis are not directly toxic to the tissues. The tissue damage appears to be due to several interacting factors. First, M. tuberculosis induces an immunoregulatory disorder of which a raised percentage of agalactosyl IgG is a marker. This is seen also in rheumatoid arthritis and Crohn's disease and is associated with tissue-damaging inflammation. Subsequently, several properties of M. tuberculosis exacerbate this disorder by triggering cytokine release, and rendering tissues sensitive to the toxicity of tumor necrosis factor (TNF). Moreover, M. tuberculosis, but not bacillus Calmette-Guérin or several Mycobacterium avium strains, produces a factor which increases the toxicity of TNF for individual cells. Thus, M. tuberculosis may distort the normal protective role of TNF so that this cytokine becomes toxic to the host. The immunoregulatory disorder associated with agalactosyl IgG appears to be susceptible to immunotherapy, so novel types of treatment for the immunopathological component of tuberculosis are being explored.

Both leprosy and tuberculosis present continuing problems in their control, especially in the developing world, despite the availability of drugs effective in producing a bacteriological cure. Improved immunoprophylaxis, and an effective immunotherapy to be used with chemotherapy are urgently required. Intradermal injection of a suspension of killed Mycobacterium vaccae promotes cell-mediated responses to antigens common to all mycobacteria, and switches off the tissue-necrotizing aspects of the Koch phenomenon. These properties led to the use of the suspensions as an improved vaccine, either alone or in combination with BCG. The same properties led to the employment of the suspension in immunotherapy as an adjunct to chemotherapy in the treatment of both leprosy and tuberculosis. The evidence leading to these conclusions is reviewed and discussed.

The development of arthritis induced in mice by intraperitoneal injection of the non-antigenic mineral oil, 2,6,10,14-tetramethylpentadecane (pristane), was shown to depend on an intact immune response possibly to a heat-shock protein (hsp) in the synovium. Initial experiments suggested that some crucial event in the development of arthritis takes place early after pristane injection. First, irradiated pristane-treated mice failed to develop arthritis unless they were reconstituted with spleen cells from normal donors within 25 days of irradiation. Second, mice irradiated up to 50 days after pristane injection, but not later, did not develop arthritis. Evidence for the involvement of an immune response to heat-shock protein (hsp) comes from the finding that mice injected with mycobacterial 65-kDa hsp prior to pristane challenge had a reduced incidence of arthritis in contrast to animals pre-immunized with the E. coli hsp equivalent GroEL or with bovine serum albumin. Other experiments revealed that T cells from mice with gross morphologically defined arthritis proliferated strongly to hsp65 and to normal joint antigens, whereas T cells from animals treated with pristane which did not develop arthritis gave much smaller responses. Mice which developed arthritis also had elevated levels of anti-hsp65 IgG in comparison with non-arthritic animals. These findings strongly suggest that autoimmune reactions to an antigen which cross-reacts with hsp65 are generated in pristane-induced arthritis. It is considered that the autoimmune response is directed to a synovial antigen and that pre-immunization with hsp65 protects the animals from the development of pristane-induced arthritis by altering the specificity or quality of the immune response to this antigen.

A simple rapid assay based on the ability of lectins to bind carbohydrates has been developed to analyse changes in the oligosaccharide chains of IgG. Bandeiraea simplicifolia lectin and Ricinus communis agglutinin have been used to detect terminal N-acetylglucosamine and galactose moieties respectively in IgG using immunodot-blotting. IgG samples (approximately 1 micrograms) were dot-blotted onto nitrocellulose followed by boiling of the blots to expose the carbohydrate moieties. The blots were then treated with biotinylated lectins followed by either streptavidin-biotin-hydrogen peroxidase conjugate or 125I-labelled streptavidin. The colour was developed using chloronaphthol and the blots read on a densitometer. The labelled blots were cut and read on a gamma counter. The use of a monoclonal antibody to N-acetylglucosamine is also discussed. The results obtained using this method are comparable to those obtained by structural analysis.

Heat shock proteins (hsp) or "stress proteins" are a group of highly conserved proteins which are important in the day to day function of all cells. Early studies by others have indicated that immunity to the 65 kDa hsp of mycobacteria is important in the development of arthritis in the adjuvant arthritis model in rats. In this paper, we review the evidence suggesting that, as for the rat model, immune reactivity to hsp is of importance in the human disease. Elevated levels of IgG antibodies to the 65 kDa hsp of mycobacteria are characteristic of rheumatoid arthritis (RA) patients. Much of this antibody cross-reacts with human 65 kDa hsp and is therefore autoreactive. The 65 kDa hsp is found in synovial fluid and is therefore a potential target for antibody. Antibodies to the 70 kDa hsp (both of mycobacterial and human origin) are elevated, but not specifically, in RA. Increased T cell responses to the 65 kDa hsp are also found in synovial fluid of RA patients. Although gamma delta T cells are present in the synovial joint of RA patients, they do not appear to be particularly increased in frequency although the subset distribution of these cells is clearly different from that seen in the circulation. In fact, the synovium looks like the "gut" with regard to these subsets!

Sixty-five patients, many of them immigrant to Kuwait, with bacteriologically proven, adult type, pulmonary tuberculosis were studied by many parameters over the 4 months following diagnosis. Twelve were infected with tubercle bacilli resistant to at least one anti-tuberculosis drug. Preliminary evidence suggested that this was frequently primary resistance in patients infected in their countries of origin. The Kuwaiti environment results in very high skin test and lymphocyte proliferative responses (LTT) to shared and species specific antigens of mycobacteria in healthy persons. In comparison, patients with tuberculosis lacked cellular responses to group i and group ii antigens, but had increased IgG and IgA binding to mycobacterial antigens in general. LTT responses to added interleukin 2, and production of alpha interferon, were normal in our patients, but induction of gamma interferon in response to phytohaemagglutinin was reduced initially, rising towards normal during treatment. Biochemical and haematological abnormalities present at the time of diagnosis rapidly corrected. The disease differed from that reported in most previous studies in that fever was uncommon, the disease was never fatal, and most tuberculin tests were not necrotising. This implied that a detrimental immunopathological component is less pronounced in those exposed to the Kuwaiti environment, and a hypothesis is put forward to explain this.

The activated macrophages present in the T cell-dependent granulomata of sarcoidosis and tuberculosis are primed for enhanced release of cytokines including tumour necrosis factor (TNF or cachectin). Release of this cytokine can induce an acute-phase response, fever, and necrosis in suitably prepared sites of inflammation; if chronic, its presence may contribute to weight loss. These clinical features are characteristic of tuberculosis, but not of sarcoidosis, though alveolar macrophages from both diseases release large quantities of TNF in vitro. We therefore postulated the presence in sarcoidosis patients of an inhibitor of TNF. We have studied levels of TNF inhibitory activity by determining the quantity of TNF required to give 50% kill of L929 cells in the presence of 20% heat-inactivated serum derived from various disease states (37 sarcoidosis, 13 tuberculosis, 13 Crohn's disease, 17 healthy donors). Normal sera used in this way do not inhibit significantly, but inhibition of TNF toxicity is caused by most sera from both sarcoidosis and tuberculosis. Used at 20%, five out of 37 sarcoidosis sera and one out of 13 tuberculosis sera caused complete inhibition of TNF, even when the latter was added at 100 times the concentration required to give 50% kill in control wells. This inhibitor may have an important physiological role.

47 patients with adult-type pulmonary tuberculosis attending the Chest Diseases Hospital in Kuwait were given a single injection of 10(9) irradiation-killed M. vaccae after 1 month of a 9-month course of chemotherapy. The patients were followed-up for 3 more months in double blind comparison with 65 patients given an injection of saline (placebo). The immunotherapeutic injection produced a small local lesion in 44/47 patients, 18 of which ulcerated and produced small scars. Immunotherapy made no measurable difference to the bacteriological, biochemical, haematological, or radiological parameters measured. However it was associated with significantly improved weight gain, reduced size of skin test response to Tuberculin, increased lymphocyte proliferation to common mycobacterial antigens, and increased antibody levels to mycobacterial antigens. The changes in skin test and LTT responses were related and occurred in 29% of patients whose recognition of common mycobacterial antigens returned to normal. The remaining patients did not differ in these respects from those receiving placebo. The proportion of patients whose responses were improved was very similar to that achieved using the same immunotherapeutic agent in a group of treated multibacillary leprosy patients.

The proportion of oligosaccharide chains on the Fc fragment of IgG which terminate with N-acetylglucosamine (GlcNAc) rather than galactose is increased in rheumatoid arthritis and tuberculosis, and in sera from patients with Crohn's disease, probably because of decreased activity of a galactosyltransferase in B lymphocytes. We have assayed the prevalence of agalactosyl oligosaccharides on IgG in sera from 67 patients with inflammatory bowel disease (32 ulcerative colitis and 35 Crohn's disease). The prevalence of agalactosyl IgG significantly increases in the majority of Crohn's patients (19/35 patients), and correlates with the level of C-reactive protein (r = 0.79), and inversely with the concentration of serum albumin. Sera from ulcerative colitis patients show less frequent (nine of 32) and less marked rises in agalactosyl IgG, and sera with high C-reactive protein values can contain normal levels. Thus in ulcerative colitis no correlation was seen between the two assays. The diseases in which the percentage of agalactosyl IgG is raised (rheumatoid arthritis, tuberculosis, Crohn's disease and some ulcerative colitis) are characterised by simultaneous T cell mediated granulomatous tissue damage, and acute phase responses. Levels are normal in less tissue damaging granulomatous conditions, including sarcoidosis, and leprosy (except during episodes of erythema nodosum leprosum). We suggest therefore that a raised percentage of agalactosyl IgG is a correlate of a particular type of T cell mediated pathology which may be relevant to the pathogenesis of inflammatory bowel disease.

We still do not understand the mechanism of immunity ty mycobacteria in man, and convincing reproducible kill of M. tuberculosis by human macrophages has not been achieved. The pathways so far elucidated, involving gamma interferon, 1,25(OH)2 vitamin D3, and TNF release seem more likely to lead to immunopathology than to protection. Meanwhile the major problem for the clinician is the existence of "persister" bacteria, which are not eliminated by the immune response, even when therapy has greatly reduced the bacterial load. It seems unlikely that it will be possible to design antibiotics which will rapidly kill dormant persister bacilli, so new strategies for therapy may need to concentrate on modulation of the host response. The objectives of such therapies would be: 1) "Reawakening" of dormant persisters. 2) Rapid immune recognition of persisters. 3) Suppression of the tissue-damaging pathway. 4) Enhancement of the optimally protective mechanism, but this has not yet been defined.

Heat-stable antigens of rodent malarial parasites induce the release of tumour necrosis factor (TNF) from mouse macrophages, in vitro and in vivo. We report here that analogous antigens of Plasmodium falciparum trigger the release of TNF from human monocytes in vitro, in conditions that exclude the effects of any contaminating endotoxin. These antigens also induced TNF release from a murine monocytic cell line and from the peritoneal macrophages of several strains of mice, including the LPS-hyporesponsive C3H/HeJ mice. Similarly, boiled soluble antigens from the rodent parasites P. yoelii and P. berghei also stimulated human monocytes. Antisera made by immunizing mice with boiled antigens of P. falciparum or P. yoelii inhibited the stimulation of TNF secretion by P. falciparum antigens. They did not block the induction of TNF by bacterial lipopolysaccharide. Thus mouse macrophages provide a convenient system for investigating the nature, cross-reactions and antigenic variation of human malarial soluble antigens. Since these are known to occur in the circulation of patients with malaria, they may be responsible for excess production of TNF, a mediator that is thought to play a central role in the pathogenesis of the disease.

Fibronectin (FN)-binding antigens are prominent components of short-term culture supernatants of Mycobacterium tuberculosis. In 3-day-old supernatants, a 30-kilodalton (kDa) protein was identified as the major FN-binding molecule. In 21-day-old supernatants, FN bound to a double protein band of 30 and 31 kDa, as well as to a group of antigens of larger molecular mass (57 to 60 kDa). FN-binding molecules in this size range, but not of 30 to 31 kDa, were also found in sonicates. We showed that the 31- and 30-kDa FN-binding bands correspond to components A and B of the BCG85 complex, previously shown to be abundant in culture supernatants of Mycobacterium bovis BCG. Thus, a polyclonal antibody to the BCG85 complex bound to the 30- and 31-kDa antigens and inhibited binding of FN to them on immunoblots of the culture filtrates. Similarly, FN bound to the purified components of the BCG85 complex, and this binding was blocked by the antibody. A monoclonal antibody, HYT27, also bound both to the BCG85 components A and B and to the 30- and 31-kDa FN-binding molecules of M. tuberculosis, but it did not block the binding of FN. Related molecules appear to be present on the surface of BCG and to mediate the binding of BCG to FN-coated plastic surfaces, since this binding could also be blocked by the polyclonal anti-BCG85 antibody and by the purified components of BCG85, particularly component A, but not by monoclonal antibody HYT27. The binding of these mycobacterial antigens to FN appears to be of very high affinity, and we suggest that this property of major secreted antigens of M. tuberculosis indicates an important role in mycobacterial disease and in the binding of BCG to tumor cells during immunotherapy of bladder cancer.

Endotoxin (LPS)-triggered release of tumour necrosis factor (TNF) from human monocytes is high for the first 3 days in culture, and then falls to a trough between Days 4 and 6. This trough is less deep if the cells are cultured in the presence of indomethacin. If monocytes are cultured in the presence of either recombinant gamma-interferon or 1,25-(OH)2 vitamin D3, their capacity for LPS-triggered TNF release is increased. Live virulent Mycobacterium tuberculosis can substitute for the LPS, and is markedly more effective as a trigger than several strains of BCG prepared in an identical manner. Since secretion of IFN-gamma and conversion of circulating 25-(OH) vitamin D3 to the active dihydroxy metabolite by interferon-activated macrophages probably occur in tuberculosis, we suggest that triggering of TNF release from the resulting activated macrophage populations might explain much of the weight loss and tissue damage that characterize this disease. IFN-gamma does not activate effective anti-mycobacterial mechanisms in human monocytes, so its role in tuberculosis may be immunopathological rather than protective.

The enzyme-linked immunosorbent assay (ELisa) was used to detect the presence of antibodies to muramyl dipeptide (MDP) in serum of patients with leprosy or tuberculosis. Using a conjugate of MDP-lysine to horse radish peroxidase, no such antibodies could be detected in sera of either patients or controls. Antibodies to a sonicate antigen of Mycobacterium tuberculosis were found in sera of all individuals tested and the binding of these antibodies to the M. tuberculosis antigen could not be inhibited by MDP. On the other hand, binding of MDP to anti-MDP antibodies, raised in rabbits, was largely inhibited by free MDP, slightly inhibited by M. tuberculosis antigen and was not inhibited by the patients' sera.

Antigens extracted from leprosy bacilli obtained from infected human and armadillo tissues have been examined by immunodiffusion analysis with serum samples from lepromatous patients and with immune sera raised in rabbits. Using the best combinations of serum and antigen extracts, 12 antigenic constituents were found in the leprosy bacilli. Six of these were antigens common to all mycobacteria and nocardiae, 4 were specific to the leprosy bacillus and the position of 2 could not be determined. Groups ii and iii antigens (i.e. those associated with the slow growing and fast growing subgenera of mycobacteria) were not found in theleprosy bacillus, suggesting some relationship with M. vaccae and similar strains, in which these antigens are also missing. Lymphocyte transformation tests performed on lymph node cells of mice infected or immunized with leprosy bacilli also showed the leprosy bacillus to have a closer relationship with M. vaccae than with other mycobacteria.

The response to soluble mycobacterial antigens of peripheral blood mononuclear cells, from most normal donors, tuberculosis patients or cases of tuberculoid leprosy (TT/BT) was enhanced by the addition of indomethacin. In contrast, indomethacin caused no enhancement of the response of cells from lepromatous leprosy (BL/LL) cases. Moreover the addition of 10-5 M prostaglandin E2 (PGE2) failed to inhibit the proliferative responses of cells from the BL/LL patients, although it markedly inhibited the responses of peripheral blood mononuclear cells from the other groups. The addition of PGE2 or indomethacin to cells which had been precultured for 48 hr had no significant effect on the proliferative responses of cells from any of the groups of donors. These results suggest that a normal, prostaglandin-dependent, indomethacin-sensitive regulatory mechanism is absent from the peripheral blood mononuclear cells of BL/LL patients.

Mice infected in the tail with M. ulcerans developed transient cell-mediated immunity which disappeared as the bacterial load increased. Lymph node cells from animals in this late phase of the disease transformed spontaneously in vitro. This transformation was inhibited by mycobacterial antigen. Lymph node cells from mice injected intravenously with 10(8) or 10(9) BCG also transformed spontaneously in vitro. Such animals did not become foot pad test-positive. Evidence is presented that the spontaneous transformation may represent an accumulation of specifically sensitized cells due to trapping in nodes overloaded with persistent antigen. The relevance of such a phenomenon to 'desensitization' in human and animal disease is discussed.

Soluble extracts of ultrasonically disrupted mycobacteria (including Mycobacterium leprae) cause strong T-cell-dependent, Cyclosporin A-sensitive, polyclonal B-cell activation, in human peripheral blood mononuclear cell preparations. It is not enhanced by cortisol. Cells from cases of lepromatous leprosy show no polyclonal B-cell activation in the presence of extracts of M. leprae, although they respond normally to other mycobacterial species. The polyclonal B-cell activation response to the M. leprae extract, of a small group of normal leprosy contacts, was reduced relative to that of non-contacts.

Tuberculosis is characterised by necrosis in the lesions and in skin-test sites, and by fever and weight loss. In contrast, other diseases with chronic T cell mediated responses, such as uncomplicated leprosy and sarcoidosis, have non-necrotising lesions with little systemic upset. Crude sonicates of M. tuberculosis and M. leprae prepare skin sites for TNF-mediated damage via a pathway which unexpectedly appears to involve CD8+ T cells, and both mycobacteria contain potent triggers of TNF release (lipoarabinomannan and peptidoglycan derivatives). These observations can partially explain the pathology of tuberculosis, but fail to explain why similar events do not normally occur in leprosy. It now seems likely that the answer lies in the existence of novel regulatory pathways. A recently recognised correlate (or consequence) of diseases characterised by T cell-dependent tissue-damaging pathology and cytokine release, is an increase in the level of agalactosyl IgG. This behaves like a T cell-dependent acute phase reactant, and is raised in tuberculosis, rheumatoid arthritis, and Crohn's disease, but not in sarcoidosis or uncomplicated leprosy. Thus it may act as a marker for a type of pathology of very broad significance, though its functional role remains obscure.