INFECrION AND IMMUNITY, Apr. 1992, p. 1441-1446 0019-9567/92/041441-06$02.00/0 Copyright © 1992, American Society for Microbiology

Vol. 60, No. 4

Tumor Necrosis Factor Production in Patients with Leprosy PETER F. BARNES,1 DELPHI

CHAT1IERJEE,2 PATRICK J. BRENNAN,2 THOMAS H. REA,3

AND ROBERT L. MODLIN4,5* Department of Medicine1 and Section of Dermatology, 3 University of Southem California School of Medicine, Los Angeles, Califomia 90033; Department of Microbiology, Colorado State University, Fort Collins, Colorado 805232; and Division of Dennatology4* and Department of Microbiology and Immunology,5 University of California at Los Angeles School of Medicine, Los Angeles, California 90024

Received 26 November 1991/Accepted 12 January 1992

The spectrum of host responses to Mycobacterium kprae provides a model for investigating the role of cytokines in the pathogenesis of mycobacterial disease. Of particular interest is tumor necrosis factor (TNF), a cytokine which may have both antimycobacterial and immunopathologic effects. To evaluate the potential role of TNF in leprosy, we measured TNF production in response to M. leprae and its defined constituents by peripheral blood mononuclear cells from patients across the spectrum of disease. The levels of TNF induced through the stimulation of cells with M. leprae or its dominant "lipopolysaccharide," lipoarabinomannan, were higher in patients with the tuberculoid form of the disease than in those with the lepromatous form. In patients with erythema nodosum leprosum (ENL), a reactional state of lepromatous leprosy, the levels of TNF release by peripheral blood mononuclear cells were higher than in any other form of the disease. Treatment of ENL patients with thalidomide reduced TNF secretion by more than 90%o. The mycolylarabinogalactanpeptidoglycan complex of Mycobacterium species, the protein-peptidoglycan complex, and muramyl dipeptide all elicited significant TNF release. Therefore, TNF release appears to be triggered by at least two major cell wall structural constituents of M. leprae, lipoarabinomannan and segments of the cell wall skeleton. The prominent TNF release in patients with the paucibacillary tuberculoid form of the disease compared with that in patients with the multibacillary lepromatous form suggests that this cytokine contributes to a resistant immune response to mycobacterial infection. However, the marked TNF release in patients with ENL indicates that TNF may also mediate immunopathologic effects, such as fever and tissue damage. The spectrum of host responses to Mycobacterium leprae provides an opportunity to evaluate and correlate immune responses to a foreign pathogen with clinical manifestations of disease. At one pole of the spectrum, patients with tuberculoid leprosy have few skin lesions, in which bacilli can rarely be identified, whereas at the opposite pole, patients with lepromatous leprosy have diffuse infiltration of skin and nerves with bacillus-laden macrophages (29, 30). Patients with tuberculoid leprosy have a strong cell-mediated immune response to M. leprae, manifested by positive skin tests (Mitsuda reaction) and marked lymphocyte proliferative responses to M. leprae in vitro. In contrast, patients with lepromatous leprosy have a depressed cell-mediated immune response to M. leprae, characterized by negative skin tests and diminished or absent lymphocyte proliferation. Imposed upon this spectrum are the so-called "reactional states": reversal reaction and erythema nodosum leprosum (ENL). The reversal reaction appears to be a naturally occurring delayed-type hypersensitivity response to M. leprae. Clinically, it is characterized by "upgrading" of the clinical picture towards the tuberculoid pole, including a reduction in the bacillary load. Immunologically, it is characterized by the development of strong skin test reactivity as well as lymphocyte responsiveness (5, 9, 17, 27, 42). The pathogenesis of ENL is thought to be related to the deposition of immune complexes in the lesions (8, 14, 43). Although there is some evidence indicating cell-mediated immunity (26, 37), significant clearance of bacilli from lesions does not

Of the cytokines which regulate the immune response to infection, tumor necrosis factor (TNF) is of particular interest in the pathogenesis of mycobacterial disease. On the one hand, TNF may participate in protective immunity by mediating granuloma formation (20) and inhibiting mycobacterial growth in vitro (6). On the other hand, the effects of TNF may result in immunopathology, such as nerve damage (34) and tissue necrosis (31). In the present study, we evaluated the potential role of TNF in leprosy by measuring its production in response to M. leprae by peripheral blood mononuclear cells (PBMC) from patients across the spectrum of disease.

MATERIALS AND METHODS Patients. Patients with leprosy were classified on the basis of the clinicopathologic criteria of Ridley and Jopling (2830). Blood was obtained from the following patient groups: (i) tuberculoid leprosy; (ii) lepromatous leprosy without a reaction; (iii) lepromatous leprosy with active ENL; (iv) lepromatous leprosy with inactive ENL (after treatment with thalidomide); and (v) reversal reaction. All patients had received less than 5 years of chemotherapy. Mycobacterial preparations. Lipoarabinomannan (LAM) was isolated and purified from M. leprae as described for the M. tuberculosis product (11). The purified LAM was redissolved in sterile pyrogen-free water, passed through a 2-ml DetoxiGel column (Pierce Chemical, Rockford, Ill.), and filtered through a 0.20-,um-pore-size sterile filter, and the filtrate was collected in a sterile pyrogen-free vial and dried. This procedure was necessary to remove any residual contaminating lipopolysaccharide (LPS). The cell wall components, protein-peptidoglycan complex

occur.

*

Corresponding author. 1441

1442

INFECT. IMMUN.

BARNES ET AL.

(PPC) (19) and mycolylarabinogalactan-peptidoglycan (mAGP) complex (12, 21), were prepared as described earlier. To remove any residual proteins from the mAGP complex (12, 21), we further treated it with proteases as follows. mAGP isolated from M. leprae (10 mg) was suspended in 1.5 ml of pronase solution, and the suspension was incubated at 37°C overnight (pronase E; type XIV; Sigma, St. Louis, Mo.; 2 mg/ml in 2% sodium dodecyl sulfate [SDS] in phosphate-buffered saline [PBS]; pH 7.8). The cell wall suspension was centrifuged for 20 min and washed extensively with 2% SDS buffer. The pronase-treated cell wall was treated further with proteinase K solution (type XXVIII; Sigma) as described above. The protein-free cell wall was then boiled in SDS-PBS buffer for 10 min. The SDS was removed by further washing with PBS, distilled water, 80% acetone, and sterile pyrogen-free water. Sonicated M. leprae, unfractionated, was prepared as previously described (19). Every effort was made to exclude endotoxin from all preparations, including the use of sterile techniques, pyrogen-free reagents, and DetoxiGel. Previous experiments with the Limulus amebocyte assay had shown that a parallel preparation of LAM from M. tuberculosis contained 0.0016 to 0.0034% endotoxin (35). Since the mAGP complex and the PPC are both insoluble, they could not be passed through the DetoxiGel to remove any possible residual endotoxin. Muramyl dipeptide (N-acetylmuramyl-L-alanyl-D-isoglutamine) was purchased from Sigma. Stimulation of cells with mycobacterial preparations. PBMC were isolated by centrifugation over Ficoll-Paque (Pharmacia Fine Chemicals, Piscataway, N.J.). Mononuclear cells (105) were suspended in 200-,ul flat-bottomed wells (Falcon, Oxnard, Calif.) containing RPMI 1640 with gentamicin (Sigma), 10% heat-inactivated sterile filtered human AB serum, and various concentrations of mycobacterial preparations. Cell culture supernatants were harvested after 24 h for the measurement of TNF concentrations. The TNF concentration induced by a given mycobacterial preparation was calculated as that measured in the appropriate supernatant minus the cytokine concentration in control wells lacking the mycobacterial stimulus. All supernatants were stored at -70°C prior to the determination of TNF levels. Measurement of cytokine concentrations. Concentrations of TNF in cell supernatants were measured by a radioimmunoassay (Centocor, Malvern, Pa.), an enzyme-linked immunosorbent assay (ELISA) (Genzyme, Boston, Mass.), or a bioassay with WEHI 164 clone 13 cells (15). We previously reported that there is excellent agreement among results obtained by these three assays (3). In the present study, the mean TNF concentrations measured in each patient group by the bioassay versus the immunoassays were not significantly different. The trends were consistent across different patient groups for each assay. Therefore, results are reported as the mean values for all patients studied. The Centocor radioimmunoassay for TNF is a solid-phase radioimmunoassay which detects as little as 10 pg of TNF per ml. Samples were added to polystyrene beads coated with murine monoclonal antibodies specific for human TNF. The amount of TNF adherent to the beads was determined by measurement of the degree of radioactivity incorporated upon the addition of 125I-labeled anti-TNF monoclonal antibodies. The cytokine concentrations in the samples were determined by comparison with radioactivity incorporated by specimens containing known TNF concentrations. The Genzyme TNF ELISA is based on the multiple-

5000

I

4000

-

3000

Cf

2000

I

1000 oF o

I

LL

RR

ENL(in) ENL(ac)

BT

FIG. 1. TNF release induced by sonicated M. leprae. PBMC were obtained from patients with lepromatous leprosy (LL: n = 11), reversal reactions (RR; n = 6), inactive ENL [ENL(in); n = 14], active ENL [ENL(ac); n = 12], and borderline tuberculoid leprosy (BT; n = 20). The mean TNF level in the presence of medium alone was 105 ± 31 pg/ml and did not differ among the patient groups.

antibody sandwich principle and can detect 12 pg of TNF per ml. Samples were incubated in microtiter plate wells coated with a mouse monoclonal antibody specific for human TNF. A rabbit anti-human TNF polyclonal antibody was added, and then biotin-conjugated goat anti-rabbit immunoglobulin G was added. Streptavidin-conjugated peroxidase, which binds to biotin, was added. Peroxide substrate and OPD chromogen were added to yield increased A492 values, indicating the presence of TNF. The bioassay for detecting TNF was based on target cell line WEHI 164 clone 13, which is extremely sensitive to the cytotoxic effects of TNF. WEHI 164 clone 13 cells were a kind gift of Genentech, Inc., South San Francisco, Calif. The assay was performed as previously described (15). In brief, samples were incubated with the WEHI cells for 22 h. The percentage of dead cells, which provides a measure of the concentration of TNF in the sample, was calculated by measuring the optical density of the culture after the addition of MTT tetrazolium as described by Mosmann (24). The concentrations of TNF were determined by comparison with a standard curve generated with specimens containing known TNF concentrations. Statistical analysis. For data that were normally distributed, the t test was used to test for statistical significance. For data that were not normally distributed, the Wilcoxon rank sum test was used. RESULTS

Production of TNF across the spectrum of leprosy. To investigate the potential role of TNF in mediating the clinical manifestations of leprosy, we measured the capacity of PBMC from patients across the spectrum of disease to produce TNF in response to M. leprae. TNF release was highest in patients with ENL, intermediate in those with tuberculoid leprosy, and lowest in patients with lepromatous leprosy, inactive ENL, and reversal reactions (Fig. 1). TNF release was significantly higher in patients with tuberculoid leprosy than in those with lepromatous leprosy (P = 0.01) and in those with active ENL than in those with inactive ENL (P = 0.01). The difference between TNF release in patients with ENL and those with tuberculoid leprosy was not statistically significant.

VOL. 60, 1992

TNF IN LEPROSY

15

5000 4000 I

010 -c

E

3000 C0.0

F

I T

E. 2000

1

2

3

4 5 6 Patient FIG. 2. TNF release induced by sonicated M. leprae in patients with ENL. PBMC were obtained from six patients with active ENL (M) and from the same patients 5 to 46 days later, when the episode of ENL had subsided (1).

Because TNF release was markedly elevated in patients with active ENL, we evaluated the dynamics of TNF release by PBMC from these patients. TNF release fell by more than 90% (from 4,756 + 1,853 pg/ml to 456 85 pg/ml; P < 0.01) as the episode of ENL subsided during treatment with thalidomide at 100 to 200 mg/day for 5 to 46 days (Fig. 2). TNF release by M. Ieprae constituents. The availability of several constituents of M. leprae in highly purified forms has enabled us to identify which components are potent inducers of TNF. We therefore evaluated the capacity of purified LAM derived from M. leprae to elicit TNF production across the spectrum of leprosy. TNF release paralleled the response to sonicated M. leprae, being highest in patients with active ENL and next highest in individuals with tuberculoid leprosy (Fig. 3). The levels of TNF release induced by LAM were similar to those induced by sonicated M. leprae but did not achieve statistical significance. Recent research on the elucidation of the fundamental structure of the cell wall core or skeleton of Mycobacterium species has resulted in the availability of this core, or parts of it, in quantities such that their biological properties can be examined. The cell wall core or skeleton may be given the chemical name mAGP, since it consists of a covalently linked complex of mycolic acids, arabinogalactan, and peptidoglycan (12, 21). In addition, several subfractions of the r

4000 annn 4uuu

Cf

I

Lr

2000 1

I

1000

0

5000

1443

000

0

LL

RR

ENL(In) ENL(ac)

FIG. 3. TNF release induced by M. leprae LAM. PBMC were obtained from patients with lepromatous leprosy (LL; n = 11), reversal reactions (RR; n = 6), inactive ENL [ENL(in); n = 14], active ENL [ENL(ac); n = 12], and borderline tuberculoid leprosy (BT; n = 20).

O

_

MLS mAGP PPC MDP FIG. 4. TNF release induced by M. leprae cell wall components. PBMC were obtained from five patients in whom the TNF level in response to sonicated M. Ieprae was higher than 400 pg/ml. Sonicated M. Ieprae (MLS), the mAGP complex, the PPC, and muramyl dipeptide (MDP) were studied.

cell wall core, notably the PPC, have been generated. Because some of these fractions are potent stimuli for T lymphocyte proliferation (4, 22) and because the arabinan components of both mAGP and LAM are structurally very similar (11), we wished to evaluate their capacity to induce TNF release. The mAGP complex and the PPC were as potent as sonicated M. leprae in eliciting TNF production (Fig. 4). In addition, the muramyl dipeptide of mycobacterial cell walls induced TNF secretion, although to a lesser extent than the other constituents of M. leprae. These results indicate that components of the core of mycobacterial cell walls are also capable of inducing TNF release.

DISCUSSION The immunologic role of TNF in leprosy is not clearly defined, although TNF is thought to be a major factor in the pathogenesis of mycobacterial infection. TNF may augment the immune response towards the elimination of the pathogen and/or mediate the pathologic manifestations of the disease. In the present study, we correlated TNF release by M. leprae with the clinical manifestations of leprosy. We found evidence supporting a role for TNF in the resistant immune response to infection, since higher TNF release was evident in patients resistant to M. leprae (patients with the paucibacillary tuberculoid form of the disease) than in those susceptible to M. leprae (patients with the multibacillary lepromatous form). Several previous studies evaluated TNF production in patients with leprosy, specifically the tuberculoid and lepromatous forms, to gain insight into the role of this cytokine in the pathogenesis of disease. Serum TNF levels, as measured by a bioassay, were reported to be higher in patients with the tuberculoid form than in those with the lepromatous form (36). However, serum TNF concentrations, as measured by an ELISA, were higher in individuals with the lepromatous form (25). This discrepancy has led to speculation that TNF inhibitors in the serum of patients with the lepromatous form interfere with the bioassay for TNF, decreasing the measured amount and perhaps the biologic effect in these individuals (33). Inhibitors of TNF bioactivity have been described in the serum of patients with tuberculosis and sarcoidosis (16). Investigations of TNF release by PBMC and in the lesions of patients have clearly demonstrated higher cytokine pro-

1444

BARNES ET AL.

duction in patients with the tuberculoid form than in those with the lepromatous form. Stimulation of PBMC with LPS (36) and with M. leprae (present study) induced significantly higher release of TNF in the former patients than in the latter. Immunohistochemical studies of skin biopsies with anti-TNF monoclonal antibodies revealed higher numbers of TNF-containing cells in tuberculoid than in lepromatous leprosy tissue lesions (2). Similarly, higher concentrations of TNF mRNA in tissue were found in tuberculoid leprosy lesions, as detected by in situ hybridization (38) and by polymerase chain reaction amplification (44). These data suggest that the more marked M. leprae-induced production of TNF by mononuclear cells in patients with tuberculoid leprosy results in elevated levels of TNF in tissue lesions. However, because the total number of skin lesions and the total number of mononuclear cells exposed to M. leprae are far higher in lepromatous leprosy patients, the total TNF secretory capacity of lepromatous leprosy patients may result in higher serum TNF levels, although the biologic effect may be diminished by circulating inhibitors. The sum of these findings could be interpreted to indicate that the net biologic effect of TNF correlates with resistance to mycobacterial disease. A role for TNF in protective immunity is further suggested by its capacity to inhibit mycobacterial growth in murine and human macrophages (6). TNF is necessary for granuloma formation and mycobacterial elimination in mice infected with M. bovis BCG (20). Furthermore, TNF mRNA is abundant in tissues of patients with resistant immune responses to other mycobacterial infections, such as tuberculous pleuritis (3). Monocytes from patients with chronic refractory tuberculosis released significantly lower amounts of TNF than did those from patients with newly diagnosed tuberculosis (39). TNF may contribute to protective immunity against mycobacteria and other intracellular organisms through the production of reactive nitrogen intermediates. TNF enhances the production of nitric oxide by murine macrophages (13). Nitric oxide has been shown to inhibit the growth of and/or to kill leishmaniae (18), M. leprae (1), and M. tuberculosis (10). In addition to its potential contribution to antimycobacterial defense mechanisms, TNF may mediate the pathologic manifestations of mycobacterial disease. TNF administered to experimental animals results in fever, weight loss, and cachexia (7) and may cause these characteristic clinical features of mycobacterial disease. The administration of TNF to animals primed with M. tuberculosis results in necrotic reactions (31), raising the possibility that this cytokine contributes to the caseation characteristic of tuberculosis. TNF production within leprosy lesions may result in nerve damage and deformities (34, 40). A surprising and striking finding of the present study was the extremely high levels of TNF induced by sonicated M. leprae in PBMC from patients with active ENL. This TNF release decreased dramatically with the concomitant clinical response to thalidomide. These findings extend the recent observation that the addition of thalidomide to mononuclear cells in vitro suppresses TNF production induced by LPS and mycobacterial antigens (32). Serum TNF levels are also markedly higher in patients with ENL than in those with tuberculoid or lepromatous leprosy (33). Although relatively low levels of TNF mRNA are present in the skin lesions of these patients (38), careful longitudinal studies of patients may show augmentation of local TNF production. Because ENL reactions are characterized by fever, weight loss, and tissue injury but not the elimination of bacilli, elevated TNF levels in these

INFECT. IMMUN.

A. Arabinans I

*. e

0 *

FA FA

-* Io

At o * Inositol . Glcrol * . 2,6 - Manp * - 6 - Manp IA TT-Manp

I

B. Mycolyl - (Ara)n - (Gal)n - Rha - GIcNAc Mur I Gly aGlu

Ia

Dap - R - D Ala -R' FIG. 5. Structural models of mycobacterial constituents. (A) Structure of LAM from M. leprae. (B) Structure of mAGP from M. leprae. Only the muramyl (Mur) component of peptidoglycan is Dap.:-

shown. n, undetermined numbers; R and R', repeating units of peptidoglycan demonstrating the two types of cross-linking; FA, fatty acyl function.

patients support the hypothesis that TNF production may mediate the immunopathologic manifestations of the reactions. To define the constituents of M. leprae which elicit TNF secretion, we examined two major entities of the cell wall of M. leprae. We used LAM, which is now recognized as the major "LPS" of mycobacteria. LAM from M. tuberculosis has been shown to induce TNF release by human and murine macrophages (23). In addition, we used the mnAGP complex (mAGP is the chemical name for the entire insoluble substratum of cell walls) and the PPC (a subsegment of mAGP with some associated proteins). The combined results indicate that several entities within the cell walls of mycobacteria, including LAM, the mAGP complex, the PPC, and muramyl dipeptide, are capable of eliciting TNF. LAM is a well-defined molecule, marked by a phosphotidylinositol at one end (the reducing end) and linear and branched arabinans at the opposite (nonreducing) end (Fig. 5A). Recent studies have clarified the structure of the mAGP complex (Fig. 5B) (12, 21). The fact that both share similar arabinan motifs would seem to implicate arabinan in TNF evocation. However, we have shown (2a) that an intact phosphatidyl unit in LAM is required for TNF release. Likewise, the present results indicate that much of the potency of the mAGP complex lies in the peptidoglycan unit. Alternatively, since mycobacterial proteins are inducers of TNF (41), the potency of the PPC may be related to the associated proteins. Thus, it appears that the ability of mycobacterial cell walls to evoke TNF is multifaceted. In defining the role of TNF in the immunopathogenesis of leprosy, it may be overly simplistic to envision TNF as having exclusively protective or pathologic effects. Instead,

VOL. 60, 1992

TNF IN LEPROSY

the pathologic effects of TNF may be the consequence of overexuberant immunoprotective mechanisms. Identification of the constituent molecules of mycobacteria which stimulate TNF release, as well as identification and characterization of TNF inhibitors in leprosy patients, will allow a more complete definition of the role of TNF in mycobacterial infection.

13.

ACKNOWLEDGMENTS 14.

This work was supported by grants from the National Institute of Allergy and Infectious Diseases (AI-22553, AI-27285, AI-31066, and AI-18357), the UNDP/World Bank/World Health Organization Special Programme for Research and Training in Tropical Diseases (IMMLEP, THELEP, and IMMTUB), the Potts Memorial Foundation, and the Los Angeles branch of the American Lung Association. Mycobacterial cell wall products were provided through contract AI-05074 from the National Institute of Allergy and Infectious Diseases. Computational assistance was provided by the CLINFO project, funded by the National Institutes of Health (grant RR 00043). We are indebted to Bonnie Kerr, Summer Young, Lily Chen, and Sue-Jane Fong for help with the TNF assays. We are grateful to Vijay Mehra for the gift of M. leprae sonicate and for helpful scientific suggestions. We thank Georges Grau for stimulating

15.

16.

17.

18.

discussions. REFERENCES 1. Adams, L. B., S. G. Franzblau, Z. Vavrin, J. B. Hibbs, and J. L. Krahenbuhl. 1991. L-Arginine-dependent macrophage effector functions inhibit metabolic activity of Mycobacterium leprae. J. Immunol. 147:1642-1646. 2. Arnoldi, J., J. Gerdes, and H.-D. Flad. 1990. Immunohistologic assessment of cytokine production of infiltrating cells in various forms of leprosy. Am. J. Pathol. 137:749-753. 2a.Barnes, P. F., et al. Submitted for publication. 3. Barnes, P. F., S.-J. Fong, P. J. Brennan, P. E. Twomey, A. Mazumder, and R. L. Modlin. 1990. Local production of tumor necrosis factor and interferon-gamma in tuberculous pleuritis. J. Immunol. 145:149-154. 4. Barnes, P. F., V. Mehra, G. R. Hirschfield, S.-J. Fong, C. Abou-Zeid, G. A. W. Rook, S. W. Hunter, P. J. Brennan, and R. L. Modlin. 1989. Characterization of T-cell antigens associated with the cell wall protein-peptidoglycan complex of Mycobacterium tuberculosis. J. Immunol. 143:2656-2662. 5. Barnetson, R. S. C., G. Bjune, J. M. H. Pearson, and G. Kronvall. 1976. Cell mediated and humoral immunity in "reversal reactions." Int. J. Lepr. 44:267-273. 6. Bermudez, L. E. M., and L. S. Young. 1988. Tumor necrosis factor, alone or in combination with IL-2, but not IFN-gamma, is associated with macrophage killing of Mycobactenum avium complex. J. Immunol. 140:3006-3013. 7. Beutler, B., and A. Cerami. 1987. Cachectin: more than a tumor necrosis factor. N. Engl. J. Med. 316:379-385. 8. Bjorvatn, B., R. S. C. Barneston, G. Kronvall, R. H. Zubler, and P. H. Lambert. 1976. Immune complexes and complement hypercatabolism in patients with leprosy. Clin. Exp. Immunol.

26:388-396. 9. Bjune, G., R. S. C. Barnetson, D. S. Ridley, and G. Kronvall. 1976. Lymphocyte transformation test in leprosy: correlation of the response with inflammation of lesions. Clin. Exp. Immunol. 25:85-94. 10. Chan, J., Y. Xing, R. S. Magliozzo, and B. R. Bloom. Killing of virulent M. tuberculosis by the reactive nitrogen intermediates produced by activated murine macrophages. J. Exp. Med., in press. 11. Chatterjee, D., C. M. Bozic, M. McNeil, and P. J. Brennan. 1991. Structural features of the arabinan component of the lipoarabinomannan of Mycobacterium tuberculosis. J. Biol. Chem. 266:9652-9660. 12. Daffe, M., P. J. Brennan, and M. McNeil. 1990. Predominant

19.

20.

21.

22.

23.

24.

25.

26. 27.

28. 29.

30. 31.

1445

structural features of the cell wall arabinogalactan of Mycobacterium tuberculosis as revealed through characterization of oligoglycosyl alditol fragments by gas chromatography/mass spectrometry and by 1H and 13C NMR analyses. J. Biol. Chem. 265:6734-6743. Ding, A. H., C. F. Nathan, and D. J. Stuehr. 1988. Release of reactive nitrogen intermediates and reactive oxygen intermediates from mouse peritoneal macrophages. Comparison of activating cytokines and evidence for independent production. J. Immunol. 141:2407-2412. Drutz, D. J., and R. A. Gutman. 1973. Renal manifestations of leprosy: glomerulonephritis, a complication of erythema nodosum leprosum. Am. J. Trop. Med. Hyg. 22:496-502. Espevit, T., and J. Nissen-Meyer. 1986. A highly sensitive cell line, WEHI 164 clone 13, for measuring cytotoxic factor/tumor necrosis factor from human monocytes. J. Immunol. Methods 95:99-105. Foley, N., C. Lambert, M. McNicol, N. Hohnson, and G. A. W. RooL 1990. An inhibitor of the toxicity of tumour necrosis factor in the serum of patients with sarcoidosis, tuberculosis and Crohn's disease. Clin. Exp. Immunol. 80:395-399. Godal, T., B. Myrvang, D. R. Samuel, W. F. Ross, and M. Lofgren. 1973. Mechanism of reactions in borderline tuberculoid (BT) leprosy. Acta Pathol. Microbiol. Scand. 236(Suppl.):4553. Green, S. J., R. M. Crawford, J. T. Hockmeyer, M. S. Meltzer, and C. A. Nacy. 1990. Leishmania major amastigotes initiate the L-arginine-dependent killing mechanism in IFN-gamma-stimulated macrophages by induction of tumor necrosis factor-alpha. J. Immunol. 145:4290-4297. Hunter, S. W., M. McNeil, R. L. Modlin, V. Mehra, B. R. Bloom, and P. J. Brennan. 1989. Isolation and characterization of the highly immunogenic cell wall-associated proteins of Mycobacterium leprae. J. Immunol. 142:2864-2872. Kindler, V., A.-P. Sappino, G. E. Grau, P.-F. Piguet, and P. Vassalli. 1989. The inducing role of tumor necrosis factor in the development of bactericidal granulomas during BCG infection. Cell 56:731-740. McNeil, M., M. Daffe, and P. J. Brennan. 1990. Evidence for the nature of the link between the arabinogalactan and peptidoglycan of mycobacterial cell walls. J. Biol. Chem. 265:1820018206. Mehra, V., B. R. Bloom, V. K. Torigian, D. Mandich, M. Reichel, S. M. M. Young, P. Salgame, J. Convit, S. W. Hunter, M. McNeil, P. J. Brennan, T. H. Rea, and R. L. Modlin. 1989. Characterization of Mycobacterium leprae cell wall-associated proteins using T-lymphocyte clones. J. Immunol. 142:28732878. Moreno, C., J. Taverne, A. Mehlert, C. A. W. Bate, R. J. Brealey, A. Meager, G. A. W. Rook, and J. H. L. Playfair. 1989. Lipoarabinomannan from Mycobacterium tuberculosis induces the production of tumour necrosis factor from human and murine macrophages. Clin. Exp. Immunol. 76:240-245. Mosmann, T. 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65:55-63. Pisa, P., M. Gennene, 0. Soder, T. Ottenhoff, M. Hansson, and R. Kiessling. 1990. Serum tumor necrosis factor levels and disease dissemination in leprosy and leishmaniasis. J. Infect. Dis. 161:988-991. Rea, T. H., and N. E. Levan. 1980. Variations in dinitrobenzene responsivity in untreated leprosy: evidence for a beneficial role for anergy. Int. J. Lepr. 48:120-125. Rea, T. H., and C. R. Taylor. 1977. Serum and tissue lysozyme in leprosy. Infect. Immun. 18:847-856. Ridley, D. S. 1969. Reactions in leprosy. Lepr. Rev. 40:77-81. Ridley, D. S. 1974. Histological classification and the immunological spectrum of leprosy. Bull. W.H.O. 51:451-465. Ridley, D. S., and W. H. Jopling. 1966. Classification of leprosy according to immunity. A five-group system. Int. J. Lepr. 34:255-273. Rook, G. A. W., R. Al Attiyah, and N. Foley. 1989. The role of cytokines in the immunopathology of tuberculosis, and the

1446

INFECT. IMMUN.

BARNES ET AL.

regulation of agalactosyl IgG. Lymphokine Res. 8:323-328. 32. Sampaio, E. P., E. N. Sarno, R. Galilly, Z. A. Cohn, and G. Kaplan. 1991. Thalidomide selectively inhibits tumor necrosis factor alpha production by stimulated human monocytes. J. Exp. Med. 173:699-703. 33. Sarno, E. N., G. E. Grau, L. M. M. Vieira, and J. A. Nery. 1991. Serum levels of tumor necrosis factor alpha and interleukin-1l during leprosy reactional states. Clin. Exp. Immunol. 84:103108. 34. Selmaj, K. W., and C. S. Raine. 1988. Tumor necrosis factor mediates myelin and oligodendrocyte damage in vitro. Ann. Neurol. 23:339-346. 35. Sibley, L. D., L. B. Adams, and J. L. Krahenbuhl. 1990. Inhibition of interferon-gamma-mediated activation in mouse macrophages treated with lipoarabinomannan. Clin. Exp. Immunol. 80:141-148. 36. Silva, C. L., and N. T. Foss. 1989. Tumor necrosis factor in leprosy patients. J. Infect. Dis. 159:787-790. 37. Stach, J. L., M. Strobel, F. Fumoux, and J. F. Bach. 1982. Defect in the generation of cytotoxic T cells in lepromatous leprosy. Clin. Exp. Immunol. 48:633-640. 38. Sullivan, L., S. Sano, C. Pirmez, P. Salgame, C. Mueller, F. Hofman, K. Uyemura, T. H. Rea, B. R. Bloom, and R. L.

39.

40.

41.

42.

43. 44.

Modlin. 1991. Expression of adhesion molecules in leprosy lesions. Infect. Immun. 59:4154-4160. Takashima, T., C. Ueta, 1. Tsuyuguchi, and S. Kishimoto. 1990. Production of tumor necrosis factor alpha by monocytes from patients with pulmonary tuberculosis. Infect. Immun. 58:32863292. Thomson, B. M., G. R. Mundy, and T. J. Chambers. 1987. Tumor necrosis factors alpha and beta induce osteoblastic cells to stimulate osteoclastic bone resorption. J. Immunol. 138:775779. Wallis, R. S., M. Amir Tahmasseb, and J. J. Ellner. 1990. Induction of interleukin 1 and tumor necrosis factor by mycobacterial proteins: the monocyte Western blot. Proc. Natl. Acad. Sci. USA 87:3348-3352. Waters, M. F. R., J. L. Turk, and S. N. C. Wemambu. 1971. Mechanisms of reaction in leprosy. Int. J. Lepr. 39:417-428. Wemambu, S. N. C., J. L. Turk, M. F. R. Waters, and R. J. W. Rees. 1969. Erythema nodosum leprosum: a clinical manifestation of the Arthus phenomenon. Lancet ii:933-935. Yamamura, M., K. Uyemura, R. J. Deans, K. Weinberg, T. H. Rea, B. R. Bloom, and R. L. Modlin. 1991. Defining protective responses to infectious pathogens: cytokine profiles in leprosy lesions. Science 254:277-279.