INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tuberculosis, a granulomatous disorder, is an infection caused by Mycobacterium tuberculosis in which the cellular immune response plays an important role. In response to infection with M. tuberculosis, activated macrophages and CD4-positive T-lymphocytes produce and secrete proinflammatory cytokines (interleukin-1 [IL-1], IL-6, and tumor necrosis factor-), Th1 cytokines (interferon- [IFN-] and IL-2) and IL-12 (1). Elevations of IFN- have been found in the affected lung (1) and bloodstream (5) of patients with pulmonary tuberculosis. IFN- has been shown to be an important mediator of macrophage activation involved in controlling M. tuberculosis (4).

IL-18 has recently been identified as an IFN--inducing factor and it plays an important role in the Th1 response (8, 9). Upregulated IL-18 expression has been found in Th1-mediated chronic inflammatory diseases: Crohn's disease (10, 11), rheumatoid arthritis (12), and primary Sjögren syndrome (13) and acute Epstein-Barr virus induced infectious mononucleosis (14). Increased levels of IL-18 have been found in cerebrospinal fluids of patients with bacterial meningitis (15) and in sera of patients with leukemia (16) and hemophagocytic lymphohistiocytosis (17). In this study we measured serum IL-18 and IFN- levels in patients with pulmonary tuberculosis.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Population

We selected 43 patients 18 to 78 yr of age (22 men and 21 women) from approximately 120 patients with culture-proven or histologically confirmed tuberculosis. To eliminate the influence of other diseases we excluded patients with tuberculosis who had complications, including collagen vascular diseases, malignant tumors, and other apparent infections such as hepatitis and HIV. We also excluded patients who had received prior therapy with corticosteroids or other immunomodulators to eliminate the influence of immunoregulatory agents. In addition, patients who had received prior therapy with antituberculosis drugs were excluded because of the difficulty of evaluating the extent of disease on chest radiographic films. In sputum examinations, 28 patients were smear-positive and 15 were smear-negative. Forty-one patients were sputum-culture-positive, and the remaining two patients who had miliary tuberculosis were sputum-culture-negative. Thirty-two patients had positive tuberculin skin test results and 11 had negative skin test results. Thirteen patients had high-grade fever (> 38° C) and 30 did not. Chest radiographic findings were reviewed without knowledge of clinical information, the IL-18 results or the IFN- results according to standard criteria (18); 10 patients had minimal, 22 had moderately advanced, and 11 had far advanced pulmonary tuberculosis. Tuberculous cervical lymphadenitis was histologically confirmed in two patients (one minimal and one moderately advanced pulmonary tuberculosis). None of them had significant pleural effusion on chest radiographic films. In far-advanced pulmonary tuberculosis, three patients were confirmed to have miliary tuberculosis on the basis of clinical symptoms (e.g., high-grade fever and weight loss); widely spread, multiple tiny nodules evident on the chest radiographs and chest CT; histopathologic proof of caseating granulomas and/or bacteriolgic proof of M. tuberculosis, and a dramatic chemotherapy response (5). Blood samples were collected at the time of diagnosis, before therapy with antituberculosis drugs. Serum samples were cryopreserved at 80° C until use.

Healthy Control Subjects

Serum samples were obtained from 25 age- and sex-matched healthy volunteers 23 to 62 yr of age (13 men and 12 women) from among our department staff. Eighteen volunteers had positive tuberculin skin test results and seven had negative skin test results. They had no history of respiratory diseases, and their chest radiographs and pulmonary function tests showed no evidence of respiratory diseases.

Informed consent about cytokine measurements was obtained from the patients and the healthy volunteers.

IL-18 Assay

ELISA for determination of IL-18 in sera was performed according to a previously described method (16). Briefly, a maxisorp plate was coated with monoclonal antibody (MoAb) no. 125-2H to human IL-18 (20 µg/ml in phosphate-buffered solution [PBS]) at room temperature for 180 min and blocked with PBS containing 1% bovine serum albumin (BSA) at 4° C overnight. After washing with PBS containing 0.05% Tween 20 (washing buffer), 100 µl of five times diluted blood samples or standards in PBS containing 1% BSA, 5% fetal calf serum, and 1 M NaCl were then incubated at room temperature for 120 min. After washing with washing buffer, peroxidase-conjugated MoAb no. 159-12B to human IL-18 (5 µg/ml) was added and incubated at room temperature for 120 min. After washing, substrate solution containing o-phenylenediamine was added and incubated at room temperature for 30 min. The reaction was stopped with 100 µl of 1 M sulfuric acid and the absorbance was measured at 490 nm. All assays were performed in duplicate. The detectable range of this ELISA was between 10 and 1,000 pg/ml.

IFN- Assay

IFN- was measured with an ELISA (high sensitivity interferon- human ELISA system (Amersham Life Science, Buckinghamshire, UK) according to the instructions of the manufacturer. The detection limit of the assay was 0.10 pg/ml.

Statistical Analysis

Data were expressed as mean ± SD. The Mann-Whitney U-test was used to compare paired sets of data, and Spearman's rank order correlation was used to determine correlation; p < 0.05 was considered to be statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Serum IL-18 and IFN-

We measured IL-18 in sera of patients with pulmonary tuberculosis and healthy control subjects (Figure 1). We found significantly higher levels of serum IL-18 in patients with pulmonary tuberculosis (424 ± 516 pg/ml) than in healthy control subjects (140 ± 89 pg/ml; p = 0.0016). There was no significant difference in circulating IL-18 between the sexes, and there was no significant correlation between circulating IL-18 and age in healthy control subjects or in patients with pulmonary tuberculosis.

View larger version (14K): [in this window] [in a new window]   Figure 1.   Circulating interleukin-18 (IL-18) levels in patients with pulmonary tuberculosis and in healthy control subjects. Closed circles indicate patients who had tuberculous cervical lymphadenitis, and closed diamonds indicate patients with miliary tuberculosis. The horizontal line at 50 pg/ml represents the lower limit of the IL-18 assay. Serum IL-18 levels in moderately advanced (p = 0.0058) and in far advanced (p < 0.0001) pulmonary tuberculosis were significantly higher than those in control subjects.

We measured IFN- in sera of the same patients and healthy control subjects with high sensitivity ELISA (Figure 2). We found significantly higher levels of serum IFN- in patients with tuberculosis (12 ± 47 pg/ml) than in healthy control subjects (0 ± 0 pg/ml; p = 0.0010). There was no significant difference in circulating IFN- between the sexes, and there was no significant correlation between circulating IFN- and age in healthy control subjects or in patients with pulmonary tuberculosis.

View larger version (14K): [in this window] [in a new window]   Figure 2.   Circulating interferon- (IFN-) levels in patients with pulmonary tuberculosis and in healthy control subjects. Closed circles indicate patients who had tuberculous cervical lymphadenitis, and closed diamonds indicate patients with miliary tuberculosis. The horizontal line at 0.10 pg/ml represents the lower limit of the IFN- assay. Serum IFN- levels in moderately advanced (p = 0.021) and in far advanced (p < 0.0001) pulmonary tuberculosis were significantly higher than those in healthy control subjects.

Relationship of Serum IL-18 and IFN- to Clinical Parameters in Tuberculosis

There were no significant differences of serum IL-18 or IFN- between smear-positive and smear-negative groups or between tuberculin-test-positive or tuberculin-test-negative groups. Circulating IL-18 levels in patients with high-grade fever (871 ± 748 pg/ml) were significantly higher than in those without high-grade fever (230 ± 160 pg/ml; p = 0.0006) (Figure 3). Circulating IFN- levels in patients with high-grade fever (39 ± 81 pg/ml) were significantly higher than in those without high-grade fever (1 ± 1 pg/ml; p < 0.0001) (Figure 4).

View larger version (12K): [in this window] [in a new window]   Figure 3.   Circulating IL-18 levels in patients with pulmonary tuberculosis with and without high-grade fever. Closed circles indicate patients who had tuberculous cervical lymphadenitis, and closed diamonds indicate patients with miliary tuberculosis. The horizontal line at 50 pg/ml represents the lower limit of the IL-18 assay. There was a significant difference between the two groups (p = 0.0006).

View larger version (13K): [in this window] [in a new window]   Figure 4.   Circulating IFN- levels in patients with pulmonary tuberculosis with and without high-grade fever. Closed circles indicate patients who had tuberculous cervical lymphadenitis, and closed diamonds indicate patients with miliary tuberculosis. The horizontal line at 0.10 pg/ ml represents the lower limit of the IFN- assay. There was a significant difference between the two groups (p < 0.0001).

We next compared circulating IL-18 values according to radiologic stages of patients with pulmonary tuberculosis (Figure 1). In far-advanced pulmonary tuberculosis (870 ± 749 pg/ ml), the circulating IL-18 values were significantly higher than in minimal (166 ± 159 pg/ml; p = 0.0012) and moderately advanced (318 ± 326 pg/ml; p = 0.0025) pulmonary tuberculosis and healthy control subjects (p < 0.0001). In moderately advanced pulmonary tuberculosis, the circulating IL-18 values were significantly higher than those in healthy control subjects (p = 0.0058). In contrast, there was no significant difference between healthy control subjects and patients with minimal pulmonary tuberculosis or between those with minimal and moderately advanced pulmonary tuberculosis.

We compared circulating IFN- values according to radiologic stages of patients with pulmonary tuberculosis (Figure 2). In minimal pulmonary tuberculosis, the circulating IFN- value in the patient who had tuberculous cervical lymphadenitis was very high (121 pg/ml), and the average value of the remaining nine patients was 1 ± 1 pg/ml. In the patients with far-advanced pulmonary tuberculosis (31 ± 84 pg/ml), the circulating IFN- levels were significantly higher than in those with minimal (p = 0.018) or moderately advanced (2 ± 4 pg/ml; p = 0.0040) pulmonary tuberculosis and in healthy control subjects (p < 0.0001). In those with moderately advanced pulmonary tuberculosis, the circulating IFN- values were significantly higher than in the healthy control subjects (p = 0.021). In contrast, there was no significant difference between healthy control subjects and those with minimal pulmonary tuberculosis or between those with minimal or moderately advanced pulmonary tuberculosis.

Correlation Between Serum IL-18 and IFN-

We analyzed the correlation between circulating IL-18 values and circulating IFN- values. There was no significant correlation between circulating IL-18 and IFN- in healthy control subjects. In pulmonary tuberculosis, circulating IL-18 values significantly correlated with circulating IFN- values (r = 0.515, n = 43, p = 0.0004) (Figure 5). When the three miliary cases were deleted, there was a significant positive correlation between circulating IL-18 and IFN- values (r = 0.407, n = 40, p = 0.01).

View larger version (12K): [in this window] [in a new window]   Figure 5.   Correlation of circulating IL-18 levels with circulating IFN- levels in patients with tuberculosis. Closed circles indicate patients who had tuberculous cervical lymphadenitis, and closed diamonds indicate patients with miliary tuberculosis. The vertical line at 50 pg/ml and the horizontal line at 0.10 pg/ml represent the lower limit of the IL-18 and IFN- assay, respectively. There was a significant positive correlation between circulating IL-18 and IFN- levels (p = 0.0004).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental studies in animals have shown that IFN- plays a central role in protective immunity to M. tuberculosis (4, 20). We documented increased levels of circulating IFN- in patients with tuberculosis who had high-grade fever and severe disease, consistent with prior studies (5, 7). In contrast, Dlugovitzky and coworkers (6) reported that serum IFN- levels were higher in minimal and moderately advanced disease than in far-advanced disease in patients with pulmonary tuberculosis, although serum IFN- levels in patients with far-advanced disease were prominently higher than in control subjects. Sodhi and coworkers (21) found that reduced IFN- production by peripheral blood mononuclear cells stimulated with heat-killed M. tuberculosis is a marker of severe disease. M. tuberculosis antigen-specific and nonspecific IFN- production is suppressed in patients with tuberculosis (4, 22). Transforming growth factor-- and IL-10-mediated immunosuppression, apoptosis, and sequestration of cells from the bloodstream to sites of disease are implicated as components of peripheral immunosuppression /anergy (4, 22). Factors other than tuberculosis might be responsible for the difference in circulating IFN- levels among radiologic stages observed between the study groups. Additional studies are needed to evaluate circulating IFN- levels in patients with tuberculosis of varying severity.

IL-18 is a novel cytokine that synergizes with IL-12 to induce IFN- production (8, 9). IL-18 expression is upregulated in Th1-mediated chronic inflammatory diseases such as Crohn's disease and rheumatoid arthritis (10). In addition, IL-18 is expressed upon infections and plays a protective role in infections by bacteria, fungi, viruses, and intracellular parasites (23). IL-18 may also contribute to protection against mycobacteria, including M. leprae, M. bovis, and M. tuberculosis (28). In IL-18-deficient mice, M. tuberculosis infection results in reduced levels of IFN-, compared with wild-type mice despite normal IL-12 levels (31). Marked granulomatous inflammation developed in IL-18-deficient mice and was inhibited by exogenous IL-18. Thus, IL-18 contributes to IFN- production and to containment of the granulomatous response in vivo, and these activities are not compensated for by IL-12 or other cytokines.

In this study we documented increased levels of circulating IL-18 in patients with tuberculosis. The levels of circulating IL-18 paralleled disease activity, as judged by radiologic extent of disease and magnitude of fever. IL-18 concentrations also correlated with IFN- concentrations. Additional studies of IL-18 production at the site of disease in patients with tuberculosis may provide further insight into the potential contribution of IL-18 to the protective Th1 response to infection with M. tuberculosis.