Interleukin-18 expression by alveolar epithelial cells type II in tuberculosis and sarcoidosis

Authors


  • Editor: Alex van Belkum

Dmitri V. Pechkovsky, Division of Infectious Diseases, University of British Columbia, 2733 Heather St., Vancouver, British Columbia, V5Z 3J5, Canada. Tel.: +1 604 875 4111; fax: +1 604 875 4013; e-mail: petchkov@interchange.ubc.ca

Abstract

The aims of the present study were to characterize the localization of interleukin-18 (IL-18) expression in lung tissue specimens from patients with pulmonary tuberculosis, sarcoidosis and controls, and to determine whether human alveolar epithelial cells type II (AEC-II) are able to generate IL-18 in primary culture. IL-18 was determined using semiquantitative reverse-transcription-PCR and localized in lungs using in situ hybridization. IL-18 protein levels were determined using the enzyme-linked immunosorbent assay and western blot analysis. Alveolar epithelial cells type II (AEC-II) were stimulated in vitro by proinflammatory cytokines, lipopolysaccharides, and whole cell lysate from Mycobacterium tuberculosis. IL-18 mRNA expression was significantly increased in the lungs affected by tuberculosis and sarcoidosis. In situ hybridization revealed different sites of expression in tuberculosis and sarcoidosis lungs, with AEC-II as one major source of IL-18 in the alveolar compartment. Basal IL-18 expression could be detected in normal AEC-II. Whole cell lysate from M. tuberculosis, but not lipopolysaccharide, led to a strong increase of IL-18 mRNA accumulation in AEC-II. Resting AEC-II secreted only small amounts of IL-18, but intracellular IL-18 protein levels increased in a time-dependent manner during culture. Proinflammatory cytokines tumour necrosis factor-α, IL-1β, and interferon-γ altered IL-18 mRNA expression and mature protein secretion of human AEC-II. These findings indicate a possible role for AEC-II and AEC-II-derived IL-18 in pathomechanisms of granulomatous lung diseases.

Introduction

Interleukin-18 (IL-18) has been identified as an interferon-γ-inducing factor and it plays an important role in the T helper1 (Th1) response (Akira, 2000). The IL-18 gene encodes a precursor protein (pro-IL18) that is processed and cleaved by IL-1β-converting enzyme/caspase-1 into the mature bioactive form of IL-18, a 18.3-kDa polypeptide (Ushio et al., 1996; Gu et al., 1997). IL-18 has a variety of biologic effects consistent with its role in promoting Th1 development. Immunity to mycobacterial infection is closely linked to the emergence of IFN-γ-secreting Th1 cells, resulting in macrophage activation and recruitment of circulating monocytes to initiate chronic granuloma formation (Barnes & Wizel, 2000; Jo et al., 2003). In IL-18-deficient mice, Mycobacterium tuberculosis infection resulted in reduced levels of IFN-γ and the development of marked granulomatous inflammation (Sugawara et al., 1999). Serum levels of IL-18 were significantly elevated in patients with active tuberculosis and sarcoidosis, and IL-18 concentration in bronchoalveolar lavage fluid from sarcoid patients was increased in comparison with bronchial asthma and healthy controls (Shigehara et al., 2000; Yamada et al., 2000; Ho et al., 2002). However, alveolar macrophages from patients with sarcoidosis spontaneously produced the same or even lower levels of IL-18 compared with alveolar macrophages from healthy controls (Ho et al., 2002; Hauber et al., 2003).

In this context, we hypothesized that other resident cells in the lung, not obtained by bronchoalveolar lavage, are important producers of IL-18, and their IL-18 production might cause the elevated levels of IL-18 in serum and bronchoalveolar lavage fluid from patients with active tuberculosis and sarcoidosis. From studies employing human bronchial epithelial cells, it is assumed that airway epithelium is also able to express IL-18 (Cameron et al., 1999). In addition, it has been shown that IL-18 expression was significantly elevated in intestinal epithelial cells isolated from patients with Crohn's disease (Pizarro et al., 1999).

These studies led us to investigate the expression and localization of IL-18 in lung tissue specimens from patients with tuberculosis and sarcoidosis in comparison with controls. In addition, we evaluated the capacity of isolated human alveolar epithelial cells type II (AEC-II) to release IL-18 in vitro and analyzed possible regulatory mechanisms.

Our data suggest that IL-18 is constitutively expressed in AEC-II, and M. tuberculosis up-regulates IL-18 mRNA expression and protein production of these cells. Using in situ hybridization and RT-PCR techniques, IL-18 mRNA transcripts were found to be increased in tuberculosis and sarcoidosis compared with controls, and in involved vs. noninvolved lung areas. In situ hybridization on surgically resected lung tissues localized IL-18 mRNA to both interstitial and intra-alveolar mononuclear cells as well as AEC-II, in which IL-18 expression was more abundant in tuberculosis lung tissues compared with sarcoidosis and noninflamed control lung tissues. To our knowledge, this report is the first identifying an inflammatory role for AEC-II in tuberculosis pathogenesis as a main source of IL-18.

Materials and methods

Clinical data

We examined lung tissue samples from four patients with pulmonary tuberculosis and four patients with sarcoidosis. Tuberculosis diagnosis was established by histopathological discovery of granulomas with caseous necrosis in lung biopsies and by positive culture of Mycobacterium tuberculosis from biopsy specimens. All patients were HIV-negative. The tuberculosis patients who were included in this study suffered from the advanced disease: two patients have had destructive tuberculomas in the upper right lobes, and two patients suffered from caseous destructive pneumonia and chronic fibro-cavernous tuberculosis. Lung tissue specimens from patients with pulmonary tuberculosis, which were used for in situ hybridization, had been excised from parts of the resections without macroscopically evident specific lesions owing to biological safety reasons. Diagnosis of pulmonary sarcoidosis was based on clinical and radiological data with the histological confirmation on lymph nodes and lung biopsy as suggested by recent consensus statement (ATS/ERS/WASOG, 1999). Four patients with sarcoidosis type II (bilateral hilar lymphadenopathy with parenchymal involvement) were enrolled in this study. Samples of lung tissue from two patients with benign tumors and eight patients with central lung cancer who underwent surgery for diagnostic or treatment reasons served as controls. Informed consent was obtained from all patients to take part in this study. Macroscopically tumor-free lung tissue was obtained from resected lung specimens of subjects with central lung cancer and benign tumors undergoing lobectomy or pneumectomy, both without any other systemic disease. None of them was taking antibiotics or immunosuppressants at the time of surgery. Lung tissue samples were processed using a sterile technique for the isolation of primary AEC-II (see next subsection). Parts of lung tissue specimens were processed for in situ hybridization and histology. The study was approved by the local ethical committee of the institutions involved.

Isolation and primary culture of human alveolar epithelial cells type II (AEC-II)

Macroscopically tumor-free lung tissue resections were used for isolation of AEC-II as described previously (Pechkovsky et al., 2002a). In brief, the lung tissue pieces were digested with sterile dispase solution (Roche Diagnostics, Mannheim, Germany), and detached cells were removed from digested tissue by intensive washing. The resulting single-cell suspension was further processed for removal of contaminating leukocytes, alveolar macrophages, and lung stromal cells. Identity of AEC-II was confirmed by a modified Papanicolaou staining, alkaline phosphatase activity, and surfactant protein A (SP-A) mRNA expression in reverse-transcription-PCR (RT-PCR). Cell purity was assessed by immunoperoxidase staining with monoclonal antibodies directed against CD3 and CD14 as previously described in detail (Pechkovsky et al., 2002a). All AEC-II preparations included in this report were free of CD14+ and CD3+ cells as determined by immunocytochemistry, 97±1.9% of cells were identified as AEC-II by Papanicolaou staining, and 94±1.1% of cells were positive for alkaline phosphatase (Pechkovsky et al., 2002b). Viability of the AEC-II after isolation was >97% as determined by trypan blue exclusion. Cells were placed in 24-well plastic plates (NUNC, Wiesbaden, Germany) and cultured in complete medium [RPMI1640 medium with 2 mM l-glutamine, 10% heat-inactivated fetal calf serum (GIBCO, Paisley, UK), 1% penicillin/streptomycin solution, and 1% sodium pyruvate solution (Biochrom, Berlin, Germany)] in a humidified atmosphere containing 5% CO2 at 37°C for different time periods (12–72 h). Additionally, AEC-II were stimulated with human recombinant TNF-α (1–10 ng mL−1) (courtesy of Dr E. Schlick, Knoll AG, Ludwigshafen, Germany), IFN-γ (10–100 U mL−1) or IL-1β (10–100 U mL−1) (both from Strahtmann AG, Hamburg, Germany), and with bacterial products: highly purified lipopolysaccharide (LPS) (10–1000 ng mL−1) (LPS from Salmonella minnesota was kindly provided by Dr K. Brandenburg, Research Center Borstel, Germany), purified protein derivative (PPD) (1–10 μg mL−1), and a whole cell lysate of Mycobacterium tuberculosis (WCL-MTB) (1–10 μg mL−1) (WCL-MTB was kindly provided by Dr K. Dobos, NIH&NIAID Contract NO1 AI-75320, Colorado State University, Fort Collins, CO). After culture, cell viability always exceeded 95% in both nonstimulated and proinflammatory cytokine- or bacterial-products-stimulated AEC-II as determined by trypan blue exclusion. AEC-II culture supernatants were collected and cells were lysed with TRIzol Reagent (Invitrogen, Karlsruhe, Germany). Cell lysates were frozen and kept at −70°C before RNA isolation. In separate experiments, cell pellets were washed with phosphate-buffered saline, resuspended in 0.2 mL of lysis buffer [50 mM Tris HCl, pH 7.6; 100 mM NaCl; 2 mM EDTA; 2 mM EGTA; 0.1% of Triton X; and 1% of protease inhibitor cocktail (Sigma, St Louis, MO)] and lysed for 30 min on ice. AEC-II lysates were kept at −70°C before western blot or enzyme-linked immunosorbent assay (ELISA).

RT-PCR

Total RNA was extracted from 1 × 106 cells or lung tissue specimens using TRIzol Reagent and RNeasy Kit according to the manufacturer's protocols (Invitrogen and Qiagen, Hilden, Germany, respectively). Total RNA (1 μg) was reverse transcribed with SuperScript™ RT (Invitrogen) using oligo (dT)12–18 primer to produce complementary DNA (cDNA) according to the manufacturer's protocol. PCR amplification of the cDNA (2.5 μL) was carried out using oligonucleotide primer pairs specific for β-actin and IL-18. To demonstrate that RNA samples from AEC-II were not significantly contaminated by RNAs from other type of cells (alveolar macrophages or lymphocytes), CD3- and CD14-specific primers were also used. All primers were synthesized by MWG-Biotech (Ebersberg, Germany). Target cDNA was amplified using a three-temperature PCR and an automated thermocycler (Biometra, Göttingen, Germany) according to Puren et al. (1999) with primer pairs for IL-18, and according to Pechkovsky et al. (2002b) with primer pairs for β-actin, SP-A, CD3 and CD14, respectively. The numbers of cycles were the same for IL-18, CD3, CD14, SP-A, and β-actin. PCR products (length: 344, 517, 341, 622, and 309 bp for IL-18, CD3, CD14, SP-A, and β-actin, respectively) were electrophoresed on 1.5% agarose gels and stained with ethidium bromide. Gel analysis was done densitometrically with the Gel Doc 2000 gel documentation system and quantity one 4.0.3 software (Bio-Rad Laboratories, Hercules, CA). To assure the identity of the PCR-amplified fragments, the size of each amplified mRNA fragment was compared with DNA standards [100 bp DNA Ladder (Invitrogen)] electrophoresed on the same gel. All primers used in the RT-PCR were intron-spanning and therefore genomic DNA contamination did not influence the PCR results.

In situ hybridization

Hepes glutamic acid buffer mediated organic solvent protection effect (HOPE)-fixed, paraffin-embedded lung tissue specimens from patients and controls were prepared as previously described (Goldmann et al., 2002). Riboprobes for in situ hybridization were generated from IL-18 cDNA using PCR and IL-18 specific primers as described in the previous section. The products of PCR were purified and concentrated with MinElute Reaction Cleanup kit (Qiagen), and labeled with digoxigenin following the manufacturer's instructions (Dig-High-Prime; Roche Diagnostics, Mannheim, Germany). Concentration of labeled probes has been estimated by comparison with a control DNA of given concentration and direct detection of the probes spotted onto positively charged nylon membranes (Roche Diagnostics). In situ hybridization was carried out as previously described in detail (Goldmann et al., 2002). Hybridization of a probe targeting the mRNA of SP-A, a specific product of AEC-II, served as an additional positive control. For negative control, sections were hybridized with the hybridization buffer in the absence of digoxigenin-labeled probes. Counterstain was performed with Mayer's hemalum.

Western blot analysis of IL-18 protein production by AEC-II

Cell lysates obtained from primary cultured AEC-II were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (10% acrylamide) and transferred onto nitrocellulose membranes (Amersham Biosciences, Freiburg, Germany) using a semidry technique (Hoefer Semi-Dry Transfer Unit; Amersham Biosciences, Little Chalfont, UK). Recombinant human IL-18 (Biotrend, Cologne, Germany) was used as a positive control. The membranes were labeled with monoclonal mouse antihuman IL-18 antibodies (mouse IgG1, clone 74801), as suggested by the supplier (R&D Systems, Wiesbaden, Germany). The blots were developed using the enhanced chemiluminescence (ECL, Amersham Biosciences, Freiburg, Germany) technique according to the manufacturer's instructions.

IL-18 ELISA

Interleukin-18 concentrations in primary cultured AEC-II supernatants and cell lysates were measured in duplicate by commercial available human IL-18 ELISA kit (MBL International, Nagoya, Japan) according to the manufacturer's instruction. Optical density readings were obtained with a MRX Microplate Reader and analyzed with revelation 2.0 software (both from Dynex Technologies, Worthing, UK). The lower detection limit of the assay was 12.5 pg mL−1. For duplicate samples, an intra-assay coefficient of variation of <10% and interassay coefficient of variation of <20% was accepted.

Statistical analysis

Data are expressed as means±SEM. Statistical comparisons were made by anova with post hoc Fisher's protected least significant difference for each agent separately. Probability values were considered significant if they were less than 0.05. All testing was done using statview 5.0 software (SAS Institute Inc., Cary, NC).

Results

IL-18 expression pattern and mRNA localization in human lung tissue

The expression of IL-18 mRNA transcripts in paraffin-embedded sections from lungs of patients with tuberculosis, sarcoidosis, and controls is presented in Fig. 1. In noninflamed and microscopically tumor-free human lung specimens, a clear positive signal for IL-18 mRNA was detected in AEC-II, which were typically localized at alveolar corners and exhibited cuboidal morphology (Fig. 1a, arrowheads), as well as in scattered alveolar macrophages in the intra-alveolar space (Fig. 1a, arrows). Signals for IL-18 mRNA were restricted to the cytoplasm of the cells. Specific signals were not detected in control preparations, in which specific probes were substituted by hybridization buffer alone or an irrelevant probe (Fig. 1b). IL-18 mRNA transcripts were found to be more abundant in both AEC-II and alveolar macrophages localized in the surrounding areas of sarcoid granulomas (Fig. 1c) and in alveolitis areas (Fig. 1d) of lung tissues from patients with sarcoidosis. There were no positive signals for IL-18 mRNA within sarcoid granulomas, interstitial inflammatory cells identified as lymphocytes, or in the endothelial layer of small pulmonary vessels and capillaries (Fig. 1c). In contrast to the results obtained from control lungs, severely affected tuberculosis lung tissues revealed very strong positive signals for IL-18 mRNA in almost all AEC-II in the lungs from patients with destructive tuberculoma (Fig. 1e), caseous pneumonia (Fig. 1g) and chronic fibro-cavernous tuberculosis (Fig. 1h). Histological evidence of caseating granulomas was not found in all lung tissue resections used for in situ hybridization. The cellular localization of IL-18 mRNA expression was confirmed by the results from in situ hybridization for SP-A mRNA, which is specifically and constitutively expressed in AEC-II. Figure 1f demonstrates the SP-A expression pattern as it was seen for IL-18 in the same lung tissue preparation from tuberculosis patient (Fig. 1e).

Figure 1.

  Interleukin-18 (IL-18) expression pattern and mRNA localization in lung tissue biopsies obtained from patients with pulmonary sarcoidosis, tuberculosis, and controls. Panels a and b (two representative photomicrographs): in situ hybridization for IL-18 in control lung tissue; panel b demonstrates the absence of the signal in preparations where IL-18 probe was omitted (negative control). There is a high expression of IL-18 mRNA in alveolar epithelial cells type II (AEC-II) (arrowheads) and macrophages free-located within the alveoli (arrows). Panels c and d (two representative photomicrographs): a strong signal for IL-18 mRNA can be seen in intra-alveolar cells (AEC-II and alveolar macrophages) in close proximity to sarcoid granuloma (c) and in inflammatory cells within areas of sarcoid alveolitis (d). Panels e, g and h (three representative photomicrographs): a strong signal for IL-18 mRNA can be seen in alveolar compartments of lung affected by tuberculosis, in both distal part (e) and in areas of inflammation and fibrosis (g and h). Positive signals for IL-18 mRNA are mainly localized in AEC-II (arrowheads) and inflammatory cells within alveoli (arrows). Panel f demonstrates the localization of surfactant protein A (SP-A) mRNA, a specific product of AEC-II, in the same preparations of lung tissue biopsies.

Results from in situ hybridization were verified by conventional RT-PCR with RNA samples obtained from lung tissue resections of patients with tuberculosis, sarcoidosis and normal controls. Indeed, IL-18 mRNA was detected in all lung tissue preparations included in the present study (Fig. 2). The increased levels of IL-18 mRNA transcripts were observed in all tuberculosis and sarcoidosis lung tissues in comparison with controls. The message levels of constitutively expressed SP-A gene were not significantly different between all lung tissue samples, except for one sample from sarcoid lung biopsy, in which lung parenchyma was completely substituted by sarcoid granulomas (Fig. 2, lane 8). In this sample, only a weak signal for IL-18 mRNA could be obtained. The transcript levels of the housekeeping gene β-actin were not different in all RNA preparations (Fig. 2). Thus, the data from this study show that IL-18 mRNA is constitutively expressed in the human lung, and tuberculosis and sarcoidosis processes up-regulate IL-18 expression mainly in AEC-II and alveolar macrophages.

Figure 2.

  Results of RT-PCR product analyses of RNA from lung tissue biopsies of patients with tuberculosis, sarcoidosis and controls using specific primers for IL-18, SP-A, and β-actin. Lanes 1–4: cDNA samples obtained from lung tissue preparations of patients with TB; lanes 5–6: cDNA samples obtained from lung tissue preparations of controls; lanes 7–10: cDNA samples obtained from lung tissue preparations of patients with sarcoidosis. Lane C: positive control (IL-18 positive cDNA); lane M: DNA molecular weight marker (100 bp ladder).

IL-18 mRNA expression and protein release of isolated human AEC-II, and their regulation by proinflammatory cytokines

To study the mechanism(s) by which tuberculosis or sarcoidosis processes might up-regulate IL-18 expression and protein release in human AEC-II, we evaluated IL-18 mRNA and protein levels in isolated and primary cultured normal human AEC-II after stimulation with the key proinflammatory cytokines TNF-α, IFN-γ, and IL-1β. This in vitro study confirms the observation in in situ hybridization that human AEC-II constitutively express IL-18 mRNA and produce IL-18 protein in vivo. As shown in Fig. 3a, freshly isolated AEC-II contain an amount of IL-18 protein, which is detectable in cell lysates by ELISA. Normal human AEC-II spontaneously produced IL-18 in a time-dependent manner, and it accumulated as nonsecreted IL-18 precursor within the cells. In contrast to the intracellular pool of IL-18, relatively small amounts of IL-18 were spontaneously secreted by AEC-II, and the level of IL-18 secretion was stable over the 72 h culture period (Fig. 3a). The increase of intracellular IL-18 was accompanied by IL-18 mRNA accumulation in primary cultured AEC-II. As shown in Fig. 3b, the amounts of the message in 24 h cultured, nonstimulated normal AEC-II were significantly higher in comparison with noncultured freshly isolated cells. Surprisingly, proinflammatory cytokines did not significantly influence IL-18 mRNA accumulation in AEC-II cultured for 24 h in the presence of TNF-α, IL-1β, and IFN-γ. Although a trend to increased IL-18 mRNA levels was detected in AEC-II stimulated with different concentrations of TNF-α or IL-1β compared with nonstimulated cells in separate experiments, these observed differences did not reach statistical significance (Fig. 3b). Moreover, spontaneous IL-18 secretion of AEC-II was significantly inhibited in IL-1β- and TNF-α-stimulated cells compared with nonstimulated controls and IFN-γ-stimulated cells. In contrast to TNF-α and IL-1β, which exhibit a suppressive effect, no consistent influence of IFN-γ could be observed (Fig. 3c).

Figure 3.

  Interleukin-18 mRNA expression, protein production and secretion of primary cultured, normal human AEC-II. Effect of proinflammatory cytokines. Panel a shows a time course of spontaneous IL-18 production and secretion of primary cultured AEC-II. After 4 h in culture, cells were washed with fresh medium. Cells were lysed with lysis buffer and culture conditioned medium was collected after the indicated time periods. IL-18 concentrations were determined in cell lysates and culture conditioned medium by ELISA. Circles represent means of duplicate determinations from two independent experiments with two different AEC-II isolations. Panel b depicts the IL-18 mRNA analysis by RT-PCR in ACE-II, noncultured (lane 0) and cultured without (lane 1, nonstimulated) or with TNF-α (1–10 ng mL−1; lanes 2, 3 and 4), IFN-γ (10–100 U mL−1; lanes 5, 6 and 7) or IL-1β (10–100 U mL−1; lanes 8, 9 and 10) stimulation for 24 h. A representative image of IL-18 and β-actin mRNA expression in AEC-II from one of five independent experiments with five different AEC-II isolations is shown. Lane M, molecular marker. Densitometric analysis of the bands was performed as described in the Materials and methods section. The PCR products from each culture were individually quantified, and values presented are the mean percentages of β-actin density±SEM (*P<0.01 compared with cultured cells, n=5, anova with protected least significant difference (PLSD). Panel c shows spontaneous and cytokine-modified IL-18 secretion of AEC-II cultured for 24 h. Data are presented as mean±SEM from five independent experiments (*P<0.05 compared with nonstimulated cells, anova with protected least significant difference).

Whole cell lysate from Mycobacterium tuberculosis (WCL-MTB) up-regulates IL-18 mRNA expression and protein release of normal AEC-II

We investigated whether cell components of M. tuberculosis directly induce the expression of IL-18 by primary cultured normal AEC-II. AEC-II were stimulated with different concentrations of WCL-MTB, PPD, and highly purified LPS for 24 h, and the levels of IL-18 mRNA expression were determined with semiquantitative RT-PCR. As shown in Fig. 4a, only WCL-MTB at concentrations of 2 and 4μg mL−1 significantly up-regulated IL-18 mRNA expression of AEC-II. Densitometry data from a typical experiment are shown in Fig. 4b. The increase in WCL-MTB concentration above 5 μg mL−1 strongly inhibited IL-18 mRNA accumulation owing to a pronounced cytotoxic effect on primary cultured AEC-II (data not shown). In contrast with WCL-MTB, neither PPD nor LPS at the concentrations of 1–10μg mL−1 influenced IL-18 mRNA accumulation in AEC-II compared with nonstimulated controls (Fig. 4a). There was no change in β-actin levels (Figs 4a and b), which provided evidence that AEC-II did not die or get lost during the cultures and stimulations with bacterial products.

Figure 4.

  The whole cell lysate from Mycobacterium tuberculosis (WCL-MTB) up-regulates IL-18 mRNA expression of primary cultured AEC-II. Panels a and b: representative gel images and the results of semiquantitative RT-PCR from two independent experiments are shown. AEC-II were stimulated with LPS (1 μg mL−1), PPD (10 μg mL−1) and WCL-MTB (2 and 4 μg mL−1) for 24 h; RNA was isolated and reverse-transcribed to complementary DNA (cDNA). Serial dilutions of cDNA were used in PCR with primers specific for IL-18 and β-actin. The PCR products from each culture and serial dilutions were individually quantified by densitometry. The graph demonstrates the higher amount of IL-18 mRNA in AEC-II stimulated with 2 μg mL−1 WCL-MTB in comparison with nonstimulated cells. There are no differences in the amounts of β-actin mRNA in nonstimulated and WCL-MBT-stimulated cells.

Western blot analysis of AEC-II lysates obtained from 24 h cultured cells showed that the 18.3 kDa mature form of IL-18 appeared in WCL-MTB-stimulated AEC-II, whereas both nonstimulated and WCL-MTB-stimulated cells contained abundant 24 kDa pro-IL-18 (Fig. 5). These results indicate that WCL-MTB up-regulates IL-18 expression at both transcription and posttranscription levels.

Figure 5.

  Effect of whole cell lysate from Mycobacterium tuberculosis (WCL-MTB) on IL-18 processing and release of AEC-II. AEC-II from two separate isolations were stimulated with 4 μg mL−1 of WCL-MTB for 24 h. Cells were lysed with lysis buffer, and lysates were subjected to SDS-PAGE with 10% gels and western blotting with anti-IL18 antibodies as described in the Materials and methods section. Recombinant human IL-18 (rhIL-18) was used as a positive control. Pro- (pro-IL-18) and mature IL-18 (mat-IL-18) are indicated by arrows accordingly to molecular weight markers.

Discussion

In this study comparing patients with pulmonary tuberculosis and sarcoidosis with control subjects, we have demonstrated increased levels of IL-18 expression in tuberculosis- and sarcoidosis-affected lungs. In control subjects, we found that alveolar epithelium and alveolar macrophages constitutively express IL-18. Moreover, IL-18 mRNA and small amounts of IL-18 protein were detectable in freshly isolated and primary cultured human AEC-II, suggesting a basal expression of IL-18 in normal lung tissue. The study also revealed that IL-18 mRNA expression was enhanced in inflammatory cells surrounding sarcoid granulomas, but it was absent within the granulomas. A previous study reported that IL-18 mRNA and protein was observed in bronchial epithelium of normal individuals and patients with sarcoidosis (Cameron et al., 1999). Additionally, it has been demonstrated that despite IL-18 increase in bronchoalveolar lavage fluid, alveolar macrophages from patients with sarcoidosis spontaneously produced the same or even lower levels of IL-18 compared with controls (Ho et al., 2002; Hauber et al., 2003). We confirmed and extended these results, showing that AEC-II are a source of the increased IL-18 concentrations in bronchoalveolar lavage fluid from patients with sarcoidosis.

Of particular interest are our findings of the IL-18 mRNA overexpression of AEC-II in the lung of patients with sarcoidosis and tuberculosis. We show that the cellular components of Mycobacterium tuberculosis directly up-regulate IL-18 mRNA expression, IL-18 precursor production and the secretion of mature protein in human AEC-II, which shifts the micromilieu of cytokines towards a Th1 pattern. However, the proinflammatory cytokines playing a key role in pathogenesis of tuberculosis and sarcoidosis (reviewed in Muller-Quernheim, 1998; Barnes & Wizel, 2000) such as TNF-α, IFN-γ, and IL-1β have only modest effects on the regulation of IL-18 gene transcription and protein release of human AEC-II. The effects of these cytokines and the up-regulatory capacity of WCL-MTB as well as the observed IL-18 overexpression of AEC-II in our in situ hybridization experiments suggest a critical role for AEC-II in regulating or orchestrating the pulmonary cytokine network. We and other authors consider that AEC-II are involved, in addition to alveolar macrophages, in pathogenesis of several granulomatous and nongranulomatous lung diseases of unknown origin (Cameron et al., 1999; Greene et al., 2000; Kitasato et al., 2004).

There is increasing evidence that IL-18 may be a key proinflammatory cytokine as well as an important mediator of immune responses to mycobacterial infections (reviewed in Akira, 2000). The production of IL-18 is controlled both at the transcriptional and at the posttranscriptional levels. Similar to IL-1β, IL-18 is first synthesized as a biologically inactive precursor (pro-IL-18) (Ushio et al., 1996) that is cleaved by IL-1β-converting enzyme/caspase-1 to generate the mature, functional cytokine that will be released (Gu et al., 1997). In human keratinocytes and in several human epithelial cell lines, constitutive gene expression for IL-18 has been reported (Naik et al., 1999; Lu et al., 2000). This is similar to our observation that in human AEC-II, IL-18 is also constitutively expressed, and these cells contain large amounts of preformed pro-IL-18 protein, which predominantly accumulates within the cells during in vitro culture. The baseline level of actively secreted mature IL-18 is very low and does not significantly change during culture. In addition, we found that IL-18 mRNA is constitutively expressed by AEC-II in noninflamed lung tissues by in situ hybridization, which reflects the in situ pattern. We detected an increase in IL-18 mRNA of cultured AEC-II in comparison with freshly isolated cells. These findings suggest that the transcriptional activity of the IL-18 promoter is regulated by yet unidentified factor(s) normally present in the lung (Puren et al., 1999).

A recent study demonstrated that proinflammatory cytokines do not modify the spontaneous release of IL-18 in human keratinocytes and epithelial cell lines (Companjen et al., 2000; Lu et al., 2000). In contrast, we are able to show that TNF-α, IL-1β, and IFN-γ can alter IL-18 mRNA expression and mature protein secretion of human AEC-II. We found that IFN-γ augmented to some extent IL-18 secretion, but not IL-18 mRNA expression. These findings are consistent with those of Vankayalapati et al. (2001), who reported that IFN-γ did not enhance IL-18 expression of human monocytes. In contrast to IFN-γ, the proinflammatory cytokines TNF-α and IL-1β slightly up-regulated IL-18 mRNA expression, and significantly inhibited spontaneous IL-18 secretion of AEC-II. It is likely that different effects of TNF-α and IL-1β vs. IFN-γ on IL-1β-converting enzyme/caspase-1 activation cause the divergent actions of these cytokines on IL-18 mRNA expression and protein secretion of AEC-II. Indeed, it has been demonstrated that IL-1β and IFN-γ have opposing effects on caspase activation and Fas-mediated apoptosis of a human epithelial cell line (Coulter et al., 2002). Additionally, Lu et al. (2000) presented data suggesting that Chlamydiatrachomatis infection causes human epithelial cell lines to secrete mature IL-18 by IL-1β-converting enzyme/caspase-1-dependent mechanism. Interestingly, the study showed that Chlamydia infection does not alter the expression of IL-18 mRNA or IL-18 precursor protein production.

Our data demonstrate that M. tuberculosis can directly up-regulate IL-18 mRNA expression and that it promotes the cleavage of premature protein into its mature form in human AEC-II. However, we could not detect any activity of highly purified LPS on IL-18 expression and production of AEC-II. Our previous study has demonstrated that human AEC-II are unresponsive to LPS stimulation, that they express very low level of CD14 (Pechkovsky et al., 2000, 2002b) and toll-like receptor (TLR) 4 mRNA, and that they do not constitutively express the coreceptor molecule MD-2 mRNA (Pechkovsky et al., unpublished data). These findings are in line with data from Abreu et al. (2001) and Jia et al. (2004) showing that human intestinal and bronchial epithelial cells are constitutively unresponsive to LPS. Host organisms have developed a set of receptors, referred to as either pattern recognition receptors or toll-like receptors, which specifically recognize pathogen-associated molecular patterns. TLR4 and TLR2 are required for recognition of LPS and mycobacterial pathogen-associated molecular patterns, respectively (Akira et al., 2001). Recently, we have reported that AEC-II express TLR2 mRNA and protein in situ (Droemann et al., 2003). In view of these findings, AEC-II might produce and secrete biologically active IL-18 protein following exposure to mycobacterial pathogen-associated molecular patterns and, thereby, play a protective role against M. tuberculosis. The exact pathomechanisms of such a response of human AEC-II to M. tuberculosis needs to be addressed by further studies.

In summary, alveolar epithelium is a dynamic cellular layer with multiple functions, and an increasing body of evidence suggests that airway epithelial cells including AEC-II play an important role in regulating acute and chronic inflammation. To our knowledge, this is the first report of a possible contribution of AEC-II and AEC-II-derived IL-18 to processes that occur in sarcoidosis and pulmonary tuberculosis. Although most studies of the interactions between M. tuberculosis and pulmonary cells have focused on alveolar macrophages, interactions between M. tuberculosis and alveolar epithelial cells may contribute significantly to the pathogenesis of tuberculosis. Recently, we have shown that AEC-II–alveolar macrophage interactions and IFN-γ are critical in the NOS2 induction and NO production of human alveolar macrophages (Pechkovsky et al., 2002b). IL-18 constitutive and M. tuberculosis-induced levels produced by AEC-II may contribute to the skewing towards a Th1-like environment in the lung, which is apparent in tuberculosis, sarcoidosis, and other lung diseases.

Acknowledgements

This work was supported in part by grant from the Deutsche Forschungsgemeinschaft (No. Mu 692/5-5). D. V. Pechkovsky is a recipient of Research Fellowship from the INTAS and Alexander von Humboldt Foundation. The authors would like to thank H. Kühl, N. Husmann and S. Adam for excellent technical assistance.

Ancillary