Suppressive effect of TNF-α and IL-1 on alveolar fibroblast proliferation in sarcoidosis

The nature of soluble factors that regulate fibroblast proliferation have not been finally characterized. Our aim was to study the role of tumour necrosis factor α (TNF-α) and interleukin-1 (IL-1) in the suppressive activity of alveolar macrophages on autologous lung fibroblasts proliferation in sarcoidosis. We found that supernatants recovered from alveolar macrophages suppressed the proliferation of alveolar fibroblast in sarcoidosis by 35.5 ± 1.13% compared to 3 ± 16% in controls (p < 0.001 between the two groups). This suppression correlated with high content of TNF-α and IL-1 in sarcoidosis patients stage II-III (7.7 ± 2.9 ng/ml TNF-α and 157 ± 53 U/ml IL-1 compared to 3.4 ± 2.4 ng/ml TNF-α and 43 U/ml IL-1 in controls; p < 0.01 and p < 0.001, respectively). Both cytokines in sarcoidosis stage I were within the normal ranges. Exogenous TNF-α (1000-0.5 ng/ml) and IL-1 (500-0.24 ng/ml) had an additive suppressive activity on fibroblast proliferation which was partially reversed by indomethacin.


Introduction
The injury of any tissue, whether caused by trauma, microbes or a foreign antigen, initiates any inflammatory response which leads to the normal physiological process of healing. Acute pneumonias, even those with marked necrosis, can heal without excessive scarring. In contrast, interstitial disorders may result in excessive fibrosis with loss of organ function. The processes responsible for these different outcomes need further definition.
Mononuclear cells are important regulators of the fibrotic response. Their regulatory effects are at least partially mediated by soluble factors that can stimulate 1'2 or inhibit 3'4 fibroblast proliferation.
Interleukin-1 (IL-1) protein has pronounced effects on various lineages of cells and other cells of mesenchymal origin, s'6 Stimulated monocytes and macrophages also elaborate tumour necrosis factor (TNF-o 0 which is now known to have a broad range of cytoregulatory effects 7'8 including the ability to regulate cell proliferation. Moreover TNF-o and IL-1 synergistically stimulate PGE2 el6boration by confluent fibroblast. Because mononuclear cell inflammation precedes the fibrotic stage in sarcoidosis the cytokines released by alveolar macrophages and their role on fibroblast growth may have a crucial role. To test this theory we determined the suppressive effect of alveolar macrophage supernatants from patients with sarcoidosis on proliferation of autologous alveolar (C) 1992 Rapid Communications of Oxford Ltd fibroblasts and correlated it with their TNFand IL-1 content.

Materials and Methods
Study population: Patients were subdivided into three groups. Sarcoidosis patients were diagnosed by clinical and roentgenological presentation, a positive Kweim test or positive biopsy of non-caseating granuloma. According to the X-rays these patients were grouped into stage I (four untreated patients) and stage 11-Ili (nine untreated patients). For controls, seven untreated patients undergoing bronchoscopy due to unexplained persistent cough or after an episode of mild haemoptysis. All of them had chest roentgenograms within normal limits. Written informed consent was obtained from each subject before bronchoscopy and bronchoalveolar lavage. Bronchoalveolar lavage: After informed consent, bronchoscopy with bronchoalveolar lavage was performed with a flexible fibre optic bronchoscope (Olympus BF-B2) as previously described. 1 Preparation of alveolar macrophages: The recovered fluid was collected in specimen traps, filtered through sterile gauze and centrifuged at 400 x g for 15 min at 4C. The pellet obtained was washed three times with cold PBS (Biological Industries, Beit Haemek), the number of viable cells was counted and purified by adherence as previously described.

Preparation of lungflbroblasts: Alveolar fibroblasts (Afb)
were obtained from bronchoalveolar lavage cells after long-term incubation (3)(4) weeks) as previously described. 11 Control fibroblasts were derived from histologically normal areas of lungs resected for diagnostic reasons. The techniques of preparation and the proliferative characteristics of these cells have been described. 12 Preparation of alveolar macrophage supernatants: Supernatants were obtained from alveolar cells cultured as previously described. The cells were allowed to adhere for 1 h, washed vigorously and overlayed by an identical volume of complete RPMI medium (2% foetal calf serum (FCS), antibiotic-antimycotic) with or without 10 #g/ml lipopolysaccharide (LPS) (Difco, St Louis, USA 055:B55). The cells were incubated for 24 h in 5% CO2. Aliquots of the medium harvested from cultures were centrifuged, filtered and frozen for future use at -70C. Fibroblast proliferation test: Fibroblast suspension (100/d) was recovered as described previously. Briefly, cells were washed and resuspended in Dulbecco modified Eagles medium (DMEM) with 1% FCS, 2-mercaptoethanol (5 x 10 -5 M) and 1% antibiotic-antimycotic mixture at 105 cells/ml. Fibroblasts were dispensed into each well of 96-well flat-bottomed microtitre plates and allowed to attach for 1-2 h. Aliquots (100/d) of supernatants of LPS pulsed alveolar macrophages were added. Cultures were incubated in a humidified 5% COg atmosphere for 72 h, and pulsed with 1 Ci3H thymidine for the last 4h of culture. For harvesting, the supernatant from each well was aspirated and 0.1 ml trypsin-EDTA was added to each well. Detached cells were harvested and counted. The growth of fibroblasts in LPS stimulated alveolar macrophage supernatant or cytokines (rlL-1 500-0.24 ng/ml Glaxo IMB; TNF-1000-0.5 ng/ml Genentech Inc, San Francisco, CA, USA) was compared with the growth of fibroblasts in complete DMEM with and without a final concentration of 10/,g/ml LPS. Neutralization of TNFand IL-1 activity was done by anti-TNF and anti-IL-1 MoAbs (240 #g/ml, Genentech and 20/.tg/ml, Genzime respectively) and PGE2 was reversed by indomethacin (1/g/ml, Sigma, Chemical Co, St Louis Mo, USA). Assay of prostaglandin, TNF and IL-1 production: Aliquots of alveolar macrophage supernatants (24 h production) were assayed for PGE2 and IL-1 production. PGE2 was assayed using an ELISA kit (Advanced Magnetic Inc.), TNFwas measured by a biological assay on A9 target cells, and IL-1 by the C3H/HeJ thymocyte comitogenic assay, as described previously. 13 The effects of these supernatants were tested on the proliferation of alveolar fibroblasts ( Table 2). Alveolar macrophage supernatants from sarcoidosis patients suppressed the proliferation of alveolar fibroblasts by 38___ 7.13% whereas alveolar macrophage supernatants from controls induced only a slight suppression of 3 _ _ _ 16% < 0.01 between sarcoidosis patients and controls).
The suppressive activity of alveolar macrophages from stage II-III sarcoidosis patients was correlated with a marked increase in secretion of IL-1 and TNF-. In view of this fact we determined if exogenous cell-free rTNFand IL-1 have suppressive effects on fibroblasts. As shown in Table  3 and Fig. 1 both IL-1 and TNFsuppressed fibroblast proliferation. The suppressive effect of IL-1 was reversed by indomethacin ( Table 3). The specificity of suppressive activity of TNFand IL-1 was ascertained by neutralization with anti-TNF-and anti-IL-1 mAbs (Table 3). Concomitant addition of IL-1 and TNFresulted in an additive suppressive effect (Fig. 2) which was also partially reversed by indomethacin.

Discussion
The cells involved in the inflammatory response generate a variety of factors which appear to regulate the healing process through the recruitment, stimulation of growth, and matrix synthesis by connective tissue cells. 14  is known about the role of these factors in interstitial lung diseases. To understand further the role of these factors we characterized the effect of alveolar macrophage supernatants from patients with sarcoidosis on alveolar fibroblast proliferation and compared it with the control group. The experiments showed that sarcoidosis supernatants exerted an inhibitory effect whereas the control supernatants induced only a slight suppression or enhancement. These results confirmed those previously shown 21 but they reflect more closely the in vivo situation as each alveolar macrophage supernatant was incubated with the autologous fibroblasts.
Although previous studies showed already that alveolar macrophages from sarcoidosis patients secrete increased amounts of IL-122 and TNF-0, 2 the importance of the intercytokine interactions in correlation with different staging in sarcoidosis has not been investigated adequately. We demonstrate here that alveolar macrophages secrete high amounts of TNF and IL-1 and also exhibit marked suppressive activity of fibroblast proliferation only in patients with stage II-III sarcoidosis. The high secretion of TNF-o and 1L-1 was not correlated to similar increases in PGE2 secretion. It should be noted that down-regulation of PGE2 secretion by alveolar macrophages from sarcoidosis patients has been reported already. 24 The high secretion of both cytokines seems to have a definitive suppressive role as the exogenous addition of TNFand IL-1 have a net suppressive effect on fibroblast proliferation within the range secreted by alveolar macrophages in sarcoidosis patients (1-10 ng/ml for TNF and 100-500 U/ml for IL-1).
Sarcoidosis is a multi-system granulornatous disease with a benign clinical course in the majority of the patients. The disease progresses from granulomatous inflammation to fibrosis only in about 10-20% of cases. 25'26 It is possible that the high secretion of TNF-o and IL-1 is involved in limiting the fibrotic response. In stage I-II sarcoidosis we showed high secretion of IL-1 only, which can have simultaneously stimulatory and inhibitory effects on fibroblast proliferation as already reported for osteoclasts. 27 It is of interest that indomethacin partially reversed the suppressive effect of TNFand IL-1. Apparently this effect is not due to the well-known property of indomethacin as a cyclooxygenase inhibitor but may reflect a direct effect of cytokines.
These studies show that cytokines can have multiple effects on fibroblast proliferation and that the effect that is noted depends on the entire set of regulatory factors affecting the target cell.