Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Pulmonary fibrosis can be caused by several known factors, such as inhaled dusts, radiation, and certain drugs. In other cases, the etiology of this disease is unknown (idiopathic pulmonary fibrosis [IPF]). The prognosis in cases of interstitial lung diseases is poor, and current therapies for these diseases are inadequate (1). Recent data obtained both from humans and from experimental animals have led to significant advances in understanding the mechanisms leading to pulmonary fibrosis. In particular, some mediators, such as cytokines, have been implicated not only in pulmonary inflammation characterized by the recruitment of inflammatory cells into the alveolar compartment, but also in the stimulation of fibroblast proliferation and collagen synthesis. Several types of inflammatory cells, such as mast cells (2), neutrophils (3), eosinophils (4), and lymphocytes (5), have been implicated in the development of lung fibrosis, but alveolar macrophages (AM) have generally been referred to as major effectors in the disease (6). Cytokines and growth factors produced by AM, such as tumor necrosis factor- (TNF-); interleukin (IL)-1, IL-6, and IL-8; and transforming growth factor- (TGF-); platelet-derived growth factor; and insulin-like growth factor, have been directly implicated in the cascade of events leading to the development of lung fibrosis (7, 8). In addition, several studies have shown a clear association in humans (9) and in different mouse models (10) between the presence of different subtypes of lymphocytes and interstitial lung disease. However, the exact role of lymphocytes in the fibrotic process remains largely in debate. In particular, some recent studies have shown conflicting results regarding the subtype of T-helper lymphocytes (Th1 or Th2) involved in both alveolitis and lung fibrosis (11, 12).

IL-12 is a key regulatory cytokine driving Th1 cell development, and is operative in the host defense against infection with parasites, viruses, and bacteria and in the rejection of tumor cells. However, IL-12 also mediates destructive inflammatory effects in endotoxic shock, and may contribute to the pathogenesis of autoimmune diseases such as multiple sclerosis, lupus, and autoimmune uveitis. No data are available concerning the possible participation of IL-12 and related proteins in lung inflammation and fibrosis. IL-12 is considered a powerful inducer of cell-mediated immunity; it triggers production of interferon- (IFN-) (a Th1 cytokine) by both T and natural killer (NK) cells during acute inflammation, stimulates their proliferation, and enhances T and NK cell-mediated cytolytic activity (13). IL-12 is a heterodimeric protein (p70) composed of a 40-kD (p40) and a 35-kD subunit (p35), each of which is encoded by a different, specific gene. Although the single p40 and p35 chains can be produced separately, only the p70 heterodimer has full biologic activity, and coexpression of both the p40 and p35 genes in the same cell is required for production of the biologically active p70 heterodimer. IL-12 (p70) exerts its biologic activities through specific, high-affinity receptors, which also consist of two subunits: IL-12R₁ and IL-12R₂. In addition, in vitro and in vivo experimental data have identified the p40 IL-12 subunit as an antagonist of the biologic activity of p70 IL-12. Therefore, depending on the assay system used, failure to distinguish between the production of p70 and p40 IL-12 could lead to overestimation of the amount of bioactive IL-12 produced, part of which may in fact represent the secretion of an IL-12 antagonist. The possibility that p40 exerts a physiologic effect in limiting the action of p70 IL-12 must especially be considered in interpreting the contribution of Th1 and Th2 cells to the pathogenesis of certain diseases (14).

Besides cytokine production, another signature of Th1 and Th2 populations is the pattern of their immunoglobulin (Ig)G isotype synthesis. In the mouse, Th1 cells induce the switching of B-lymphocytes to produce IgG_2a, whereas a Th2 environment chiefly induces IgG₁ expression. The study of these different isotypes may help delineate better the balance between Th1 and Th2 involvement in the development of disease (15).

To study the role of IL-12 and related proteins in the process leading to lung fibrosis, we have established a mouse model that, through the use of mineral particles with different biologic activities, may help clarify why, under some circumstances, the reaction of the lung leads to a resolution of pulmonary damage and a return to normal function, whereas in other situations there is progression to an inflammatory response and fibrosis. The intratracheal administration of tungsten carbide (WC) particles does not induce any pulmonary inflammation or fibrosis. MnO₂ particles induce an inflammatory reaction without subsequent fibrosis, and silica particles induce an inflammatory process leading to the development of lung fibrosis. Using this comparative model, we have assessed the participation of both p70 and p40 IL-12 and analyzed the levels of both IgG₁ and IgG_2a, as well as IFN- in the bronchoalveolar lavage fluid (BALF) of mice challenged with the various particulate components of the model. The present study showed that a fibrotic process in the mouse lung is specifically accompanied by protracted overproduction of the p40 subunit of IL-12, but not of p70 IL-12. The resolution of alveolitis induced by MnO₂ particles was associated with an upregulation of IgG_2a and IFN-, whereas progression to fibrosis induced by silica particles was associated with dominant levels of IgG₁.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Mice

Female NMRI mice weighing 20 to 25 g were purchased from Iffa Credo (Brussels, Belgium). The animals were housed in air-conditioned, positive-pressure units (25°C, 50% relative humidity) on a 12-h light/dark cycle. A total of 20 mice was used per experimental group and per time point (3, 15, 30, and 120 d); six mice were used for bronchoalveolar lavage (BAL); six for lung homogenates; four for histopathology; and four for messenger RNA (mRNA) analysis.

Instillation Method

WC (d₅₀ = 1 µm); MnO₂ (d₅₀ = 3.7 µm); crystalline silica (DQ12; d₅₀ = 2.2 µm), or saline (controls) was introduced directly into the lungs of mice by intratracheal instillation (2.5 mg/mouse). All instillations (100 µl/mouse) were performed on anesthetized animals (sodium pentobarbital, 2 mg/mouse, given intraperitoneally) after surgical opening of the neck. To allow their sterilization and the inactivation of any trace of endotoxin, particles were heated at 200°C for 4 h immediately before suspension and administration.

BAL and Cell Culture

At selected time intervals, treated mice were killed with sodium pentobarbital (20 mg/animal, given intraperitoneally), and BAL was performed by cannulating the trachea and infusing the lungs six times with volumes of 1.5 ml sterile NaCl 0.9%. The BALF was centrifuged (1,000 × g, 10 min, 4°C), and the cell-free supernatant of this first lavage fraction was used for biochemical measurements. BAL was continued with three additional volumes (1.5 ml each) of sterile 0.9% saline. The lavage fluids within each treatment group were centrifuged and the respective cell pellets were pooled. Aliquots of the cell suspensions were used to determine cell number and viability (trypan blue exclusion method). Cell differential counts were made on cytocentrifuge preparations fixed in methanol and stained with Diff-Quik (Dade, Brussels, Belgium; 200 cells counted). The cell suspensions were adjusted to a concentration of 0.5 × 10⁶ viable BALF cells/ml of RPMI-1640 (GIBCO BRL, Merelbeke, Belgium) containing lactalbumin hydrolysate (0.1%) and antibiotics (1%). Aliquots consisting of 1 ml of the cell suspensions were seeded into 24-well culture plates. BALF cell cultures were stimulated with 1 µg/ml of lipopolysaccharide (LPS); Escherichia coli 055:B5 (Sigma, Bornem, Belgium) for 18 h. The culture supernatants were centrifuged (1,000 × g, 10 min, 4°C) and kept at 80°C until analysis.

Biochemical Analyses

Lactate dehydrogenase (LDH) activity was assayed spectrophotometrically by monitoring the reduction of nicotinamide adenine dinucleotide at 340 nm in the presence of lactate. Total proteins (TP) were determined by the pyrogallol red staining method (Technicon RA system; Bayer Diagnostics, Domont, France).

Histopathology and Immunohistochemical Staining

The whole lung was excised and fixed in Bouin's solution (Merck-Belgolabo, Overyse, Belgium). Paraffin-embedded sections were stained with hematoxylin and eosin, Masson's trichrome, or toluidine blue for light-microscopic examination. For immunohistochemical studies, dewaxed and rehydrated tissue sections were subjected to endogenous peroxidase inactivation (0.5 % H₂O₂ for 20 min). After incubation with a monoclonal rat antimouse p40 IL-12 antibody (Genzyme, Cambridge, MA; clone C 17.8, 1:500 dilution in phosphate-buffered saline [PBS] supplemented with 0.1% albumin, 1 h at 37°C), antigen-antibody complexes were detected with the antirat avidin-biotin conjugate Perox kit (Vectastain Labs, DAKO Als, Glostrup, Denmark). Peroxidase activity was revealed by the benzidine peroxide substrate (3-3'-diaminobenzidine-tetrahydrochloride [Aldrich, Beerse, Belgium] and H₂O₂). Sections were counterstained with Mayer's hematoxylin, rinsed, and mounted (DPX 360294H; BDH, Poole, UK) for histologic examination.

Preparation of Lung Homogenates for Cytokine Measurements

The excised whole lung was placed into a Falcon (Vel, Leuven, Belgium) tube chilled on ice, and 3 ml of cold PBS (GIBCO BRL) was added. The content of each tube was then homogenized for 30 s with a Polytron PT1200 homogenizer (Kinematica AG, Lucerne, Switzerland). The tubes were centrifuged at 4°C at 2,000 rpm for 10 min and the supernatants were frozen at 80°C until use.

Cytokine Measurements

The concentrations of p40 and p70 IL-12 proteins in BALF, lung homogenates, and culture media were measured with cytokine-specific enzyme-linked immunosorbent assays (ELISAs) obtained from Biosource International (Camarillo, CA) and Genzyme, respectively. The detection limits of these ELISAs were 2 pg/ml and 5 pg/ ml, respectively. The p70 ELISA recognizes the protein through an anti-p40 capture antibody and anti-p35 detection antibody. IFN- was also measured, with an ELISA from Genzyme having a detection limit of 5 pg/ml.

Semiquantitative Polymerase Chain Reaction Analysis

RNA from whole lung and BAL cells obtained from saline or particle-treated animals were isolated by the Tryzol method (GIBCO BRL). Reverse transcription (RT) was performed on total RNA (2.5 µg) with an oligodeoxythymidine primer, and amplification (16, 17) was done for 25 and 40 polymerase chain reaction (PCR) cycles for -actin and for p35 and p40, respectively, with the following specific primers: for p35 IL-12, sense: 5'-AAGACATCACACGGGACCAAACCA-3'; antisense: 5'-CGCAGAGTCTCGCCATTATGATTC-3'. For p40 IL-12, sense: 5'-CCACTCACATCTGCTGCTCCACAAG-3' antisense: 5'-ACTTCTCATAGTCCCTTTGGTCCAG-3'. For -actin, sense: 5'-ATGGATGACGATATCGCTGC-3'; antisense: 5'-GCTGGAAGGTGGACAGTGAG-3'.

An aliquot of the PCR reaction product was electrophoresed in a 1% agarose gel stained with ethidium bromide. Gel analysis was done densitometrically with the Bio1D V 6.11d software system. Results were expressed as arbitrary units, which were calculated by integrating the intensity for each pixel over the spot area. To assure the identity of the PCR-amplified fragments, the size of each amplified gene fragment was compared with DNA standards (Standard XIV; Boehringer, Brussels, Belgium) electrophoresed on the same gel (not shown). As positive control, mRNA from the lungs of mice treated with 100 µg of E. coli LPS (IP) was used in all PCR assays (not shown).

IgG₁ and IgG_2a Determination

IgG subclass levels in BALF were measured through ELISA. Briefly, polystyrene plates (Grenier, Nurtingen, Germany) were coated overnight with affinity-purified goat antibodies to rabbit IgG, followed by rabbit antibodies specific for mouse IgG subclasses (18). After 2 h of incubation at 37°C with samples serially diluted in Tris-buffered saline (10 mM Tris, 10 mM merthiolate, 130 mM NaCl, pH 7.4) supplemented with 5% fetal calf serum, biotinylated monoclonal antibodies of the IgG₁ and IgG_2a subclasses were added for 2 h at 37°C. The fixation of labeled antibodies was measured with an avidin-peroxidase complex.

Statistics

Treatment-related differences were evaluated through one-way analysis of variance followed by pairwise comparison with the Newman-Keuls test. Values of P < 0.05 were considered statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Characterization of the Comparative Mouse Models

To characterize the biologic responses induced by the chemical particles used in this study, we measured the levels of LDH and proteins, and the number of inflammatory cells in the BALF at different time intervals (3, 15, 30, and 120 d after treatment). Changes in pulmonary architecture after particle treatment were assessed histopathologically after 120 d.

Biochemical and Cellular Parameters

Table 1 shows that instillation of WC particles did not induce significant changes in LDH, protein levels, or number of cells in BALF. In contrast, at 3 d and 15 d after treatment, MnO₂ led to a marked pulmonary inflammation characterized by a significant increase in LDH and protein levels, as well as an important recruitment of polymorphonuclear neutrophils (PMN) in the lung (73% of BAL cells after 3 d). This alveolitis was completely resolved after 30 d as demonstrated by inflammatory parameters that returned to control values. After 30 and 120 d, PMN represented only 2% and 3%, respectively, of the lavaged cells. Lymphocytes were also observed after treatment with MnO₂, mainly after 15 and 30 d. Instillation of silica particles induced a permanent inflammatory lung reaction, as shown by an increase in LDH, protein levels, and number of cells throughout the study period. Recruitment of PMN was the first cellular change in this reaction (50% of BAL cells after 3 d), followed by accumulation of lymphocytes (up to 23% of total cells) at 120 d.

Histopathology

One hundred and twenty days after exposure to WC, an accumulation of particles was observed in the lung parenchyma, without any structural modification. The histologic appearance of the lungs after WC treatment was similar to that of controls (Figure 1B). MnO₂ particles were cleared from the lung after 30 d. After 30 d and 120 d, no inflammatory or fibrotic reaction was observed in the lungs of animals treated with MnO₂ (Figure 1C). As shown by cellular analyses, silica particles induced a significant and permanent recruitment of PMN. At 120 d, clear silicotic nodules, characterized by an accumulation of mesenchymal cells and collagen deposition, were observed in the lungs of all animals treated with silica particles (Figure 1D). Silica particles persisted in the lung for up to 120 d after their administration.

View larger version (130K): [in this window] [in a new window]   Figure 1.   Representative lung sections 120 d after instillation of (A) saline, (B) WC, (C ) MnO2, and (D) crystalline silica particles in NMRI mice. Note the presence of fibrotic nodules only after silica treatment. Masson's trichrome stain; original magnification: ×100.

Overall, BAL and histopathologic analyses demonstrated the different nature of the pulmonary reactions induced by the three types of particles used in the study, which will hereafter be called noninflammatory (NI: WC), resolutive alveolitis (RA: MnO₂), and fibrosing alveolitis (FA: silica).

Levels of p40 and p70 IL-12 in BALF and Lung Tissue in the Different Models

Figure 2 shows that at 3 d after treatment, only the inflammatory reactions (RA and FA) were characterized by a significant increase in p40 IL-12 levels in BALF when compared with the control and NI models. This effect was transient in the RA model, with p40 IL-12 returning to control levels after 30 d. In contrast, in the FA model a persisting increase in p40 IL-12 was noted for up to 120 d. In lung tissue homogenates, a significant increase in p40 IL-12 was present at 3 d after treatment in all models. This overexpression of the p40 subunit was progressively attenuated in the NI and RA models, and returned to control values after 15 d and 120 d, respectively. As observed in BALF, increased levels of p40 IL-12 were maintained for up to 120 d in the FA model. No significant changes in p70 IL-12 could be demonstrated in any model, either in BALF or in lung tissue homogenates.

View larger version (26K): [in this window] [in a new window]   Figure 2.   Levels of p40 IL-12 in BALF (A) and lung tisue (B), and of p70 IL-12 in BALF (C ) and lung tissue (D) in control (saline), NI, RA, and FA models. Bars represent means ± SD for at least four animals. *P < 0.05 versus control group, Student-Newman-Keuls multiple comparison test.

Analysis of mRNAs Encoding the p40 and p35 Subunits of IL-12 at 4 mo after Treatment with Particles

The expression levels of the mRNAs for IL-12 and its subunits followed the same pattern as those for the respective proteins in the three models. A potent induction of expression of the mRNA for p40 IL-12 but not of that for p35 IL-12 was observed in the FA model. A slight increase in p35 and an upregulation of the p40 subunit was observed in the RA model. In the NI model, mRNA levels for the two IL-12 subunits did not differ from those of controls (Figure 3).

View larger version (31K): [in this window] [in a new window]   Figure 3.   Expression of p35 and p40 IL-12 mRNAs in lung homogenates (Day 120) from control (saline), NI, RA, and FA models. Shown are photographs of ethidium bromide-stained gels containing the RT-PCR products for p35 and p40 IL-12, and -actin (right panel ). Relative expression of p35 (A) and p40 (B), as quantified by densitometric scanning and expressed as a percent of the corresponding expression of -actin, is shown in the left panel.

Expression of p40 and p70 IL-12 Proteins and mRNAs in Explanted BAL Cells

To determine whether inflammatory cells recruited during alveolitis and/or fibrosis can participate in the production of IL-12, we cultured BAL cells and assessed their IL-12- subunit expression (Figure 4).

View larger version (24K): [in this window] [in a new window]   Figure 4.   Ex vivo release of p40 (A) and p70 (B) IL-12 by BALF cells obtained from control (saline), NI, RA, and FA models. Cells were cultured for 18 h without LPS stimulation, and levels of p40 and p70 protein were determined in the culture media. Relative expression is shown of p40 (C ) and p35 (D) IL-12 mRNAs by BALF cells freshly isolated from treated mice, as quantified by densitometric scanning and normalized for -actin expression.

BAL cells obtained from the NI, RA, and FA models after 3 d produced much more p40 IL-12 than did control cells. At the same time and in the same models, a similar increase was noted in p70 IL-12 synthesis, suggesting that the main subunit of IL-12 present in the supernatants of BAL cell cultures was the p70 subunit. Progressively, in cells obtained from the NI and RA models, p70 and p40 levels decreased and reached control levels after 30 d. In the FA model, a significant increase in the p40 and p70 subunits was still noted after 15 d and 30 d. Because the levels of the p40 subunit were more pronounced than those of p70 IL-12, the difference represented specific production of the p40 IL-12 subunit. At the end of the observation period, no clear induction of either subunit was observed. The same pattern of responses was observed in LPS-stimulated BAL cells (data not shown).

Expression of the mRNAs for the p40 and p35 subunits of IL-12 by BAL cells was increased in cells obtained from the NI model at Days 3 and 15. This induction returned to control values at later time points. In the same manner but more intensively, an upregulation of mRNA expression for both the p40 and p35 subunits was observed in cells obtained from the RA model when compared with controls. A rapid and marked overexpression of p40 subunit mRNA was observed in cells obtained from the FA model. This activation was downregulated as fibrosis progressed. In contrast to the findings for the RA model, a slight and constant upregulation of p35 mRNA was noted in explanted BAL cells from the FA model up to Day 120.

Cellular Localization of p40 IL-12 Production in the Lung

To determine the localization and cellular sources of the observed increase in p40 IL-12 in the FA model, we evaluated lung tissue sections obtained at Day 120 through immunohistochemistry. This analysis confirmed the presence of numerous cells producing p40 IL-12 in the FA model (Figure 5). p40 IL-12-positive cells were found exclusively in silicotic areas of the lung; unaffected zones were free of any detectable p40 IL-12-expressing cells (Figure 5). Interstitial macrophages were the major type of cells expressing the p40 IL-12 protein. Using polarized light and high magnification, we found that only macrophages that had phagocytosed silica particles expressed p40 IL-12 (Figure 5). Lymphocytes did not express p40 IL-12.

View larger version (82K): [in this window] [in a new window]   Figure 5.   (A) Immunohistochemical analysis of lung p40 IL-12 in the FA model at Day 120. Original magnification: ×400. (B) Positive cells are specifically confined in the silicotic nodules and contain silica particles. Original magnification: ×400, polarized light. (C ) p40 IL-12-positive macrophages activated by silica particles (arrows). Original magnification: ×1,000.

IgG₁, IgG_2a, and IFN- Levels in BALF in the Different Models

At all times at which they were analyzed, levels of IgG₁ in BALF were not different in the NI model than in the control group (Figure 6). A significant increase in IgG₁ was observed in the RA model at Days 3 and 15 as compared with controls. After that, levels of IgG₁ in BALF from the RA model were progressively reduced. In contrast, the FA model was associated with an increase in IgG₁ levels, which progressed with time to reach a maximum at Day 120. 

View larger version (17K): [in this window] [in a new window]   Figure 6.   IgG1 (A) and IgG2a (B) levels in BALF from control (saline), NI, RA, and FA models. Bars represent means ± SD for at least four animals. *P < 0.05 versus control group, Student- Newman-Keuls multiple comparison test.

At all times of analysis, no effect on IgG_2a production was observed in the NI model as compared with control values. At Day 3, a strong effect on IgG_2a synthesis was found in both inflammatory models. This induction of IgG_2a production was proportionally more important than that of IgG₁. At later time points, a constant stimulation of IgG_2a was noted in the FA model up to Day 120, but this was proportionally less important than stimulation of IgG₁. In contrast, a high level of expression of IgG_2a was specifically observed in the RA model on Day 15; this was significantly different from that in the FA model and proportionally more important than the production of IgG₁. On Days 30 and 120, levels of IgG_2a in BALF from the RA model had returned to control levels.

Significant levels of IFN- were observed only in BALF from the RA model at Days 15 and 120 (6.5 ± 1.6 pg/ml and 7.9 ± 0.6 pg/ml, respectively). IFN- was not detected in BALF from the FA or NI models or from controls.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

It is well established both in the mouse and in human systems that CD4⁺ T cells can be divided in two subsets, designated Th1 and Th2. These subsets are defined by the profile of cytokines that they secrete. Typically, Th1 cells are associated with the production of IFN-, IL-2, and TNF-, whereas Th2 cells characteristically produce a panel of cytokines including IL-4, IL-5, IL-6, IL-10, and IL-13. Th1 responses predominantly mediate cellular immunity, and are accompanied in the mouse by the synthesis of IgG_2a immunoglobulins, whereas Th2 cells are responsible for allergic and humoral reactions, which are characterized by high IgG₁ and IgE levels and by the presence of eosinophilia and mastocytosis. IL-12 and IL-4 are the key factors mediating the development of Th1 and Th2 cells, respectively. Clinical data and experimental models clearly indicate the important role of T-helper cell dichotomy in the pathogenesis of diseases such as leishmaniasis, protozoal infections, and autoimmune diseases. The resistance or susceptibility of the host to a disease depends strongly on the Th1 or Th2 type of immune response (15). However, for a large number of pathologic conditions, including fibrosing alveolitis, the relative Th1 versus Th2 contribution is not clearly defined or is still debated.

It is known that inhalation of certain inorganic dusts can induce an irreversible inflammatory and fibrotic lung response, or can be associated with a reversible pulmonary inflammation, whereas other particles can be considered innocuous (19). We have taken advantage of these different biologic responses to establish a mouse model that allows delineation of the changes that are specifically associated with a fibrogenic lung response. The involvement of IL-12, which is the key cytokine driving Th1-cell development, was investigated in this system.

Although instillation of silica resulted in a significant and persisting upregulation of mRNA expression for p40 IL-12 and of p40 protein synthesis in vivo, no induction of an identical amount of secreted p70 protein or of p35 IL-12 mRNA was observed during this fibrotic reaction. Recent in vitro and in vivo data (20) have shown that p40 IL-12 can antagonize the biologic activity of p70 IL-12 via competition for its receptor. Because this antagonism can suppress Th1-mediated immune responses, probably by limiting IFN- expression (21), some authors support the view that expression of p40 IL-12 is a mechanism for restoring inflammatory homeostasis (22). Interestingly, p40 IL-12 was shown to induce a Th2 cytokine response in a model of cardiac allograft rejection (23), confirming the opposite activities of p70 and p40 IL-12. Therefore, the increased production of p40 IL-12 observed in our model of fibrosing alveolitis may conceivably represent the secretion of an endogenous antagonist of p70 IL-12 that exerts a Th2-promoting activity. That the protracted production of p40 IL-12 led to a Th2-dominant environment in our study was confirmed by the measured levels of IgG₁ in BALF. In the fibrotic model, IgG₁ levels were significantly increased and, at 4 mo after treatment, were twice the levels of IgG_2a. In contrast, in a model of RA induced by the administration of MnO₂ particles, only a transient upregulation of p40 protein and IgG₁ was observed at Day 15. Interestingly, the downregulation of the early Th2 response observed in the RA model was accompanied by a strong and transient overproduction of IgG_2a at Day 15, as well as by high levels of IFN-, which reflect Th1 stimulation. No effect on BALF IL-12 or IgG levels was observed in the NI model produced with WC particles.

A number of experimental and clinical studies have examined Th1/Th2 profiles in fibrotic diseases. In vitro experiments have investigated the effects of IFN- and IL-4 on the collagen synthesis of murine lung fibroblast subsets. The results have shown that IL-4 (Th2) increases collagen expression in fibroblast cultures, whereas IFN- (Th1) significantly reduces collagen production (24). Animal experiments have shown that lung fibrosis induced by irradiation or bleomycin is associated with an overproduction of Th2 cytokines, such as IL-4 (12) or IL-5 (25), respectively. In addition, pulmonary treatment with adenovirus constructs containing genes for IL-4 or active TGF- (also considered a Th2-promoting cytokine) was associated with marked progression of granulomas and fibrotic lung reaction, respectively (26, 27). The pulmonary response induced by silica particles was found to be accompanied by increased expression of IL-10, and IL-10-deficient mice have shown a significant reduction in magnitude of the silicotic reaction as compared with their wild-type counterparts, which supports a profibrotic activity of IL-10, a Th2 cytokine (28). Wahl and colleagues (29), using a mouse model of liver fibrosis induced by schistosomes, have also reported that the fibrogenic process was accompanied by Th2 cytokines, and that neutralizing Th2 cytokines with antibodies could alleviate the fibrotic reaction. Interestingly, in a pulmonary model of fibrosis induced by bleomycin, treatment with IFN- limited the expansion of lung fibrosis, supporting the idea that a Th1-like response can control excessive collagen synthesis and the fibrotic process (30). On the other hand, Sharma and colleagues (11), working with a mouse model of bleomycin-induced lung fibrosis, have reported opposite results in terms of the Th1/ Th2 response pattern. They found that after intratracheal administration of bleomycin, IL-2 and IFN- production by lung lymphocytes was significantly increased as compared with that in saline-treated animals, whereas IL-4 synthesis in treated mice was decreased. These conflicting results could conceivably be explained by the use of different mouse strains, which is a key factor in the development of Th1 and Th2 responses (15), or by differences in the stage of disease analyzed.

To the best of our knowledge, no human data are available on levels of IL-12 and its subunits in fibrosing alveolitis. Several studies have found that Th2 cytokines such as IL-4, IL-5, and IL-10 were predominantly present in interstitial lung diseases, crediting a Th2-predominant profile in human pulmonary fibrosis (31, 32). However, another human study was unable to show a clear Th1 or Th2 dichotomy in IPF, because IL-5 but also IL-2 and IFN- were found to be significantly increased in IPF patients, which may suggest that clinical situations are more complex than experimental ones (33). Interestingly, in humans, impaired IFN- release by pulmonary lymphocytes was considered a potentiating factor in the pathogenesis of fibrosing lung disease (34).

In silicosis (35) and IPF (36), significantly higher levels of IgG were found in the blood, but homologous subclasses of IgG were not specified. Such fibrotic conditions are also accompanied by an overproduction of IgA and IgM, and in certain studies by IgE (37). In addition to this hypergammaglobulinemia, findings have confirmed the presence of additional humoral immunologic parameters, such as autoantibodies, rheumatoid factor, and immune complexes in silicosis (35). Altogether, these data demonstrate that patients with fibrotic lung reactions have increased humoral immunity, which is consistent with the notion that the fibrogenic process is accompanied by a Th2-predominant cell response.

Inflammatory cells harvested from the FA model in our study produced significant amounts of p40 IL-12 mRNA and protein at Days 15 and 30 in vitro, suggesting a contribution of these cells in the overproduction of p40 IL-12 observed in BALF and lung homogenates. At Day 120 (Figure 4), p40 synthesis by BAL cells from the FA model returned to control values, whereas p40 contents in BALF and lung homogenates were still upregulated. These apparently contradictory data are explained by the failure of cells harvested by BAL to reflect the population of inflammatory cells present in the interstitium of alveolar walls and silicotic nodules (38). Because p40 IL-12 was mainly expressed by interstitial macrophages trapped in silicotic nodules, as demonstrated by immunohistochemistry, these cells could not be recovered by the lavage procedure and were not represented in the cultures examined at Day 120.

BAL cells obtained from the FA model expressed more p35 IL-12 mRNA than did cells from the control group at each time of analysis. This upregulation of p35 mRNA was not accompanied by an increased production of p70 IL-12 protein. This might be explained by an absence of translation of p35 mRNA. Such post-transcriptional inhibition is well documented for other cytokines, such as TNF- (39), and may represent an additional mechanism for limiting the biologic activity of cytokines.

Additionally, during the acute phase following particle exposure (Day 3), we noted a significant increase in p70 IL-12 in culture supernatants of BAL cells not only in the FA model but also in the RA and NI models. This nonspecific activation of p70 synthesis by particles probably represented the stimulation of dust phagocytosis by macrophages and neutrophils recruited into the lung (40).

These findings for BAL cell expression of IL-12 and related proteins support the idea that BAL cell cultures can provide interesting data but may not always represent the global expression of a studied cytokine in the lung. The interpretation of such data must therefore take into account both mRNA and protein levels of the cytokine, as well as in vivo (e.g., in BALF) measurements of cytokines.

In conclusion, our observations show that the fibrosing alveolitis induced by silica particles in NMRI mice is accompanied by a Th2 pattern of immune response characterized by a protracted overproduction of p40 IL-12 and IgG₁ in the lung. In the RA model induced by administration of MnO₂ particles, the p40 response was only transient, and a Th1 response was identified from IFN- and dominant levels of IgG_2a in BALF. Further studies will be needed to determine whether Th2 cytokines are actively involved in or simply accompany the fibrotic reaction. A better clarification of this process may be particularly important in understanding the basis of the apparent association between pulmonary fibrotic diseases (silicosis) and manifestations of Th2-driven effects in such diseases, including exaggerated humoral immunity.