IL-1R1/MyD88 signaling and the inflammasome are essential in pulmonary inflammation and fibrosis in mice

Neutrophil and lymphocyte recruitment in the bronchoalveolar space are MyD88 and IL-1R1 dependent.

MyD88, a critical adaptor molecule for most members of the TLR/IL-1R family, has been reported to be involved in TLR-mediated acute lung inflammation (14, 31). To further analyze the implications of the different MyD88-dependent pathways in acute lung injury, we evaluated the airway inflammatory cell response by enumerating the cells recruited in the bronchoalveolar lavage fluid (BALF) after i.n. BLM administration. Total cell counts were increased at days 1, 7, and 11 upon BLM administration in C57BL/6 mice but were lower in MyD88^–/– and IL-1R1^–/– mice (Figure 1A). BLM elicited acute neutrophil recruitment in C57BL/6 mice within 24 hours, which persisted over 7 days and normalized on day 11 (Figure 1B) and was largely reduced in the absence of MyD88 and IL-1R1. Lymphocyte counts were augmented at days 7 and 11 in WT mice but were significantly lower in MyD88^–/– and IL-1R1^–/– mice (Figure 1C). Alveolar macrophages recovered in BALF were augmented at day 11 in WT, but not in MyD88^–/– or IL-1R1^–/– mice, although the difference did not reach statistical significance (Figure 1D). Therefore, BLM-induced neutrophil and lymphocyte influx into the bronchoalveolar space is dependent upon IL-1R1/MyD88 signaling.

BLM-induced neutrophil recruitment in the lung is TLR2, -3, -4, and -9 independent.

Since neutrophil recruitment was abrogated in the absence of MyD88, the common adaptor of TLRs, we investigated the role of individual TLRs in BLM-induced acute lung inflammation. Mice deficient for TLR2, TLR4, TLR9, as well as TLR3, the only MyD88-independent TLR, showed abundant neutrophil recruitment 24 hours after BLM administration, with no significant difference in comparison with WT mice (Supplemental Figure 1A; supplemental material available online with this article; doi:10.1172/JCI32285DS1). These data demonstrate that the TLRs tested are dispensable for MyD88-mediated acute neutrophil influx into bronchoalveolar space. To exclude a potential compensation of TLR2 and TLR4, we tested double TLR2/4-deficient mice and found no significant difference from C57BL/6 mice (Supplemental Figure 1B). Identical results were obtained in the absence of the common adaptor protein for TLR2 and -4, TIRAP, using TIRAP-deficient mice (data not shown). We further examined myeloperoxidase (MPO) activity (Supplemental Figure 1C), expression of neutrophil chemotactic factor keratinocyte-derived chemokine (KC) and IL-6 (Supplemental Figure 1, D and E), and expression of thymus activation–regulated chemokine (TARC), a lymphocyte chemotactic factor (Supplemental Figure 1F), in the lung after BLM-induced lung injury. No significant difference in lung MPO, KC, IL-6, or TARC responses between TLR2/4-deficient and C57BL/6 mice was found. Similar data were obtained for KC, IL-6, and TARC levels in BALF (data not shown). These data demonstrate that TLR-deficient mice, and in particular double TLR2/4-deficient mice, are not protected from BLM-induced lung inflammation, in contrast to MyD88-deficient mice. Therefore, the resistance of MyD88-deficient mice to BLM-induced lung injury appears to be TLR independent, leaving IL-1R–related signaling as a potential pathogenic pathway.

Reduced acute inflammation induced by BLM in IL-1R1– but not in IL-18R–deficient mice.

Acute stress and injury may activate the inflammasome promoting the processing and secretion of proinflammatory cytokines such as IL-1 and IL-18 (32, 33), and IL-1 is increased upon BLM injury (34). Since resistance of MyD88-deficient mice to BLM-induced lung injury was TLR independent, we studied further the implication of the IL-1R family in BLM-induced acute lung injury. In fact, IL-1R1–deficient mice display a dramatic reduction of neutrophil recruitment into the bronchoalveolar space at day 1 after BLM administration (Figure 2A). Pulmonary MPO activity at day 7 was elevated only in C57BL/6 mice but in neither IL-1R1– nor MyD88-deficient mice (Figure 2B). No neutrophils were found in either mouse group after saline administration.

Since IL-1 and TNF-α share proinflammatory properties, we investigated the participation of TNF-α, a cytokine that does not use the MyD88 adaptor protein for signaling, using TNF-α–deficient mice (35). The inflammatory response of TNF-α–deficient mice to BLM was not significantly different from that of C57BL/6 mice in terms of increased total cell influx into BALF, induced neutrophil influx into BALF and lung, or KC- and IL-6–enhanced production in lung, suggesting that TNF-α is dispensable for the inflammatory response to BLM (Supplemental Figure 2, A–E).

Further, IL-1β is produced in the lung upon BLM administration (Figure 2D), while IL-18 was not detectable by ELISA (data not shown). IL-1 and IL-18 signal through IL-1R1 and IL-18R, respectively, which belong to the IL-1 receptor subfamily and use the adaptor MyD88 protein (36). To exclude any contribution by the IL-18R pathway in the MyD88-dependent BLM-induced lung injury, we investigated the response of IL-18R^–/– mice. Unlike IL-1R1^–/– and MyD88^–/– mice, IL-18R^–/– mice responded with a strong neutrophil influx to BLM (Figure 2C). Therefore, IL-18R is dispensable for acute neutrophil influx into the bronchoalveolar space, showing that IL-1R1, but not IL-18R, signaling is critical in the MyD88-mediated pulmonary inflammatory response to BLM injury.

The induction of cytokines and chemokines involved in neutrophil recruitment was evaluated in BALF and lung homogenates of C57BL/6 WT and MyD88- and IL-1R1–deficient mice. IL-1β was augmented in the lung at 24 hours after BLM administration (Figure 2D) in C57BL/6 control and IL-1R1–deficient mice but not in MyD88-deficient mice, which displayed baseline levels. These data suggest that the production of IL-1β is also dependent on MyD88 pathways other than IL-1R1. Other IL-1–related family receptors, in particular ST2 (the IL-33 receptor) and IL-1Rrp2, also signal through MyD88 and could play a role in BLM-induced IL-1β production and inflammation (37, 38). MyD88 was also recently implicated in IFN-γR–induced inflammation (39).

The production of the neutrophil chemotactic factor KC (Figure 2E) and another inflammatory cytokine, IL-6 (Figure 2F), were induced in response to BLM in the lung of C57BL/6 mice but were dramatically reduced in MyD88^–/– and IL-1R1^–/– mice. Comparable results were obtained in the BALF (data not shown). Therefore, BLM-induced IL-1β secretion and IL-1R1 and MyD88 signaling are required for the production of KC and IL-6 and for neutrophil recruitment into the lung tissue and bronchoalveolar space.

Resident cells mediate BLM-induced acute neutrophil recruitment as well as IL-1, IL-6, and KC production into the lung.

Since IL-1R1/MyD88 is expressed on a broad range of cells, we next addressed the respective role of bone marrow–derived cells and lung resident cells in the BLM response using mixed bone marrow chimeric mice (31). C57BL/6 WT bone marrow reconstitution of MyD88^–/– mice (WT→KO) did not restore neutrophil recruitment to the bronchoalveolar lavage (BAL) in response to BLM (Figure 3A), while WT mice reconstituted with MyD88^–/– bone marrow (KO→WT) displayed a significant but reduced neutrophil influx in BAL (Figure 3A). Furthermore, WT bone marrow reconstitution of MyD88^–/– (WT→KO) did not restore neutrophil recruitment into the lung parenchyma as assessed by MPO activity in lung homogenate after BLM (Figure 3B). By contrast, WT mice reconstituted with MyD88^–/– bone marrow (KO→WT) had increased MPO activity, although it was reduced in comparison with irradiated and reconstituted WT mice (WT→WT) (Figure 3B). Therefore, neutrophil recruitment in the lung tissue and bronchoalveolar space in response to BLM largely depends on MyD88 signaling in radioresistant resident lung cells, but some contribution of hemopoietic cells is needed for optimal neutrophil influx.

We next determined the chemokine and cytokine production in lung homogenates from chimeric mice. WT bone marrow reconstitution of MyD88-deficient (WT→KO) mice after BLM administration did not restore IL-1β production in lung tissue (Figure 3C), which remained at the level of MyD88^–/– mice (KO→KO). In contrast, WT mice reconstituted with MyD88^–/– bone marrow (KO→WT) had IL-1β production in lung tissue comparable with that of WT controls (WT→WT) (Figure 3C), indicating that the radioresistant resident cells are responsible for most IL-1β production in response to BLM administration. Furthermore, WT bone marrow reconstitution of MyD88^–/– mice (WT→KO) after BLM administration did not restore KC or IL-6 production in lung tissue (Figure 3, D and E), which remained at the level of MyD88^–/– mice (KO→KO). WT mice reconstituted with MyD88^–/– bone marrow (KO→WT) had increased KC and IL-6 production in lungs, which was superior to that of MyD88^–/– mice but slightly lower than that of WT controls (WT→WT) (Figure 3, D and E). Altogether, the results from mixed bone marrow chimera suggest that BLM induced neutrophil recruitment in the lung and that IL-1β, KC, and IL-6 productions are mediated by IL-1R1/MyD88 signaling mainly by a radioresistant, nonhemopoietic cell type that cannot be restored by bone marrow cell reconstitution.

Remodeling process induced by BLM injury is IL-1R1 and MyD88 dependent.

We extended our analysis of BLM-induced injury and fibrosis by evaluating remodeling factors during the repair process. The gelatinase activities were assessed by the measurement of MMP-9 (gelatinase A) and MMP-2 (gelatinase B) by zymography. MMP-9 (99 kDa) is largely produced by neutrophils, and its activity is associated with neutrophil recruitment (40). MMP-9 activity was significantly upregulated in BALF at day 1 in C57BL/6 mice but not in MyD88^–/– mice, and only partially in IL-1R1^–/– mice (Figure 4A). These data suggest that the increase in MMP-9 may also be dependent on MyD88 pathways other than IL-1R1. MMP-9 activity returned to baseline level at day 11 (data not shown), whereas MMP-2 (76 kDa) activity was elevated only at day 11 in C57BL/6 mice but not in MyD88^–/– and IL-1R1^–/– mice (Figure 4B). Since the balance of tissue inhibitors of metalloproteinase (TIMPs) and MMPs is an important factor in the fibrotic process, we analyzed the production of TIMP-1. TIMP-1 activity was upregulated in BALF (data not shown) and lung homogenates of C57BL/6 and TLR2/4^–/–, but not MyD88^–/– or IL-1R1^–/–, mice within 24 hours (Figure 4, C and D). Therefore, the data suggest that IL-1R1/MyD88 signaling is required for MMP-9, MMP-2, and TIMP-1 activation and contributes to fibrosis.

Attenuation of BLM-induced lung fibrosis in IL-1R1– and MyD88-deficient mice.

The net result of BLM-induced tissue injury is chronic inflammation and progressive pulmonary fibrosis. Microscopic investigations at 7 days after BLM administration showed acute inflammatory reactions in the lungs and signs of fibrosis in C57BL/6 mice. The hallmarks of the pathology are severe alveolitis with abundant neutrophils in the alveoli, recruitment of mononuclear cells, and destruction and repair with thickening of the alveolar septae. On day 11, the lung sections also showed extensive fibrotic areas with abundant collagen production (Figure 5B), as compared with normal alveolar structure of the saline controls (Figure 5A). Cellular infiltrates, alveolar wall destruction, and collagen deposition were significantly reduced in IL-1R1^–/– (Figure 5D) and MyD88^–/– mice (Figure 5E), but not in TLR2/4^–/– mice (Figure 5C). The inflammatory and fibrotic morphological alterations as assessed semiquantitatively were significantly reduced in the absence of IL-1R1/MyD88 (data not shown). Moreover, biochemically significant collagen deposition in the lung tissue of C57BL/6 mice, but not IL1-R1^–/– (Figure 5F) or MyD88^–/– mice (data not shown), was found. TGF-β1, a critical mediator of remodeling and fibrotic responses in the lung, was evaluated at different time points after BLM administration. Latent TGF-β1 was detected after activation in BALF from WT mice 7 days after BLM administration, but was not detected in BALF from MyD88- or IL-1R1–deficient mice (Figure 5G), and persisted until day 14 in WT mice (53 pg/ml). Active TGF-β1 was undetectable in BALF from WT and deficient mice probably due to lower production (data not shown). Therefore, IL-1R1/MyD88 signaling is required for the development of pulmonary inflammation, fibrosis, and collagen deposition in response to BLM.

Exogenous IL-1β causes acute and late lung inflammation, remodeling, and fibrosis.

In order to confirm that IL-1β is central in BLM-induced acute inflammatory response, we first analyzed the effects of exogenous recombinant murine IL-1β (rmIL-1β). Mice treated with rmIL-1β presented a very transient and modest body weight loss in comparison with BLM-treated mice. A single i.n. rmIL-1β administration resulted in acute neutrophil recruitment into BALF at 24 hours in WT mice (Figure 6A), while other cell types were essentially not affected (data not shown). As expected, neutrophil recruitment in the BALF of IL-1R1– and MyD88-deficient mice was impaired after i.n. rmIL-1β, whereas it was dramatically increased in TLR2/4-, and TNF-α–deficient mice, suggesting that TLR2/4 signaling and TNF-α regulate inflammatory response to IL-1β (Figure 6A). Neutrophil influx into lung measured by lung MPO activity was impaired in IL-1R1– and MyD88-deficient mice after i.n. rmIL-1β but not in TLR2/4- and TNF-α–deficient mice (Figure 6B). Administration of rmIL-1β also enhanced the expression of neutrophil chemotactic factor CXCL1 (KC), IL-6, and TARC at 24 hours (Figure 6, C–E) in the lung of C57BL/6, TLR2/4^–/–, and TNF-α^–/– mice but not in IL-1R1^–/– or MyD88^–/– mice. Therefore, exogenous IL-1β induced acute inflammation with neutrophil influx, KC, IL-6, and TARC expression dependent on IL-1R1 and MyD88 but not on TLR2/4 or TNF-α receptor signaling, which resembles the pathology induced by BLM.

To further assess the role of IL-1β in the development of lung fibrosis, we analyzed the effect of exogenous rmIL-1β on the remodeling process. Indeed, 24 hours after IL-1β instillation, MMP-9 activity was upregulated in the BALF of C57BL/6, TLR2/4^–/–, and TNF-α^–/– mice but not MyD88^–/– or IL-1R1^–/– mice (Figure 6F). TIMP-1 activity, the hallmark of evolution to pulmonary fibrosis, was dramatically upregulated by rmIL-1β within 24 hours in lung homogenates of C57BL/6, TLR2/4^–/–, and TNF-α^–/– mice but not MyD88^–/– or IL-1R1^–/– mice (Figure 6G). Similar results were obtained in the BALF of these mice (data not shown). Therefore, exogenous IL-1β induced MMP-9 and TIMP-1 activation, which depended on IL-1R1/MyD88 but not TLR2/4 signaling or the presence of TNF-α. TIMP-1 production in WT mice was more transient after a single exogenous IL-1β administration than after BLM and was reduced at day 7 (data not shown). Moreover, we were unable to detect latent TGF-β1 in BALF from WT mice after i.n. rIL-1β at day 1, 7, and 14, whereas latent TGF-β1 was detected 7 days after BLM administration (Figure 5G) and was still present at day 14. A single i.n. rmIL-1β administration resulted in significant lymphocyte recruitment into BALF in WT mice at 7 days as seen with BLM (Figure 7A). Moreover, the pulmonary collagen deposition was increased by rmIL-1β, although to a lower level than after BLM treatment (Figure 7B). Histological analysis of the lung revealed that within 7 days of administration IL-1β caused tissue injury with marked tissue destruction, disruption of alveolar architecture, inflammation, and fibrosis (although less pronounced than that obtained after BLM treatment; Figure 7, D and E) that was absent in saline controls (Figure 7C) and in the lungs of IL-1R1^–/– and MyD88^–/– mice (data not shown). Further collagen accumulation was observed 14 days after rmIL-1β and after BLM administration (Figure 7, G and H), in comparison with saline treatment (Figure 7F). Therefore, rmIL-1β caused a clear but moderate pulmonary fibrosis, suggesting a critical role for endogenous IL-1 in BLM-induced inflammation and fibrosis.

Neutralization of IL-1 reduces acute lung inflammation and remodeling induced by BLM.

To confirm that IL-1β is a critical mediator of BLM-induced lung pathology, IL-1β neutralizing antibodies and IL-1Ra were given i.p. at the time of BLM administration. The IL-1β neutralizing antibodies reduced neutrophil recruitment in the BALF as well as KC and IL-6 production in the lung at day 1 (data not shown). IL-1Ra dramatically reduced neutrophil influx into BALF at day 1 (Figure 8A). Interestingly, IL-1Ra was even more effective than IL-1β neutralizing antibodies in inhibiting neutrophil recruitment in the BALF at day 1 (Figure 8A) and day 14 (Figure 8B). Total cells were also significantly reduced at days 1 and 14 by IL-1Ra (data not shown). IL-1β, KC, and IL-6 levels in lung were also inhibited by IL-1Ra at day 1 (Figure 8, C–E). Importantly, TIMP-1 was also dramatically reduced in lung (Figure 8F) and BALF (data not shown) after IL-1Ra treatment. Therefore, IL-1 is a key effector cytokine of the BLM-induced fibrotic lung pathology and may represent a critical therapeutic target.

BLM-induced acute inflammation is attenuated in ASC-deficient mice.

Acute stress and injury induced by BLM may activate the inflammasome promoting the processing and secretion of proinflammatory cytokines such as IL-1β. We showed that IL-1β is increased and essential in BLM-induced lung inflammation and fibrosis. Therefore, we hypothesized that the inflammasome may be critical in BLM-induced acute lung injury. In fact, mice deficient for the adaptor molecule ASC, shared by NALP1–3, display an important reduction of total cell influx (Figure 9A) and neutrophil recruitment (Figure 9B) into the bronchoalveolar space at day 1 after BLM administration. Pulmonary MPO activity at day 1 was elevated only in ASC^+/+ mice but was significantly reduced in ASC^–/– mice (Figure 9C). Interestingly, reduction of neutrophil influx was less important in ASC^–/– than in IL-1R1^–/– mice, suggesting that IL-1α may also play a role in BLM-induced inflammation. IL-1β production into the lung was attenuated in ASC^–/– in comparison with ASC^+/+ control littermates 24 hours after BLM administration (Figure 9D). Since both pro-IL-1β and IL-1β can be detected by this method, residual amount of cytokine measured in ASC^–/– mice might correspond to the pro-IL-1β form. IL-6 production into the lung 24 hours after BLM treatment was attenuated in ASC^–/– in comparison with ASC^+/+ control littermates (Figure 9E). Comparable reductions of IL-1β and IL-6 were obtained in the BALF (data not shown). Therefore, BLM-induced cellular inflammation and IL-1β and IL-6 productions are dependent upon the ASC inflammasome adaptor molecule, which is required for optimal lung inflammation in response to BLM.

Mice.

MyD88^–/– (77), IL-1R1^–/– (78), IL-18R^–/– (79), TLR4^–/– (80), TLR2^–/– (81), TLR2^–/–TLR4^–/–, TLR3^–/– (82), TLR9^–/– (83), TRIF^–/– (84), TNF-α^–/– (35), or ASC^–/– (75) mice were used. MyD88^–/–, TNF-α^–/–, TLR2^–/–, TLR4^–/–, and double-deficient TLR2^–/–TLR4^–/– mice were backcrossed 10 times on the WT C57BL/6 genetic background and 7 times for IL-1R1^–/– and TLR3^–/– mice. Littermate controls were used for ASC^–/– backcrossed only 3 times on the WT C57BL/6 genetic background. All mice, including control C57BL/6 (MyD88^+/+), were bred in our animal facility at the Transgenose Institute (CNRS, Orleans, France). For experiments, adult (6–10 weeks old) animals were kept in sterile, isolated ventilated cages. All animal experiments complied with the French Government’s ethical and animal experiment regulations and were approved by the Comité Régional d’Ethique pour l’Expérimentation Animale (Tours, France). All animals had access to water and food ad libitum before and after exposure.

Administration of BLM, IL-1β, IL-1 neutralizing antibody, or IL-1Ra.

BLM sulfate (300 μg or 15 mg/kg) from Bellon Laboratories in saline, mr IL-1β (0.25 μg or 1 μg) from R&D Systems in saline, or saline alone were given through the airways by nasal instillation in a volume of 40 μl under light ketamine-xylasine anesthesia. The IL-1β neutralizing antibody (provided by H. Gram Novartis Pharma, Basel, Switzerland) and the IL-1Ra (Amgen) were injected at 100 μg i.p. at the time of BLM administration.

BALF.

After incision of the trachea, a plastic cannula was inserted and airspaces were washed using 0.3 ml of PBS solution heated to 37°C. The rib cage was then gently massaged to enable maximum cell collection. The fluid was collected by careful aspiration. This procedure was performed 10 times, and the recovery of the total lavage exceeded 95%. The samples collected were dispatched in 2 fractions: the first (1 ml corresponding to the 2 first lavages) was used for mediator measurement, and the second for the cell determination (4 ml). The first fraction was centrifuged (600 g for 10 minutes), and supernatant was fractionated and kept at –80°C until mediator determination. The cell pellet was then resuspended in 0.4 ml PBS, pooled with the second fraction, and maintained at 4°C until cell determination.

Lung homogenization.

After BAL was performed, the whole lung was removed and placed inside a microtube (Lysing matrix D; Q Bio Gene) with 1 ml PBS, total lung tissue extract was prepared using a Fastprep system (FP120, Q Bio Gene), the extract was then centrifuged and the supernatant stored at –80°C before mediator measurement, MPO, or collagen assay with Sircol Collagen Assay (France Biochem Division).

MPO activity in lung.

Lung tissue MPO activity was evaluated as described (85). In brief, the right heart ventricle was perfused with saline to flush the vascular content, and lungs were frozen at –20°C until use. Lung was homogenized by polytron and centrifuged, and the supernatant was discarded. The pellets were resuspended in 1 ml PBS containing 0.5% hexadecyltrimethyl ammonium bromide (HTAB) and 5 mM EDTA. Following centrifugation, 50 μl of supernatants were placed in test tubes with 200 μl PBS-HTAB-EDTA, 2 ml Hanks’ balanced salt solution (HBSS), 100 μl o-dianisidine dihydrochloride (1.25 mg/ml), and 100 μl H₂O₂ 0.05%. After 15 minutes of incubation at 37°C in an agitator, the reaction was stopped with 100 μl NaN₃ 1%. The MPO activity was determined as absorbance at 460 nm against medium.

Cell count and determination.

Total cell count was determined in BALF using a particle counter (Z2 Coulter; Beckman Coulter). Differential cell counts were performed on cytospin preparations (Cytospin 3; Thermo Shandon) after staining for 4 minutes in May-Grünwald stain (MG-1L; Sigma-Aldrich) and 8 minutes in 95% Giemsa stain (GS-500; Sigma-Aldrich). Differential cell counts were made on 100 cells using standard morphological criteria.

Mediator measurements.

IL-1β, KC, IL-6, TARC, and TIMP-1 levels in BALF or lung homogenate were determined using ELISA assay kits (Mouse DuoSet for IL-1β, KC, IL-6, TARC, and TIMP-1; R&D Systems), except for the experiments described in Figures 2 and 3, according to manufacturer’s instructions. For experiments described in Figures 2 and 3, IL-1β, KC, and IL-6 were measured by multiplex ELISA cytokine arrays with a limit detection of 1 pg/ml (Upstate Ltd.). TGF-β1 levels in BALF were determined using an ELISA kit (Quantikine, mouse, rat, porcine, canine TGF-β1 Immunoassay; R&D Systems).

Bone marrow transplantation to obtain mixed chimera.

Recipient mice underwent a lethal total-body irradiation as reported before (86, 87). Fresh, unseparated bone marrow cells (2 × 10⁶ per mouse) were injected into the lateral tail vein of the irradiated recipient mice 24 hours after lethal irradiation. Blood was collected in EDTA-containing tubes at regular intervals, and the hematological parameters were determined with a Technikon H1E analyzer. The reconstituted mice were used 3 months after bone marrow transplantation. Further, in some experiments we used GFP-expressing donor cells (88), in order to control the extent of reconstitution, as previously described (31). The current protocol yielded more than 95% of blood circulating leukocytes of donor origin as assessed by green fluorescence of the peripheral blood using flow cytometry.

Zymographic analysis of MMPs.

MMP-2 and MMP-9 levels were determined by gelatin zymography. Briefly, nonreduced supernatant samples of BALF (15 μl) and standards (161-0305; Bio-Rad) were loaded onto 7% polyacrylamide gels (wt/vol) incorporating 0.1% (wt/vol) gelatin substrate. Proteins were subjected to electrophoresis at 20–30 mA for 3 hours. The gel was then washed twice in 2.5% Triton (vol/vol), rinsed 3 times quickly with distilled water, and placed in distilled water 3 times for 20 minutes. Each wash was performed using gentle stirring. Gels were incubated overnight at 37°C in 50 mM Tris buffer (containing 5 mM CaCl₂ and 2 μM ZnCl₂). Finally, gels were stained in Coomassie Blue and then destained progressively until bands of lysis (enzyme activity) in the gels appeared as regions of negative staining. The areas of lysis were analyzed using a densitometric analyzer (Bio-profil; Vilbert Lourmat), images were taken, and band densities were measured.

Lung collagen measurements.

Aliquots of lung homogenate (50 μl) were then assayed for lung collagen levels and compared with a standard curve prepared from bovine skin using the Sircol collagen dye binding assay according to the manufacturer’s instructions (Biocolor Ltd.).

Histology.

After BAL and lung perfusion, the large lobe was fixed in 4% buffered formaldehyde for standard microscopic analysis. Sections (3 μm) were stained with H&E or Chromotrope Aniline Blue as described previously (89). The severity of the morphological changes (infiltration by neutrophils and mononuclear cells and destruction and thickening of the alveolar septae and fibrosis) were assessed semiquantitatively using a score of 0–5 by 2 independent observers.

Statistics.

Statistical evaluation of differences between the experimental groups was determined by Mann-Whitney U test using Prism software. P values of less than 0.05 were considered statistically significant.