Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Idiopathic pulmonary fibrosis (IPF) is a chronic, usually fatal disorder characterized by dyspnea and impaired oxygen transfer, alveolar collapse, and interstitial and intraalveolar fibrosis (1). These changes result from injury to the cellular components of the alveolar walls and the accumulation of mesenchymal cells and their connective tissue products within alveolar walls and alveolar air spaces (2, 3). The occurrence of intraluminal fibrosis in the pathogenesis of IPF relates to an important concept: There are defects in the alveolar epithelium lining and basement membrane through which mesenchymal cells migrate from the interstitium into the intraluminal compartment (4).

Several lines of evidence support the concept that alveolitis is tightly associated with fibrotic lung disease (3, 5- 7). Alveolar macrophages may play a central role in this process by releasing mediators that amplify the inflammation by recruiting other inflammatory cells such as mast cells, neutrophils, and lymphocytes (8). Together, the mass of activated inflammatory cells in the lung tissue causes progressive injury and stimulates mesenchymal cells to accumulate. Because the features of IPF are damage to alveolar epithelium and breaks in alveolar basement membranes (9), it is reasonable to hypothesize that matrix metalloproteinases (MMPs) play a major role in the pathogenesis of IPF.

MMPs are a family of extracellular matrix-degrading, zinc-dependent enzymes comprising at least 18 members with different, albeit overlapping, substrate specificities. MMPs are secreted in latent form. During activation by other enzymes, including serine proteinases, and by autocatalytic cleavage, their propeptides are cleaved, converting the enzyme to lower molecular forms. Catalytic activity is specifically inhibited by the tissue inhibitors of metalloproteinases (TIMPs) (10, 11). Among the MMPs produced by the alveolar macrophages, MMP-9, also named gelatinase 92 kD or gelatinase B (12, 13), possesses the capacity to degrade preferentially collagen type IV and entactin (14) (components of alveolocapillary basement membranes), collagen type VII, anchoring fibrils connecting collagen type IV and interstitial collagens, as well as collagen type V (15, 16) in interaction with interstitial fibrillar collagens. Also, gelatinase B degrades parenchymal elastin (17).

In view of these data, the present study was undertaken to evaluate the expression of gelatinase B in IPF by studying cultured AMs and bronchoalveolar lavage fluid (BALF) from untreated patients with IPF, patients with IPF under treatment with steroid and immunosuppressive agents, and control subjects. The results showed that (1) gelatinase B expression is increased in AM in culture from IPF tissue both at the cellular and extracellular level and in BALF from patients with IPF; and that (2) treatment with steroid and immunosuppressive agents normalizes the gelatinase B expression by AM from patients with IPF.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Study Population

Seventeen consecutive patients (women and men, mean age 59 ± 3 yr, three former smokers and 14 nonsmokers) with IPF were included in this study. The diagnosis of IPF was based on the presence of progressive breathlessness, bilateral dry crackles, widespread bilateral infiltrates on chest roentgenogram with honeycombing, and pulmonary function tests showing a reduced total lung capacity, reduced forced vital capacity, decreased single-breath carbon monoxide diffusion capacity, and resting hypoxemia or exercise-induced hypoxemia. Patients with a collagen-vascular disorder or with a clinical history of environmental exposure, hypersensitivity pneumonitis, drug-induced pulmonary disease, or chronic pulmonary infection were excluded. Histologic confirmation of the diagnosis was obtained before treatment in 14 out of 17 patients by video-assisted thoracoscopic lung biopsy. All biopsy specimens showed varying degrees of interstitial fibrosis. Interstitial and intraalveolar cellular infiltration were present in most, consisting of macrophages, lymphocytes, neutrophils, and occasionally eosinophils or histiocytes. None of the specimens had granulomas, significant inorganic material found by polarized light microscopy, infection, or malignancy.

Twelve patients with IPF were untreated at the time of the study; five were treated with steroids (prednisone 0.5 mg/kg/d) associated with cyclophosphamide (2 mg/kg/d) or azathioprine (2 mg/kg/d). Sequential evaluation was performed in four patients who were untreated at the time of the first evaluation and treated with immunosuppressive drugs at the time of the second evaluation.

Follow-up evaluation included detailed history of pulmonary symptoms and breathlessness, physical examination, posteroanterior and lateral chest roentgenogram, and pulmonary function tests. Ten nonsmoking healthy subjects were included in the study as a control group. Oral informed consent was obtained from all patients.

Bronchoalveolar Lavage

Bronchoalveolar lavage (BAL) was performed after premedication with atropine under local anesthesia with lignocaine using a wedged fiberoptic bronchoscope using 250 ml of sterile saline solution in aliquots of 5 × 50. The aspirated fluid was collected into sterile siliconized containers and transported on ice immediately to the laboratory. The different aliquots were pooled.

Cell Isolation and AM Cultures

BAL was filtered through sterile surgical gauze and separated into its fluid and the cellular components by centrifugation at 400 × g for 10 min at 4°C. After one washing, the pellet was resuspended in RPMI 1640 in the presence of 2 mM L-glutamine, penicillin and streptomycin, and neomycin (Life Technologies, Eragny, France), the total and differential cell counts performed and the cells adjusted at a cell concentration of 1 × 10⁶ AM/ml. The differential cell counts were determined on cytocentrifuged smears stained with May Grünwald Giemsa (Sigma Co., St. Louis, MO). Cells were allowed to adhere to plastic culture dishes (2 ml in a 35-mm diameter well) for 2 h at 37°C. The nonadherent cells were removed by three washings with RPMI. Adherent cells contained more than 95% AM and less than 1% lymphocytes. The viability of the AM was evaluated by trypan blue exclusion and was never less than 90%. AMs were activated by addition of phorbol myristate acetate (100 nM PMA) (Sigma Co., St. Louis, MO). Supernatants from stimulated and unstimulated cells were collected after 24 h culture, centrifuged, and frozen in aliquots at 70°C. Adherent cells were washed with RPMI and lysed with 2 ml of 50 nM Tris-HCl buffer (pH 7.4) containing 0.1% Triton X-100. After centrifugation, cell lysates were frozen in aliquots at 80°C.

The lavage fluid volumes were normalized to the volume of epithelial lining fluid (ELF) by using the urea method (18).

Aliquots of BALF from patients with IPF or from healthy patients were dialyzed against distilled water for 24 h at 4°C, lyophilized, reconstituted in distilled water (1/ 100 of initial volume), aliquoted, and stored at 80°C.

Urea and Albumin Assays

Urea in concentrated BALF was determined by spectrophotometric absorbance at 600 nm using Berthelot's reaction (Boehringer Mannheim, Mannheim, Germany). Albumin in concentrated BALF was determined by spectrophotometric absorbance at 628 nm by using Bromocresol green (Sigma).

Protease Zymography

AM supernatant, AM cell lysate, and BALF underwent sodium dodecyl sulfate/polyacrylamide gel electrophoresis (SDS-PAGE) in polyacrylamide gels containing 1 mg/ml gelatin, under nonreducing conditions. After electrophoresis, gels were washed twice in 2.5% Triton X-100 for 1 h, rinsed briefly, and incubated at 37°C for 48 h in a buffer containing 100 mM Tris-HCl (pH 7.4) and 10 mM CaCl₂. After incubation, the gels were stained with the Coomassie Brilliant Blue R250 (Sigma) and destained in a solution of 7.5% acetic acid and 5% methanol. Zones of enzymatic activity were evident as clear bands against a blue background.

To determine the inhibition profile of the protease activity found in the samples, zymograms were also incubated in the above Tris-HCl buffer containing one of the following inhibitors: ethylenediaminetetraacetic acid (EDTA) (10 mM), N-ethylmaleimide (NEM) (2 mM), or phenylmethylsulfonyl fluoride (PMSF) (2 mM).

Gelatinolytic activities in the gel slabs were quantified using a semiautomated image analysis program (National Institutes of Health image 1.52), which quantifies both the surface and intensity of lysis bands after scanning of the gels. Results were expressed in arbitrary units: AU/48 h/ 10⁶ cells. To check that this method for measuring enzymatic activity on zymograms was linear over the range of activities in unknown samples, we evaluated activities for increasing volumes of culture medium and found that arbitrary units obtained with the image analysis system increased linearly with volume of the samples (r = 1.00) (19).

Reverse Zymography

Gelatinase-inhibitory activity in conditioned media was detected using reverse zymography (19). Briefly, aliquots of samples were resolved by 11.5% SDS-PAGE in the presence of 1 mg/ml gelatin. The standard zymographic method was modified after the removal of SDS from the gel by incubating the gel for 1 h at 37°C in conditioned medium from phorbol myristate acetate (PMA)-activated rabbit skin fibroblasts that contained activated gelatinases. Then the gel was incubated in the reaction buffer as described earlier and stained/destained as in standard zymography. Protection of the gelatin in the gel by the presence of TIMPs led to the appearance of relatively dark bands against a lighter background. Recombinant TIMP-1 (Valbiotech, France) was used as a reference.

Gelatinase Assays on Radiolabeled Gelatin

Free gelatinase activity was assayed using gelatin radiolabeled with [³H]acetic anhydride according to Cowston and Barett (20). Aliquots of AM-collected supernatants and BALF were tested with or without 1 mM aminophenylmercuric acetate (APMA) (incubation at 37°C for 2 h) in the presence of 2 mM NEM and PMSF. The proteolytic reaction was allowed to proceed for 48 h at 37°C and pH 7.4 in the presence of toluene to prevent bacterial contamination, and gelatinase assays were performed as previously described (24).

Western Blot Analysis

Aliquots of AM collected supernatants and AM cell lysates were separated by SDS-PAGE and transferred to an immobilon-P filter (polyvinylidene difluoride, 0.45 µm). Nonspecific staining was blocked by incubating the transfers for 90 min in Tris-buffered saline (TBS) containing 5% nonfat dry milk. The transfers were then incubated with rabbit polyclonal antiserum against 92 kD gelatinase (Valbiotech) diluted 1:500 in TBS. The blots were washed three times in TBS, 0.05% Tween 20 and incubated for 90 min with biotinylated goat antirabbit immunoglobulin G (IgG) diluted 1:1,000 as the secondary antibody. The blots were visualized using alkaline phosphatase and Fast fast red TR/naphthol AS-MX (Sigma).

Macrophage 92 kD Gelatinase Immunolocalization

Immunocytochemical studies. BALF freshly harvested cells were cytocentrifuged (10⁵ cells in RPMI 1640 per smear), fixed for 5 min in 4% formaldehyde and permeabilized or not with 0.1% Triton X-100 in PBS for 5 min. The cells were incubated for 2 h at room temperature with antihuman gelatinase B developed in rabbit affinity isolated antibodies (Valbiotech), diluted 1:50 in phosphate-buffered saline (PBS) containing nonimmune sheep serum and 0.02% NaN₃. After three washes, the second antibody (sheep antirabbit fluorescein isothiocyanate; Valbiotech) diluted in PBS was added and incubated for 1 h. Slides were mounted in Citifluor (UKC Chemical Laboratory, Canterbury, UK) and observed under a confocal microscope (LSM 410; Zeiss, Jena, Germany). Photographs were taken with Agfachrome 1000 ASA film uprated to 2000 ASA.

Immunohistochemical studies. Paraffin-embedded pulmonary tissue blocks from two IPF patients and two patients without lung disease were acquired from the surgical pathology archives of Hospital A. Calmette (Lille, France). Tissues had been fixed in 10% buffered formalin and routinely processed. First, antigen retrieval was performed in 1 mM EDTA (pH 8) and processed in a microwave oven for three cycles as previously described. Immunolocalization of gelatinase B was determined using antigelatinase B mouse monoclonal antibody (Oncogene Research Products, Calbiochem, Meudon, France), for 1 h at room temperature. IgG1k was used as a control antibody. Immunolocalization of gelatinase B was detected by alkaline phosphatase antialkaline phosphatase technique (Dako, Trappes, France). The enzyme was visualized with a Fast Red substrate system (Dako).

RNA Extraction

Total RNA (RNA_T) was extracted from AMs using Trizol reagent (Life Technologies). RNA_T was quantified at 260/280 nm, and the integrity of the samples was checked by 1.5% agarose gel electrophoresis. Reproducibly, 5 to 10 µg of RNA_T were obtained from 10⁶ × cells, and aliquots were stored in sterile microcentrifuge tubes at 80°C.

Quantitative Reverse Transcription/Polymerase Chain Reaction for Evaluation of 92 kD Gelatinase mRNA

Primer design and synthesis. Sense and antisense primers for human 92 kD gelatinase were designed as previously described (24): 92 kD gelatinase primers: sense, 5' GTGCTGGGCTGCTGCTTTGCTG3'; antisense, 5' GTCGCCCTCAAAGGTTTGGAAT3'.

Reverse transcriptase step. Reverse transcription was carried out using RNA_T as previously described (22). Briefly, to a final volume of 25 µl, the following compounds were added: 3 µl of 10× polymerase chain reaction (PCR) buffer (200 mM Tris-HCl, pH 8.3, 500 mM KCl, 15 mM MgCl₂, 1 mg/ml gelatin), 10 µl dilution buffer for RNA (1 M Tris pH 8.3 10 µl, 0.1 M DTT 20 µl, RNAsine 1 µl, bovine serum albumin 100 µl, H₂O 870 µl), 10 µl RNA_T (10 ng RNA_T) obtained from cultured human bronchial epithelial cells or AMs, and 2 µl of downstream primer (10 pmol). After heating for 2 min at 80°C in the thermocycler to break up secondary structures, the tubes were equilibrated at 42°C. Each sample was supplemented with 25 µl of reverse transcriptase (RT) mix containing 2.5 µl of 10× PCR buffer, 1.25 mM of each dNTP 16 µl, 100 mM MgCl₂ 1.5 µl, and 100 mM dithiothreitol (DTT) 4 µl with or without 200 U of Moloney Murine Leukemia Virus reverse transcriptase (Life Technologies). The final volume was 50 µl. The RT reaction lasted 45 min and was carried out at 42°C to prevent excessive mispriming and possible RNA refolding. After completion of RT, the temperature was raised to 96°C for 30 s to inactivate the enzyme and denature the RNA-DNA hybrid. The temperature was then equilibrated at 80°C.

Quantitative PCR. Quantitative RT-PCR assays were previously described (22) and required the availability of a specific internal DNA standard corresponding to the 92 kD gelatinase RNA_T target.

Statistical Analysis

Values were expressed as means ± standard errors of the mean (SEM). Comparison of untreated IPF, treated IPF, and healthy patients were performed using the nonparametric Mann-Whitney U test.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Cellularity

The total cell number recovered by BAL was increased 3-fold in the IPF patients compared to healthy controls (Table 1). However, AMs remained the predominant cells in the three groups. In untreated IPF patients, the percentages of neutrophils, eosinophils, and lymphocytes were three to four times more elevated than those observed in healthy patients. Steroid and immunosuppressive treatment appeared to reduce the number of inflammatory cells other than AMs, particularly the lymphocyte number.

Variation of ELF and Albumin

ELF volumes in BAL of untreated and treated IPF patients (1.7 ± 0.6 and 1.3 ± 0.2 ml/100 ml, respectively) were not significantly different compared to healthy controls (1.5 ± 0.3). Also, the albumin concentration in ELF was equivalent between the three (0.32 ± 0.06 µg/ml ELF in treated and untreated IPF patients versus 0.41 ± 0.05 µg/µl ELF in healthy controls), indicating similar alveolocapillary permeability in IPF patients compared with healthy controls.

Zymography

Zymography on SDS-gelatin was used to determine the levels of gelatinase activity released into AM culture media (Figures 1 and 2, top panel ) or associated with AM lysates (Figures 1 and 2, bottom panel ). The level of secreted 92 kD gelatinase proform was barely detectable in AM culture medium from healthy control subjects. By contrast, gelatinase B was prominent in AM culture media from untreated IPF subjects compared with control subjects (4.1 ± 1.7 versus 0.3 ± 0.2 10⁵ AU/10⁴ AM for the 92-kD proform). Gelatinase activity tended to normalize in treated IPF patients (Figures 1 and 2, top panel ). The presence of 100 nM PMA in culture media induced only a discrete potentiation of 92 kD gelatinase production by AM, whatever the patient group. The level of 88 kD gelatinase active form was scarcely detectable in AM culture medium from any patient, with or without the presence of PMA.

View larger version (48K): [in this window] [in a new window]   Figure 1.   Gelatinase production by cultured AMs and gelatinase secretion in BALF. Evaluation was carried out by zymography. Cultured AMs were incubated for 24 h in the presence of 2 mM L-glutamine and in the absence of fetal calf serum. Each culture medium or AM lysate (10 µl corresponding to 104 × AM), as well as 10 µl of BALF, were subjected to SDS-PAGE/gelatin electrophoresis. Lanes 1, 6, and 11: AM culture media from treated, IPF patient (IPF/T), untreated IPF patient, and healthy subject (control), respectively. Lanes 2, 7, and 12: the same in presence of PMA during incubation time. Lanes 3, 8, and 13: AM lysates from treated-IPF, IPF, and healthy control subject, respectively. Lanes 4, 9, and 14: the same in presence of PMA during incubation time. Lanes 5, 10, and 15: BALF from treated-IPF patient, IPF, and healthy control subject, respectively.

View larger version (12K): [in this window] [in a new window]   View larger version (31K): [in this window] [in a new window]   Figure 2.   Quantification of gelatinase B activities by gelatin zymography. Quantitative image analysis of the surface and intensity of lysis bands obtained from AM-conditioned media and AM lysates for each subject was carried out. Cultures were with (full circles) or without (empty circles) PMA. Results are expressed as gelatinase B activity arbitrary units (AU)/24 h/ 10 µl culture medium corresponding to 104 × AMs (top panel ) or as 92 and 88 kD gelatinase activity AU/24 h/10 µl AM lysate corresponding to 104 × AMs (bottom panel ). Shaded bars = PMA in the cultures.

In AM lysates the relationship between progelatinase B (92 kD) and active enzyme (88 to 84 kD) was assessed. In AM lysates from healthy control subjects, low levels of pro and active gelatinase B forms were barely detectable (Figures 1 and 2, bottom panel ). These levels were increased approximately 4-fold in IPF patients compared with healthy subjects, whereas in treated patients, levels had a tendency to return to normal values. The inclusion of 100 nM PMA in culture media induced a slight potentiation of the 92/88 kD gelatinase associated with AM lysates. No other gelatinase was detected by zymography in supernatants and lysates from AM collected in patients and in control subjects.

In the three IPF patients studied before treatment and then while under treatment, the results were even more persuasive about the effect of therapy than using whole groups. The mean gelatinase B expression by AM was reduced 94 ± 5% for the three IPF patients as their own controls, whereas the decrease was 70 ± 18% when comparison was performed between all of the IPF patients.

In BALF (Figure 1 and Table 2), significant increases of both gelatinase B and gelatinase A (72 kD gelatinase) levels were observed in IPF patients compared with the healthy controls, whereas normal levels were measured in treated patients. EDTA completely inhibited the activity of all forms of gelatinases produced by cultured AM or secreted in BALF, whereas PMSF and NEM did not, consistent with previous findings that the gelatinases B and A belong to the matrix metalloproteinase family (Figure 3).

View larger version (54K): [in this window] [in a new window]   Figure 3.   Identification of gelatinases as metalloproteinases. After SDS-PAGE/gelatin electrophoresis of AM-conditioned media, AM lysates, or BALF, gelatin substrate gels were incubated in 100 mM Tris buffer without inhibitor (control) or in the presence of 10 mM EDTA, 2 mM PMSF, or 2 mM NEM as metallo, serine, or cysteine proteinase inhibitors, respectively. Gels were then studied as described. All gelatinases including gelatinases B and A were EDTA-inhibitable but resistant to inhibition by other proteinase inhibitors. Prestained molecular mass markers (MW) are in the first lane on the right. Lane 1: conditioned medium from AM. Lane 2: AM lysates. Lane 3: BALF.

Gelatinase Assays on Radiolabeled Gelatin

Gelatinase assays in AM culture media. Using ³[H]-gelatin, only barely detectable free metallogelatinolytic activity was found in AM culture media (CM) performed with or without the presence of PMA from controls (Figure 4A). This result demonstrated the presence of metalloproteinase inhibitors such as TIMPs sufficient to prevent free forms of activated gelatinase. However, when gelatinase assays were performed in the presence of 1 mM APMA, a 5- and 3-fold significant increase of free metallogelatinolytic activity was observed in culture media with or without PMA, respectively, indicating that the amount of MMP inhibitor was not sufficient to counterbalance activated gelatinase forms. In IPF patients, about a 2-fold increase (P < 0.05) of free metallogelatinolytic activity was observed for each experimental condition (CM alone, CM from PMA-treated AM, CM + APMA, or CM + APMA from PMA-treated AM) compared with the same conditions for AM cultures from control subjects. The free metallogelatinolytic activity in IPF-treated patients was only slightly lower than that of IPF patients for each experimental condition, but the loss of significant difference compared to control subjects suggests some restrictive effect of steroid and immunosuppressive treatment. The assay against radiolabeled gelatin does not distinguish between gelatinase B and gelatinase A, but since no gelatinase A was produced by AM in our experimental conditions, the assay mainly reflects the free gelatinolytic activity from gelatinase B. 

View larger version (37K): [in this window] [in a new window]   Figure 4.   Gelatinase assays on [3H]- gelatin. Culture media (A) or AM lysates (B) from AMs cultured with or without 100 nM PMA were investigated for free gelatinolytic activity. Assays were performed with or without prior gelatinase activation by 1 mM APMA and in the presence of 2 mM PMSF and NEM as serine and cysteine proteinase inhibitors, respectively. Bars = SD (*P < 0.05). Results are expressed as cpm of [3H]-gelatin hydrolyzed per 48 h/100 µl of AM culture medium or lysate corresponding to 105 × cultured AM.

Gelatinase assays in AM lysates. Minimal free metallogelatinolytic activity was associated with AM lysates from healthy control subjects whatever the experimental conditions used (AM alone, PMA-treated AM, AM + APMA, or PMA-treated AM + APMA) (Figure 4B). By contrast, there were large increases in activity in untreated IPF AM compared to the lysates of AM from healthy control subjects. Normal values were observed in IPF patients receiving corticotherapy and immunosuppressive treatment.

Gelatinase assays in BALF. No difference was detected between levels of free metallogelatinolytic activity in BALF between the three groups, indicating the presence of metalloproteinase inhibitors such as TIMPs sufficient to prevent free forms of activated gelatinase (data not shown).

Reverse Zymography

In accordance with the previously given data, TIMP-1 was readily detected by reverse zymography in AM culture media from IPF patients, whereas TIMP-1 was undetectable in AM culture media from control subjects and treated patients (Figure 5). Thus, increased content of TIMP-1 in AM-conditioned media from IPF patients partially counterbalanced the excess of activated gelatinase B or the TIMP-1 content regulated somewhat gelatinase B activation. TIMP-2 was undetectable in AM culture media of any group.

View larger version (46K): [in this window] [in a new window]   Figure 5.   Reverse zymography of conditioned media from cultured AMs. Culture media were subjected to 11.5% SDS-PAGE/ gelatin electrophoresis. The standard zymographic method was modified after the removal of SDS by incubating the gel for 1 h at 37° in the conditioned medium issued from PMA-activated rabbit skin fibroblasts. This medium provides a source of activated gelatinases able to degrade gelatin in the gel extensively. The gel was then processed as in standard zymography. Protection of the gelatin in the gel by the presence of TIMPs led to the appearance of relatively dark bands against a lighter background. Prestained molecular mass markers are in the first lane on the right (MW). Lanes 1, 2, and 3: AM culture media from untreated IPF patients. Lanes 4, 5, and 6: AM culture media from treated-IPF patients (IPF-T). Lanes 7, 8, and 9: AM culture media from heathy subjects (C). Lane 10: 150 ng recombinant TIMP-1 (rTIMP).

Western Blot Analysis and Immunolocalization

AM culture media and AM lysates from IPF patients were recognized by antibody against human gelatinase B (Figure 6). Moreover, in AM lysates, two others bands (88 and 84 kD, respectively) corresponding to autoactivation or degradation products were also recognized. This result, together with that obtained by zymography, clearly showed the presence of active forms of gelatinase B associated with the AM lysates.

View larger version (39K): [in this window] [in a new window]   Figure 6.   Western blot analysis. Two aliquots corresponding to AM culture medium and AM lysate from one IPF patient were analyzed by immunoblotting to ensure the metalloproteinase property of 92 kD gelatinase. Lane 1: 200 ng of purified human gelatinase B as positive reference. Lanes 2 and 3: 10 µl of AM culture medium and lysate, respectively, were immunoblotted under reduced conditions with rabbit polyclonal antiserum against human gelatinase B. The specificity of the 92/88/84 bands recognized by the immune serum was confirmed by the negative result obtained with the preimmune serum (lane 4).

Immunolocalization with confocal microscopy (Figure 7) demonstrated gelatinase B both inside and at the surface of freshly harvested AM from treated patients. By contrast, barely detectable immunofluorescent staining was visualized in AM from healthy control subjects. Intermediate staining intensity was associated with AM from patients receiving steroid and immunosuppressive treatment.

View larger version (158K): [in this window] [in a new window]   Figure 7.   Immunolocalization of gelatinase B in human AMs (immunocytochemical studies). AMs were cultured in chamber slides, fixed with 4% paraformaldehyde and incubated with sheep antihuman gelatinase B IgG. The second antibody was the fluorescein isothiocyanate-conjugated (FITC) pig antisheep F(ab'). Slides were mounted in Citifluor and observed under a Zeiss confocal photomicroscope III. (A) Negative control, in presence of sheep nonimmune serum. (B) AMs from a heathy subject, demonstrating slight immunofluorescent staining. (C) AMs from untreated IPF patient, showing strong pericellular immunofluorescent staining. (D) AM from treated IPF patient, showing moderate immunofluorescent staining.

RT-PCR for Gelatinase B mRNA

We were successful in performing RT-PCR for gelatinase B from freshly harvested AM. In every case, we obtained a single band of expected size 303 bp. PCR performed directly on RNA_T not subjected to the reverse transcription step, and run in parallel with the test samples, was negative. RT-PCR control, consisting of total RNA harvested from human bronchial epithelial cells (24), was positive (Figure 8).

View larger version (21K): [in this window] [in a new window]   Figure 8.   Quantitative RT-PCR for gelatinase B mRNA in human AMs. Ten nanograms of RNAT from AMs and from human bronchial epithelial cells used as positive reference were reverse transcribed and coamplified with a constant amount (104 × molecules) of the specific internal DNA standard. Two serial bands of the expected sizes (491 bp for the internal DNA standard and 303 bp for AM cDNA) were observed. The bands of the internal standard were approximately the same across the range of amplified samples, indicating a constant amplification rate. Bands for gelatinase B from untreated IPF patients were clearly enhanced (lanes 2, 3, and 4) compared to treated IPF patients (lanes 5, 6, and 7). Lane 8: negative control excluding the RNAT. Lane 9: negative control excluding the reverse-transcription step. Lane 10: PCR with the specific internal standard alone.

Using semiquantitative RT-PCR and scanning analysis of autoradiograms showed that the level of gelatinase B mRNA was increased 5-fold (P < 0.005) in untreated IPF patients compared with treated patients (1.6 ± 0.4 versus 0.3 ± 0.1 10⁴ mRNA molecules). This result, together with data showing increased gelatinase B expression at the protein level, indicates AM gelatinase B mRNA overexpression in IPF patients and its negative modulation by corticotherapy.

Immunohistochemical Analysis

Lungs from the two IPF patients showed characreristic areas of dense fibrosis and structural remodeling. The regions adjacent to the honeycomb change showed interstitial inflammation associated with collagen deposition in the alveolar septa and small clusters of AMs in the alveolar spaces. The restructured air spaces were lined by cuboidal epithelial cells. In general, the reaction for gelatinase B was stronger in the regions with predominant inflammatory cell infiltrate than those with end-stage honeycomb change. Cuboidal epithelial cells and myofibroblasts of the alveolar septa and sparse endothelial cells were barely positive, whereas AMs were strongly positive (Figures 9A and 9C). By contrast, only some AM were weakly reactive in control lungs (Figure 9D). Neutrophils were strongly positive in normal and IPF lungs, providing a positive immunoreactive reference. No immunoreactivity was observed with control antibody in IPF or normal pulmonary tissues (Figure 9B).

View larger version (114K): [in this window] [in a new window]   Figure 9.   Immunohistochemical localization of gelatinase B. Gelatinase B was detected by using monoclonal antigelatinase B antibody and alkaline phosphatase antiphosphatase (APAAP) technique. Panels A and C: immunoreactive gelatinase B is mainly observed in AM of pulmonary tissues from IPF patients. Panel B: no staining was seen in the same specimen incubated with control antibody IgG1k. Panel D: no staining of AM was detected in normal lung, except for a few immunoreactive neutrophils. All sections were counterstained with hematoxylin. (Original magnification: panels A and B: ×250, panels C and D: ×400.)

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

As previously shown, AMs were the major inflammatory cells recruited into alveolar spaces in patients with IPF. Neutrophils also accumulated in the airspaces but at lower densities (about 5 to 15%) in agreement with other works (2, 23, 24), suggesting that the moderate neutrophil influx is a constant feature throughout the course of the disease. Nevertheless, AMs from IPF subjects are currently considered to be key players in directing the fibrotic response and differ functionally and metabolically from AMs obtained from normal subjects. References include production of reactive oxygen intermediates (25), recruitment and local proliferation of AMs (26), spontaneous production of the fibrogenic molecules platelet-derived growth factor (PDGF) and T-cell growth factor  (27) and spontaneous release of chemotactic factors for neutrophils (28). In this study, we have demonstrated that AM also expresses increased gelatinase B during the course of IPF.

Overexpression of 92 kD Gelatinase by AM from IPF Patients

Cultures of AM from healthy subjects secreted barely detectable level of gelatinase B, suggesting the quiescent state of these AMs. By contrast, cultures of AMs from IPF subjects exhibited a net overexpression of gelatinase B both at the protein and mRNA level, pointing out that AMs could synthesize this metalloproteinase during the course of IPF. Elevated gelatinase B expression is in accordance with recent immunohistochemical studies of gelatinase B in the human lungs demonstrating that AMs from IPF patients reacted more intensely than AMs from healthy subjects (29).

We observed a low but significant level of free metallogelatinolytic activity in [³H]-gelatin assays on AM-conditioned media from IPF patients, as well as its graduate increase in response to PMA stimulation, APMA activation, or both PMA-APMA, respectively. This strongly suggests that the amount of gelatinase inhibitors as TIMP-1, even if clearly increased in AM culture media as we demonstrated by reverse zymography, does not fully compensate for the overexpression of the gelatinase and for APMA-induced progelatinase activation.

Gelatinase B increases associated with AMs in IPF were found in AM culture media and, more surprisingly, in cell lysates. Indeed, the current concept is that AMs synthesize and secrete, rather than store the gelatinolytic activity (30). Our results, however, clearly argue in favor of intra- and pericellular AM gelatinase B forms in IPF: (1) these forms represent about 25 to 50% of the total gelatinase B product; (2) they are distributed between the 92-kD proform (50%) and the 88- to 84-kD activated forms (50%) recognized by specific 92 kD antibodies; (3) they exhibit a net free metallogelatinolytic activity against [³H]- gelatin, which is increased in response to PMA stimulation or following APMA activation; (4) immunocytochemical studies demonstrate its net pericellular membrane localization besides a widespread intracellular distribution; and (5) immunohistochemical studies show a strong positive reaction associated with AM clusters in the lungs of IPF patients. These results raise the possibility that a substantial part of the gelatinase B produced by AMs from IPF patients is directly involved in AM migration potential and lung matrix remodeling by AM.

Possible biologic roles for AM gelatinase B during the course of IPF. The enhanced secretion of gelatinase B by macrophages has been already observed during chronic inflammation. AMs isolated from patients with sarcoidosis secrete greater amounts of gelatinase B than do normal AMs (31). Likewise, peritoneal macrophages isolated from patients with peritonitis also secrete elevated levels of gelatinase B (32). This observed gelatinase secretion is likely the result of macrophage activation at inflammatory foci, as a similar upregulation of gelatinase release has been demonstrated in vitro following stimulation of macrophages with phorbol 12-myristate 13-acetate or lipopolysaccharide (33).

Modulation of AM gelatinase B expression by corticosteroid and immunosuppressive treatment during the course of IPF. Our data showed that corticosteroid and immunosuppressive treatment tended to normalize all parameters as follows: total cell influx, gelatinase expression B by AM both at the protein and mRNA level, free metallogelatinolytic activity released into conditioned media and associated with AM lysates, and TIMP-1 production by AMs. These results are in accordance with previous studies (43) demonstrating that corticosteroids selectively and coordinately inhibit expression of gelatinase B as well as TIMP by human AM cultured in vitro, and block the effects of one of the most potent known signals, lipopolysaccharide (LPS) for upregulation of gelatinase B production. The mechanisms of glucocorticoid-induced transcriptional signal appears to involve direct binding of the glucocorticoid receptor complex to the activating protein jun (44, 45). Glucocorticoids also downregulate the production of cytokines, such as tumor necrosis factor and IL-1, that are able to induce gelatinase B (46). Accordingly, corticosteroids may affect gelatinase B expression in several ways. Nevertheless, corticosteroid-induced gelatinase B-downregulation in IPF did not appear correlated with clinical improvement for the three patients studied before and after treatment.