Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The matrix metalloproteinases (MMPs) comprise a class of proteolytic enzymes that (1) are secreted in latent form and activated by removal of the amino-terminal propeptide-sequence; (2) contain a zinc ion at the active site, and thus are inhibited in the presence of ethylenediamenetetraacetic acid (EDTA); (3) have activity against components of the extracellular matrix; and (4) are inhibitable by tissue inhibitor of metalloproteinase (TIMPs) (1). Macrophages are known to produce a number of MMPs, but the spectrum of enzymes elaborated by these cells differs from species to species and from site to site, and is influenced by the state of activation and differentiation of the elaborating cells. Human alveolar macrophages produce MMP-1 (interstitial collagenase), MMP-2 (72-kD gelatinase A), MMP-3 (stromelysin 1), MMP-9 (92-kD gelatinase B), and MMP-12 (metalloelastase) (4). Human monocytes but not alveolar macrophages also produce MMP-7 (matrilysin) (4). Exposure of human alveolar macrophages, but not peripheral blood monocytes, to stimuli such as bacterial lipopolysaccharide (LPS) and phorbol myristate acetate (PMA) upregulates production of metalloproteinases, especially MMP-9 (7).

At present, little is known about the MMPs elaborated by lung macrophages from the rat. Although it might seem reasonable to hypothesize a similar protease profile for the rat alveolar macrophage as for the human, our recent studies with rat neutrophils suggest caution in this regard. Rat neutrophils show both qualitative and quantitative differences in protease secretion as compared with human neutrophils (8, 9). Given the observation that rat alveolar macrophage MMPs appear to play a direct "effector" role in some forms of acute lung injury (see following article, Reference 10), characterization of the spectrum of MMPs secreted by rat alveolar macrophages would seem to be a necessary prerequisite to identifying those enzymes that participate in the progression of the disease and to determine how they function to promote tissue injury. The present report describes efforts to characterize and identify MMPs elaborated by rat alveolar macrophages, and to compare these activities with those activities present in bronchoalveolar lavage fluid (BALF) during macrophage-dependent injury.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals and Reagents

Male Long-Evans rats were purchased from Charles River Breeding Company (Wilmington, MA) and housed under specific pathogen-free conditions in sterile cages under laminar flow. Experiments performed on the animals were carried out after review and approval by the University Committee on Use and Care of Animals (University of Michigan, Ann Arbor, MI). Recombinant human TIMP-2 was a gift of Dr. Keith Langley of Amgen Pharmaceuticals (Thousand Oaks, CA). Mouse monoclonal antibody to human MMP-9 and antimouse horseradish peroxidase (HRP)-conjugated secondary antibody were purchased from Oncogene Science (Cambridge, MA). Sheep polyclonal antibody to human MMP-2 and antisheep HRP secondary antibody were purchased from The Binding Site (Birmingham, UK). Rabbit polyclonal antibody to mouse MMP-12 was a generous gift of Dr. Steve Shapiro (Washington University, St. Louis, MO). Rabbit polyclonal antibody to human MMP-13 was obtained from Chemicon (Temecula, CA). Mouse MMP-12 and -elastin were purchased from Elastin Products (Pacific, MO). Tritiated collagen, deoxycytidine 5'-triphosphate ([³²P]dCTP), Random Primer labeling kit, Hybond membrane, and the Enhanced Chemiluminescence detection kit were purchased from Amersham (Arlington Heights, IL). QuickHyb hybridization solution and NucTru-Push columns were purchased from Stratagene (La Jolla, CA). Bovine serum albumin (BSA), LPS, PMA, gelatin, E64, pepstatin, phenylmethylsulfonyl fluoride (PMSF), -casein, Triton X-100, EDTA, aminophenyl mercuric acetate (APMA), dimethyl sulfoxide (DMSO), and Tris-HCl were purchased from Sigma (St. Louis, MO). Other reagents are noted below in the respective protocols.

Preparation of Rat Alveolar Macrophages and Other Cells

Alveolar macrophages were isolated from BALF as described in detail previously (11). After the rats were killed by lethal injection of Ketamine, the lungs were lavaged in situ with 10 ml of sterile normal saline. Cells collected in the lavage fluid were pelleted by 10 min centrifugation at 400 × g and plated in minimal essential medium Eagle with Earle's salts, supplemented with 0.02% BSA, and penicillin/streptomycin. Normally, 1 × 10⁶ cells were plated per well of a 24-well dish. After allowing the cells to adhere to the plates for 2 h, nonadherent cells were removed with two washes. Morphologic examination and staining for nonspecific esterase or with the macrophage-specific lectin Griffonia simplicifolia-1 (12) indicated that cells prepared in this manner were routinely > 95% alveolar macrophages. Alveolar macrophages prepared in this manner were determined to be quiescent by finding negligible tumor necrosis factor release (13) and minimal hydrogen peroxide production (14). HT1080 and P388D1 cells were purchased from ATCC (Rockville, MD). Rat lung fibroblasts and culture-differentiated human monocytes were prepared as described previously (15). Rat peritoneal neutrophils were elicited by intraperitoneal glycogen administration as described previously (9).

Unstimulated control cells and cells stimulated with LPS (10 µg/ml), PMA (100 nM), or immunoglobulin (Ig)A immune complexes (10 µg of preformed complexes) were incubated for 18 h. At the end of the incubation period, cellular debris was removed by centrifugation at 400 × g for 10 min and the conditioned medium examined as described below. In some instances the conditioned medium was concentrated 10- to 50-fold using Centricon concentrators (Amicon, Danvers, MA) before assay. In these cases, volume-to-volume concentration was strictly controlled.

Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis Substrate-Embedded Enzymography

Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) substrate-embedded enzymography (zymography) was carried out by a modification of the method of Heussen and Dowdle (16). Zymography was used in these studies as the initial approach to identifying and characterizing the MMPs. Briefly, SDS-PAGE gels were prepared for minigels from 30:1 acrylamide/bis with the incorporation of either gelatin (1 mg/ml), -casein (1 mg/ml), or -elastin (1 mg/ml) before casting. The gelatin gels were routinely 7.5% acrylamide, whereas the casein and elastin gels were routinely cast at 10% acrylamide. Various denatured but nonreduced samples and standards were then run into the gels at constant voltage of 150 V under nonreducing conditions. When the dye fronts reached a point approximately 0.5 cm from the bottom of the gels, the gels were removed and subjected to the following washing protocol: twice for 15 min each time in 50 mM Tris buffer (containing 1 mM Ca²⁺ and 0.5 mM Zn²⁺) with 2.5% Triton X-100; once for 5 min in Tris buffer alone; and finally overnight in Tris buffer with 1% Triton X-100. Inhibitors were added as desired to the overnight wash, as indicated later. The gels were stained the following morning with Coomassie Brilliant Blue 250-R. After destaining, zones of enzyme activity showed up as regions of negative staining. Relevant controls included samples incubated with TIMP-2 or EDTA (MMP inhibitors), PMSF (serine proteinase inhibitor), E64 (cysteine proteinase inhibitor), and pepstatin (aspartic proteinase inhibitor). Additionally, some samples were incubated with APMA during electrophoresis, to show the band shift that occurs when latent MMPs are activated through cleavage of the N-terminal domain (17).

Native Substrate Degradation Assay

The same culture fluids were tested for their ability to degrade type I collagen as described previously (9). Briefly, 0.2 µCi of [³H]collagen (Collaborative Research, Bedford, MA) were combined with 50 µl of the solution to be tested, brought to a total volume of 100 µl in Tris buffer, and incubated overnight at 22°C. The nondegraded protein was then precipitated in 10% trichloroacetic acid (TCA) and the remaining soluble phase dissolved in scintillation fluid for counting (Beckman Instruments, Irvine, CA). Bacterial collagenase (Worthington Biochemicals, Freehold, NJ) and enzyme released from PMA-stimulated human neutrophils were used as positive controls.

Western Blot Analysis and Immunoprecipitation

Conditioned medium was resolved by SDS-PAGE as described previously, then electroblotted onto polyvinylindene fluoride (PVDF) membranes at 100 V for 1 h in a pH-8.3 transfer buffer of 25 mM Tris, 192 mM glycine, and 20% vol/vol methanol. The resultant blots were blocked overnight in Tris-buffered saline (20 mM Tris, pH 7.6) with 0.1% Tween (TBS-T) and 5% powdered milk (or 1% gelatin when the sheep antihuman MMP-2 polyclonal antibody was used, because that antibody cross-reacted with a component of powdered milk). After the blocking, the blots were washed three times for 10 min each time in TBS-T, and then incubated for 1 h with an appropriate dilution of primary antibody. This was followed by another three 10-min washes with TBS-T and 30 min incubation with an appropriate dilution of HRP-labeled secondary antibody. Finally, the blots were washed five times for 10 min each time with TBS-T, bathed with chemiluminescence detection substrate, and exposed to film.

Reverse Transcription and Polymerase Chain Reaction Complementary DNA Probe Preparation

[³²P]-labeled complementary DNA (cDNA) probes were prepared to MMP-9, MMP-2, MMP-12, MMP-7, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) by reverse transcriptase-polymerase chain reaction (RT-PCR) and random primer labeling as described previously (15). Briefly, alveolar macrophages were isolated and cultured as described above, with and without stimulation. Cells were then lysed and the RNA purified with TriZol reagent (GIBCO BRL, Grand Island, NY). A sample of the total RNA was checked for purity, yield, and degradation, and then used as a template for reverse transcription. The amount of 1.5 µg of total RNA was mixed with 17 base oligo-dT primers and RT and incubated at 37°C for 90 min. The resultant cDNA was excised from the gel and then used as the template for PCR (Perkin-Elmer Cetus, Norwalk, CT). In the case of MMP-12, PCR primers were prepared to the published mouse sequence (18). The resultant 440-base pair (bp) cDNA fragment of rat metalloelastase was random primer-labeled and -tested against messenger RNA (mRNA) prepared from P388D1 cells as a positive control by Northern blot analysis. Unincorporated radioactivity was removed with NucTru push columns (Stratagene) and [³²P] incorporation was checked by scintillation counting. Preparation of the 595-bp rat MMP-9 probe, the 850-bp rat MMP-2 probe, the 425-bp cDNA fragment of rat matrilysin (MMP-7), and the 900-bp GAPDH probe have previously been described elsewhere (15, 19, 20).

Northern Blot Analysis

A sample of RNA, purified as described previously, was denatured at 65°C in 37% formaldehyde, and then run into 1.5% agarose (5 µg/lane). Fifteen-day cultured human macrophage total RNA was purified and used as a positive control lane. The gel was then transferred overnight by capillary action onto Hybond-N membrane. The blots were then probed with either the cDNA fragment of rat MMP-9, MMP-2, MMP-12, or the 425-bp cDNA fragment of rat matrilysin (MMP-7) described previously. Equal loading was confirmed by probing the blot for GAPDH. Hybridization was in QuickHyb Hybridization Solution (Stratagene), followed by two high-stringency washes in 0.1× saline sodium citrate/0.1% SDS at 65°C for 30 min. The blots were then exposed to X-ray film (Fuji, Tokyo, Japan) and developed.

Spectrophotometric Assay for Serine Proteinase Activity

Elastase and cathepsin G activities were measured spectrophotometrically by the cleavage of specific paranitroaniline chromogenic substrates (MeO-Suc-Ala-Ala-Pro-Val-p-nitroanilide for elastase and Suc-Ala-Ala-Pro-Phe-p-nitroanilide for cathepsin G). Both substrates were obtained from Calbiochem (La Jolla, CA). Briefly, in a cuvette were combined 50 mM Tris buffer (containing 1 mM Ca²⁺ and 0.5 mM Zn²⁺) (volume adjusted to make 1.2 ml total volume), 100 to 500 µl of the sample to be tested, any inhibitors as indicated below, and 100 µl of the appropriate substrate in DMSO (for a final substrate concentration of 1 mM). The solutions were mixed thoroughly and absorbance measured over time at 410 nm. The slopes of the linear portion of the resultant time curves were taken to determine the rate of product formation for the sample, and the amount of product determined using 8,800 cm¹ mM¹ for the extinction coefficient. The resulting activities were expressed as nanomoles per min per sample-quantity.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Zymography of Conditioned Medium from Rat Alveolar Macrophages

In the first series of experiments, conditioned medium from rat alveolar macrophages was characterized by gelatin, -casein, and -elastin zymography. The left panel of Figure 1 shows that rat alveolar macrophages produced a band of gelatinolytic activity at 92 kD when stimulated for 18 h with PMA (Figure 1, left panel, lane 1). This activity was virtually undetectable in unstimulated, quiescent cells from the same preparation (Figure 1, left panel, lane U). The 92-kD product underwent the characteristic MMP band shift when exposed to APMA before zymography, and all of the gelatinolytic activities were resistant to PMSF, E64, and pepstatin but were inhibited by TIMP-2 and EDTA (not shown). An additional faint band of gelatinolytic activity could be seen at 72 kD. These data are consistent with the presence of MMP-2 (gelatinase A) and MMP-9 (gelatinase B) in the macrophage culture fluids.

View larger version (21K): [in this window] [in a new window]   Figure 1.   Gelatinolytic, caseinolytic, and elastinolytic activities in rat alveolar macrophage culture fluids. Macrophages were isolated from healthy rats and placed in culture as described in MATERIALS AND METHODS. The cells were allowed to condition the medium for 18 h and then the conditioned media were tested for activity by zymography. Left panel: Gelatin: conditioned medium from unstimulated macrophages is illustrated in lane U. Lane 1 represents conditioned medium from cells stimulated with PMA (100 nM). Middle panel: -Casein: conditioned medium from unstimulated macrophages is illustrated in lane U. Lane 1 represents culture fluid from cells stimulated with PMA. Lane 2 (inset) shows a portion of a gel from a 10-fold concentrated sample from cells stimulated with PMA; the 22-kD activity is apparent. Right panel: -Elastin: lane U represents elastin zymography of conditioned medium from unstimulated rat alveolar macrophages. Lane 1 represents elastin zymography of conditioned medium from 100 nM PMA-stimulated macrophages.

Further bands on the gelatin zymogram were observed at 35 and 22 kD in culture fluids from stimulated macrophages (Figure 1, left panel, lane 1). The 35-kD band also appeared on the -casein zymogram (Figure 1, middle panel, lane 1). Upon activation, there was a loss of 35-kD activity, presumably as a result of band shift to a lower molecular weight (not shown). Although the 22-kD product was too faint to be clearly seen in the casein gel without concentration, it could be seen when samples were concentrated 10-fold before use (Figure 1, middle panel, inset lane 2). Both the 35-kD and 22-kD activities were resistant to PMSF, E64, and pepstatin, but were inhibited by EDTA and TIMP-2 (not shown). The 22-kD activity was also active against -elastin (Figure 1, right panel, lane 1). For these experiments, purified mouse MMP-12 was used as a control. The 22-kD activity against -elastin from the rat alveolar macrophage comigrated exactly with the purified mouse enzyme (not shown). Of note is a 35-kD intermediate form that was also present in the purified mouse MMP-12 and showed activity against gelatin but not -elastin. These data are consistent, therefore, with the activity of MMP-12 (metalloelastase) (21) and represent the first identification of this protein in the rat.

These studies demonstrate that rat alveolar macrophages release a number of MMPs into the culture fluid upon stimulation with PMA. Next, studies were performed to determine whether the two injury-inducing agonists (e.g., LPS, and IgA immune complexes) (see 10) also stimulated elaboration of MMPs by alveolar macrophages. Cells were treated with either PMA (100 nM), LPS (10 µg/ml), or preformed IgA immune complexes (10 µg/ml) and allowed to condition the culture medium for 18 h. Conditioned media were then subjected to analysis by gelatin zymography. Culture fluids from macrophages stimulated by all three agonists exhibited an identical zymographic profile. Quantification of zymographic bands by densitometry showed no significant differences among the agonists (Figure 2). Likewise, there were no significant differences between the intensities of individual bands (not shown).

View larger version (62K): [in this window] [in a new window]   Figure 2.   Comparison of PMA-, LPS-, and IgA-containing immune complexes as stimulators of MMP production by rat alveolar macrophages. Gelatin zymograms of conditioned medium from rat alveolar macrophages stimulated with either 100 nM PMA (lane PMA), 10 µg/ml LPS (lane LPS), 10 µg/ml pre-formed IgA immune complexes (lane IgA), or unstimulated controls were assessed for MMP activity by combined densitometry of all the bands. Data are expressed as fold-induction over unstimulated controls (n = 8 for all cases except IgA, where n = 6).

Confirmation of the 92-kD Protein as MMP-9

The zymographic data presented previously are sufficient to show that the identified activities belong to the MMP family and suggest tentative identifications. The fact that the rat alveolar macrophage MMPs have not previously been characterized, however, means that further effort must be taken to confirm these tentative zymographic identifications. Not only could the activities represent new proteins with sizes similar to known proteins, but in the case of the 22-kD activity, there are at least two known enzymese.g., MMP-7 (matrilysin) and MMP-12 (metalloelastase)that have 22-kD active forms.

An antihuman MMP-9 monoclonal antibody was available for use in the immunologic identification of the 92-kD activity. The antibody, which was raised against MMP-9 secreted by human HT1080 fibrosarcoma cells (22), reacted strongly with culture fluid from HT1080 cells (Figure 3, lane HT). The antibody reacted with a (92-kD) protein from stimulated rat alveolar macrophages (Figure 3, lane AM+) that comigrates with the 92-kD proteins from HT1080 cells. The same protein was not detected in culture fluids of unstimulated cells (Figure 3, lane AM), consistent with the zymographic evidence. The macrophage 92-kD protein, therefore: (1) has gelatinolytic activity; (2) is inhibitable by TIMP-2 and EDTA but not by PMSF, E64, or pepstatin; (3) is activated by APMA; (4) comigrates with human MMP-9; and (5) cross-reacts with antihuman MMP-9 antibodies. This evidence confirms the identification of the 92-kD enzyme as rat MMP-9.

View larger version (52K): [in this window] [in a new window]   Figure 3.   Western blot of alveolar macrophage-conditioned medium with an antihuman-MMP-9 monoclonal antibody. Conditioned media from control and PMA-stimulated alveolar macrophages were subjected to SDS-PAGE and transferred to a PVDF membrane as described in MATERIALS AND METHODS. The resulting membrane was blotted with a mouse antihuman MMP-9 monoclonal antibody. Lane AM is the media from unstimulated (negative control) macrophages; lane AM+ is 18-h conditioned medium from rat alveolar macrophages stimulated with 100 nM PMA; and lane HT is conditioned medium from HT1080 cells (against which the antibody was raised; used as a positive control).

Confirmation of the 72-kD Protein as MMP-2

The first step taken to characterize the 72-kD activity was comparing its migration with that of MMP-2 from cultured rat fibroblasts. Figure 4 (upper panel) shows that the constitutively expressed band from rat alveolar macrophages (Figure 4, upper panel, lane AM/APMA) comigrates with the latent form of rat MMP-2 (upper panel, lane Fib/ APMA). It can also be seen from Figure 4 that exposure of the fibroblast culture fluids to APMA converted the latent form of the enzyme to the active product (upper panel, lane Fib/+APMA). Due to its initial scarcity and to successive autolytic inactivation of MMP-2 (23), it was not possible to obtain clear evidence for production of the active form of the macrophage enzyme upon APMA treatment. We therefore concentrated culture medium from PMA-stimulated macrophages 50-fold before zymography and exposed the concentrated culture medium to APMA. Under these conditions, an activation product was detected (Figure 4, upper panel, lane AM/+APMA).

View larger version (29K): [in this window] [in a new window]   Figure 4.   Characteristics of the 72-kD activity in rat alveolar macrophage culture fluids. Upper panel: Evidence for activation with APMA. Conditioned media were collected and subjected to gelatin zymography as described in MATERIALS AND METHODS. Lanes Fib/APMA and AM/APMA are conditioned media from rat fibroblasts and alveolar macrophages stimulated with 100 nM PMA. The corresponding duplicate lanes (+APMA) were exposed to APMA for 90 min before electrophoresis. Middle panel: Western blot using an antihuman-MMP-2 polyclonal antibody. Conditioned media were subjected to SDS-PAGE and transferred to a PVDF membrane as described in MATERIALS AND METHODS. The resulting membrane was blotted with a sheep antihuman MMP-2 monoclonal antibody. Lane FIB is 18-h conditioned medium from rat lung fibroblasts, and lane AM is conditioned medium from rat alveolar macrophages that have been stimulated with 100 nM PMA. Lower panel: Immunoprecipitation of alveolar macrophage-conditioned medium using an anti- MMP-2 polyclonal antibody. Conditioned media were stripped of 72-kD activity by immunoprecipitation with excess antihuman MMP-2 antibody, and the bound enzyme from the pellet was resolubilized and run on a gelatinogram, as described in MATERIALS AND METHODS. Lane Fib is activity at 72 kD from the washed (three times) pellet after immunoprecipitation of 18-h conditioned medium from rat lung fibroblasts (included as a positive control). Lane 1 is conditioned medium from rat alveolar macrophages that have been stimulated with 100 nM PMA. Lane 2 is the same macrophage-conditioned medium after stripping with antibody, and lane 3 is the activity that was precipitated in the pellet and resolubilized.

A sheep antihuman MMP-2 polyclonal antibody was tested for cross-reactivity with the rat 72-kD band by Western blotting. The middle panel of Figure 4 shows that the antihuman MMP-2 antibody demonstrated a positive response with the rat fibroblast 72-kD protein (middle panel, lane Fib). The reactivity of the rat fibroblast culture fluid with the antihuman MMP-2 appeared to be weak, given the strong response in gelatin zymography. Not surprisingly in light of this, the rat macrophage culture fluids also demonstrated a very weak reactivity with the antihuman MMP-2 antibody by Western blotting (Figure 4, middle panel, lane AM). Although these Western blot studies support identification of the rat macrophage 72-kD activity as MMP-2, further studies were warranted to provide convincing identification. The concentrative effects of immunoprecipitation were used for this (Figure 4, lower panel). Conditioned medium from rat alveolar macrophages (Figure 4, lower panel, lane 1) was stripped of 72-kD activity by immunoprecipitation with excess antihuman MMP-2 antibody (lane 2), and the bound enzyme from the pellet was resolubilized and run on a gelatinogram in parallel. Lane 3 shows that the antihuman MMP-2 bound and brought down the rat 72-kD activity. Thus, the 72-kD protein: (1) has gelatinolytic activity; (2) is inhibitable by TIMP-2 and EDTA but not by PMSF, E64, or pepstatin; (3) is activated by APMA; (4) comigrates with rat fibroblast MMP-2; and (5) cross-reacts with antihuman MMP-2 antibodies. This evidence confirms the identification of the 72-kD enzyme as rat MMP-2.

Confirmation that the 22-kD Protein is MMP-12

An antimouse MMP-12 antibody was available for use in the immunologic identification of the pair of activities that were present as a latent form of 35 kD and an active form of 22 kD on zymograms of conditioned medium from the rat alveolar macrophages. The antibody, which was raised against MMP-12 secreted by mouse P388D1 cells (18, 21), reacted strongly with culture fluid from P388D1 cells (not shown). The antibody reacted with a protein from stimulated rat alveolar macrophages (Figure 5, lane AM+) that comigrated with the 22-kD active form of mouse MMP-12 from P388D1 cells. The same protein was not detected in culture fluids of unstimulated cells (Figure 5, lane AM), consistent with the zymographic evidence. The macrophage 22-kD protein, therefore: (1) has elastinolytic activity; (2) is inhibitable by TIMP-2 and EDTA but not by PMSF, E64, or pepstatin; (3) comigrates with mouse MMP-12; and (4) cross-reacts with antimurine MMP-12 antibodies. This evidence confirms the identification of the 22-kD enzyme as rat MMP-12.

View larger version (61K): [in this window] [in a new window]   Figure 5.   Western blot of alveolar macrophage-conditioned medium with an antihuman-MMP-12 monoclonal antibody. Conditioned medium from PMA-stimulated macrophages was subjected to SDS-PAGE and transferred to a PVDF membrane as described in MATERIALS AND METHODS. The resulting membrane was blotted with a rabbit antimouse MMP-12 monoclonal antibody. Lane AM+ is 18-h conditioned medium from rat alveolar macrophages stimulated with 100 nM PMA; lane AM is the media from unstimulated control macrophages.

Identification of Corresponding mRNAs for Detected MMPs

Total RNA from alveolar macrophages was prepared and reverse transcribed to make the corresponding cDNA. The cDNA was used as a template for PCR with primers generated to unique portions of the published rat or mouse sequences, as described in MATERIALS AND METHODS. Primers generated against the rat GAPDH sequence were used as a loading control. The resulting sequences have all been previously characterized except MMP-12. The PCR product from the rat alveolar macrophages examined with MMP-12 primers comigrated with the product from the P388D1 cells (not shown). It was cut out of the gel and sequenced. The resultant 440-bp sequence exactly matched that of mouse metalloelastase (18).

Figure 6 shows data from Northern analysis of mRNA from unstimulated control (column C) and PMA-stimulated (column PMA) rat alveolar macrophages. Strong hybridization to message was evident for MMP-9 and MMP-12 in stimulated but not unstimulated alveolar macrophages. In addition, faint hybridization to MMP-2 was seen in stimulated alveolar macrophages. GAPDH hybridization showed equal loading.

View larger version (48K): [in this window] [in a new window]   Figure 6.   Northern blot analysis of rat alveolar macrophage mRNA for MMP-9, MMP-2, MMP-12, and MMP-7. Column C represents corresponding portions of Northern blots of RNA isolated from unstimulated control rat alveolar macrophages, and column PMA represents portions of blots from rat alveolar macrophages stimulated with 100 nM PMA. Each row represents blots probed for the corresponding message: MMP-9, MMP-2, MMP-12, MMP-7, or GAPDH.

Because the active form of MMP-7 (matrilysin) is also a 22-kD enzyme, and this enzyme has caseinolytic activity (5), it was possible that the 22-kD caseinolytic activity might represent MMP-7 as well as MMP-12. With this in mind, a previously described 425-bp cDNA for rat MMP-7 (15) was used to probe mRNA prepared from the same rat alveolar macrophages and human culture-differentiated monocytes. No hybridization was found with the rat alveolar macrophages (either unstimulated or after exposure to 100 ng/ml PMA) (Figure 6). Culture-differentiated human monocytes demonstrated a positive response in the same blot (not shown).

Proteinases Present in Low or Undetectable Amounts

If interstitial collagenase (MMP-1, 54 kD latent and 45 kD active), collagenase-3 (MMP-13, 54 kD latent and 41 kD active), or neutrophil-collagenase (MMP-8, 75 kD latent and 64 kD active) (1) were present, prominent bands on caseinograms should have been observed. Because they were not, this suggests that enzymes with collagenolytic activity were present in low amounts, if at all, in the culture fluid from PMA-stimulated rat alveolar macrophages. To confirm that rat alveolar macrophages do not express significant levels of collagenase activity, culture fluids from PMA-stimulated cells were examined for ability to degrade ³H-collagen into TCA-soluble fragments. Under conditions in which degranulation products from 10⁴ human neutrophils degraded 42 ± 7% of the native collagen, the activated conditioned medium from 5 × 10⁷ PMA- stimulated alveolar macrophages degraded only 7 ± 8% (which was not statistically different from the buffer control value of 0 ± 3%; n = 9 for all treatments) (Figure 7, top panel). The lack of significant collagenolytic activity was confirmed by experiments in which native type I collagen and heat-denatured (60°C for 5 min) type I collagen were exposed to culture fluids from PMA-stimulated rat alveolar macrophages. Although the macrophage-conditioned medium contained activity sufficient to completely degrade the gelatin, there was no measurable degradation of native collagen, even after 18 h of incubation (data not shown). It should be noted that a faint band of reactivity was observed in 50-fold-concentrated culture fluids from PMA-stimulated rat alveolar macrophages when examined by Western blotting with an antibody to MMP-13.

View larger version (29K): [in this window] [in a new window]   Figure 7.   Collagenolytic and serine proteinase activities in culture fluids from PMA-stimulated alveolar macrophages. Top panel: Conditioned media were tested for the ability to degrade tritiated native type I rat collagen in vitro as described in MATERIALS AND METHODS. Degranulation products from 104 human neutrophils degraded 42 ± 7% of the native collagen. Activated conditioned medium from 5 × 107 PMA-stimulated alveolar macrophages degraded only 7 ± 8%, which was not statistically different from the buffer control value of 0 ± 3% (n = 9 for all treatments). Bacterial collagenase (1 mg/ml) was used as an additional positive control. Bottom panel: Conditioned medium from either unstimulated cells or cells stimulated for 18 h with PMA was tested for the ability to degrade either the elastase-specific substrate MeO-Suc-Ala-Ala-Pro-Val-p-nitroanilide or the cathepsin G-specific substrate Suc-Ala-Ala-Pro-p-nitroanilide over time. The resulting rates of cleavage are shown from a representative experiment. Note that no activity could be obtained from rat alveolar macrophages, even when 5 × 107 cells were stimulated for up to 5 d with a variety of agonists and the assay carried out for 48 h. The experiment was repeated five times, with similar results.

In a final set of studies, rat alveolar macrophages were examined for elaboration of serine proteinaseselastase and cathepsin Gusing chromogenic synthetic substrate assays (Figure 7, bottom panel). Human peripheral blood neutrophils and rat peritoneal neutrophils were used as controls. Rat alveolar macrophages failed to produce detectable levels of either elastase or cathepsin G, even when 50-fold-concentrated samples from 5 × 10⁷ cells were used. These data confirm that rat alveolar macrophages, like their human counterparts (4), do not produce either of the serine proteinases.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the following article (10), evidence was presented for the involvement of MMPs in acute lung inflammation induced in rats by IgA immune complexes and by LPS. Oxidants are known from past studies to play important roles in the pathogenesis of the disease in both experimental models (24), but this is the first direct evidence that proteinases (in this case MMPs) are also involved. On the basis of the nature of our studies (i.e., using intratracheal instillation of TIMP-2 to reduce injury), it was not possible to identify the specific MMP(s) responsible for injury or their source(s). It was shown, however, that the BALF from injured animals contained increased amounts of several MMPs and that the spectrum of bronchoalveolar lavage activities seen on gelatin and casein zymograms paralleled the activities produced by alveolar macrophages isolated from the same animals or by alveolar macrophages isolated from uninjured animals and stimulated in culture. Other cell types isolated from the lung parenchyma, including type II epithelial cells, fibroblasts, and pulmonary endothelial cells, did not demonstrate a similar spectrum of enzymes. On this basis, we suggest that the alveolar macrophages themselves are a likely source of at least some of the proteinases responsible for injury. The studies described here were carried out to characterize MMPs elaborated by rat alveolar macrophages, and specifically to identify those activities that are present in BALF during macrophage-dependent injury.

Using a combination of gelatin, -casein, and -elastin zymography, at least three activities were detected. On the basis of apparent molecular sizes, substrate specificities, band shift upon activation with APMA, and sensitivity to various inhibitors, it was concluded that these activities represent MMPs. They were tentatively identified as MMP-2 (gelatinase A; 72-kD gelatinase), MMP-9 (gelatinase B; 92-kD gelatinase), and MMP-12 (metalloelastase). Subsequent immunochemical and molecular approaches confirmed the presence of MMP-2, MMP-9, and MMP-12. We concluded tentatively, on the basis of previous work (21), on direct comparison with purified murine MMP-12, and on the loss of 35-kD expression following activation, that the 35-kD activity is a form of MMP-12. Alternatively, this activity may reflect a novel eznyme. Of equal interest are the enzymatic activities that were not detected or were present in extremely low amount. These include enzymes with collagenolytic activity as well as stromelysin-1 (on the basis of the lack of appropriate-sized activities on casein gels) and matrilysin (on the basis of the lack of detectable mRNA and a lack of appropriate-sized activities on casein gels). Also not detected were two serine proteinases, elastase and cathepsin G.

The mRNA data from Northern blot studies confirm the protein data. Stimulated rat alveolar macrophages produced high levels of message for MMP-9 and MMP-12, but these messages were not detectable in RNA from control cells. Likewise, MMP-2 was barely detectable at the protein level (in either unstimulated or stimulated cells) and MMP-2 mRNA was also barely detectable under both conditions. It is interesting that the lack of detectable MMP-7 message effectively rules out the production of matrilysin (MMP-7) by rat alveolar macrophagesa possible second source of the 22-kD activity. Indirectly, therefore, this provides added support to the theory that the 35-kD activity, which has activity against gelatin and casein but not elastin, is an intermediate form of MMP-12.

These results demonstrate parallels between rat and human alveolar macrophages. However, there are differences. Human alveolar macrophages are known to produce detectable levels of MMP-1 (4). In the rat, MMP-1 is not present; its function is performed by MMP-13 (e.g., collagenase-3) (30). However, the fact that there was minimal collagenase activity in the rat alveolar macrophage culture fluids suggests that MMP-13 is present in small amounts. The lack of significant collagenolytic activity in the alveolar macrophages of the rat may not be too surprising. A past study (9) showed that rat neutrophils were also deficient in collagenase activity relative to their human counterparts. Another potential difference is in expression of MMP-3 (stromelysin-1). Previous studies have identified stromelysin-1 in human alveolar macrophages (4). The possibility that rat alveolar macrophages also make this enzyme is not ruled out by the present study. However, the lack of detectable MMP-3 at the protein level, given the relative sensitivity of zymography (to 10 pg), suggests that if stromelysin is produced, it is not expressed at a high level.

In summary, the present study demonstrates that alveolar macrophages from rats produce a spectrum of MMPs that is similar (though not identical) to the spectrum of enzymes produced by their human counterparts. With regard to MMPs, rat and human alveolar macrophages appear to be more similar than are rat and human neutrophils (8, 9). Thus, our observation that macrophage MMPs play a role in acute lung inflammation in rats might suggest a similar role in human lung disease.