Plasminogen Activator Inhibitor Type 1 Production by Rat Type II Pneumocytes in Culture Lance A. Parton, David Warburton, and Walter E. Laug Developmental Lung Cell and Molecular Biology Research Center, Division of Neonatology and Pediatric Pulmonology and Division of Hematology-Oncology, Childrens Hospital of Los Angeles, and Department of Pediatrics, University of Southern California School of Medicine, Los Angeles, California, and Division of Neonatology, Department of Pediatrics, SUNY at Stony Brook School of Medicine, Stony Brook, New York
Co-secretion of plasminogen activator inhibitor type 1 (PAl-I) and urokinase-type plasminogen activator was identified in short-term cultures of primary type II pneumocytes isolated from adult rats. After separation by sodium dodecyl sulfate (SDS)-PAGEand reverse fibrin autography (reverse FA) of serum-free conditioned medium (SFCM), cellular lysate, and extracellular matrix (ECM), the inhibitor was seen as a zone of spared lysis at an apparent molecular mass of 46 to 48 kD. The plasminogen activator (PA) activity could only be visualized when human instead of bovine fibrin was used in the indicator gel. It presented as a single band of lysis at an apparent molecular mass of 45 kD when tested by regular FA and was found adjacent to PAI-I when examined by reverse FA. Immunoblot analysis oftype II pneumocyte SFCM, cellular lysate, and ECM revealed two bands at 46 and 48 kD, consistent with the apparent molecular masses (M,) reported for rat PAI-I from HTC hepatoma cells. Type II pneumocyte PAI-I formed SDS-resistant complexes with tissue-type and urokinase-type plasminogen activator and was found to be stable to acid, to short-term exposure to heat, and to the denaturants guanidine HCl and SDS, while being sensitive to treatment with alkali and urea. When levels of type II pneumocyte PAI-I activity were monitored over time during short-term culture conditions, the level of PAI-I in SFCM remained stable, whereas activity in the lysate accumulated and activity in the ECM declined. The PAI-I present in SFCM remained active over prolonged periods of time, and only minimal latent form, which could be activated by detergent, was found. We speculate that co-secretion of stabilizing factors such as negatively charged phospholipids or the association with vitroneetin keep the secreted PAI-I in active form. PAI-I is a novel type II pneumocyte product that may play an important role in the pathogenesis of pulmonary fibrin deposition.
The plasminogen activator (PA)-plasmin system plays an important role in many physiologic and pathologic processes including ovulation, tissue remodeling, inflammation, tumor cell invasion, and metastasis (for review, see reference 1). Two genetically and immunologically distinct types of PA, namely urokinase PA (uPA) and tissue type PA (tPA), have been described which activate the zymogen plasminogen to its active form plasmin. Multiple substrates including fibrin (Received in original form August 27. 1990 and infinalform July 22, 1991) Address correspondence to: Lance A. Parton, M.D., Department of Pediatrics, Division of Neonatology, SUNY at Stony Brook School of Medicine, HSC T-ll, Stony Brook, NY 11794-8111. Abbreviations: adult respiratory distress syndrome, ARDS; Dulbecco's modified Eagle's medium, DMEM; disaturated phosphatidylcholine, DSPC; extracellular matrix, ECM; fibrin autography, FA; plasminogen activator, PA; polyacrylamide gel electrophoresis, PAGE; plasminogen activator inhibitor, PAl; phosphatidylcholine, PC; soldium dodecyl sulfate, SDS; serum-free conditioned medium, SFCM; solid-phase fibrin-tPA activity assay, SOFIA-tPA; tissue-type plasminogen activator, tPA; urokinase-type plasminogen activator, uPA. Am. J. Respir. Cell Mol. BioI. Vol. 6. pp. 133-139, 1992
and extracellular matrix (ECM) proteins have been described for plasmin. The extracellular activities of the PAare tightly controlled by specific inhibitors released by many different cell types. Three distinct PA inhibitors (PAl) have been isolated and characterized, namely PAl-I, PAI-2, and protease-nexin I (for review, see reference 2). PAl-I, an almost ubiquitous inhibitor is produced by many different cell types and released in soluble form into the pericellular space to control local proteolysis induced by either uPA or tPA. In addition, anchorage-dependent cells also integrate PAI-I into the ECM, where it remains active for prolonged periods of time (3, 4). The secreted PAl-I, in contrast, is rapidly converted into a latent, inactive form that can be reactivated by exposure to detergents (5). Increased plasma levels of PAI-I have been found in various disease states such as thrombotic vascular diseases as well as in myocardial infarction (2). In fact, transgenic mice containing the human endothelial cell PAI-I complementary DNA under the control of the murine metallothionein I promoter were recently found to develop venous thrombi causing necrotic tails and swollen hind feet (6).
134
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
The lung is a highly vascularized organ that maintains a tenuous balance in its regulation of extravascular fibrinolysis, as exemplified by the number of pulmonary disease processes that result in alveolar fibrin deposition, such as pneumonitis, interstitial lung diseases (7-9), adult respiratory distress syndrome (ARDS) (10), and hyaline membrane disease (11, 12). Recent reports indicate a decreased bronchoalveolar uPA activity in patients with ARDS due to increased production of PAI-1 and antiplasmins (13). The source of PAI-1 in these disease states has not yet been determined, whereas type II pneumocytes have recently been shown to be a source of uPA (14, 15). In the present study, we document the production of PAI-1 in addition to uPA by rat type II pneumocytes in short-term primary culture. PAI-1 was found in the lysate of type II pneumocytes, in the serum-free culture medium (SFCM) of these cells, and in the ECM deposited on the culture dish by these pneumocytes. Thus, type II pneumocytes may be one of the PAI-1 sources responsible for impaired fibrin dissolution in ARDS.
Materials and Methods Materials were obtained from the following sources. Pathogen-free rats were obtained from Charles River Laboratories (Boston, MA), elastase from Worthington Laboratories (Freehold, NJ), and Dulbecco's modified Eagle's medium (DMEM) from Sigma Chemical Co. (St. Louis, MO). Plasticware was purchased from Falcon (Oxnard, CA). Fetal calf serum was purchased from Tissue Culture Biologicals (Thlare, CA), nitrocellulose membranes from Bio-Rad Laboratories (Richmond, CA), uPA from Abbott Laboratories (North Chicago, IL), thrombin from Parke-Davis (Morris Plains, NJ), and tPA from American Diagnostica (Greenwich, CT). Chromogenic substrate (D-val-Ieu-Iys-pNA) was from Sigma. Rat HTC hepatoma cell conditioned medium as well as antibody generated against HTC hepatoma cell PAI-1 were kind gifts from Drs. Ron Zeheb and Thomas D. Gelehrter (University of Michigan Medical School, Ann Arbor, MI). Purified human PAI-1 was kindly provided by Dr. AnglesCano (lNSERM U-152, Paris, France). Recombinant tPA was a gift from Genentech (South San Francisco, CA). All other biochemicals of the highest grade available were obtained from Sigma. Primary Culture of Type II Pneumocytes Type II pneumocytes were isolated from male 250- to 300-g, pathogen-free rats following the protocol of Dobbs and coworkers (16), which includes lung lavage followed by elastase digestion and mechanical dissociation of the cells. After differential IgG adherence, type II pneumocytes were harvested with> 90% viability as shown by trypan blue staining and with> 90% purity as demonstrated with phosphene 3R fluorescent staining. Alveolar macrophages comprised < 5 % of the cells plated. The cells were seeded at 2.5 to 3.5 X 1Q5 cells/well in 12-well plates and grown in DMEM containing 10% fetal calf serum, penicillin (100 ttg/ml), and streptomycin (50 U/ml), and fed every other day. On the days indicated, the cultures were washed twice with phosphate-buffered saline and kept for 24 h in serum-free medium. After harvesting the SFCM, cellular lysates and ECM were obtained. SFCM was collected and centrifuged at 500 x g for 4 min to remove
nonadherent cells and frozen at -80 0 C until testing. Cell counts were obtained after trypsinization of the cultures. ECM were obtained after lysis of the cells with 0.025 M ammonium hydroxide as reported (17). Cell Iysates were the postnuclear fraction of cultures lysed in 10 mM Tris (pH 8.0) containing 0.5 % deoxycholate and 1 mM phenylmethylsulfonyl fluoride as described (18). Synthesis and Release of Phosphatidylcholine (PC) by Type II Pneumocytes Cells were incubated for 24 h with PH]choline 1.0 ttCi/ well, then washed 6 times with serum-free DMEM and further incubated for 3 h at 370 C. The medium was collected, centrifuged at 500 X g for 4 min to pellet loose cells, and frozen at -70 0 C. Phospholipids were extracted from cells and media as described by Bligh and Dyer (19). PH] dpm associated with phospholipid was shown to be > 98 % PC, and the percent release of PC approximated the percentage of dipalmitoyl PC per well (data not shown) (20). Cell viability was measured by lactate dehydrogenase release into the medium (LDL-20 kit; Sigma) and levels measured were < I% of the total cellular enzyme activity in all studies reported herein. Reverse Fibrin Autography (Reverse FA) In essence, the method of Loskutoff and associates was used (21). Bovine fibrinogen (Type I-S; Sigma) was purified as described (22), and plasminogen was isolated form outdated human plasma by the method of Deutsch and Mertz (23). The samples were separated on sodium dodecyl sulfate (SDS)-polyacrylamide mini-gels with 4 % stacking and 8% separating gels (24). The SDS was removed by incubation with 2.5% Triton X-IOO before reverse FA (21). PAl activity was detected as a zone of spared lysis. Human fibrinogen (Type I, Fraction I; Sigma) was used instead of bovine material to test for the presence of uPA. Reverse Solid-phase Fibrin-tPA Activity Assay (Reverse SOFIA-tPA) Quantitation of free inhibitor present in samples was performed using a variation of the SOFIA-tPA assay as described by Angeles-Cano (25, 26). Glutaraldehyde cross-linking of fibrinogen to U-bottom wells provides the substrate for thrombin to convert this to a linked fibrin. After preincubation of sample with a known amount of tPA, this mixture was added to the fibrin-linked well. Residual tPA complexed with this linked fibrin so that upon addition of plasminogen, plasmin was generated, which subsequently converted the chromogenic substrate (D-val-Ieu-Iys-pNA) to a product visible at 405 nm. Initial rates of reaction were obtained by plotting A405 versus minutes squared. Known tPA samples were assayed in triplicate over a range of values. Inhibitory activity of samples were interpolated by a shift in the half-maximal saturation point between inhibitor and tPA (where 1.4 nM = 45 IV/ml). One PAI-l unit was defined as the amount of inhibitor required to shift this curve by 1 U of tPA activity at the point of half-maximal binding. Fibrin Plate Analysis Relative values of fibrinolysis inhibition were determined by the fibrin plate method modified from that described by Laug
Parton, Warburton, and Laug: Plasminogen Activator Inhibitor Type I Production
(27). Samples were preincubated with 0.05 U of uPA followed by determination of fibrinolysis measured by 1251 release. Percent inhibition of fibrinolysis was determined by comparing the radioactivity released from pre incubated samples with that released by 0.05 U of uPA alone (= 0% inhibition). Background counts released by preincubation with sample buffer were subtracted before calculation of percentage. The samples of SFCM, ECM, and cellular lysate were normalized for cell number. Casein Overlay After separation of the samples by SDS polyacrylamide gel electrophoresis (PAGE), and removal of the SDS with Triton X-lOO, the gels were placed on a casein overlay as described by Vassalli and colleagues (28). Immunoblotting After separation of the samples by SDS-PAGE, the proteins were electrophoretically transferred onto nitrocellulose membranes. The presence of PAl-I was visualized with rabbit anti-rat PAl-I serum (29) (developed against rat HTC hepatoma PAl-I by Drs. Zeheb and Gelehrter) followed by goat-anti-rabbit antibody conjugated to alkaline phosphatase. Color development was achieved with nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate (p-toluidine salt). Denaturations for SDS-PAGE Aliquots of SFCM were either heated to 80 0 C for 5 min; acidified with HCI to pH 2 for 60 min, followed by a return to pH 7; treated with NaOH to pH 12 for 60 min, followed by a return to pH 7; made 4 M with guanidine HCI for 60 min; treated with I M urea for 60 min; or made 1% with respect to SDS. After treatment, samples (including untreated ones) were dialyzed exhaustively against phosphate-buffered saline at 4 0 C. Samples containing equal amounts of total protein (Bio-Rad protein assay; Bio-Rad) were then analyzed by SDS-PAGE followed by reverse FA. Additionally, samples of SFCM were treated with denaturants and analyzed in triplicate over a range of protein concentrations by reverse SOFIA-tPA as done by Angles-Cano and co-workers (25, 26).
Results The morphologic appearance of rat type II pneumocytes in short-term cultures is shown in Figure 1. The cells were initially rounded in appearance (days 1 and 2), then spread out to cover the dish and became cuboidal (days 3 to 5). Thereafter, the cells dedifferentiated to spindle cells and they lost the ability to convert PC into saturated forms. These observations are similar to those reported previously by others (30). To assess proper functional activity of the cells, [3H]choline incorporation into cellular PC and disaturated phosphatidylcholine (DSPC) (Figure 2) and the percent release of PC and DSPC into the medium were determined. Culture on plastic for longer than 7 days resulted in a morphologic change from cuboidal to fibroblastic appearance as well as a decline of [3H]choline incorporation into PC (data not shown). Therefore, all experiments were done on primary cultures between days I and 5 after seeding. Screening for PAl activity from type II pneumocyte cultures was performed by reverse FA on SFCM, ECM, and cell
135
lysate. Broad bands of inhibition of lysis were found in SFCM and ECM with considerably smaller amounts in the cell Iysates with an approximate molecular mass of 48 kD when bovine fibrinogen was used in the indicator gel (data not shown). No PA activities were detectable by either reverse or regular FA. Because bovine fibrinogen has been found to be a poor substrate for rat PA (1. 1. Emeis, Leiden, The Netherlands), human fibrinogen was used in the indicator gel. Clear bands of lysis at a molecular mass of 45 kD were found in SFCM and cell lysate when tested by regular fibrin overlay (indicator gel without urokinase) as shown in Figure 3A. When tested by reverse FA, lysis bands appeared within 4 h of incubation in SFCM and cell lysate in front of the broad bands of inhibition (Figure 3B). After further incubation (total 8 h), persistence of zones of inhibition was documented with the remainder of the gel completely cleared (Figure 3C). No PA activities were detected in the ECM. The relative distribution of functional PA inhibitory activity between SFCM, ECM, and celllysates of type II pneumocytes is shown in Figure 4. There was a fairly constant inhibitory activity released into the SFCM by the cells over a 4-day period as measured by either the reverse SOFIA-tPA or the fibrin plate test. In contrast, the inhibitory activity in cell Iysates quadrupled over 4 days whereas the level found in ECM decreased almost 3-fold over the same time period. Because of the low protein content in the ECM preparation of these short-term cultures, this decrease of PAl activity in ECM has to be interpreted with caution. The PAl activity in SFCM treated with denaturants appeared unchanged when samples were analyzed by SDSPAGE followed by reverse FA (Table I). This could have been due to the maximal activation of PAl in all of the samples following exposure to SDS. Because these characteristics suggested the presence of PAl-I, SFCM, ECM, and cellular lysate were analyzed by Western blotting techniques using specific anti-rat PAl-I antiserum. The SFCM, cellular lysate, the ECM revealed bands at 46 and 48 kD, with the 48-kD species predominating (Figure 5). A more diffuse background of crossreactivity was visualized in the lane of cellular lysate, as this lane was overloaded with protein. The bands at 48 and 46 kD correspond to the zone of inhibition observed with reverse FA (data not shown). To further substantiate these findings, SFCM of rat II pneumocytes was incubated with either uPA or recombinant-type tPA and tested for SDS-resistant PA-PAI-I complex formation using the casein overlay method (28). Rat HTC hepatoma cell SFCM was included as a positive control. PA-PAI-I complexes retain partial proteolytic activity allowing for their detection after SDS-PAGE by fibrin or casein overlay (27, 28). The lytic activity of recombinant tPA was found at an M, of 60 kD whereas complexes of rat PA with PAI-I were detected at an approximate M, of 105 and 100 kD (Figure 6, upper panel). Complexes were found after incubation with low-molecular-mass uPA at approximately 84 kD (Figure 6, lower panel). Similar results were obtained when radioiodinated PA preparations were used (data not shown). To test for the presence of latent PAl-I secreted from type II pneumocytes, SFCM from type II pneumocytes was exposed to a number of denaturants, which have previously been shown to activate PAl-I (5) found in other sources
136
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
Figure 1. Cell morphology of adult rat type II pneumocytes. Cells were isolated as described, seeded at 3 X 105 cells/well in a l2-well plate, and grown for up to 5 days.
(Table 1). After dialysis, quantitative analysis by reverse SOFIA-tPA failed to reveal supra-activation of PAl treated with denaturants, but showed resistance to degradation after acid, heat, and treatment with guanidine HCl and SDS, while alkalinization and treatment with urea diminished PAl activity by about half. In addition, repeated thawing and freezing of samples for prolonged periods of time (3 yr) did not change the PAl activity (data not shown), indicating an absence oflatent PAl-lor an absence in the conversion of active PAl into a latent form.
•
c1
[
o cc w
CD
::>
::>
\\
\
~
1C
I
~
T
~
Co-secretion of plasminogen activator inhibitor type 1 (PAl-I) and urokinase-type plasminogen activator was identified in short-term cultures of primary type II pneumocytes isolated from adult rats. After separation by sodium dodecyl sulfate (SDS)-PAGEand reverse fibrin autography (reverse FA) of serum-free conditioned medium (SFCM), cellular lysate, and extracellular matrix (ECM), the inhibitor was seen as a zone of spared lysis at an apparent molecular mass of 46 to 48 kD. The plasminogen activator (PA) activity could only be visualized when human instead of bovine fibrin was used in the indicator gel. It presented as a single band of lysis at an apparent molecular mass of 45 kD when tested by regular FA and was found adjacent to PAI-I when examined by reverse FA. Immunoblot analysis oftype II pneumocyte SFCM, cellular lysate, and ECM revealed two bands at 46 and 48 kD, consistent with the apparent molecular masses (M,) reported for rat PAI-I from HTC hepatoma cells. Type II pneumocyte PAI-I formed SDS-resistant complexes with tissue-type and urokinase-type plasminogen activator and was found to be stable to acid, to short-term exposure to heat, and to the denaturants guanidine HCl and SDS, while being sensitive to treatment with alkali and urea. When levels of type II pneumocyte PAI-I activity were monitored over time during short-term culture conditions, the level of PAI-I in SFCM remained stable, whereas activity in the lysate accumulated and activity in the ECM declined. The PAI-I present in SFCM remained active over prolonged periods of time, and only minimal latent form, which could be activated by detergent, was found. We speculate that co-secretion of stabilizing factors such as negatively charged phospholipids or the association with vitroneetin keep the secreted PAI-I in active form. PAI-I is a novel type II pneumocyte product that may play an important role in the pathogenesis of pulmonary fibrin deposition.
The plasminogen activator (PA)-plasmin system plays an important role in many physiologic and pathologic processes including ovulation, tissue remodeling, inflammation, tumor cell invasion, and metastasis (for review, see reference 1). Two genetically and immunologically distinct types of PA, namely urokinase PA (uPA) and tissue type PA (tPA), have been described which activate the zymogen plasminogen to its active form plasmin. Multiple substrates including fibrin (Received in original form August 27. 1990 and infinalform July 22, 1991) Address correspondence to: Lance A. Parton, M.D., Department of Pediatrics, Division of Neonatology, SUNY at Stony Brook School of Medicine, HSC T-ll, Stony Brook, NY 11794-8111. Abbreviations: adult respiratory distress syndrome, ARDS; Dulbecco's modified Eagle's medium, DMEM; disaturated phosphatidylcholine, DSPC; extracellular matrix, ECM; fibrin autography, FA; plasminogen activator, PA; polyacrylamide gel electrophoresis, PAGE; plasminogen activator inhibitor, PAl; phosphatidylcholine, PC; soldium dodecyl sulfate, SDS; serum-free conditioned medium, SFCM; solid-phase fibrin-tPA activity assay, SOFIA-tPA; tissue-type plasminogen activator, tPA; urokinase-type plasminogen activator, uPA. Am. J. Respir. Cell Mol. BioI. Vol. 6. pp. 133-139, 1992
and extracellular matrix (ECM) proteins have been described for plasmin. The extracellular activities of the PAare tightly controlled by specific inhibitors released by many different cell types. Three distinct PA inhibitors (PAl) have been isolated and characterized, namely PAl-I, PAI-2, and protease-nexin I (for review, see reference 2). PAl-I, an almost ubiquitous inhibitor is produced by many different cell types and released in soluble form into the pericellular space to control local proteolysis induced by either uPA or tPA. In addition, anchorage-dependent cells also integrate PAI-I into the ECM, where it remains active for prolonged periods of time (3, 4). The secreted PAl-I, in contrast, is rapidly converted into a latent, inactive form that can be reactivated by exposure to detergents (5). Increased plasma levels of PAI-I have been found in various disease states such as thrombotic vascular diseases as well as in myocardial infarction (2). In fact, transgenic mice containing the human endothelial cell PAI-I complementary DNA under the control of the murine metallothionein I promoter were recently found to develop venous thrombi causing necrotic tails and swollen hind feet (6).
134
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
The lung is a highly vascularized organ that maintains a tenuous balance in its regulation of extravascular fibrinolysis, as exemplified by the number of pulmonary disease processes that result in alveolar fibrin deposition, such as pneumonitis, interstitial lung diseases (7-9), adult respiratory distress syndrome (ARDS) (10), and hyaline membrane disease (11, 12). Recent reports indicate a decreased bronchoalveolar uPA activity in patients with ARDS due to increased production of PAI-1 and antiplasmins (13). The source of PAI-1 in these disease states has not yet been determined, whereas type II pneumocytes have recently been shown to be a source of uPA (14, 15). In the present study, we document the production of PAI-1 in addition to uPA by rat type II pneumocytes in short-term primary culture. PAI-1 was found in the lysate of type II pneumocytes, in the serum-free culture medium (SFCM) of these cells, and in the ECM deposited on the culture dish by these pneumocytes. Thus, type II pneumocytes may be one of the PAI-1 sources responsible for impaired fibrin dissolution in ARDS.
Materials and Methods Materials were obtained from the following sources. Pathogen-free rats were obtained from Charles River Laboratories (Boston, MA), elastase from Worthington Laboratories (Freehold, NJ), and Dulbecco's modified Eagle's medium (DMEM) from Sigma Chemical Co. (St. Louis, MO). Plasticware was purchased from Falcon (Oxnard, CA). Fetal calf serum was purchased from Tissue Culture Biologicals (Thlare, CA), nitrocellulose membranes from Bio-Rad Laboratories (Richmond, CA), uPA from Abbott Laboratories (North Chicago, IL), thrombin from Parke-Davis (Morris Plains, NJ), and tPA from American Diagnostica (Greenwich, CT). Chromogenic substrate (D-val-Ieu-Iys-pNA) was from Sigma. Rat HTC hepatoma cell conditioned medium as well as antibody generated against HTC hepatoma cell PAI-1 were kind gifts from Drs. Ron Zeheb and Thomas D. Gelehrter (University of Michigan Medical School, Ann Arbor, MI). Purified human PAI-1 was kindly provided by Dr. AnglesCano (lNSERM U-152, Paris, France). Recombinant tPA was a gift from Genentech (South San Francisco, CA). All other biochemicals of the highest grade available were obtained from Sigma. Primary Culture of Type II Pneumocytes Type II pneumocytes were isolated from male 250- to 300-g, pathogen-free rats following the protocol of Dobbs and coworkers (16), which includes lung lavage followed by elastase digestion and mechanical dissociation of the cells. After differential IgG adherence, type II pneumocytes were harvested with> 90% viability as shown by trypan blue staining and with> 90% purity as demonstrated with phosphene 3R fluorescent staining. Alveolar macrophages comprised < 5 % of the cells plated. The cells were seeded at 2.5 to 3.5 X 1Q5 cells/well in 12-well plates and grown in DMEM containing 10% fetal calf serum, penicillin (100 ttg/ml), and streptomycin (50 U/ml), and fed every other day. On the days indicated, the cultures were washed twice with phosphate-buffered saline and kept for 24 h in serum-free medium. After harvesting the SFCM, cellular lysates and ECM were obtained. SFCM was collected and centrifuged at 500 x g for 4 min to remove
nonadherent cells and frozen at -80 0 C until testing. Cell counts were obtained after trypsinization of the cultures. ECM were obtained after lysis of the cells with 0.025 M ammonium hydroxide as reported (17). Cell Iysates were the postnuclear fraction of cultures lysed in 10 mM Tris (pH 8.0) containing 0.5 % deoxycholate and 1 mM phenylmethylsulfonyl fluoride as described (18). Synthesis and Release of Phosphatidylcholine (PC) by Type II Pneumocytes Cells were incubated for 24 h with PH]choline 1.0 ttCi/ well, then washed 6 times with serum-free DMEM and further incubated for 3 h at 370 C. The medium was collected, centrifuged at 500 X g for 4 min to pellet loose cells, and frozen at -70 0 C. Phospholipids were extracted from cells and media as described by Bligh and Dyer (19). PH] dpm associated with phospholipid was shown to be > 98 % PC, and the percent release of PC approximated the percentage of dipalmitoyl PC per well (data not shown) (20). Cell viability was measured by lactate dehydrogenase release into the medium (LDL-20 kit; Sigma) and levels measured were < I% of the total cellular enzyme activity in all studies reported herein. Reverse Fibrin Autography (Reverse FA) In essence, the method of Loskutoff and associates was used (21). Bovine fibrinogen (Type I-S; Sigma) was purified as described (22), and plasminogen was isolated form outdated human plasma by the method of Deutsch and Mertz (23). The samples were separated on sodium dodecyl sulfate (SDS)-polyacrylamide mini-gels with 4 % stacking and 8% separating gels (24). The SDS was removed by incubation with 2.5% Triton X-IOO before reverse FA (21). PAl activity was detected as a zone of spared lysis. Human fibrinogen (Type I, Fraction I; Sigma) was used instead of bovine material to test for the presence of uPA. Reverse Solid-phase Fibrin-tPA Activity Assay (Reverse SOFIA-tPA) Quantitation of free inhibitor present in samples was performed using a variation of the SOFIA-tPA assay as described by Angeles-Cano (25, 26). Glutaraldehyde cross-linking of fibrinogen to U-bottom wells provides the substrate for thrombin to convert this to a linked fibrin. After preincubation of sample with a known amount of tPA, this mixture was added to the fibrin-linked well. Residual tPA complexed with this linked fibrin so that upon addition of plasminogen, plasmin was generated, which subsequently converted the chromogenic substrate (D-val-Ieu-Iys-pNA) to a product visible at 405 nm. Initial rates of reaction were obtained by plotting A405 versus minutes squared. Known tPA samples were assayed in triplicate over a range of values. Inhibitory activity of samples were interpolated by a shift in the half-maximal saturation point between inhibitor and tPA (where 1.4 nM = 45 IV/ml). One PAI-l unit was defined as the amount of inhibitor required to shift this curve by 1 U of tPA activity at the point of half-maximal binding. Fibrin Plate Analysis Relative values of fibrinolysis inhibition were determined by the fibrin plate method modified from that described by Laug
Parton, Warburton, and Laug: Plasminogen Activator Inhibitor Type I Production
(27). Samples were preincubated with 0.05 U of uPA followed by determination of fibrinolysis measured by 1251 release. Percent inhibition of fibrinolysis was determined by comparing the radioactivity released from pre incubated samples with that released by 0.05 U of uPA alone (= 0% inhibition). Background counts released by preincubation with sample buffer were subtracted before calculation of percentage. The samples of SFCM, ECM, and cellular lysate were normalized for cell number. Casein Overlay After separation of the samples by SDS polyacrylamide gel electrophoresis (PAGE), and removal of the SDS with Triton X-lOO, the gels were placed on a casein overlay as described by Vassalli and colleagues (28). Immunoblotting After separation of the samples by SDS-PAGE, the proteins were electrophoretically transferred onto nitrocellulose membranes. The presence of PAl-I was visualized with rabbit anti-rat PAl-I serum (29) (developed against rat HTC hepatoma PAl-I by Drs. Zeheb and Gelehrter) followed by goat-anti-rabbit antibody conjugated to alkaline phosphatase. Color development was achieved with nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate (p-toluidine salt). Denaturations for SDS-PAGE Aliquots of SFCM were either heated to 80 0 C for 5 min; acidified with HCI to pH 2 for 60 min, followed by a return to pH 7; treated with NaOH to pH 12 for 60 min, followed by a return to pH 7; made 4 M with guanidine HCI for 60 min; treated with I M urea for 60 min; or made 1% with respect to SDS. After treatment, samples (including untreated ones) were dialyzed exhaustively against phosphate-buffered saline at 4 0 C. Samples containing equal amounts of total protein (Bio-Rad protein assay; Bio-Rad) were then analyzed by SDS-PAGE followed by reverse FA. Additionally, samples of SFCM were treated with denaturants and analyzed in triplicate over a range of protein concentrations by reverse SOFIA-tPA as done by Angles-Cano and co-workers (25, 26).
Results The morphologic appearance of rat type II pneumocytes in short-term cultures is shown in Figure 1. The cells were initially rounded in appearance (days 1 and 2), then spread out to cover the dish and became cuboidal (days 3 to 5). Thereafter, the cells dedifferentiated to spindle cells and they lost the ability to convert PC into saturated forms. These observations are similar to those reported previously by others (30). To assess proper functional activity of the cells, [3H]choline incorporation into cellular PC and disaturated phosphatidylcholine (DSPC) (Figure 2) and the percent release of PC and DSPC into the medium were determined. Culture on plastic for longer than 7 days resulted in a morphologic change from cuboidal to fibroblastic appearance as well as a decline of [3H]choline incorporation into PC (data not shown). Therefore, all experiments were done on primary cultures between days I and 5 after seeding. Screening for PAl activity from type II pneumocyte cultures was performed by reverse FA on SFCM, ECM, and cell
135
lysate. Broad bands of inhibition of lysis were found in SFCM and ECM with considerably smaller amounts in the cell Iysates with an approximate molecular mass of 48 kD when bovine fibrinogen was used in the indicator gel (data not shown). No PA activities were detectable by either reverse or regular FA. Because bovine fibrinogen has been found to be a poor substrate for rat PA (1. 1. Emeis, Leiden, The Netherlands), human fibrinogen was used in the indicator gel. Clear bands of lysis at a molecular mass of 45 kD were found in SFCM and cell lysate when tested by regular fibrin overlay (indicator gel without urokinase) as shown in Figure 3A. When tested by reverse FA, lysis bands appeared within 4 h of incubation in SFCM and cell lysate in front of the broad bands of inhibition (Figure 3B). After further incubation (total 8 h), persistence of zones of inhibition was documented with the remainder of the gel completely cleared (Figure 3C). No PA activities were detected in the ECM. The relative distribution of functional PA inhibitory activity between SFCM, ECM, and celllysates of type II pneumocytes is shown in Figure 4. There was a fairly constant inhibitory activity released into the SFCM by the cells over a 4-day period as measured by either the reverse SOFIA-tPA or the fibrin plate test. In contrast, the inhibitory activity in cell Iysates quadrupled over 4 days whereas the level found in ECM decreased almost 3-fold over the same time period. Because of the low protein content in the ECM preparation of these short-term cultures, this decrease of PAl activity in ECM has to be interpreted with caution. The PAl activity in SFCM treated with denaturants appeared unchanged when samples were analyzed by SDSPAGE followed by reverse FA (Table I). This could have been due to the maximal activation of PAl in all of the samples following exposure to SDS. Because these characteristics suggested the presence of PAl-I, SFCM, ECM, and cellular lysate were analyzed by Western blotting techniques using specific anti-rat PAl-I antiserum. The SFCM, cellular lysate, the ECM revealed bands at 46 and 48 kD, with the 48-kD species predominating (Figure 5). A more diffuse background of crossreactivity was visualized in the lane of cellular lysate, as this lane was overloaded with protein. The bands at 48 and 46 kD correspond to the zone of inhibition observed with reverse FA (data not shown). To further substantiate these findings, SFCM of rat II pneumocytes was incubated with either uPA or recombinant-type tPA and tested for SDS-resistant PA-PAI-I complex formation using the casein overlay method (28). Rat HTC hepatoma cell SFCM was included as a positive control. PA-PAI-I complexes retain partial proteolytic activity allowing for their detection after SDS-PAGE by fibrin or casein overlay (27, 28). The lytic activity of recombinant tPA was found at an M, of 60 kD whereas complexes of rat PA with PAI-I were detected at an approximate M, of 105 and 100 kD (Figure 6, upper panel). Complexes were found after incubation with low-molecular-mass uPA at approximately 84 kD (Figure 6, lower panel). Similar results were obtained when radioiodinated PA preparations were used (data not shown). To test for the presence of latent PAl-I secreted from type II pneumocytes, SFCM from type II pneumocytes was exposed to a number of denaturants, which have previously been shown to activate PAl-I (5) found in other sources
136
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
Figure 1. Cell morphology of adult rat type II pneumocytes. Cells were isolated as described, seeded at 3 X 105 cells/well in a l2-well plate, and grown for up to 5 days.
(Table 1). After dialysis, quantitative analysis by reverse SOFIA-tPA failed to reveal supra-activation of PAl treated with denaturants, but showed resistance to degradation after acid, heat, and treatment with guanidine HCl and SDS, while alkalinization and treatment with urea diminished PAl activity by about half. In addition, repeated thawing and freezing of samples for prolonged periods of time (3 yr) did not change the PAl activity (data not shown), indicating an absence oflatent PAl-lor an absence in the conversion of active PAl into a latent form.
•
c1
[
o cc w
CD
::>
::>
\\
\
~
1C
I
~
T
~
