Bleomycin Regulation of Transforming Growth Factor-S mRNA in Rat Lung Fibroblasts Ellen Breen, Susan Shull, Sandra Burne, Marlene Absher, Jason Kelley, Sem Phan, and Kenneth R. Cutroneo Departments of Biochemistry and Medicine, College of Medicine, University of Vermont, Burlington, Vermont, and Department of Pathology, University of Michigan, School of Medicine, Ann Arbor, Michigan
Pulmonary fibrosis is a well-known toxic response to bleomycin treatment. Here we demonstrate the direct effects of bleomycin on lung fibroblasts that resulted in a marked increase of collagen synthesis as compared with total noncollagen protein synthesis. Bleomycin treatment of rat lung fibroblast cultures resulted in an increase of total cellular transforming growth factor-S (TGF-{3) mRNA and increased secretion of TGF-{3 protein into the conditioned media. (32-Microglobulin was measured as an mRNA that did not increase with bleomycin treatment. The bleomycin-induced increase of TGF-13 mRNA was decreased by cells cultured in the presence of either cycloheximide, an inhibitor of protein synthesis, or 2-mercapto-l-(13-4pyridethyl) benzimidazole, an inhibitor of RNA synthesis. To assess the mechanism underlying increased steady-state mRNA levels, the nuclear fraction was isolated from bleomycin-treated cells and the TGF-13 transcripts were determined. Transcription of TGF-{3 mRNA was increased 12h after bleomycintreatment, whereas the transcription of type I procollagen, type III procollagen, and l3-actin mRNAs were increased after 48 h of bleomycin treatment. f32-Microglobulin mRNA synthesis was not increased within this time frame. These results suggest bleomycin regulation of TGF-13 at both the mRNA and protein levels. Rat lung fibroblasts were separated by cell sorting into two subpopulations. One population of fibroblasts demonstrated increased procollagen type I mRNAs, whereas fibroblasts in the other population had increased procollagen type III mRNA. Following bleomycin treatment, TGF-13 mRNA was shown to be located more prominently in those fibroblasts that contain primarily collagen type I mRNAs.
Bleomycin, an antineoplastic glycopeptide isolated from Streptomyces verticil/us, is used to treat several neoplasms including squamous cell carcinoma of the head and neck, lymphomas, and testicular carcinomas (1). Pulmonary fibrosis is a major side effect of bleomycin treatment (2-7). Fibrosis occurs in four phases, which include initial injury, an inflammatory phase, a proliferative phase of connective tissue cells and other cells, and a remodeling-repair phase (2-7). Bleomycin-induced lung fibrosis is a complex process involving cell-cell interactions, stimulatory and inhibitory factors, morphologic changes, immunologic changes, biochemical changes, and the participation of many cell types. Increased synthesis and deposition of collagen result in increased lung collagen content and distortion of the pulmonary structural architecture (2-7). Increased collagen deposition in lung results from al(Received in original form January 25. 1991 and in final form August 1. 1991) Address correspondence to: Kenneth R. Cutroneo, Ph.D., Department of Biochemistry, College of Medicine, University of Vermont, Health Science Complex, Given Building, Burlington, VT 05405. Abbreviations: 2-mercapto-l-(iJ-4-pyridethyl) benzimidazole, MPB; sodium dodecyl sulfate, SDS; transforming growth factor-S, TGF-iJ. Am. J. Respir. Cell Mol. BioI. Vol. 6. pp. 146-152, 1992
terations of both the synthesis and degradation of newly synthesized collagen (8). In bleomycin-induced fibrotic lung, collagen synthesis is selectively increased while collagen degradation is decreased (8). Procollagen type I and procollagen type III steady-state mRNAs are elevated in bleomycininduced interstitial pulmonary fibrosis (9, 10). These changes in type I mRNA accumulation are associated with an increase in the rates of procollagen gene transcription (10). Fibroblasts cultured in the presence of bleomycin demonstrate increased collagen synthesis (11-13), and decreased cell growth (14-16). In addition, type III collagen is selectively decreased as compared with type I collagen (11). Five isoforms of transforming growth-S (TGF-{3) and various receptors are synthesized by many different cell types (17). TGF-{3 acts at a variety of cellular sites, displaying various biologic activities such as regulation of protein synthesis (17). In this study, the effects of TGF-{3, on collagen synthesis are examined. TGF-{3 may stimulate human dermal fibroblast transcription, secretion, and processing of total collagen. This processing may involve the co-translational and post-translational events during collagen synthesis (18). The elevation of colla- . gen synthesis by TGF-{3 is associated with increased cellular steady-state levels of both type I and type III collagen mRNAs in human dermal fibroblasts (19, 20) and in human
Breen, Shull, Burne et al.: Bleomycin and TGF-J3 mRNA
147
lung fibroblasts (21). The production of type I and type III collagens is also significantly increased in human lung fibroblasts treated with TGF-J3 (22). In NRK-49F rat fibroblasts, TGF-J3 increased the levels of type I procollagen and fibronectin mRNAs, which were inhibited by actinomycin D but not by cycloheximide (23). In NIH 3T3 mouse fibroblasts, however, cycloheximide inhibits the TGF-J3 stimulation of type I procollagen mRNAs (24). Furthermore, during confluent growth, pro« 1(1) collagen mRNA stability occurs, whereas in subconfluent cells this mRNA stability is not evident (24). Finally, TGF-J3 increases the activity of the proa2(I) procollagen promoter at least partially through the NF-l site (25). In relation to bleomycin-induced pulmonary fibrosis, TGF-J3 mRNA is increased in lung homogenates before type I and type III procollagen mRNAs (26, 27). Khalil and coworkers (28) demonstrated that when TGF-J3 is primarily associated with the extracellular matrix, there is maximal collagen synthesis in explanted fibroblasts. However, early in the bleomycin-induced injury there is immunohistochemical evidence of TGF-J3 being localized in alveolar macrophages, coinciding with a minimal increase in collagen synthesis. As yet, the cell(s) responsible for the bleomycin-induced increase of TGF-J3 mRNA have not been identified. The objective of the present studies was to determine the ability of bleomycin to increase TGF-J3 mRNA and to demonstrate the secretion of TGF-J3 from rat lung fibroblasts. These studies were designed to determine whether bleomycin directly increased TGF-J3 gene expression in tandem with procollagen gene expression. A second objective was to determine the source of TGF-J3 mRNA and TGF-J3 protein in fibroblasts of bleomycin-induced fibrotic lung (26, 27). Bleomycin was added to fibroblasts that mainly secrete type I collagen. Bleomycin was added to fibroblast cultures to determine the effect on TGF-J3 mRNA without interference of other cell types. The dose in vivo was greater because bleomycin was localized in the lung by endotracheal injection. The maximum dose used to treat fibroblast cultures in the present studies was 3 1Lg/ml of bleomycin for 24 h, the dose that gives the maximal increase of TGF-J3 mRNA.
determine the levels of noncollagen and procollagen synthesis in the cell layer.
Materials and Methods Each experiment was performed at least twice with similar results and minimal variability from experiment to experiment. Collagen and Noncollagen Protein Synthesis Rat lung fibroblasts (RL87) were isolated as described by Breen and associates (29). The cells were supplemented daily with 10-5 M ascorbic acid. Confluent rat lung fibroblasts cultured for 4 days were incubated with various doses of porcine TGF-J3I (R & D Systems, Minneapolis, MN), ranging from 0.1 to 10 ng/ml for 24 h. Three hours before the cells were harvested with 0.25 % trypsin:phosphatebuffered saline solution, L-[5-3H]proline (Amersham Corp., Arlington Heights, IL) was added to each culture at a concentration of 10 1LCi/mi. Harvested cells were digested with bacterial collagenase form III (Advance Biofactures, Lynbrook, NY) as described by Newman and Cutroneo (30) to
Recombinant DNA The porcine TGF-J3, cDNA recombinant plasmid, pTGF1333, was a generous gift of Dr. Michael Sporn (31). The mouse J32-microglobulin cDNA plasmid, pBRcB-3, was a gift of Dr. Jane R. Parnes (32). The ratproal(I) and proa2(I) recombinant plasmids, palRI and pa2R2, respectively, were generous gifts of Dr. David Rowe (33). The mouse genomic proo l(III) plasmid, pMSC-l, was a gift of Dr. Benoit de Crombrugghe (34). The recombinant plasmids were characterized by restriction mapping and sequence analysis. For slot blot hybridizations, the plasmids were digested with the appropriate restriction enzymes and electrophoresed in 0.8% low melting point agarose with 0.5 ng/ml ethidium bromide. The insert bands were excised from the gel, boiled, and stored at -20° C. The DNAs were subsequently oligolabeled by the method of Feinberg and Vogelstein (35). Recombinant DNA techniques were performed under PI containment. Isolation of Total Cellular RNA The single-step method of RNA isolation using acid guanidinium thiocynate-phenol-chloroform extraction was used to isolate total cellular RNA from rat lung fibroblasts (36). Slot Blot Hybridization One microgram of total cellular RNA was directly spotted onto nitrocellulose using a Minifold II Slot Blot System (Schleicher & Schuell, Keene, NH). The nitrocellulose was baked for 2 h at 80° C and hybridized. The (32P]dCTP oligolabeled probes had a specific activity > 1.0 x 109 cpm/ug. Blots were washed under high stringency conditions and autoradiographed. Band intensities were quantified by densitometric scanning using a dual-wavelength thin-layer chromatography scanner (Model CS-930; Shimadzu Corp., Kyoto, Japan). The RNA synthesis inhibitor 2-mercapto-l(J3-4-pyridethyl) benzimidazole (MPB) (37, 38) was used at 10-5 M, and cellular protein was determined (39). Cycloheximide, at 10-4 M, was used to inhibit protein synthesis. Northern Blot Analysis of TGF-J3 mRNA from Control and Bleomycin-treated Rat Lung Fibroblasts Northern blot analysis was performed to determine if bleomycin increased the TGF-J3 mRNA in rat lung fibroblasts. Other cells were treated with desamido bleomycin, which was kindly supplied by Dr. John Lazo. This bleomycin metabolite had no significant effect on increasing TGF-J3 mRNA in rat lung fibroblasts (data not shown). Total cellular RNA (15 1Lg) was denatured and electrophoresed in a 0.8% agarose/formaldehyde gel, which was stained with 0.5 ng/ml ethidium bromide to determine intact 28S rRNA and l8S rRNA. The RNA is then transferred to nitrocellulose, baked at 80° C for 2 h, and hybridized to nick-translated pTGFJ333 plasmid (40). The samples were prehybridized at 42° C for 24 h in a buffer containing 50% (vol/vol) deionized formamide, 5x SSPE (pH 7.4), 0.025% (wt/vol) of each bovine serum albumin, Ficoll, and polyvinylpyrrolidine, 0.1% (wt/vol) sodium dodecyl sulfate (SDS), 250 1Lg/ml of dena-
148
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
tured salmon sperm DNA, and 1 Itg/ml poly(A). The blots were hybridized for 48 h at 42 C in a buffer containing 50 % (vol/vol) formamide, 5 x SSPE, 0.01 % (wt/vol) of bovine serum albumin, Ficoll, and polyvinylpyrrolidine, 0.1% (wt/vol) SDS, 10% (wt/vol) dextran sulfate, and 125 ng/ml of denatured salmon sperm DNA. The buffer also contained 5.4 x 10' cpm of (32P]dCTP-labeled cDNA at a specific activity of 1.1 x 108 cpm/ug. The blot was washed with 2 x SSC at room temperature and then with 0.1 x SSC, 0.1% (wt/vol) SDS 3 times at 55 0 C for 15 min and autoradiographed.
82 Microglobulin mRNA
0.4
0
Nuclear Run-on Transcription The transcription assay was done as described by Kindy and Sonenshein (41). Fifteen micrograms of plasmid probe was spotted and baked onto nitrocellulose using a slot blot apparatus. The filters were prehybridized and then hybridized with buffer containing 1.5 to 7.0 x 105 cpm/ml of 32p_ labeled transcripts. Flow Cytometry and Cell Sorting The method for isolating rat lung fibroblast subpopulations using antibodies specific for cell surface-associated collagens was previously described by Breen and associates (29). Both normal fibroblasts and fibroblasts derived from bleomycin-treated rat lungs were isolated (42). Before fibroblast isolation, pulmonary fibrosis was induced in certain rats by endotracheal injection of bleomycin after a tracheostomy. Animals used as controls received sterile saline. Fibroblasts isolated from normal saline or bleomycin-treated animals were analyzed by flow cytometry with a rat type I collagen antiserum (DMI, Westbrook, ME) and sorted. Presumably, this cell-sorting technique separates fibroblast subpopulations by type I antibody adhering to type I collagen on the cell membrane. Two fibroblast cell populations were separated by cell sorting and had different amounts of steady-state levels of type I and type III collagen mRNAs. This assay system is used to separate lung fibroblasts to determine which fibroblast cell type is most active in bleomycin-induced lung fibrosis. Cell sorting resulted in high-intensity and lowintensity cells. The high-intensity fibroblasts had an increased amount of type I procollagen mRNA, whereas the low-intensity fibroblasts had an increased amount of type III procollagen mRNA. One microgram of each RNA was hybridized to excess (32P]dCTP-oligolabeled pTGF,B33 as described. Determination of TGF-,B in Conditioned Media The amount of TGF-,B was determined by the CCl-64 mink lung cell growth inhibitor assay (43) as modified by Phan and colleagues (44). To confirm TGF-,B activity, the assay was done in the presence and absence of anti-TGF-,B antibody. Fibroblast-conditioned media was used as a source ofTGF-,B protein. Conditioned media was collected from the cell layer at 24 h.
Results The direct effect of bleomycin on TGF-,B gene transcription and TGF-,B synthesis was examined as it relates to effects of TGF-,B on collagen mRNA and collagen synthesis in rat lung fibroblasts. A link was indicated among bleomycin, TGF-,B,
Q)
0.3
g III
~
02
.c
-c 0.1
o Control
0.6
3.0
TGF-13 mRNA
0.4
Q)
0.06
0.3
o
c:
III
-eo 0.2
Pulmonary fibrosis is a well-known toxic response to bleomycin treatment. Here we demonstrate the direct effects of bleomycin on lung fibroblasts that resulted in a marked increase of collagen synthesis as compared with total noncollagen protein synthesis. Bleomycin treatment of rat lung fibroblast cultures resulted in an increase of total cellular transforming growth factor-S (TGF-{3) mRNA and increased secretion of TGF-{3 protein into the conditioned media. (32-Microglobulin was measured as an mRNA that did not increase with bleomycin treatment. The bleomycin-induced increase of TGF-13 mRNA was decreased by cells cultured in the presence of either cycloheximide, an inhibitor of protein synthesis, or 2-mercapto-l-(13-4pyridethyl) benzimidazole, an inhibitor of RNA synthesis. To assess the mechanism underlying increased steady-state mRNA levels, the nuclear fraction was isolated from bleomycin-treated cells and the TGF-13 transcripts were determined. Transcription of TGF-{3 mRNA was increased 12h after bleomycintreatment, whereas the transcription of type I procollagen, type III procollagen, and l3-actin mRNAs were increased after 48 h of bleomycin treatment. f32-Microglobulin mRNA synthesis was not increased within this time frame. These results suggest bleomycin regulation of TGF-13 at both the mRNA and protein levels. Rat lung fibroblasts were separated by cell sorting into two subpopulations. One population of fibroblasts demonstrated increased procollagen type I mRNAs, whereas fibroblasts in the other population had increased procollagen type III mRNA. Following bleomycin treatment, TGF-13 mRNA was shown to be located more prominently in those fibroblasts that contain primarily collagen type I mRNAs.
Bleomycin, an antineoplastic glycopeptide isolated from Streptomyces verticil/us, is used to treat several neoplasms including squamous cell carcinoma of the head and neck, lymphomas, and testicular carcinomas (1). Pulmonary fibrosis is a major side effect of bleomycin treatment (2-7). Fibrosis occurs in four phases, which include initial injury, an inflammatory phase, a proliferative phase of connective tissue cells and other cells, and a remodeling-repair phase (2-7). Bleomycin-induced lung fibrosis is a complex process involving cell-cell interactions, stimulatory and inhibitory factors, morphologic changes, immunologic changes, biochemical changes, and the participation of many cell types. Increased synthesis and deposition of collagen result in increased lung collagen content and distortion of the pulmonary structural architecture (2-7). Increased collagen deposition in lung results from al(Received in original form January 25. 1991 and in final form August 1. 1991) Address correspondence to: Kenneth R. Cutroneo, Ph.D., Department of Biochemistry, College of Medicine, University of Vermont, Health Science Complex, Given Building, Burlington, VT 05405. Abbreviations: 2-mercapto-l-(iJ-4-pyridethyl) benzimidazole, MPB; sodium dodecyl sulfate, SDS; transforming growth factor-S, TGF-iJ. Am. J. Respir. Cell Mol. BioI. Vol. 6. pp. 146-152, 1992
terations of both the synthesis and degradation of newly synthesized collagen (8). In bleomycin-induced fibrotic lung, collagen synthesis is selectively increased while collagen degradation is decreased (8). Procollagen type I and procollagen type III steady-state mRNAs are elevated in bleomycininduced interstitial pulmonary fibrosis (9, 10). These changes in type I mRNA accumulation are associated with an increase in the rates of procollagen gene transcription (10). Fibroblasts cultured in the presence of bleomycin demonstrate increased collagen synthesis (11-13), and decreased cell growth (14-16). In addition, type III collagen is selectively decreased as compared with type I collagen (11). Five isoforms of transforming growth-S (TGF-{3) and various receptors are synthesized by many different cell types (17). TGF-{3 acts at a variety of cellular sites, displaying various biologic activities such as regulation of protein synthesis (17). In this study, the effects of TGF-{3, on collagen synthesis are examined. TGF-{3 may stimulate human dermal fibroblast transcription, secretion, and processing of total collagen. This processing may involve the co-translational and post-translational events during collagen synthesis (18). The elevation of colla- . gen synthesis by TGF-{3 is associated with increased cellular steady-state levels of both type I and type III collagen mRNAs in human dermal fibroblasts (19, 20) and in human
Breen, Shull, Burne et al.: Bleomycin and TGF-J3 mRNA
147
lung fibroblasts (21). The production of type I and type III collagens is also significantly increased in human lung fibroblasts treated with TGF-J3 (22). In NRK-49F rat fibroblasts, TGF-J3 increased the levels of type I procollagen and fibronectin mRNAs, which were inhibited by actinomycin D but not by cycloheximide (23). In NIH 3T3 mouse fibroblasts, however, cycloheximide inhibits the TGF-J3 stimulation of type I procollagen mRNAs (24). Furthermore, during confluent growth, pro« 1(1) collagen mRNA stability occurs, whereas in subconfluent cells this mRNA stability is not evident (24). Finally, TGF-J3 increases the activity of the proa2(I) procollagen promoter at least partially through the NF-l site (25). In relation to bleomycin-induced pulmonary fibrosis, TGF-J3 mRNA is increased in lung homogenates before type I and type III procollagen mRNAs (26, 27). Khalil and coworkers (28) demonstrated that when TGF-J3 is primarily associated with the extracellular matrix, there is maximal collagen synthesis in explanted fibroblasts. However, early in the bleomycin-induced injury there is immunohistochemical evidence of TGF-J3 being localized in alveolar macrophages, coinciding with a minimal increase in collagen synthesis. As yet, the cell(s) responsible for the bleomycin-induced increase of TGF-J3 mRNA have not been identified. The objective of the present studies was to determine the ability of bleomycin to increase TGF-J3 mRNA and to demonstrate the secretion of TGF-J3 from rat lung fibroblasts. These studies were designed to determine whether bleomycin directly increased TGF-J3 gene expression in tandem with procollagen gene expression. A second objective was to determine the source of TGF-J3 mRNA and TGF-J3 protein in fibroblasts of bleomycin-induced fibrotic lung (26, 27). Bleomycin was added to fibroblasts that mainly secrete type I collagen. Bleomycin was added to fibroblast cultures to determine the effect on TGF-J3 mRNA without interference of other cell types. The dose in vivo was greater because bleomycin was localized in the lung by endotracheal injection. The maximum dose used to treat fibroblast cultures in the present studies was 3 1Lg/ml of bleomycin for 24 h, the dose that gives the maximal increase of TGF-J3 mRNA.
determine the levels of noncollagen and procollagen synthesis in the cell layer.
Materials and Methods Each experiment was performed at least twice with similar results and minimal variability from experiment to experiment. Collagen and Noncollagen Protein Synthesis Rat lung fibroblasts (RL87) were isolated as described by Breen and associates (29). The cells were supplemented daily with 10-5 M ascorbic acid. Confluent rat lung fibroblasts cultured for 4 days were incubated with various doses of porcine TGF-J3I (R & D Systems, Minneapolis, MN), ranging from 0.1 to 10 ng/ml for 24 h. Three hours before the cells were harvested with 0.25 % trypsin:phosphatebuffered saline solution, L-[5-3H]proline (Amersham Corp., Arlington Heights, IL) was added to each culture at a concentration of 10 1LCi/mi. Harvested cells were digested with bacterial collagenase form III (Advance Biofactures, Lynbrook, NY) as described by Newman and Cutroneo (30) to
Recombinant DNA The porcine TGF-J3, cDNA recombinant plasmid, pTGF1333, was a generous gift of Dr. Michael Sporn (31). The mouse J32-microglobulin cDNA plasmid, pBRcB-3, was a gift of Dr. Jane R. Parnes (32). The ratproal(I) and proa2(I) recombinant plasmids, palRI and pa2R2, respectively, were generous gifts of Dr. David Rowe (33). The mouse genomic proo l(III) plasmid, pMSC-l, was a gift of Dr. Benoit de Crombrugghe (34). The recombinant plasmids were characterized by restriction mapping and sequence analysis. For slot blot hybridizations, the plasmids were digested with the appropriate restriction enzymes and electrophoresed in 0.8% low melting point agarose with 0.5 ng/ml ethidium bromide. The insert bands were excised from the gel, boiled, and stored at -20° C. The DNAs were subsequently oligolabeled by the method of Feinberg and Vogelstein (35). Recombinant DNA techniques were performed under PI containment. Isolation of Total Cellular RNA The single-step method of RNA isolation using acid guanidinium thiocynate-phenol-chloroform extraction was used to isolate total cellular RNA from rat lung fibroblasts (36). Slot Blot Hybridization One microgram of total cellular RNA was directly spotted onto nitrocellulose using a Minifold II Slot Blot System (Schleicher & Schuell, Keene, NH). The nitrocellulose was baked for 2 h at 80° C and hybridized. The (32P]dCTP oligolabeled probes had a specific activity > 1.0 x 109 cpm/ug. Blots were washed under high stringency conditions and autoradiographed. Band intensities were quantified by densitometric scanning using a dual-wavelength thin-layer chromatography scanner (Model CS-930; Shimadzu Corp., Kyoto, Japan). The RNA synthesis inhibitor 2-mercapto-l(J3-4-pyridethyl) benzimidazole (MPB) (37, 38) was used at 10-5 M, and cellular protein was determined (39). Cycloheximide, at 10-4 M, was used to inhibit protein synthesis. Northern Blot Analysis of TGF-J3 mRNA from Control and Bleomycin-treated Rat Lung Fibroblasts Northern blot analysis was performed to determine if bleomycin increased the TGF-J3 mRNA in rat lung fibroblasts. Other cells were treated with desamido bleomycin, which was kindly supplied by Dr. John Lazo. This bleomycin metabolite had no significant effect on increasing TGF-J3 mRNA in rat lung fibroblasts (data not shown). Total cellular RNA (15 1Lg) was denatured and electrophoresed in a 0.8% agarose/formaldehyde gel, which was stained with 0.5 ng/ml ethidium bromide to determine intact 28S rRNA and l8S rRNA. The RNA is then transferred to nitrocellulose, baked at 80° C for 2 h, and hybridized to nick-translated pTGFJ333 plasmid (40). The samples were prehybridized at 42° C for 24 h in a buffer containing 50% (vol/vol) deionized formamide, 5x SSPE (pH 7.4), 0.025% (wt/vol) of each bovine serum albumin, Ficoll, and polyvinylpyrrolidine, 0.1% (wt/vol) sodium dodecyl sulfate (SDS), 250 1Lg/ml of dena-
148
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
tured salmon sperm DNA, and 1 Itg/ml poly(A). The blots were hybridized for 48 h at 42 C in a buffer containing 50 % (vol/vol) formamide, 5 x SSPE, 0.01 % (wt/vol) of bovine serum albumin, Ficoll, and polyvinylpyrrolidine, 0.1% (wt/vol) SDS, 10% (wt/vol) dextran sulfate, and 125 ng/ml of denatured salmon sperm DNA. The buffer also contained 5.4 x 10' cpm of (32P]dCTP-labeled cDNA at a specific activity of 1.1 x 108 cpm/ug. The blot was washed with 2 x SSC at room temperature and then with 0.1 x SSC, 0.1% (wt/vol) SDS 3 times at 55 0 C for 15 min and autoradiographed.
82 Microglobulin mRNA
0.4
0
Nuclear Run-on Transcription The transcription assay was done as described by Kindy and Sonenshein (41). Fifteen micrograms of plasmid probe was spotted and baked onto nitrocellulose using a slot blot apparatus. The filters were prehybridized and then hybridized with buffer containing 1.5 to 7.0 x 105 cpm/ml of 32p_ labeled transcripts. Flow Cytometry and Cell Sorting The method for isolating rat lung fibroblast subpopulations using antibodies specific for cell surface-associated collagens was previously described by Breen and associates (29). Both normal fibroblasts and fibroblasts derived from bleomycin-treated rat lungs were isolated (42). Before fibroblast isolation, pulmonary fibrosis was induced in certain rats by endotracheal injection of bleomycin after a tracheostomy. Animals used as controls received sterile saline. Fibroblasts isolated from normal saline or bleomycin-treated animals were analyzed by flow cytometry with a rat type I collagen antiserum (DMI, Westbrook, ME) and sorted. Presumably, this cell-sorting technique separates fibroblast subpopulations by type I antibody adhering to type I collagen on the cell membrane. Two fibroblast cell populations were separated by cell sorting and had different amounts of steady-state levels of type I and type III collagen mRNAs. This assay system is used to separate lung fibroblasts to determine which fibroblast cell type is most active in bleomycin-induced lung fibrosis. Cell sorting resulted in high-intensity and lowintensity cells. The high-intensity fibroblasts had an increased amount of type I procollagen mRNA, whereas the low-intensity fibroblasts had an increased amount of type III procollagen mRNA. One microgram of each RNA was hybridized to excess (32P]dCTP-oligolabeled pTGF,B33 as described. Determination of TGF-,B in Conditioned Media The amount of TGF-,B was determined by the CCl-64 mink lung cell growth inhibitor assay (43) as modified by Phan and colleagues (44). To confirm TGF-,B activity, the assay was done in the presence and absence of anti-TGF-,B antibody. Fibroblast-conditioned media was used as a source ofTGF-,B protein. Conditioned media was collected from the cell layer at 24 h.
Results The direct effect of bleomycin on TGF-,B gene transcription and TGF-,B synthesis was examined as it relates to effects of TGF-,B on collagen mRNA and collagen synthesis in rat lung fibroblasts. A link was indicated among bleomycin, TGF-,B,
Q)
0.3
g III
~
02
.c
-c 0.1
o Control
0.6
3.0
TGF-13 mRNA
0.4
Q)
0.06
0.3
o
c:
III
-eo 0.2
