Proliferation of Rat Pleural Mesothelial Cells in Response to Hepatocyte and Keratinocyte Growth Factors

Transcription

1 Proliferation of Rat Pleural Mesothelial Cells in Response to Hepatocyte and Keratinocyte Growth Factors Ian Y. R. Adamson, Julita Bakowska, and Heather Prieditis Department of Pathology, University of Manitoba, Winnipeg, Manitoba, Canada (Received in original form January 20, 2000 and in revised form March 28, 2000) Address correspondence to: Dr. I. Y. R. Adamson, Dept. of Pathology, University of Manitoba, D Bannatyne Ave., Winnipeg, MB R3E 0W3, Canada. Abbreviations: Dulbecco s modified Eagle s medium, DMEM; enzymelinked immunosorbent assay, ELISA; fetal bovine serum, FBS; hepatocyte growth factor, HGF; 3 H thymidine, 3 HT; keratinocyte growth factor, KGF; mesothelial cell, MC; platelet-derived growth factor, PDGF; pleural lavage, PL; tumor necrosis factor, TNF. Am. J. Respir. Cell Mol. Biol. Vol. 23, pp , 2000 Internet address: The proliferative response of cultured pulmonary mesothelial cells (MCs) to epithelial cell mitogens such as keratinocyte growth factor (KGF) and hepatocyte growth factor (HGF) is investigated. A cell line of rat pleural MCs and freshly prepared rat visceral and parietal MCs were studied. Both KGF and HGF stimulated thymidine uptake in the cell line when cultured for 2 d in serum-free conditions; the growth increase was magnified when tumor necrosis factor (TNF)- was also added to the cultures. Adding asbestos fibers alone to MCs in culture did not enhance DNA synthesis by these cells. The MCs were also shown to synthesize significant amounts of HGF but much less KGF when cultured for 2 d. When freshly prepared MCs were examined, normal cell growth was more rapid in the parietal cells, which also had a more epithelialtype morphology. The addition of HGF and KGF resulted in increased DNA synthesis in each cell type, but no effect of added TNF- was found. The results indicate that pulmonary MCs have the potential to proliferate in response to cytokines such as HGF and KGF that are usually associated with epithelial cell regeneration after injury. An early proliferative response is seen in various cell types after the deposition of asbestos in the lung, with most of the cell division occurring in the epithelial and interstitial populations at the bronchiolar-alveolar duct regions, the main site of fiber deposition (1, 2). However, there is also a rapid proliferative burst seen in pleural mesothelial cells (MCs) a few days after either intratracheal or inhaled asbestos exposure (3 5). While it is possible that some fibers may reach the pleural surface in this short time span, it has also been suggested that growth factors produced in central areas of the lung in the cellular response to asbestos may diffuse to the pleura and stimulate MCs proliferation (3, 6). Various cytokines have been demonstrated in the lung after asbestos exposure, most of these are macrophage derived and stimulate fibroblast proliferation (7 9). In an earlier study, we collected bronchoalveolar lavage (BAL) and pleural lavage (PL) fluids after instilling crocidolite asbestos in rat lung. Two distinct cytokine effects were seen, one responsible for fibroblast growth, which was most likely due to platelet-derived growth factor (PDGF), and the other was associated with MC proliferation (6). This effect could be partially blocked by antibodies to keratinocyte growth factor (KGF), which is a known lung epithelial cell mitogen. The fact that MCs share properties of both mesenchymal and epithelial cells makes it possible that lung MCs may respond to any epithelial cell mitogens that are present in the lung after asbestos-induced injury. We now wish to determine whether lung MCs proliferate in vitro in response to two potent mitogens for pulmonary epithelium, KGF and hepatocyte growth factor (HGF). These factors have been found in lung lavage fluids at a time of epithelial cell proliferation after other types of injury (10), and both stimulate normal alveolar type 2 cell division (11 13). We now examine the effects of pure KGF and HGF on lung MC growth in culture using both a cell line and also freshly prepared rat parietal and visceral MCs. The possibility that MCs may also be a source of these growth factors is studied by measuring the KGF and HGF levels produced by MCs in culture using enzyme-linked immunosorbent assay (ELISA) methods. Materials and Methods Cell Line Rat pleural MCs (CCL 216) were obtained from the American Type Culture Collection (Rockville, MD). These cells were subcultured for use between passages 15 and 22. Cells were freshly prepared for each experimental run and were seeded in 24-well plates at cells/well in Dulbecco s modified Eagle s medium (DMEM) with 0.5% fetal bovine serum (FBS). After allowing 2 h for attachment, the medium was changed to serum-free conditions and various factors were added. These included KGF, HGF, tumor necrosis factor (TNF)- (R&D Systems, Minneapolis, MN) and crocidolite asbestos; control cultures had either no addition or 5% FBS added as a positive control for growth. All cell culture supplies were purchased from Life Technologies (Burlington, ON, Canada). Preparation of Fresh Parietal and Visceral MCs Cells were prepared from 250-g male Sprague Dawley rats. Normal animals were killed by intraperitoneal barbiturate overdose, and vascular perfusion was carried out with 20 ml saline containing ethylenediaminetetraacetic acid. A tracheotomy was performed and the lungs were inflated with 5 ml of a solution containing 4.3 U/ml elastase (14) (Worthington Biochemicals, Freehold, NJ). The lungs were tied off, removed, and immersed in 50 ml of the elastase solution for 15 min at room temperature. The pleural surface of each lobe was then gently scraped with a single-edge razor blade and the surface rinsed with culture medium (DMEM and 5% FBS); the rinsed cells plus medium were collected in a 60-mm tissue culture dish that was then set in the incubator at 37 C with 5% CO 2. Meanwhile, a large section of the thoracic wall had been cut out, rinsed in saline, and placed in a dish containing 50 ml of 0.05% trypsin for 20 min. The inner surface was scraped with a rubber policeman and rinsed with DMEM plus FBS into a 60-mm culture dish that was then placed in the incubator, giving a preparation of parietal MCs from the same animal as the visceral MCs.

2 346 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL Figure 3. Percentage change in 3 HT uptake by cell-line MCs cultured 2 d with asbestos added. SF serumfree control. *P 0.05 SF. Figure 1. Percentage change in 3 HT uptake by pleural MCs from a cell line cultured 2 d with HGF and TNF- added. SF serumfree control. *P 0.05 SF. Both MC types were maintained for 1 d, gently washed and rinsed, and then cultured to confluence. This took about 1 wk for parietal cells and 1 to 2 wk for visceral MCs. Cells were used for proliferation assays when nearing confluence or at the same stage after passage 1. They were seeded at cells/well in 24-well plates. In about 10% of the original cultures, fibroblast contamination was observed with rapidly growing, spindle-shaped cells predominating. These cultures were discarded. Cell Proliferation Freshly prepared cells or those from the cell line were seeded in medium containing 0.5% serum for 2 h, then changed to serum free for 2 d with various cytokine doses added. For the final 24 h, tritiated thymidine ( 3 HT) at 0.1 Ci/ml was added to each well. Doses of HGF and KGF were added at 5, 15, and 30 ng/ml, and the effect of additional TNF- at 15 or 30 ng was also assessed. Because it is also possible that fibers translocated to the pleural surface in vivo could directly stimulate DNA synthesis in MCs, we added crocidolite asbestos at 1 or 10 g/ml (0.5 or 5 g/cm 2 ) to MCs for 2 d to look for any direct stimulatory effect on growth. In a separate test, viability of MCs exposed to various doses of asbestos was measured by trypan blue exclusion. At the end of the culture period with 3 HT, the wells were rinsed twice then cells scraped off for scintillation counting. For every experimental condition, four wells per treatment were assayed and each experiment was repeated 4 to 6 times. The serum-free incorporation level by MCs was taken as 100% for each experimental run and the percentage change in 3 HT uptake was calculated. The mean percentage standard error of growth stimulation was calculated per group. Experimental groups were compared with serum-free controls using the t test; P 0.05 was considered significant. In addition, dose effects were assessed using one-way analysis of variance. KGF and HGF Production This was determined on the MC line using standard ELISA methods. The antibodies and growth factor standards were obtained from R&D Systems. These assays were designed specifically to measure human growth factors, but we have shown previously that they can be used to measure KGF and HGF in rat lung (10). Rat MCs were cultured for 2 d under normal culture conditions, then the cells were rinsed, removed, and counted before measurement of KGF and HGF. Results were calculated as picograms/10 6 cells. As a comparison, levels of KGF and HGF were measured in a few cultures of freshly prepared visceral and parietal MCs. Results Mesothelial Cell Line In each experimental run, 3 HT uptake for unstimulated controls (serum free) ranged from 10 to 16,000 disintegrations per minute (dpm). The addition of 5% FBS always induced a significant increase in incorporation and was used as a positive control. The effect of adding HGF to MCs in culture is shown in Figure 1. Doses of 15 and 30 ng/ ml added to serum-free cultures resulted in a 50% increase in 3 HT uptake measured as dpm. When TNF- alone was added to MCs at 30 ng, a 25% increase in uptake was seen; however, when TNF- was added together with 30 mg HGF, a further increase in uptake was found compared with the effect of HGF alone (Figure 1). Similar effects were produced when KGF was added to the MCs (Figure 2). Doses of 15 and 30 ng induced significant increases in DNA synthesis over serum-free conditions, and the effect was augmented when TNF- was also added. Using the same culture system, different doses of asbestos fibers were added to MCs. At doses up to and including 1.0 g/ml (0.5 g/cm 2 ), MC viability was in the range of 95 to 98% at the end of the culture period and no increase in 3 HT incorporation was found (Figure 3). At the higher dose of 10 g/ml (5 g/cm 2 ), viability was reduced to 80% and DNA synthesis was also lower. Figure 2. Percentage change in 3 HT uptake by cell-line MCs cultured 2 d with KGF and TNF- added. SF serum-free control. *P 0.05 SF. Figure 4. Levels of HGF (dotted bars) and KGF (solid bars) produced by cell-line MCs cultured with and without TNF-.

3 Adamson, Bakowska, and Prieditis: Mesothelial Cell Growth 347 Figure 5. Phase micrograph of freshly prepared rat parietal MCs cultured 7 d. Cells are confluent and show cobblestone -type morphology. Original magnification: 400. Figure 7. Percentage change in 3 HT uptake by fresh visceral MCs cultured 2 d with KGF or HGF. SF serum-free control. *P 0.05 SF. Production of HGF and KGF The levels of these cytokines produced by the rat MC line was determined by ELISA. After 2 d in culture, the MCs contained 47.5 pg/10 6 cells of HGF and these levels were not increased when TNF- was added to the culture medium (Figure 4). These cells made only small amounts of KGF measured at 2.1 pg/10 6 cells, and the level was also unchanged by adding TNF- (Figure 4). On a small number of cultures, visceral and parietal cells also produced HGF at equal levels around 40 pg/10 6 cells. Freshly Prepared MCs Differences were seen in the growth and morphology of visceral and parietal MCs from the same rat under normal culture conditions. The parietal cells tended to grow faster and showed a more regular cobblestone appearance (Figure 5). These cells reached confluence usually in about 1 wk, whereas the visceral MCs usually took 10 to 14 d to reach confluence. These cells were less regular in morphology and appeared to be more spread out on the culture dish (Figure 6). When grown for 2 d serum free, the visceral MCs showed increased 3 HT uptake at the higher dose of HGF and KGF, with HGF producing a slightly greater increase in DNA synthesis (Figure 7). The HGF and KGF also stimulated 3 HT incorporation in parietal MCs, which responded particularly to both doses of HGF (Figure 8). The addition of TNF- to either set of MC cultures did not result in any significant change in 3 HT uptake, with or without the growth factors present (data not shown). Discussion Pleural MCs form a single epithelial cell layer that covers the lung (visceral MCs) and the diaphragm and thoracic wall (parietal MCs). Increased proliferation of these cells and eventual tumor formation are most commonly associated with asbestos exposure. In most cases, the abnormal cell division has been related to fiber translocation to the pleural surface where mitotic damage to MCs with altered gene function occurs (9, 15). It may be expected that considerable time would be required for sufficient fibers to reach the pleura and cause significant MC damage, but there is evidence that MC proliferation can occur soon after fiber deposition in the lung (4 6). This occurred at the same time as epithelial repair in bronchioles and alveoli in direct response to fiber damage (3, 6). As a result, it was postulated that epithelial cell mitogens generated early in the response to asbestos may diffuse through the lung to activate MCs. These cells have properties of both fibroblasts and epithelia (9, 16), and so may respond to epithelial-type growth factors in vivo. To determine if MCs have this potential, we chose to study cells in vitro. Cultured MCs respond to several cytokines, mostly the types that also stimulate fibroblasts such as PDGF and transforming growth factor- (9, 16). These cells also proliferate faster in response to TNF- (17), an effect that was also found in the present study. Some MCs proliferate in the presence of epidermal growth factor (EGF), and receptors for this factor have also been demonstrated in rat cells (18). However, in another study, EGF was only effective in vitro and had no effect on Figure 6. Phase micrograph of freshly prepared rat visceral pleural MCs cultured 12 d. Cells are near confluence and many appear more spread out than the parietal MCs. Original magnification: 400. Figure 8. Percentage change in 3 HT uptake by fresh parietal MCs cultured 2 d with KGF or HGF. SF serum-free control. *P 0.05 SF.

4 348 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL growth of human MCs in vivo (19). In an earlier study on the pulmonary response to crocidolite, we found that alveolar lavage and PL fluids collected a few days after fiber deposition stimulated rat MC growth in vitro; this effect was not blocked by anti-pdgf, but it was partially inhibited by anti-kgf (6). This led to the hypothesis that wellestablished pulmonary epithelial growth factors such as KGF and HGF may stimulate lung MC division. The present experiments show that KGF is a mitogen for the rat MC line in a dose-dependent manner. KGF is produced by lung fibroblasts and is a known mitogen for alveolar type 2 cells and also bronchiolar epithelia (11, 12). It has also been shown that KGF is secreted in the lung after epithelial cell damage such as that induced by bleomycin (10), and we have previously found evidence for KGF in PL fluid a few days after asbestos deposition (6). It is possible that lung injury and the generation of cytokines stimulate fibroblasts to upregulate KGF production. We found that the presence of TNF- together with KGF caused an additional increase in DNA synthesis by the cultured MCs. This is relevant to asbestos exposure, which induces TNF- production by macrophages (20). Both of these cytokines, although produced largely in the central regions of the lung, could diffuse to the pleural cavity and stimulate the MC population. In a similar manner, HGF may be produced in the lung after asbestos-induced epithelial injury as has been shown after bleomycin administration (10). In the present study, HGF also stimulates MC growth as shown by increased thymidine uptake, and the addition of TNF- caused a further increase in DNA synthesis by these cells. There is some literature evidence that mesothelioma cell lines have the receptor for HGF but not normal cells (21) and that HGF is mitogenic for the tumor cells in culture. In another study, mesothelioma cells with epithelial morphology showed increased proliferation and mobility in response to HGF, whereas cells with a fibroblast-like appearance showed no increase in growth (22). The MC response to HGF then may depend on the phenotype of the particular cells under study. Because the properties of our MC line may have changed as passage number increased, freshly prepared MCs were also evaluated to confirm the above mitogenic effects. Both the parietal and visceral cells isolated from the same rats showed increased DNA synthesis in response to HGF and to KGF. The two MC preparations were different in growth even under normal conditions with slower growth seen with the visceral MCs. These cells were more elongated as compared with the parietal MCs, which maintained a more epithelial-like morphology. However, the results show that both HGF and KGF can induce a proliferative response in both of these freshly prepared MC populations. The fibroblast is reportedly the sole source of KGF, whereas HGF is produced by various mesenchymal cells (11, 12, 23, 24). In the present study, we found that the MCs in culture can synthesize both of these cytokines, though much greater levels of HGF were found. The production of HGF and KGF by MCs was not increased when TNF- was added to the cells in culture. Although TNF- may not have a direct effect on cytokine production by MCs, it stimulates MC proliferation and augments the growth effects of HGF and KGF. In vivo, the initial interaction of particles with macrophages and epithelial cells can generate chemotactic factors and cytokines such as TNF- (8, 20) that may diffuse across the interstitium or reach the pleura through blood or lymphatics. It has been shown, for example, that a Clara cell secretory protein can diffuse across the visceral pleura in injured rat lung (25). Molecules such as TNF- may diffuse through the lung and then induce the production of interleukin-8 by MCs, resulting in pleural inflammation (9, 15). In addition, TNF- could also upregulate fibroblasts (26), and possibly other cells, to produce cytokines such as HGF and KGF that are capable of inducing MC proliferation. The relationship of the proliferative response of MCs in vitro treated with KGF and HGF to the neoplastic process in vivo is not known. The rapid increase in MC growth seen shortly after asbestos deposition may not be due to a direct effect of translocated fibers because adding asbestos directly to MCs did not increase DNA synthesis. It seems more likely that cytokine production in the lung may result in factors such as KGF and HGF reaching the pleural cavity. There is evidence for KGF in PL fluid 1 wk after asbestos (6), and both KGF and HGF are present in BAL fluid in injured lung 3 d after bleomycin administration (10). Although translocation of a large number of fibers may require a longer time period, some fibers could reach the pleural cavity soon after inhalation. For example, ceramic fibers were found in the pleural cavity after 4 wk when increased MC proliferation was also detected; growth was higher in parietal rather than visceral MCs (27). When the inhaled dose was increased, fibers were found in the pleural cavity after five daily exposures (28), raising the possibility that growth factors and fibers may reach the pleural space at the same time. Taken together, these results suggest that ongoing lung injury in response to the inhalation of fibers such as asbestos may result in the continued presence of MC mitogens such as HGF and KGF in the pleural space. These factors could promote continued cell proliferation and possibly render these activated MCs more susceptible to DNA damage that may be induced by any translocated fibers. Acknowledgments: This research project was supported by grant MT 3878 from the Medical Research Council of Canada. References 1. Brody, A. R., and L. H. Overby Incorporation of tritiated thymidine by epithelial and interstitial cells in bronchiolar-alveolar regions of asbestos exposed rats. Am. J. Pathol. 134: Bowden, D. H., and I. Y. R. Adamson Bronchiolar and alveolar lesions in the pathogenesis of crocidolite-induced pulmonary fibrosis in mice. J. Pathol. 147: Adamson, I. Y. R., J. Bakowska, and D. H. Bowden Mesothelial cell proliferation after instillation of long or short asbestos fibers into mouse lung. Am. J. Pathol. 142: Warheit, D. B., A. M. Hartsky, and S. R. Frame Pulmonary effects in rats inhaling size-separated chrysotile asbestos fibers or p-aramid fibrils. Toxicol. Lett. 88: Sekhon, H., J. Wright, and A. Churg Effects of cigarette smoke on airway, vascular and mesothelial cell proliferation. Int. J. Exp. Pathol. 75: Adamson, I. Y. R., H. Prieditis, and L. Young Lung mesothelial cell and fibroblast responses to pleural and alveolar macrophage supernatants and to lavage fluids from crocidolite-exposed rats. Am. J. Respir. Cell Mol. Biol. 16: Brody, A. R., J. C. Bonner, L. H. Overby, A. Badgett, V. Kalter, R. K. Ku-

5 Adamson, Bakowska, and Prieditis: Mesothelial Cell Growth 349 mar, and R. A. Bennett Interstitial pulmonary macrophages produce platelet-derived growth factor that stimulates rat lung fibroblast proliferation in vitro. J. Leukoc. Biol. 51: Rom, W. N., W. D. Travis, and A. R. Brody Cellular and molecular basis of asbestos-related diseases. Am. Rev. Respir. Dis. 143: Kuwahara, M., and E. Kagan The mesothelial cell and its role in asbestos-induced pleural injury. Int. J. Exp. Pathol. 76: Adamson, I. Y. R., and J. Bakowska Relationship of keratinocyte growth factor and hepatocyte growth factor levels in rat lung lavage fluid to epithelial cell regeneration after bleomycin. Am. J. Pathol. 155: Ulich, T. R., E. S. Yi, K. Longmuir, S. Yin, R. Blitz, C. F. Morris, R. M. Housely, and G. F. Pierce Keratinocyte growth factor is a growth factor for type II pneumocytes in vivo. J. Clin. Invest. 93: Panos, R. J., J. S. Rubin, S. A. Aaronson, and R. J. Mason Keratinocyte growth factor and hepatocyte growth factor/scatter factor are heparin-binding growth factors for alveolar type II cells in fibroblast conditioned medium. J. Clin. Invest. 92: Mason, R. J., K. McCormick-Shannon, J. S. Rubin, T. Nakamura, and C. C. Leslie Hepatocyte growth factor is a mitogen for alveolar type II cells in rat lavage fluid. Am. J. Physiol. 271:L46 L Dobbs, L. G., R. Gonzalez, and M.C. Williams An improved method for isolating type II cells in high yield and purity. Am. Rev. Respir. Dis. 134: Gerwin, B. I Asbestos and the mesothelial cell: a molecular trail to mitogenic stimuli and suppressor gene suspects. Am. J. Respir. Cell Mol. Biol. 11: Jaurand, M. C., J. Fleury-Feith, J. F. Bernaudin, and J. Bignon Pleural mesothelial cells. In The Lung: Scientific Foundations, 2nd ed. R. G. Crystal, J. B. West, P. J. Barnes, and E. R. Weibel, editors. Lippincott- Raven, Philadelphia Owens, M. W., and S. R. Grimes Pleural mesothelial response to inflammation: tumor necrosis factor induced mitogenesis and collagen synthesis. Am. J. Physiol. 265:L382 L Bermudez, E., J. Everitt, and C. Walker Expression of growth factor and growth factor receptor RNA in rat pleural mesothelial cells in culture. Exp. Cell Res. 190: Mutsaers, S. E., R. J. McAnulty, G. J. Laurent, M. A. Versnel, D. Whitaker, and J. M. Papadimitriou Cytokine regulation of mesothelial cell proliferation in vitro and in vivo. Eur. J. Cell Biol. 72: Bissonnette, E., and M. Rola-Pleszczynki Pulmonary inflammation and fibrosis in a murine model of asbestosis and silicosis: possible role of tumor necrosis factor. Inflammation 13: Klominek, J., B. Baskin, Z. W. Liu, and D. Hauzenberger Hepatocyte growth factor/scatter factor stimulates chemotaxis and growth of malignant mesothelioma cells through c-met receptor. Int. J. Cancer 76: Harvey, P., A. Warn, S. Dobbin, N. Arakaki, Y. Daikuhara, M. C. Jaurand, and R. M. Warn Expression of HGF/SF in mesothelioma cell lines and its effects on cell mobility, proliferation and morphology. Br. J. Cancer 77: Rubin, J. S., A. M. Chan, D. P. Bottaro, W. H. Burgess, W. G. Taylor, A. C. Cech, D. W. Hirschfield, J. Wong, T. Miki, P. W. Finch, and S. A. Aaronson A broad spectrum human lung fibroblast-derived mitogen is a variant of hepatocyte growth factor. Proc. Natl. Acad. Sci. USA 88: Ohmichi, H., U. Koshimizu, K. Matsumoto, and T. Nakamura Hepatocyte growth factor (HGF) acts as a mesenchyme-derived morphogenic factor during fetal lung development. Development 125: Hermans, C. O., O. Lesur, B. Weynand, T. Pieters, M. Lambert, and A. Bernard Clara cell protein (CCIG) in pleural fluids. Am. J. Respir. Crit. Care Med. 157: Brauchle, M., K. Angermeyer, G. Hubner, and S. Werner Large induction of keratinocyte growth factor expression by serum growth factors and pro-inflammatory cytokines in cultured fibroblasts. Oncogene 9: Everitt, J. I., T. R. Gelzleichter, E. Bermudez, J. B. Mangum, B. A. Wong, D. B. Janszen, and O. R. Moss Comparison of pleural responses of rats and hamsters to subchronic inhalation of refractory ceramic fibers. Environ. Health Perspect. 105(Suppl. 5): Gelzleichter, T. R., E. Bermudez, J. B. Mangum, B. A. Wong, J. I. Everitt, and O. R. Moss Pulmonary and pleural responses in Fischer 344 rats following short term inhalation of a synthetic vitreous fiber: I. Quantitation of lung and pleural fiber burdens. Fundam. Appl. Toxicol. 30:31 38.