- Email: [email protected]
Angiotensin II and Basic Fibroblast Growth Factor Induce Neonatal Bladder Stromal Cell Mitogenesis
~022-5347/96/1562-0593$03.00/0
THEJOURKAL O F UROLOGY Copyright 0 1996 by AMERICGUROLOCICAL ASSOCIATION, INC.
Vol. 156, 593-597, August 1996 Printed in U.S.A.
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR INDUCE NEONATAL BLADDER STROMAL CELL MITOGENESIS EARL Y. CHENG,“ TOM GRAMMATOPOULOS, CHUNG LEE, JULIA SENSIBAR, ROBERT DECKER, WILLIAM E. KAPLAN, MAX MAIZELS AND CASIMIR F. FIRLIT From the Division of Urology, Children’s Memorial Hospital and Departments of Urology and Medicine, Northwestern University Medical School, Chicago, Illinois
ABSTRACT
Purpose: Our aims were t o establish primary stromal cell cultures from t h e neonatal rabbit bladder and investigate t h e potential mitogenic effects of angiotensin I1 and basic fibroblast growth factor on these cells. Materials a n d Methods: Primary bladder stromal cell cultures were obtained from 3-day-old rabbits, plated at a density of 3 X lo4cells p e r ml. a n d allowed to grow for 24 hours. Subconfluent cells were growth arrested in serum deficient (0.25%newborn calf serum) or serum-free media for 24 h o u r s and then stimulated with lod7 M. angiotensin I1 or 10 ng./ml. basic fibroblast growth factor for an additional 48 hours. Cell counts and c3H1 thymidine incorporation were done to m e a s u r e cellular proliferation a n d deoxyribonucleic acid synthesis. Results: Angiotensin I1 and basic fibroblast growth factor each stimulated neonatal bladder stromal cell proliferation and r3H1 thymidine incorporation u n d e r s e r u m deficient conditions. Angiotensin I1 provoked an average 26% increase in cell number ( p <0.01) and 35% increase in [3H]thymidine incorporation ( p ~0.01) compared to control values. Basic fibroblast growth factor w a s an even more potent mitogen with a 47% increase in cell number ( p CO.01) a n d 180% increase in [3H]thymidine incorporation ( p <0.01) compared to controls. In contrast, angiotensin I1 and basic fibroblast growth factor each failed t o have significant stimulatory effects under serum-free conditions. Conclusions: Angiotensin I1 a n d basic fibroblast growth factor induce a mitogenic response to neonatal bladder stromal cells in vitro. These mitogenic effects require t h e presence of serum factors. Whether angiotensin I1 and basic fibroblast growth factor are involved in the i n vivo regulation of bladder growth associated with obstructive uropathy requires further investigation. KEYWORDS:bladder; angiotensin 11; fibroblast growth factor, basic; stromal cells Bladder neck obstruction may occur in utero (for example posterior urethral valves) and postnatally, as in men with benign prostatic hyperplasia. Regardless of when it occurs, outlet obstruction results in pathological changes in the bladder, which can ultimately alter the neurophysiology of the lower urinary tract. Histopathological changes include bladder muscle hypertrophykyperplasia and increased collagen production. It has been proposed that these changes have a role in altered compliance of the obstructed hufnan and animal bladder.1-4 Fortunately when obstruction is relieved, these changes are partially reversible in the majority of cases, resulting in normalization of bladder compliance and clinical improvement in obstructive voiding symptoms. Difficulty arises in treating patients who have adequate relief of obstruction but still have persistent bladder dysfunction. This pathophysiological condition likely exists secondary to irreversible obstructive changes in the architecture of the bladder. Currently little information is available in regard to the cellular and molecular mechanisms that regulate smooth muscle growth and collagen production in the bladder. Even less is known about factors that govern the reversibility of these processes. Further understanding of these mechanisms 1s critical for the development of better clinical diagnostic and therapeutic tools to treat bladder neck obstruction in the Pediatric and adult populations. Clinically the response of bladder smooth muscle to outlet
* Requests for reprints: Division of Urolo Children’s Memorial Hospital, 2300 Children’s Plaza, Chicago, Il%ois 60614.
obstruction is similar to that of cardiac muscle to increased load (for example hypertension or aortic coarctation). As in the bladder, outflow obstruction in the heart causes an initial compensatory response in the myocardium consisting of muscle hypertrophykyperplasia and altered collagen production. When obstruction is unrelieved, subsequent decompensation and pump failure ensue. In vitro studies using smooth muscle cells from the heart and aorta have demonstrated that angiotensin I1 and basic fibroblast growth factor may be responsible for increased smooth muscle growth in this pathological Also, angiotensin I1 may be involved in the regulation of collagen production in these cells.8 Since the bladder responds to obstruction in a fashion similar to the heart, we hypothesized that the same regulatory factors and mechanisms responsible for controlling stromal cell growth and function in the heart are also present in the bladder. To investigate this hypothesis further we attempted to establish and characterize primary stromal cell cultures from the neonatal rabbit bladder. We then investigated the role of angiotensin I1 and basic fibroblast growth factor in modulating the mitogenic properties of these cultured cells. MATERIALS AND METHODS
Establishment of primary cell cultures. Three to 5-day-old New Zealand rabbits were used for the establishment of primary neonatal bladder stromal cell cultures. Routinely bladders from 2 animals were used for each primary culture. Animals were sacrificed with an overdose of pentobarbital. Bladders were then harvested and immediately placed into
5193
594
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR STROMAL CELL MITOGENESIS
37C modified M199 media with 10%newborn calf serum, as described by Baskin et al.9 The epithelium was removed from the muscle portion of the bladder wall with a scalpel blade and the bladders were processed as previously described.I0 Bladders were minced into 1 mm.3 pieces with curved scissors. The minced tissue was placed in 50 ml. 1.0 mM. dithiothreitol in phosphate buffered saline. The tube was subjected to a rotator speed of 30 cycles per minute at 37C for 30 minutes. The supernatant was discarded and the tissue pellet was resuspended with 10 ml. dissociation solution containing 1% deoxyribonuclease type I and 0.28%collagenase type I in RPMI-1640 with 10% fetal bovine serum and rotated at 37C for 30 minutes. The supernatant fraction containing the separated individual cells was then passed through a series of 3 sieves with 230 to 26 pm. pore sizes for the recovery of cells. The remaining tissue pieces were subjected to another round of treatment with dissociation solution and the resultant supernatant was again sieved and pooled with the first supernatant fraction. The combined cell suspension was then spun at 1,000 revolutions per minute for 5 minutes and the supernatant was discarded. The pellet was resuspended in 10 ml. modified MI99 media and washed an additional time. The resultant cells were again resuspended in modified M199 media and plated at a density of 3 X lo6 cells per 75 cm.2 flask. Cultures were incubated at 37C in a humidified atmosphere of 5% carbon dioxide in air. Following initial establishment of the primary culture cells became 90 to 100% confluent and required passaging (1:3 split) every 4 to 7 days. The cells used for this investigation were between passages 2 and 6. Zmmunohistochernistry.Bladder stromal cells were grown on slides (4 x lo4 cells per ml.), which were pre-coated with 10% poly-1-lysine for 3 to 4 days until they were 60 to 70% confluent. Excess media were gently blotted off and the cells were washed twice with phosphate buffered saline for 5 minutes. The cells were then h e d with ice-cold methanol for 4 minutes. Immunohistochemical staining was performed as previously described" using a commercial mouse immunoglobulin G staining kit. The specific antibodies used were a-smooth muscle actin at a 1:1,000 dilution, vimentin at a 1:200 dilution and cytokeratin 8.13 at a 1:200 dilution. Mouse ascites at the appropriate dilution was used as a negative control for each monoclonal antibody. L3Hl thymidine incorporation assays. Bladder stromal cells were plated at a density of 3 X lo4 cells per ml. in modified M199 media with 10% newborn calf serum in standard 12 well tissue culture plates. After attachment and growth for 24 hours the medium was removed and the cells were washed once with phosphate buffered saline. The cells were then incubated with serum deficient (0.25%newborn calf serum) or serum-free media and allowed to become quiescent for 24
hours. Quiescent cells were then stimulated with M. angiotensin 11 or 10 ng./ml. basic fibroblast growth factor individually or in combination for a n additional 48 hours. Because of possible angiotensin I1 degradation by an endogenous angiotensinase in the culture media? angiotensin 11 was added every 12 hours during this period. [3Hl thymidine (5 pCi./ml.) was added for the last 24 hours of stimulation. After stimulation cells were harvested and precipitated in ice-cold 20%trichloroacetic acid over 2 hours. Solubilization was accomplished with 1 M. sodium hydroxide at 37C overnight. Relative [3H] thymidine incorporation was determined by liquid scintillation counting. Cell proliferation assays. Bladder stromal cells were plated and stimulated with angiotensin I1 and basic fibroblast growth factor exactly as described for the L3H1 thymidine incorporation assays, except for the addition of t3Hl thymidine. These assays were performed in parallel t o the aforementioned mitogenic assays with the same cells and under the same conditions. At the end of stimulation cells were recovered using 0.25% trypsin and cell counts were performed using a Coulter counter. Statistical analysis. Data were analyzed using computer software and are expressed as means plus or minus standard error of mean. Statistical analysis was performed using the unpaired Student t test and the difference between means was considered significant at p 50.05. RESULTS
Primary cell culture characteristics. Bladder stromal cells were reliably obtained from the neonatal rabbit bladder using the aforementioned methodology. Usually 3 to 5 million cells were harvested from each bladder. Cells grew rapidly with an approximate doubling time of 24 hours. Cell proliferation slowed with confluence, suggesting some element of contact inhibition. Subsequent passaging resulted in progressively slower growth rates and cells ultimately became senescent by passages 8 to 10. Examination of cell morphology revealed 2 cell shapes, which appeared to depend on cell to cell contact and degree of confluence. In the nonconfluent state with minimal cell t o cell contact cells were triangular with emanating pseudopods, which extended outward. As cells came in contact with neighboring cells and attained confluence, a more spindle-shaped cell resulted. Confluent spindle-shaped cells then grew in a whorl-like fashion (fig. 1). Immunohistochemistry demonstrated that stromal cells were consistently positive for a-smooth muscle actin and vimentin. Cells did not stain positive for cytokeratin 8.13. Staining with cytokeratin 8.13 allows for the differentiation between epithelial and stromal cells, since it binds to determinants in a large number of cytokeratins in epithelial cells
FIG. 1. Neonatal rabbit bladder stromal cells in culture. A, cells in subconfluent state are triangular with emanating pseudopods. B , after cells attain confluence they are more spindle-shaped and grow in whorl-like fashion. Reduced from ~ 2 0 0 .
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR STROMAL CELL MITOGENESIS
but not to stromal cells. This includes cytokeratin polypeptide numbers 5, 7, 8 and 18, which are present in bladder epithelium. l 2 Cultures from early passages suggested some epithelial contamination, as evidenced by cytokeratin 8.13 positive cells. Subsequent passaging resulted in diminution of the epithelial cells, so that by passage 3 cell cultures comprised greater than 90% pure stromal cells (fig. 2). Angiotensin 11 and basic fibroblast growth factor induce stromal cell mitogenesis. After establishment and characterization of the primary stromal cell culture system, the effects of angiotensin I1 and basic fibroblast growth factor were examined by stimulating the bladder stromal cells with each factor alone and in combination during a 48-hour period. This step was performed in serum deficient (0.25%newborn calf serum) and serum-free conditions. When mitogenic and proliferation assays were performed in serum deficient media, the administration of M. angiotensin I1 or 10 ng./ml. basic fibroblast growth factor resulted in significant increases in ['HI thymidine incorporation and cell counts compared to baseline control values. With respect to L3H1 thymidine incorporation cumulative data from several individual experiments demonstrated that angiotensin I1 administration resulted in a 35% increase (p <0.01) in L3H1 thymidine incorporation and basic fibroblast growth factor administration caused a 180% increase (p <0.01) compared to control values (fig. 3, A). Results of cell proliferation assays showed similar significant increases in response to angiotensin I1 and basic fibroblast growth factor. Angiotensin I1 induced a 26% increase (p <0.01) in cell number, while basic fibroblast growth factor resulted in a 47% increase (p <0.01, fig. 3, B). No synergistic activity was observed between angiotensin I1 and basic fibroblast growth factor in either assay when the 2 were added together. In contrast, when these mitogenic and proliferation assays were performed under serum-free conditions, angiotensin I1 and basic fibroblast growth factor failed to have any stimulatory effect.
595
these cells respond to physical strain with increased production of type I11 collagen.13 In that system smooth muscle cells were identified by morphology, growth characteristics, anatomical isolation and phenotypic expression of a-smooth muscle actin. Using these identification criteria the cells established in this study could also be considered smooth muscle cells, since they were isolated from the same anatomical location and were similar in all of these criteria except for having a quicker doubling time and some element of contact inhibition. While it is likely that the cells in our system are indeed smooth muscle cells, the possibility that they may represent another cell type from the stroma (for example fibroblasts or myofibroblasts) cannot be entirely excluded at this time. As demonstrated in numerous other systems, the differentiated state between these cell types can be difficult to maintain in cell culture. The expression of a-smooth muscle actin and vimentin can be observed in all 3 of these cell types and it can also be modified by different culture conditions.14 For these reasons we are currently designating these cells as stromal cells until further characterization studies can more clearly delineate their identity. However, recent studies demonstrating that vimentin and a-smooth muscle actin are expressed in smooth muscle cells from the intact neonatal rabbit bladder provide further evidence and support that the stromal cells in our culture system are smooth muscle cells.15 Following characterization of this primary cell culture system additional experiments were performed to investigate the potential mitogenic effects of angiotensin I1 and basic fibroblast growth factor. The results of these experiments demonstrated that both factors were mitogenic to neonatal bladder stromal cells under serum deficient conditions but not under serum-free conditions. With respect to angiotensin I1 these results are important since they provide the first evidence that angiotensin I1 may act as a trophic factor in the wall of the bladder and be involved in stimulating stromal cell growth. Previous studies have shown that angiotensin I1 receptors exist in the detrusor musculature and that it is able DISCUSSION to initiate contractile activity.16.17 The additional role of anIn this study we established and characterized primary giotensin I1 as a growth factor regulated by a local tissue stromal cell cultures from the neonatal rabbit bladder. Neo- renin-angiotensin system has been suggested but it is not yet natal rabbits were used to evaluate bladder cells from an proved.18 The fact that angiotensin I1 is able to stimulate animal in early development as opposed t o those from a bladder stromal cell growth in vitro is not entirely surprising, mature and fully developed adult. The establishment of this since it has previously been shown to have similar effects on culture system is important, since it will enable further study cardiac myocytes5 and vascular smooth muscle cells.7 It is of the behavior of developing bladder cells under different in now apparent that angiotensin I1 may act as a local growth vitro conditions. Other in vitro models of primary stromal cell factor in an autocrine and/or paracrine fashion to regulate cultures from the bladder have been described previously. cell function and growth via a local tissue renin-angiotensin Baskin et a1 first demonstrated the ability to culture fetal system in the cardiovascular system.19.20 Using cultured carbovine smooth muscle cells in vitro9 and also showed that diac myocytes and fibroblasts from the neonatal rat heart Sadoshima and Izumo demonstrated that angiotensin I1 administration causes hypertrophy of myocytes and hyperplasia of fibroblasts.5 They have also shown that angiotensin I1 causes up regulation and induction of many immediate-early genes, such as c-fos, c-myc and c-jun.21 In addition, other experiments have shown that increased load has similar effects, leading to the speculation that obstruction causes the local production of angiotensin 11.22 Other studies have also produced similar findings using cardiac myocytes23 as well as vascular smooth muscle cells.24 From these data it is clear that angiotensin I1 has a role in the regulation of smooth muscle cell growth in the heart and aorta. In the present study preliminary steps have been taken to demonstrate that a similar regulatory process of smooth muscle cell growth may exist in the bladder by showing that angiotensin I1 induces mitogenesis in neonatal bladder stromal cells in vitro. The observation that basic fibroblast growth factor is a Dotent mitoaen to neonatal bladder stromal cells is also not FIG. 2. Neonatal rabbit bladder stromal cells in vitro stain positive for a-smooth muscle actin. Rare e ithelial cell (arrow) stains surprising. Recent work in the adult rabbit has shown that increased levels of basic fibroblast growth factor are present negative for a-smooth muscle actin. Refuced from X200.
-
596
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR STROMAL CELL MITOGENESIS
1 I
n cm(rm
All
bFGF
A U+bFGF
1
ConM
All
bFGF
AIWFGF
FIG.3. Angiotensin I1 (A 11)and basic fibroblast growth factor (WGF)stimulation during 48 hours. A, 35 and 180%increases, respectively, in r3H1 thymidine incorporation compared to control values (p 4 . 0 1 [*I). Results of cumulative data from 3 expefiments !each,with triplicate wells) are expressed as average of percent control value plus or minus standard error of mean. No synergstic actinty 1s noted when angiotensin I1 and basic fibroblast growth factor are added. B, 26 and 47%increases in cell number, respectively, compared to control values (p <0.02 ["I). Cell number experiments were run parallel to [3H] thymidine incorporation assays with same passage cells and under same experimental conditions.
in the bladder 24 hours after bladder neck obstruction.25 As with angiotensin 11, basic fibroblast growth factor has been shown to be mitogenic to several cell types, including smooth muscle cells, fibroblasts and endothelial cells.26 Fibroblast growth factors are a family of growth factors, which includes basic and acidic fibroblast growth factors. Basic fibroblast growth factor is 10 to 100-fold more potent than its acidic counterpart in in vitro assays, and it is thought to be more physiologically active in vivo. Although the exact function of basic fibroblast growth factor is not known, current evidence suggests that it has a major role in the composition of connective tissue following injury or stress. I t is also likely that basic fibroblast growth factor is involved in the maintenance of various cell types in their normal differentiated state. Also, basic fibroblast growth factor has been implicated in the development of pathological states, such as atherosclerosis and tumor formation.27 Given these known actions of basic fibroblast growth factor the demonstration that it is mitogenic to bladder stromal cells in vitro suggests that it may also have a role in stromal cell proliferation after outlet obstruction. The fact that basic fibroblast growth factor induced a greater mitogenic response than angiotensin I1 in this study does not necessarily translate to this being the case in the intact animal or human. The regulation of tissue availability of basic fibroblast growth factor is poorly understood but it seems to involve a complex interaction between the extracellular matrix, cell surface receptors and posttranslational modification of basic fibroblast growth factor. Also, the relationship between these 2 growth factors may be that 1factor causes up regulation of the other. For example, angiotensin I1 has been shown to cause up regulation of secondary growth factors in vascular smooth muscle cells.28 Further clarification of this regulation will provide new insight into the relevance of these in vitro results to responses in the intact animal and human. The finding that angiotensin I1 and basic fibroblast growth factor require serum in the media to exert their mitogenic effects is an intriguing phenomenon. This observation strongly suggests that there are factors in serum working in conjunction with these 2 growth factors to stimulate cell growth or that the presence of serum in the culture media alters the cells in some fashion (for example up or down regulation of cell surface receptors) to become responsive to these 2 growth factors. Serum is known to contain numerous growth factors, hormones, proteins and other factors that could act in this fashion. For example, it is well known that heparin sulfate is involved in binding basic fibroblast growth factor to its respective cell surface receptors. Heparin is
present in serum and may be necessary in this in vitro assay to stimulate cell growth. Identification of the factorb) in serum required for the mitogenic effects of angiotensin I1 and basic fibroblast growth factor is extremely important if we are to understand fully the mechanisms by which both factors stimulate cell division. CONCLUSIONS
To our knowledge our study is the first to develop and characterize neonatal rabbit bladder stromal cells in vitro. We have also demonstrated that angiotensin I1 and basic fibroblast growth factor induce mitogenesis in these cells in the serum deficient but not serum-free state. This cell culture system appears appropriate for the further study of the mechanisms by which angiotensin I1 and basic fibroblast growth factor stimulate cell division in bladder stromal cells as well as the effect of other growth factors, such as plateletderived growth factor and transforming growth factor+. Furthermore, this system will also be useful to study the effect of angiotensin I1 and other growth factors on the production of collagen. Precise knowledge of the regulatory pathways governing stromal cell growth and collagen production may allow us to diagnose and treat better the child with obstructive uropathy involving the lower urinary tract. REFERENCES
1. Kim, K. M., Kogan, B. A,, Massad, C. A. and Huang, Y.-C.:
Collagen and elastin in the obstructed fetal bladder. J. Urol., 146 528, 1991. 2. Shapiro, E., Becich, M. J., Perlman, E. and Lepor, H.: Bladder wall abnormalities in myelodysplastic bladders: a computer assisted morphometric analysis. J . Urol., 145: 1024, 1991. 3. Ewalt, D. H., Howard, P. S., Blyth, B., Snyder, H. M., 111, Duckett, J. W., Levin, R. M. and Macarak, E. J.: Is lamina propria matrix responsible for normal bladder compliance? J. Urol., 148: 544, 1992. 4. Landau, E. H., Jayanthi, V. R., Churchill, B. M., Shapiro, E., Gilmour, R. F., Khoury, A. E., Macarak, E. J., McLorie, G. A., Steckler, R. E. and Kogan, B. A.: Loss of elasticity in dysfunctional bladders: urodynamic and histochemical correlation. J. Urol., part 2, 152 702, 1994. 5. Sadoshima, J. and Izumo, S.: Molecular characterization of angiotensin 11-induced hypertrophy of cardiac myocytes and hyperplasia of cardiac fibroblasts. Critical role of the AT1 receptor subtype. Circ. Res., 73: 413, 1993. 6. Fkidy, M. A.: Growth factors and arterial smooth muscle cell proliferation. Ann. N. Y. Acad. Sci., 714: 225, 1994. 7. Weber, H., Taylor, D. S. and Molloy, C. J.: Angiotensin I1 induces delayed mitogenesis and cellular proliferation in rat aortic
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR STROMAL CELL MlTOGENESIS smooth muscle cells. Correlation with the expression of specific endogenous growth factors and reversal by suramin. J. Clin. Invest., 93: 788, 1994. 8. Kato, K., Monson, F. C., Lmghurst, P. A., Wein, A. J., Haugaard, N. and Levin, R. M.: The functional effects of long-term outlet o b struction on the rabbit urinary bladder. J. Urol., 143 600, 1990. 9. Baskin, L. S., Howard, P. S., Duckett, J . W., Snyder, H. M. and Macarak, E. J.: Bladder smooth muscle cells in culture: I. Identification and characterization. J . Urol., 149: 190, 1993. 10. Ilio, K. Y., Sensibar, J. A. and Lee, C.: Effect of TGF-01, TGF-a, and EGF on cell proliferation and cell death in rat ventral prostatic epithelial cells in culture. J. Androl., 16:482, 1995. 11. Lee, C., Sensibar, J. A,, Dudek, S. M., Hiipakka, R. A. and Liao, S. T.: Prostatic ductal system in rats: regional, variation in morphological and functional activities. Biol. Reprod., 43: 1079, 1990. 12. Cintorino, M., Del Vecchio, M. T., Bugnoli, M., Petracca, R. and Leoncini, P.: Cytokeratin pattern in normal and pathological bladder urothelium: immunohistochemical investigation using monoclonal antibodies. J. Urol., 139 428, 1988. 13. Baskin, L., Howard, P. S. and Macarak, E.: Effect of physical forces on bladder smooth muscle and urothelium. J. Urol., part 2, 150 601, 1993. 14. Sappino, A. P., Schurch, W. and Gabbiani, G.: Differentiation repertoire of fibroblastic cells: expression of cytoskeletal proteins as marker of Dhenotmic modulations. Lab. Invest.. 63: “. 144, 1990. 15. Chiaveeato. A,. Scatena. M.. Roelofs. M.. Ferrarese. P.. Pauletto. P., Passerini-Glazel, G., Pagano, F. and Sartore,’S.:’Cytoskel: eta1 and cytocontractile protein composition of smooth muscle cells in developing and obstructed rabbit bladder. Exp. Cell Res., 207: 310, 1993. 16. Anderson, G. F., Barraco, R. A,, Normille, H. J . and Rosen, T. N.: Evidence for angiotensin I1 receptors in the urinary bladder of the rabbit. Canad. J. Physiol. Pharmacol., 62: 390, 1984. 17. Tanabe, N., Ueno, A. and Tsujimoto, G.:Angiotensin I1 receptors in the rat urinary bladder smooth muscle: type 1 subtype receptors mediate contractile responses. J . Urol., 150 1056, 1993.
597
18. Lindberg, B. F., Nilsson. 1,. G . , Hedlund, H., Stahl, M. and Andersson, K. E.: Angiotensin 1 is converted to angiotensin 11 by a serine protease in human detrusor smooth muscle. Amer. J . Physiol., part 2, 266 R1861, 1994. 19. Dzau, V. J . and Pratt, R. E.: Cardiac, vascular and intrarenal renin angiotensin system in normal physiology and disease. In: The Renin-Angiotensin System. Edited by J. I. S. Robertson and M. G. Nicholls. London: Gower Medical Publishing, pp. 42.1-42.8, 1993. 20. Baker, K. M., Booz, G. W. and Dostal, D. E.: Cardiac actions of angiotensin 11: role of an intracardiac renin-angiotensin system. Ann. Rev. Physiol., 54: 227, 1992. 21. Sadoshima, J. and Izumo, S.: Signal transduction pathways of angiotensin 11-induced c-fos gene expression in cardiac myocytes in vitro. Roles of phospholipid-derived second messengers. Circ. Res., 73: 424, 1993. 22. Sadoshima, J., Jahn, L., Takahashi, T., Kulik, T. J . and Izumo, S.: Molecular characterization of the stretch-induced adaption of cultured cardiac cells. An in vitro model of load-induced cardiac hypertrophy. J. Biol. Chem., 267: 10551, 1992. 23. Baker, K. M. and Aceto, J . F.: Angiotensin I1 stimulation of protein synthesis and cell growth in chick heart cells. Amer. J. Physiol., part 2, 2 5 9 H610, 1990. 24. Gibbons, G. H., Pratt, R. E. and Dzau, V. J.: Vascular smooth muscle cell hypertrophy vs. hyperplasia. Autocrine transforming growth factor-beta 1 expression determines growth response to angiotensin 11. J. Clin. Invest., 90: 456, 1992. 25. Buttyan, R., Jacobs, B., Blaivas, J. and Levin, R.: Early molecular response to rabbit bladder outlet obstruction. Neurourol. Urodynam., 11: 225, 1992. 26. Baird, A. and Walicke, P. A,: Fibroblast growth factors. Brit. Med. Bull., 45:438, 1989. 27. Basilica, C. and Mostacelli, D.: The FGF family of growth factors and oncogenes. Adv. Cancer Res., 59 115, 1992. 28. Naftilan, A. J., Pratt, R. E. and Dzau, V.: Induction of plateletderived growth factor A-chain and c-myc gene expressions by angiotensin I1 in cultured rat vascular smooth muscle cells. J. Clin. Invest., 83: 1419, 1989.
THEJOURKAL O F UROLOGY Copyright 0 1996 by AMERICGUROLOCICAL ASSOCIATION, INC.
Vol. 156, 593-597, August 1996 Printed in U.S.A.
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR INDUCE NEONATAL BLADDER STROMAL CELL MITOGENESIS EARL Y. CHENG,“ TOM GRAMMATOPOULOS, CHUNG LEE, JULIA SENSIBAR, ROBERT DECKER, WILLIAM E. KAPLAN, MAX MAIZELS AND CASIMIR F. FIRLIT From the Division of Urology, Children’s Memorial Hospital and Departments of Urology and Medicine, Northwestern University Medical School, Chicago, Illinois
ABSTRACT
Purpose: Our aims were t o establish primary stromal cell cultures from t h e neonatal rabbit bladder and investigate t h e potential mitogenic effects of angiotensin I1 and basic fibroblast growth factor on these cells. Materials a n d Methods: Primary bladder stromal cell cultures were obtained from 3-day-old rabbits, plated at a density of 3 X lo4cells p e r ml. a n d allowed to grow for 24 hours. Subconfluent cells were growth arrested in serum deficient (0.25%newborn calf serum) or serum-free media for 24 h o u r s and then stimulated with lod7 M. angiotensin I1 or 10 ng./ml. basic fibroblast growth factor for an additional 48 hours. Cell counts and c3H1 thymidine incorporation were done to m e a s u r e cellular proliferation a n d deoxyribonucleic acid synthesis. Results: Angiotensin I1 and basic fibroblast growth factor each stimulated neonatal bladder stromal cell proliferation and r3H1 thymidine incorporation u n d e r s e r u m deficient conditions. Angiotensin I1 provoked an average 26% increase in cell number ( p <0.01) and 35% increase in [3H]thymidine incorporation ( p ~0.01) compared to control values. Basic fibroblast growth factor w a s an even more potent mitogen with a 47% increase in cell number ( p CO.01) a n d 180% increase in [3H]thymidine incorporation ( p <0.01) compared to controls. In contrast, angiotensin I1 and basic fibroblast growth factor each failed t o have significant stimulatory effects under serum-free conditions. Conclusions: Angiotensin I1 a n d basic fibroblast growth factor induce a mitogenic response to neonatal bladder stromal cells in vitro. These mitogenic effects require t h e presence of serum factors. Whether angiotensin I1 and basic fibroblast growth factor are involved in the i n vivo regulation of bladder growth associated with obstructive uropathy requires further investigation. KEYWORDS:bladder; angiotensin 11; fibroblast growth factor, basic; stromal cells Bladder neck obstruction may occur in utero (for example posterior urethral valves) and postnatally, as in men with benign prostatic hyperplasia. Regardless of when it occurs, outlet obstruction results in pathological changes in the bladder, which can ultimately alter the neurophysiology of the lower urinary tract. Histopathological changes include bladder muscle hypertrophykyperplasia and increased collagen production. It has been proposed that these changes have a role in altered compliance of the obstructed hufnan and animal bladder.1-4 Fortunately when obstruction is relieved, these changes are partially reversible in the majority of cases, resulting in normalization of bladder compliance and clinical improvement in obstructive voiding symptoms. Difficulty arises in treating patients who have adequate relief of obstruction but still have persistent bladder dysfunction. This pathophysiological condition likely exists secondary to irreversible obstructive changes in the architecture of the bladder. Currently little information is available in regard to the cellular and molecular mechanisms that regulate smooth muscle growth and collagen production in the bladder. Even less is known about factors that govern the reversibility of these processes. Further understanding of these mechanisms 1s critical for the development of better clinical diagnostic and therapeutic tools to treat bladder neck obstruction in the Pediatric and adult populations. Clinically the response of bladder smooth muscle to outlet
* Requests for reprints: Division of Urolo Children’s Memorial Hospital, 2300 Children’s Plaza, Chicago, Il%ois 60614.
obstruction is similar to that of cardiac muscle to increased load (for example hypertension or aortic coarctation). As in the bladder, outflow obstruction in the heart causes an initial compensatory response in the myocardium consisting of muscle hypertrophykyperplasia and altered collagen production. When obstruction is unrelieved, subsequent decompensation and pump failure ensue. In vitro studies using smooth muscle cells from the heart and aorta have demonstrated that angiotensin I1 and basic fibroblast growth factor may be responsible for increased smooth muscle growth in this pathological Also, angiotensin I1 may be involved in the regulation of collagen production in these cells.8 Since the bladder responds to obstruction in a fashion similar to the heart, we hypothesized that the same regulatory factors and mechanisms responsible for controlling stromal cell growth and function in the heart are also present in the bladder. To investigate this hypothesis further we attempted to establish and characterize primary stromal cell cultures from the neonatal rabbit bladder. We then investigated the role of angiotensin I1 and basic fibroblast growth factor in modulating the mitogenic properties of these cultured cells. MATERIALS AND METHODS
Establishment of primary cell cultures. Three to 5-day-old New Zealand rabbits were used for the establishment of primary neonatal bladder stromal cell cultures. Routinely bladders from 2 animals were used for each primary culture. Animals were sacrificed with an overdose of pentobarbital. Bladders were then harvested and immediately placed into
5193
594
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR STROMAL CELL MITOGENESIS
37C modified M199 media with 10%newborn calf serum, as described by Baskin et al.9 The epithelium was removed from the muscle portion of the bladder wall with a scalpel blade and the bladders were processed as previously described.I0 Bladders were minced into 1 mm.3 pieces with curved scissors. The minced tissue was placed in 50 ml. 1.0 mM. dithiothreitol in phosphate buffered saline. The tube was subjected to a rotator speed of 30 cycles per minute at 37C for 30 minutes. The supernatant was discarded and the tissue pellet was resuspended with 10 ml. dissociation solution containing 1% deoxyribonuclease type I and 0.28%collagenase type I in RPMI-1640 with 10% fetal bovine serum and rotated at 37C for 30 minutes. The supernatant fraction containing the separated individual cells was then passed through a series of 3 sieves with 230 to 26 pm. pore sizes for the recovery of cells. The remaining tissue pieces were subjected to another round of treatment with dissociation solution and the resultant supernatant was again sieved and pooled with the first supernatant fraction. The combined cell suspension was then spun at 1,000 revolutions per minute for 5 minutes and the supernatant was discarded. The pellet was resuspended in 10 ml. modified MI99 media and washed an additional time. The resultant cells were again resuspended in modified M199 media and plated at a density of 3 X lo6 cells per 75 cm.2 flask. Cultures were incubated at 37C in a humidified atmosphere of 5% carbon dioxide in air. Following initial establishment of the primary culture cells became 90 to 100% confluent and required passaging (1:3 split) every 4 to 7 days. The cells used for this investigation were between passages 2 and 6. Zmmunohistochernistry.Bladder stromal cells were grown on slides (4 x lo4 cells per ml.), which were pre-coated with 10% poly-1-lysine for 3 to 4 days until they were 60 to 70% confluent. Excess media were gently blotted off and the cells were washed twice with phosphate buffered saline for 5 minutes. The cells were then h e d with ice-cold methanol for 4 minutes. Immunohistochemical staining was performed as previously described" using a commercial mouse immunoglobulin G staining kit. The specific antibodies used were a-smooth muscle actin at a 1:1,000 dilution, vimentin at a 1:200 dilution and cytokeratin 8.13 at a 1:200 dilution. Mouse ascites at the appropriate dilution was used as a negative control for each monoclonal antibody. L3Hl thymidine incorporation assays. Bladder stromal cells were plated at a density of 3 X lo4 cells per ml. in modified M199 media with 10% newborn calf serum in standard 12 well tissue culture plates. After attachment and growth for 24 hours the medium was removed and the cells were washed once with phosphate buffered saline. The cells were then incubated with serum deficient (0.25%newborn calf serum) or serum-free media and allowed to become quiescent for 24
hours. Quiescent cells were then stimulated with M. angiotensin 11 or 10 ng./ml. basic fibroblast growth factor individually or in combination for a n additional 48 hours. Because of possible angiotensin I1 degradation by an endogenous angiotensinase in the culture media? angiotensin 11 was added every 12 hours during this period. [3Hl thymidine (5 pCi./ml.) was added for the last 24 hours of stimulation. After stimulation cells were harvested and precipitated in ice-cold 20%trichloroacetic acid over 2 hours. Solubilization was accomplished with 1 M. sodium hydroxide at 37C overnight. Relative [3H] thymidine incorporation was determined by liquid scintillation counting. Cell proliferation assays. Bladder stromal cells were plated and stimulated with angiotensin I1 and basic fibroblast growth factor exactly as described for the L3H1 thymidine incorporation assays, except for the addition of t3Hl thymidine. These assays were performed in parallel t o the aforementioned mitogenic assays with the same cells and under the same conditions. At the end of stimulation cells were recovered using 0.25% trypsin and cell counts were performed using a Coulter counter. Statistical analysis. Data were analyzed using computer software and are expressed as means plus or minus standard error of mean. Statistical analysis was performed using the unpaired Student t test and the difference between means was considered significant at p 50.05. RESULTS
Primary cell culture characteristics. Bladder stromal cells were reliably obtained from the neonatal rabbit bladder using the aforementioned methodology. Usually 3 to 5 million cells were harvested from each bladder. Cells grew rapidly with an approximate doubling time of 24 hours. Cell proliferation slowed with confluence, suggesting some element of contact inhibition. Subsequent passaging resulted in progressively slower growth rates and cells ultimately became senescent by passages 8 to 10. Examination of cell morphology revealed 2 cell shapes, which appeared to depend on cell to cell contact and degree of confluence. In the nonconfluent state with minimal cell t o cell contact cells were triangular with emanating pseudopods, which extended outward. As cells came in contact with neighboring cells and attained confluence, a more spindle-shaped cell resulted. Confluent spindle-shaped cells then grew in a whorl-like fashion (fig. 1). Immunohistochemistry demonstrated that stromal cells were consistently positive for a-smooth muscle actin and vimentin. Cells did not stain positive for cytokeratin 8.13. Staining with cytokeratin 8.13 allows for the differentiation between epithelial and stromal cells, since it binds to determinants in a large number of cytokeratins in epithelial cells
FIG. 1. Neonatal rabbit bladder stromal cells in culture. A, cells in subconfluent state are triangular with emanating pseudopods. B , after cells attain confluence they are more spindle-shaped and grow in whorl-like fashion. Reduced from ~ 2 0 0 .
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR STROMAL CELL MITOGENESIS
but not to stromal cells. This includes cytokeratin polypeptide numbers 5, 7, 8 and 18, which are present in bladder epithelium. l 2 Cultures from early passages suggested some epithelial contamination, as evidenced by cytokeratin 8.13 positive cells. Subsequent passaging resulted in diminution of the epithelial cells, so that by passage 3 cell cultures comprised greater than 90% pure stromal cells (fig. 2). Angiotensin 11 and basic fibroblast growth factor induce stromal cell mitogenesis. After establishment and characterization of the primary stromal cell culture system, the effects of angiotensin I1 and basic fibroblast growth factor were examined by stimulating the bladder stromal cells with each factor alone and in combination during a 48-hour period. This step was performed in serum deficient (0.25%newborn calf serum) and serum-free conditions. When mitogenic and proliferation assays were performed in serum deficient media, the administration of M. angiotensin I1 or 10 ng./ml. basic fibroblast growth factor resulted in significant increases in ['HI thymidine incorporation and cell counts compared to baseline control values. With respect to L3H1 thymidine incorporation cumulative data from several individual experiments demonstrated that angiotensin I1 administration resulted in a 35% increase (p <0.01) in L3H1 thymidine incorporation and basic fibroblast growth factor administration caused a 180% increase (p <0.01) compared to control values (fig. 3, A). Results of cell proliferation assays showed similar significant increases in response to angiotensin I1 and basic fibroblast growth factor. Angiotensin I1 induced a 26% increase (p <0.01) in cell number, while basic fibroblast growth factor resulted in a 47% increase (p <0.01, fig. 3, B). No synergistic activity was observed between angiotensin I1 and basic fibroblast growth factor in either assay when the 2 were added together. In contrast, when these mitogenic and proliferation assays were performed under serum-free conditions, angiotensin I1 and basic fibroblast growth factor failed to have any stimulatory effect.
595
these cells respond to physical strain with increased production of type I11 collagen.13 In that system smooth muscle cells were identified by morphology, growth characteristics, anatomical isolation and phenotypic expression of a-smooth muscle actin. Using these identification criteria the cells established in this study could also be considered smooth muscle cells, since they were isolated from the same anatomical location and were similar in all of these criteria except for having a quicker doubling time and some element of contact inhibition. While it is likely that the cells in our system are indeed smooth muscle cells, the possibility that they may represent another cell type from the stroma (for example fibroblasts or myofibroblasts) cannot be entirely excluded at this time. As demonstrated in numerous other systems, the differentiated state between these cell types can be difficult to maintain in cell culture. The expression of a-smooth muscle actin and vimentin can be observed in all 3 of these cell types and it can also be modified by different culture conditions.14 For these reasons we are currently designating these cells as stromal cells until further characterization studies can more clearly delineate their identity. However, recent studies demonstrating that vimentin and a-smooth muscle actin are expressed in smooth muscle cells from the intact neonatal rabbit bladder provide further evidence and support that the stromal cells in our culture system are smooth muscle cells.15 Following characterization of this primary cell culture system additional experiments were performed to investigate the potential mitogenic effects of angiotensin I1 and basic fibroblast growth factor. The results of these experiments demonstrated that both factors were mitogenic to neonatal bladder stromal cells under serum deficient conditions but not under serum-free conditions. With respect to angiotensin I1 these results are important since they provide the first evidence that angiotensin I1 may act as a trophic factor in the wall of the bladder and be involved in stimulating stromal cell growth. Previous studies have shown that angiotensin I1 receptors exist in the detrusor musculature and that it is able DISCUSSION to initiate contractile activity.16.17 The additional role of anIn this study we established and characterized primary giotensin I1 as a growth factor regulated by a local tissue stromal cell cultures from the neonatal rabbit bladder. Neo- renin-angiotensin system has been suggested but it is not yet natal rabbits were used to evaluate bladder cells from an proved.18 The fact that angiotensin I1 is able to stimulate animal in early development as opposed t o those from a bladder stromal cell growth in vitro is not entirely surprising, mature and fully developed adult. The establishment of this since it has previously been shown to have similar effects on culture system is important, since it will enable further study cardiac myocytes5 and vascular smooth muscle cells.7 It is of the behavior of developing bladder cells under different in now apparent that angiotensin I1 may act as a local growth vitro conditions. Other in vitro models of primary stromal cell factor in an autocrine and/or paracrine fashion to regulate cultures from the bladder have been described previously. cell function and growth via a local tissue renin-angiotensin Baskin et a1 first demonstrated the ability to culture fetal system in the cardiovascular system.19.20 Using cultured carbovine smooth muscle cells in vitro9 and also showed that diac myocytes and fibroblasts from the neonatal rat heart Sadoshima and Izumo demonstrated that angiotensin I1 administration causes hypertrophy of myocytes and hyperplasia of fibroblasts.5 They have also shown that angiotensin I1 causes up regulation and induction of many immediate-early genes, such as c-fos, c-myc and c-jun.21 In addition, other experiments have shown that increased load has similar effects, leading to the speculation that obstruction causes the local production of angiotensin 11.22 Other studies have also produced similar findings using cardiac myocytes23 as well as vascular smooth muscle cells.24 From these data it is clear that angiotensin I1 has a role in the regulation of smooth muscle cell growth in the heart and aorta. In the present study preliminary steps have been taken to demonstrate that a similar regulatory process of smooth muscle cell growth may exist in the bladder by showing that angiotensin I1 induces mitogenesis in neonatal bladder stromal cells in vitro. The observation that basic fibroblast growth factor is a Dotent mitoaen to neonatal bladder stromal cells is also not FIG. 2. Neonatal rabbit bladder stromal cells in vitro stain positive for a-smooth muscle actin. Rare e ithelial cell (arrow) stains surprising. Recent work in the adult rabbit has shown that increased levels of basic fibroblast growth factor are present negative for a-smooth muscle actin. Refuced from X200.
-
596
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR STROMAL CELL MITOGENESIS
1 I
n cm(rm
All
bFGF
A U+bFGF
1
ConM
All
bFGF
AIWFGF
FIG.3. Angiotensin I1 (A 11)and basic fibroblast growth factor (WGF)stimulation during 48 hours. A, 35 and 180%increases, respectively, in r3H1 thymidine incorporation compared to control values (p 4 . 0 1 [*I). Results of cumulative data from 3 expefiments !each,with triplicate wells) are expressed as average of percent control value plus or minus standard error of mean. No synergstic actinty 1s noted when angiotensin I1 and basic fibroblast growth factor are added. B, 26 and 47%increases in cell number, respectively, compared to control values (p <0.02 ["I). Cell number experiments were run parallel to [3H] thymidine incorporation assays with same passage cells and under same experimental conditions.
in the bladder 24 hours after bladder neck obstruction.25 As with angiotensin 11, basic fibroblast growth factor has been shown to be mitogenic to several cell types, including smooth muscle cells, fibroblasts and endothelial cells.26 Fibroblast growth factors are a family of growth factors, which includes basic and acidic fibroblast growth factors. Basic fibroblast growth factor is 10 to 100-fold more potent than its acidic counterpart in in vitro assays, and it is thought to be more physiologically active in vivo. Although the exact function of basic fibroblast growth factor is not known, current evidence suggests that it has a major role in the composition of connective tissue following injury or stress. I t is also likely that basic fibroblast growth factor is involved in the maintenance of various cell types in their normal differentiated state. Also, basic fibroblast growth factor has been implicated in the development of pathological states, such as atherosclerosis and tumor formation.27 Given these known actions of basic fibroblast growth factor the demonstration that it is mitogenic to bladder stromal cells in vitro suggests that it may also have a role in stromal cell proliferation after outlet obstruction. The fact that basic fibroblast growth factor induced a greater mitogenic response than angiotensin I1 in this study does not necessarily translate to this being the case in the intact animal or human. The regulation of tissue availability of basic fibroblast growth factor is poorly understood but it seems to involve a complex interaction between the extracellular matrix, cell surface receptors and posttranslational modification of basic fibroblast growth factor. Also, the relationship between these 2 growth factors may be that 1factor causes up regulation of the other. For example, angiotensin I1 has been shown to cause up regulation of secondary growth factors in vascular smooth muscle cells.28 Further clarification of this regulation will provide new insight into the relevance of these in vitro results to responses in the intact animal and human. The finding that angiotensin I1 and basic fibroblast growth factor require serum in the media to exert their mitogenic effects is an intriguing phenomenon. This observation strongly suggests that there are factors in serum working in conjunction with these 2 growth factors to stimulate cell growth or that the presence of serum in the culture media alters the cells in some fashion (for example up or down regulation of cell surface receptors) to become responsive to these 2 growth factors. Serum is known to contain numerous growth factors, hormones, proteins and other factors that could act in this fashion. For example, it is well known that heparin sulfate is involved in binding basic fibroblast growth factor to its respective cell surface receptors. Heparin is
present in serum and may be necessary in this in vitro assay to stimulate cell growth. Identification of the factorb) in serum required for the mitogenic effects of angiotensin I1 and basic fibroblast growth factor is extremely important if we are to understand fully the mechanisms by which both factors stimulate cell division. CONCLUSIONS
To our knowledge our study is the first to develop and characterize neonatal rabbit bladder stromal cells in vitro. We have also demonstrated that angiotensin I1 and basic fibroblast growth factor induce mitogenesis in these cells in the serum deficient but not serum-free state. This cell culture system appears appropriate for the further study of the mechanisms by which angiotensin I1 and basic fibroblast growth factor stimulate cell division in bladder stromal cells as well as the effect of other growth factors, such as plateletderived growth factor and transforming growth factor+. Furthermore, this system will also be useful to study the effect of angiotensin I1 and other growth factors on the production of collagen. Precise knowledge of the regulatory pathways governing stromal cell growth and collagen production may allow us to diagnose and treat better the child with obstructive uropathy involving the lower urinary tract. REFERENCES
1. Kim, K. M., Kogan, B. A,, Massad, C. A. and Huang, Y.-C.:
Collagen and elastin in the obstructed fetal bladder. J. Urol., 146 528, 1991. 2. Shapiro, E., Becich, M. J., Perlman, E. and Lepor, H.: Bladder wall abnormalities in myelodysplastic bladders: a computer assisted morphometric analysis. J . Urol., 145: 1024, 1991. 3. Ewalt, D. H., Howard, P. S., Blyth, B., Snyder, H. M., 111, Duckett, J. W., Levin, R. M. and Macarak, E. J.: Is lamina propria matrix responsible for normal bladder compliance? J. Urol., 148: 544, 1992. 4. Landau, E. H., Jayanthi, V. R., Churchill, B. M., Shapiro, E., Gilmour, R. F., Khoury, A. E., Macarak, E. J., McLorie, G. A., Steckler, R. E. and Kogan, B. A.: Loss of elasticity in dysfunctional bladders: urodynamic and histochemical correlation. J. Urol., part 2, 152 702, 1994. 5. Sadoshima, J. and Izumo, S.: Molecular characterization of angiotensin 11-induced hypertrophy of cardiac myocytes and hyperplasia of cardiac fibroblasts. Critical role of the AT1 receptor subtype. Circ. Res., 73: 413, 1993. 6. Fkidy, M. A.: Growth factors and arterial smooth muscle cell proliferation. Ann. N. Y. Acad. Sci., 714: 225, 1994. 7. Weber, H., Taylor, D. S. and Molloy, C. J.: Angiotensin I1 induces delayed mitogenesis and cellular proliferation in rat aortic
ANGIOTENSIN I1 AND BASIC FIBROBLAST GROWTH FACTOR STROMAL CELL MlTOGENESIS smooth muscle cells. Correlation with the expression of specific endogenous growth factors and reversal by suramin. J. Clin. Invest., 93: 788, 1994. 8. Kato, K., Monson, F. C., Lmghurst, P. A., Wein, A. J., Haugaard, N. and Levin, R. M.: The functional effects of long-term outlet o b struction on the rabbit urinary bladder. J. Urol., 143 600, 1990. 9. Baskin, L. S., Howard, P. S., Duckett, J . W., Snyder, H. M. and Macarak, E. J.: Bladder smooth muscle cells in culture: I. Identification and characterization. J . Urol., 149: 190, 1993. 10. Ilio, K. Y., Sensibar, J. A. and Lee, C.: Effect of TGF-01, TGF-a, and EGF on cell proliferation and cell death in rat ventral prostatic epithelial cells in culture. J. Androl., 16:482, 1995. 11. Lee, C., Sensibar, J. A,, Dudek, S. M., Hiipakka, R. A. and Liao, S. T.: Prostatic ductal system in rats: regional, variation in morphological and functional activities. Biol. Reprod., 43: 1079, 1990. 12. Cintorino, M., Del Vecchio, M. T., Bugnoli, M., Petracca, R. and Leoncini, P.: Cytokeratin pattern in normal and pathological bladder urothelium: immunohistochemical investigation using monoclonal antibodies. J. Urol., 139 428, 1988. 13. Baskin, L., Howard, P. S. and Macarak, E.: Effect of physical forces on bladder smooth muscle and urothelium. J. Urol., part 2, 150 601, 1993. 14. Sappino, A. P., Schurch, W. and Gabbiani, G.: Differentiation repertoire of fibroblastic cells: expression of cytoskeletal proteins as marker of Dhenotmic modulations. Lab. Invest.. 63: “. 144, 1990. 15. Chiaveeato. A,. Scatena. M.. Roelofs. M.. Ferrarese. P.. Pauletto. P., Passerini-Glazel, G., Pagano, F. and Sartore,’S.:’Cytoskel: eta1 and cytocontractile protein composition of smooth muscle cells in developing and obstructed rabbit bladder. Exp. Cell Res., 207: 310, 1993. 16. Anderson, G. F., Barraco, R. A,, Normille, H. J . and Rosen, T. N.: Evidence for angiotensin I1 receptors in the urinary bladder of the rabbit. Canad. J. Physiol. Pharmacol., 62: 390, 1984. 17. Tanabe, N., Ueno, A. and Tsujimoto, G.:Angiotensin I1 receptors in the rat urinary bladder smooth muscle: type 1 subtype receptors mediate contractile responses. J . Urol., 150 1056, 1993.
597
18. Lindberg, B. F., Nilsson. 1,. G . , Hedlund, H., Stahl, M. and Andersson, K. E.: Angiotensin 1 is converted to angiotensin 11 by a serine protease in human detrusor smooth muscle. Amer. J . Physiol., part 2, 266 R1861, 1994. 19. Dzau, V. J . and Pratt, R. E.: Cardiac, vascular and intrarenal renin angiotensin system in normal physiology and disease. In: The Renin-Angiotensin System. Edited by J. I. S. Robertson and M. G. Nicholls. London: Gower Medical Publishing, pp. 42.1-42.8, 1993. 20. Baker, K. M., Booz, G. W. and Dostal, D. E.: Cardiac actions of angiotensin 11: role of an intracardiac renin-angiotensin system. Ann. Rev. Physiol., 54: 227, 1992. 21. Sadoshima, J. and Izumo, S.: Signal transduction pathways of angiotensin 11-induced c-fos gene expression in cardiac myocytes in vitro. Roles of phospholipid-derived second messengers. Circ. Res., 73: 424, 1993. 22. Sadoshima, J., Jahn, L., Takahashi, T., Kulik, T. J . and Izumo, S.: Molecular characterization of the stretch-induced adaption of cultured cardiac cells. An in vitro model of load-induced cardiac hypertrophy. J. Biol. Chem., 267: 10551, 1992. 23. Baker, K. M. and Aceto, J . F.: Angiotensin I1 stimulation of protein synthesis and cell growth in chick heart cells. Amer. J. Physiol., part 2, 2 5 9 H610, 1990. 24. Gibbons, G. H., Pratt, R. E. and Dzau, V. J.: Vascular smooth muscle cell hypertrophy vs. hyperplasia. Autocrine transforming growth factor-beta 1 expression determines growth response to angiotensin 11. J. Clin. Invest., 90: 456, 1992. 25. Buttyan, R., Jacobs, B., Blaivas, J. and Levin, R.: Early molecular response to rabbit bladder outlet obstruction. Neurourol. Urodynam., 11: 225, 1992. 26. Baird, A. and Walicke, P. A,: Fibroblast growth factors. Brit. Med. Bull., 45:438, 1989. 27. Basilica, C. and Mostacelli, D.: The FGF family of growth factors and oncogenes. Adv. Cancer Res., 59 115, 1992. 28. Naftilan, A. J., Pratt, R. E. and Dzau, V.: Induction of plateletderived growth factor A-chain and c-myc gene expressions by angiotensin I1 in cultured rat vascular smooth muscle cells. J. Clin. Invest., 83: 1419, 1989.