Inhibitory Effect of Tranilast on Activation and Transforming Growth Factor β1 Expression in Cultured Rat Stellate Cells

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.

227, 322–327 (1996)

1508

Inhibitory Effect of Tranilast on Activation and Transforming Growth Factor b1 Expression in Cultured Rat Stellate Cells Hitoshi Ikeda,* Miki Inao,† and Kenji Fujiwara†,1 *First Department of Internal Medicine, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan; and †Third Department of Internal Medicine, Saitama Medical School, 38 Moro-hongo, Moroyama-chou, Iruma-gun, Saitama 350-04, Japan Received August 29, 1996 Stellate cells, the primary extracellular matrix-producing cells in the liver, undergo activation characterized by fibrogenesis, proliferation and smooth muscle a-actin expression, in hepatic fibrosis or when cultured on plastic. TGFb1 is known to have a pivotal role in fibrogenesis. Tranilast, a drug used for allergic diseases with anti-inflammatory effects, is known to inhibit collagen synthesis by cultured fibroblasts. Thus, effects of tranilast on activation and TGFb1 expression in stellate cells was investigated in vitro. Tranilast reduced collagen synthesis in a dose-related manner up to 50.8% of the control. This effect was reversible after tranilast withdrawal. The mobility of procollagen on gel electrophoresis and the ratio of intracellular procollagen to extracellular collagen concentrations were not affected by tranilast. Tranilast decreased DNA synthesis and increased smooth muscle a-actin expression. mRNA expressions of procollagen and TGFb1 were reduced by tranilast. Tranilast with anti-fibrogenic and anti-inflammatory actions merits consideration as a candidate for therapeutic agent of hepatic fibrosis. q 1996 Academic Press, Inc.

Liver cirrhosis is a chronic disease characterized by excessive collagen deposition in the liver. A variety of therapeutic candidates for this hepatic fibrosis have been proposed to reduce collagen synthesis. However, there is no established agent because of general toxic effects in vivo. Stellate cells in the liver (lipocytes, fat-storing or Ito cells) are major effector cells involved in hepatic fibrogenesis (1,2). The cells undergo activation characterized by fibrogenesis, proliferation and smooth muscle a-actin expression in the process of hepatic fibrosis and when cultured on uncoated plastic (3-9). Transforming growth factor b1 (TGFb1) stimulates collagen synthesis in cultured stellate cells (10,11), and hepatic overexpression of TGFb1 in transgenic mice is shown to cause hepatic fibrosis (12), suggesting that TGFb1 has a pivotal role in fibrogenesis. Tranilast, N-(3,4-dimethoxyciannamoyl)-anthranilic acid, is an anti-allergic drug clinically used, because it inhibits in vitro release of chemical mediators from mast cells (13), and release of cytokines from macrophages (14). Tranilast is also shown to inhibit collagen synthesis by cultured vascular smooth muscle cells (15) and fibroblasts (16), and collagen accumulation in rats with hypersensitive granulomatous inflammation (17). Thus this agent might be a therapeutic candidate for liver cirrhosis where fibrogenesis and inflammation commonly coexist, if it could inhibit activation of stellate cells. In this study, we examined the effect of tranilast on activation and TGFb1 expression in cultured stellate cells. MATERIALS AND METHODS Tranilast solution. Tranilast was provided by Kissei Pharmaceutical Co., Ltd. (Nagano, Japan). It was dissolved in 1% NaHCO3 solution at 1 mg/ml, diluted and added to the culture medium. Final concentration of NaHCO3 was 0.01%. Controls were run with NaHCO3 alone. 1

To whom correspondence should be addressed. Fax: 81-492-94-8404. 322

0006-291X/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

BBRC 5495

/

690d$$$$61

09-26-96 08:28:18

bbrca

AP: BBRC

Vol. 227, No. 2, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Isolation and culture of stellate cells. Stellate cells were isolated from male Sprague-Dawley rats (Shizuoka Laboratory Animal Center, Shizuoka, Japan), weighing 300 to 500 g, which were fed a standard pelleted diet and water ad libitum, as previously described (18,19) with some modifications (20). Isolated cells were viable at more than 90% as assessed by trypan blue exclusion and consisted of more than 90% of stellate cells as determined by staining with a horseradish peroxidase-coupled anti-desmin antibody. The cells were seeded on uncoated plastic tissueculture dishes at a starting density of 0.625 to 2.5 1 105 cells/cm2. The medium was changed 24 hours later and then every other day. All animal study protocols conformed to National Research Council criteria for humane care. Measurement of collagen synthesis. Stellate cells were incubated in DMEM containing 10% fetal bovine serum for 7 days, and then used for determination of collagen synthesis as described previously (20). The ratio of collagen synthesis to total protein synthesis was calculated according to the method of Peterkofsky and Diegelmann (21). To examine the intracellular procollagen, the medium and cells were precipitated separately and collagenase digestable radioactivity in the cells was measured as intracellular procollagen. In another experiment, to examine the reversibility of tranilast effect on collagen synthesis by stellate cells, the cells cultured for 7 days were incubated with 50 mg/ml tranilast for 24 hours, then the medium was discarded and the cells were incubated with the regular medium for another 24 hours. Collagen synthesis was measured similarly. Immunoblotting of pro a1(I) chain of procollagen. Seven days after plating, the culture medium of cofluent cells was changed to DMEM without serum and tranilast was added to the medium. Twenty-four hours later, the medium was collected and precipitated with 10% trichloroacetic acid. The pellet was dissolved in 2% sodium dodecyl sulfate (SDS) solution containing 60 mM Tris, pH 6.8, 100 mM dithiothreitol and 10% glycerol and heated at 100 7C for 5 min. odIdentification of pro a1(I) chain of procollagen was accomplished by immunoblotting as previously described (20) using a polyclonal antibody specific for the carboxy-terminal propeptide of pro a1(I) chain, which was provided by Dr. Catherine H. Wu, University of Connecticut School of Medicine. Isolation of RNA. Confluent stellate cells cultured for 7 days were incubated for 24 hours with ascorbic acid and tranilast. Total RNA was isolated by the guanidine thiocyanate/phenol/chloroform extraction method (22). Preparation of cDNA probe. With the use of a random primer synthesis method (23), the cDNA fragments of human a1(I) procollagen (24), rat TGFb1 (25) and human glyceraldehyde-3 phosphate dehydrogenase (G3PDH; Clontech Laboratories, Inc., Palo Alto, CA) were radiolabeled with [a-32P]dCTP (New England Nuclear, Boston, MA) up to a specific activity of 1 1 109 cpm/mg DNA. Northern blot. Ten micrograms of total RNA were subjected to electrophoresis through agarose gels containing formaldehyde and transferred onto nylon filters by capillary elution (26). The filters were treated with the UV crosslinker and hybridization was performed as previously described (20). Hybridization signals were quanitated by an image analyzer (Fuji BAS 2000, Tokyo, Japan). Results are relative to G3PDH expression, which was used as an internal standard. Measurement of DNA synthesis. Three days or seven days after plating, subconfluent stellate cells were incubated with tranilast for 24 hours, and labeled with 5 mCi/ml [methyl-3H]thymidine (20 Ci/mmol; New England Nuclear, Boston, MA) for the last 2 hours of incubation. In harvest, the cells were washed and the trichloroacetic acidprecipitable radioactivity was measured as DNA synthesis as previously described (20). Immunoblotting of smooth muscle a-actin expression. Seven days after plating, tranilast was added to the culture medium of confluent stellate cells. The medium was discarded 24, 48 and 72 hours later. The cells were incubated in 0.5% SDS solution containing 60 mM Tris, pH 6.8, boiled for 5 min and then disrupted by sonication. Identification and quantitation of smooth muscle a-actin were accomplished by immunoblotting as described above using monoclonal anti-smooth muscle a-actin (BioMakor, Rehovot, Israel; 20). Statistical analysis. Numerical data were expressed as the mean { S.D. of individual experiments (nÅ3). Paired data were analyzed by Student’s t-test. Each individual experiment was performed with the cells isolated from the same animal.

RESULTS AND DISCUSSION

Because stellate cells cultured on uncoated plastic are known to begin to proliferate actively after 3 days and to be activated after 7 days of incubation (5,9,27), the DNA synthesis was determined in the cells cultured for 3 days and 7 days, and the determinations related to collagen production, smooth muscle a-actin expression and TGFb1 expression were done using the cells cultured for 7 days. As shown in Table 1, tranilast reduced collagen synthesis by stellate cells in a dose-related manner (põ0.01, by one way ANOVA). Tranilast at 50 mg/ml reduced collagen synthesis to 50.8% of the untreated control. Although tranilast also reduced noncollagenous protein synthesis, the relative ratio of collagen synthesis to noncollagenous protein synthesis reduced by 50 mg/ml of tranilast was 71.0% of the untreated control, indicating that tranilast action was rather 323

AID

BBRC 5495

/

690d$$$$62

09-26-96 08:28:18

bbrca

AP: BBRC

Vol. 227, No. 2, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

TABLE 1 Effect of Tranilast on Collagen and Noncollagenous Protein Synthesis by Stellate Cells

Tranilast (mg/ml)

Collagen synthesis (dpm/mg protein 1 1002)

Experiment 1 0 10 50 100 Experiment 2 0 50 Experiment 3 0 50

3.27 2.74 1.66 1.31

{ { { {

Noncollagenous protein synthesis (dpm/mg protein 1 1002)

0.31 0.41 0.52* 0.36**

14.78 13.79 10.96 8.80

{ { { {

Relative ratio of collagen synthesis (%)

3.09 1.51 3.17 1.26*

4.0 3.5 2.7 2.7

{ { { {

0.6 0.2 0.1* 0.4*

1.45 { 0.11 0.67 { 0.04**

6.50 { 0.70 3.87 { 0.30**

4.0 { 0.2 3.1 { 0.1**

0.77 { 0.07 0.41 { 0.05**

2.06 { 0.17 1.64 { 0.12*

6.5 { 0.1 4.4 { 0.1*

* Põ0.05, **Põ0.01 compared to 0 mg/ml.

specific for collagen synthesis. When tranilast was removed from the medium after 24 hours incubation, the collagen and noncollagenous protein synthesis 24 hours later became not different from those of the untreated control cells (Table 2). Thus the reduction caused by cytotoxic effects of tranilast was less likely. It is well known that the inhibition of collagen synthesis at the posttranslational level such as blocking hydroxylation of prolyl or lysyl residues produces active intracellular degradation of underhydroxylated procollagen which in turn are cleaved extracellularly (28), and alters the mobility of procollagen on gel electrophoresis (29,30), as well as the ratio of intracellular procollagen to extracellular collagen concentration (31,32). As shown in Figure 1, there was no difference in the electrophoretic patterns on SDS-PAGE of pro a1(I) chain of procollagen in the culture medium between the presence and absence of 50 mg/ml tranilast. The ratio of intracellular procollagen to extracellular collagen concentration was 0.41:0.59 for the untreated cells and 0.45:0.55 for the cells incubated with 50 mg/ml tranilast, showing no difference. On the other hand, tranilast at 50 mg/ml decreased type I procollagen mRNA level by 44.1% of the untreated control (Figure 2). Thus it would be reasonable to assume that tranilast inhibited collagen synthesis at the pretranslational level. As shown in Table 3, tranilast reduced the DNA synthesis by the cells cultured for 3 days in a dose-related manner (põ0.01, by one way ANOVA); the reduction by 50 mg/ml tranilast was 40.1% of the untreated control levels. Although the reduction was less, tranilast also

TABLE 2 Collagen and Noncollagenous Protein Synthesis by Stellate Cells after Withdrawal of Tranilast

Treatment

Collagen synthesis (dpm/mg protein 1 1002)

Noncollagenous protein synthesis (dpm/mg protein 1 1002)

Control Tranilast withdrawal

2.91 { 0.45 3.10 { 0.32

12.73 { 0.91 11.58 { 0.19

Note. Stellate cells cultured for 7 days were incubated with 50 mg/ml tranilast for 24 hours, then the medium was discarded and the cells were incubated with the regular medium for another 24 hours. Collagen synthesis was measured by collagenase digestion assay. 324

AID

BBRC 5495

/

690d$$$$62

09-26-96 08:28:18

bbrca

AP: BBRC

Vol. 227, No. 2, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 1. (a) Immunoblot analysis of pro a1(I) chain of procollagen in the medium of cultured stellate cells in the presence (/) or absence (0) of tranilast. Proa1(I) indicates a band of pro a1(I) chain of procollagen and pCa1(I), a band of pro a1(I) chain of which aminoterminal propeptide is cleaved. (b) Immunoblot analysis of smooth muscle a-actin expression in stellate cells cultured in the presence (/) or absence (0) of tranilast for 48 h and 72 h. SMaA indicates a band of smooth muscle a-actin.

reduced the DNA synthesis by the cells cultured for 7 days in a dose-related manner (põ0.01, by one way ANOVA). The expression of smooth muscle a-actin was unchanged after 24 hours of incubation (data not shown), but increased after 48 and 72 hours of incubation with 50 mg/ ml tranilast (Figure 1). As shown in Figure 2, tranilast at 50 mg/ml decreased TGFb1 mRNA level by 50.4% of the untreated control. In hepatic fibrosis, stellate cells have been reported to be activated to undergo collagen synthesis, proliferation and smooth muscle a-actin expression with increased hepatic TGFb1 expression (7,9,33). Activation of stellate cells is also seen in vitro when the cells are cultured on plastic dishes (9). In cultured stellate cells, however, TGFb1 is shown to stimulate collagen synthesis at the pretranslational level (10,11) and to inhibit smooth muscle a-actin expression (34). On the other hand, TGFb1 is known to be expressed highly in myofibroblasts (7), endothelial cells (35) and Kupffer cells (36) in pathological states, suggesting that the regulatory mechanisms of TGFb1 expression in stellate cells or major origins of TGFb1 involved in fibrogenesis might differ in vivo and in vitro. Tranilast decreased the DNA synthesis by cultured stellate cells in a dose-related manner as well (Table 3). TGFb1 has been suggested to activate stellate cells (37). Thus it is conceivable that the action of tranilast on collagen synthesis, proliferation and smooth muscle a-actin expression was the

FIG. 2. Northern blot analysis of type I procollagen, TGFb1 and G3PDH mRNA in stellate cells cultured with tranilast. No addition and addition of tranilast at 50 and 100 mg/ml are indicated as 0, 50 and 100, respectively. 325

AID

BBRC 5495

/

690d$$$$62

09-26-96 08:28:18

bbrca

AP: BBRC

Vol. 227, No. 2, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 3 Effect of Tranilast on DNA Synthesis by Liver Stellate Cells DNA synthesis (dpm/mg protein 1 1002)

Tranilast (mg/ml) The cells cultured for 3 days Experiment 1 0 10 50 100 Experiment 2 0 50 100 Experiment 3 0 50 The cells cultured for 7 days 0 50 100

4.02 3.63 1.22 1.04

{ { { {

0.16 0.05* 0.14** 0.31**

4.33 { 0.35 2.08 { 0.05** 1.98 { 0.14** 10.95 { 1.48 4.61 { 0.48** 1.99 { 0.10 1.65 { 0.15* 1.32 { 0.24*

*Põ0.05, **Põ0.01 compared to 0 mg/ml.

result of suppressed expression of TGFb1 operating in autocrine manner. This should be investigated in the future. Tranilast is shown to have anti-inflammatory action through reduction of release of chemical mediators and cytokines as well (17). In the strategy for treatment of liver cirrhosis, antiinflammation is also important, as initial events of fibrogenesis in vivo include participation of inflammatory cells such as Kupffer cells and hepatic macrophages by interacting with stellate cells in collagen production (36,38,39). Tranilast is already used clinically for allergic disorders as an oral drug with minimal toxicity. Although smooth muscle a-actin expression stimulated by tranilast might be an adverse effect on portal hypertension through enhancing stellate cell contractility (40), tranilast with anti-collagenous and anti-inflammatory actions merits consideration as a candidate for therapeutic agent of hepatic fibrosis. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

Ramadori, G. (1991) Virchows Archiv. B Cell Pathol. 61, 147–158. Friedman, S. L. (1993) N. Engl. J. Med. 328, 1828–1835. Enzan, H. (1985) Acta Pathol. Jpn. 35, 1301–1308. Shiratori, Y., Ichida, T., Geerts, A., and Wisse, E. (1987) Dig. Dis. Sci. 32, 1281–1289. Friedman, S. L., Roll, F. J., Boyles, J., Arenson, D. M., and Bissell, D. M. (1989) J. Biol. Chem. 264, 10756– 10762. Milani, S., Herbst, H., Schuppan, D., Kim, K. Y., Riecken, E. O., and Stein, H. (1990) Gastroenterology 98, 175– 184. Nakatsukasa, H., Nagy, P., Evarts, R. P., Hsia, C-C., Marsden, E., and Thorgeirsson, S. S. (1990) J. Clin. Invest. 85, 1833–1843. Maher, J. J., and McGuire, R. F. (1990) J. Clin. Invest. 86, 1641–1648. Rockey, D. C., Boyles, J. K., Gabbiani, G., and Friedman, S. L. (1992) J. Submicrosc. Cytol. Pathol. 24, 193– 203. Davis, B. H. (1988) J. Cell Physiol. 136, 547–553. Matsuoka, M., Pham, N-T., and Tsukamoto, H. (1989) Liver 9, 71–78. Sanderson, N., Factor, V., Nagy, P., Kopp, J., Kondaiah, P., Wakefield, L., Roberts, A. B., Sporn, M. B., and Thorgeirsson, S. S. (1995) Proc. Natl. Acad. Sci. USA 92, 2572–2576. 326

AID

BBRC 5495

/

690d$$$$62

09-26-96 08:28:18

bbrca

AP: BBRC

Vol. 227, No. 2, 1996 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40.

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Azuma, H., Banno, K., and Yoshimura, T. (1976) Br. J. Pharmacol. 58, 483–488. Suzawa, H., Kikuchi, S., Ichikawa, K., and Koda, A. (1992) Jpn. J. Pharmacol. 60, 85–90. Tanaka, K., Honda, M., Kuramochi, T., and Morioka, S. (1994) Atherosclerosis 107, 179–185. Yamada, H., Tajima, S., Nishikawa, T., Murad, S., and Pinnell, S. R. (1994) J. Biochem. 116, 892–897. Isaji, M., Aruga, N., Naito, J., and Miyata, H. (1994) Life Sci. 55, 287–292. Knook, D. L., Seffelaar, A. M., and De Leeuw, A. M. (1982) Exp. Cell. Res. 139, 468–471. De Leeuw, A. M., McCarthy, S. P., Geerts, A., and Knook, D. L. (1984) Hepatology 4, 392–403. Ikeda, H., and Fujiwara, K. (1995) Hepatology 21, 1161–1166. Peterkofsky, B., and Diegelmann, R. (1971) Biochemistry 10, 988–994. Chomcynski, P., and Sacchi, N. (1987) Anal. Biochem. 162, 156–159. Feinberg, A. P., and Vogelstein, B. A. (1983) Anal. Biochem. 132, 6–13. Chu, M-L., Myers, J. C., Bernard, M. P., Ding, J-F., and Ramirez, F. (1982) Nucleic Acids Res. 10, 5925–5934. Kaname, S., Uchida, S., Ogata, E., and Kurokawa, K. (1992) Kidney Int. 42, 1319–1327. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) in Molecular Cloning: A Laboratory Manual (J. Sambrook, E. F. Fritsch, and T. Maniatis, Eds.), Vol. I, pp. 37–55. Cold Spring Harbor, New York, NY. Friedman, S. L., and Roll, F. J. (1987) Anal. Biochem. 161, 207–218. Prockop, D. J., Kivirikko, K. I., Tuderman, L., and Guzman, N. A. (1979) N. Engl. J. Med. 301, 13–23. Steinmann, B., Bruckner, P., and Superti-Furga, A. (1991) J. Biol. Chem. 266, 1299–1303. Torre-Blanco, A., Adachi, E., Hojima, Y., Wootton, J. A. M., Minor, R. R., and Prockop, D. J. (1992) J. Biol. Chem. 267, 2650–2655. Blanck, T. J. J., and Peterkofsky, B. (1975) Arch. Biochem. Biophys. 171, 259–267. Schwarz, R. I., Mandell, R. B., and Bissell, M. J. (1983) Mol. Cell. Biol. 1, 843–853. Geerts, A., Lanzou, J. M., De Bleser, P., and Wisse, E. (1992) Hepatology 13, 1193–1202. Davis, B. H., Rapp, U. R., and Davidson, N. O. (1991) Biochem. J. 278, 43–47. Bissell, D. M., Wang, S-S., Jarnagin, W. R., and Roll, F. J. (1995) J. Clin. Invest. 96, 447–455. Matsuoka, M., and Tsukamoto, H. (1990) Hepatology 11, 599–605. Bachem, M. G., Meyer, D., Melchior, R., Sell, K-M., and Gressner, A. (1992) J. Clin. Invest. 89, 19–27. Shiratori, Y., Geerts, A., Ichida, T., Kawase, T., and Wisse, E. (1986) J. Hepatol. 3, 294–303. Friedman, S. L., and Arthur, M. J. P. (1989) J. Clin. Invest. 84, 1780–1785. Rockey, D. C., Housset, C. N., and Friedman, S. L. (1983) J. Clin. Invest. 92, 1795–1804.

327

AID

BBRC 5495

/

690d$$$$62

09-26-96 08:28:18

bbrca

AP: BBRC