Liu P et al / Acta Pharmacol Sin 2002 Aug; 23 (8): 733-738

Effect of salvianolic acid B on collagen production and mitogen-activated protein kinase activity in rat hepatic stellate cells¹

LIU Ping², LIU Cheng-Hai, WANG Hai-Nan, HU Yi-Yang, LIU Cheng

Institute of Liver Disease, Shanghai University of Traditional Chinese Medicine, Shanghai 200032, China

¹ Project supported by the National Science Foundation for Outstanding Youth, ¡í 39585128.

² Correspondence to Dr LIU Ping. Phn 86-21-5423-1109. Fax 86-21-6403-6889. E-mail Liuliver@online.sh.cn

Received 2001-08-13 Accepted 2002-05-30

KEY WORDS Salviae miltiorrhizae; salvianolic acid B; liver cirrhosis; cell division; collagen; transforming growth factor beta; mitogen-activated protein kinases; signal transduction

ABSTRACT

AIM: To investigate the mechanism of salvianolic acid -B (SA-B) action against liver fibrosis relating to mediating hepatic stellate cell (HSC) activation and transforming growth factor-¦Â1 (TGF-¦Â1) intracellular signal transduction. METHODS: HSC was isolated from normal rat through in situ perfusion of liver with pronase E and density-gradient centrifugation with 11 % nycondenz, then cells were subcultured. Cell proliferation was observed by [³H]TdR uptake. Cellular collagen deposition was measured with Ponceau S stain and semi-quantified with image analytic system. Type I collagen secretion in the supernatant was detected with ELISA. The gene expression of type I pro-collagen was analyzed by RT-PCR. The supernatant was acidified and active TGF-¦Â1 contents were assayed with ELISA. Mitogen-activated protein kinase (MAPK) activity was analyzed with immunoprecipitation and Western blot. Results: SA-B 0.1, 1, 10, and 100 ¦Ìmol/L suppressed HSC proliferation concentration-dependently as determined by [³H]TdR uptake by 94.1 %, 82.4 %, 62.7 %, and 4 % of the control respectively (P<0.05 or P<0.01). SA-B 1, 10, and 100 ¦Ìmol/L inhibited soluble type I collagen secretion by 75.3 %, 69.8 %, and 63.5 % of the control and decreased the matrix collagen deposition to 86.2 %, 75.4 %, and 73.4 % (P<0.05 or P< 0.01). SA-B 1 and 10 ¦Ìmol/L decreased the cell active TGF-¦Â1 secretion by 63.3 % and 15.6 % of the control, down-regulated pro-collgen ¦Á₁(I) mRNA expression to 77.0 % and 51.8 % respectively (P<0.05). SA-B 1 and 10 ¦Ìmol/L also inhibited MAPK activity by 1 to 2 fold respectively. Conclusion: SA-B inhibited HSC proliferation and collagen production as well as decreased the cells' TGF-¦Â1 autocrine and MAPK activity, which might contri-bute to the mechanism of SA-B action against hepatic fibrosis.

INTRODUCTION

Liver fibrosis, characterized by the overproduction and accumulation of hepatic extracellular matrix (ECM), was the common and important pathological changes in the chronic liver diseases, through which the chronic hepatitis developed into cirrhosis. Hepatic stellate cells (HSC), located at Disse's space in liver, was widely recognized as the main resource of hepatic ECM and the cellular basis for liver fibrosis. Transforming growth factor-¦Â1 (TGF-¦Â1) was one of the most potent pro-fibrogenetic cytokines and can stimulate HSC activation through paracrine or/and autocrine by mediating Ras/mitogen-activated protein kinase (MAPK) and other signal transduction pathways^[1].

Radix salviae Miltiorrhizae, one of frequently used Chinese herbs, had good action against liver fibrosis^[2]. Salvianolic acid B (SA-B), one of water soluble component from Radix salviae Miltiorrhizae, has molecular formula C₃₆H₃₀O₁₆ with M_r 718. In our previous study^[3,4], it is found that SA-B can effectively reverse liver fibrosis induced by tetrachloride carbon (CCl₄) or dimethylnitrosamine (DMN). But the mechanism of SA-B action against liver fibrosis was still unclear. In the study, we investigated SA-B effects on HSC collagen production, TGF-¦Â1 expression, and TGF-¦Â1-mediated MAPK activity, to elucidate the mechanism of SA-B effect on liver fibrosis.

MATERIALS AND METHODS

Animals Male Wistar rats, weighing (360±20) g (SPF, Certificate ¡í 980004), were obtained from the Animal Center of Shanghai University of Traditional Chinese Medicine, and were fed with food and water ad libitum.

Chemical structure of SA-B. C₃₆H₃₀O₁₆, M_r 718.

Drugs and reagents SA-B was extracted and identified by Shanghai Institute of Materia Medica, Chinese Academy of Sciences, and diluted into the concentrations of 0.1-100 ¦Ìmol/L with Medium 199 (M199). Pronase E, type IV collagenase, DNase I, Modified Eagle's Medium (MEM), and nycodenz were purchased from Sigma Co. M199 and Dulbecco's Modified Eagle's Medium (DMEM) were from Gibco/BRL. New bovine serum (NBS) was from Hyclone Co. [³H]TdR was from Shanghai Atom Co. TGF-¦Â1 immunoassay system kit, PCR marker, and Access RT-PCR Kit were from Promega Co. Rabbit-anti mouse type I collagen serum (1:500) and human recombinant TGF-¦Â1 were from Calbiochem Co. Trizol reagent was from Shanghai Sangon Co.

HSC isolation and culture HSC was isolated as described previously^[5]. Briefly, the liver was perfused in situ with a Ca²⁺/Mg²⁺-free MEM followed with pronase E and collagenase. The digested liver suspension was filtered through double layer gauze and was gradiently centrifuged (1450×g, 4 ¡æ, 20 min) with 11 % (w/v) nycodenz. HSC was obtained from the interface of gradient solution and seeded onto 100-mm uncoated plastic culture plates at a density of 1×10⁹ /L and incubated at 37 ¡æ in a 5 % CO₂-95 % air humidified atmosphere. The medium was changed with M199 containing 10 % NBS after 48 h and every other day thereafter. The cells were trypsinized and passaged after they grew confluent and applied for the following experiments as the subcultured HSC grew confluent.

Cell proliferation assay HSC proliferation was measured by [³H]TdR uptake^[6]. Confluent HSC in 24-well plates was incubated with SA-B for 48 h (the following as the same), and impulsed with 74 kBq/well [³H]TdR among the last 24 h. Then the cells were harvested by trypsinization and collected onto 0.45 mm membrane. The radioactivity of intracellular [³H]TdR incorporation was counted by Beckman Wallac 1410 liquid-scintillator.

Collagen deposition measurement After treated with SA-B, cells on the 24 well-plate were washed with phosphate buffer saline (PBS), fixed with 10 % formalin, dehydrated with 70 % ethanol, and stained with Victorial blue for 15 min. After washing with distilled water twice, cells were stained with Ponceau S for 2 min and dehydrated with ethanol twice. The collagen deposition was analyzed according to its optical density with computer MPIAS-500 image analytic system.

Type I collagen and active TGF-¦Â1 content assay After drug treatment, cells were incubated with serum free M199 for 24 h and the supernatants were collected. Type I collagen contents were assayed with ELISA according to Rennard's method^[7]. For active TGF-¦Â1 content assay, the supernatants were acidified with HCl 1 mol/L and then measured by ELISA with the TGF-¦Â1 immunoassay system kit according to the manufacturer's instructions.

Pro-collagen type I mRNA level analysis Total RNA was extracted from the cells by the acid-guandium -phenol-chloroform method as described^[8]. mRNA levels of TGF-¦Â1 and pro-collagen type I were analyzed by reverse-transcription polymerase chain reaction (RT-PCR), while house keeping gene-"¦Â-actin" was used as the internal standard. The left and right primers for ¦Â-actin were respectively as follows: 5'-ACA TCT GCT GGA AGG TGG AC-3' and 5'-GGT ACC ACC ATG TAC CCA GG-3' (the expected product size 163 bp)^[9]. The left and right primers for type I pro-collagen ¦Á₁(I): 5'-CAC CCT CAA GAG CCT GAG TC-3' and 5'-GTT CGG GCT GAT GTA CCA GT-3' (the expected product size 253 bp)^[10]. The RT-PCR reaction was carried on with Access RT-PCR kit. The products were fractionated in 2 % agarose electrophoresis containing ethidium bromide (EB) 0.5 mg/L and photographed under UV transilluminator. Their density was semi-quantitatively analyzed with MPIAS-500 image system.

MAPK activity assay HSC was treated with TGF-¦Â1 5 mg/L for 1 h and then incubated with SA-B. The cytoplasmic proteins were lysed and immuno-precipitated with p44/p42 MAPK antibody. After reacting with EIK-1 and ATP substrate, the immunoprecipitated antigen was fractionated by SDS/polyacrylamide gel electrophoresis, and were transferred onto PVDF membranes. After blocking with 5 % skim milk, the blot was incubated with p44/p42 MAPK primary antibody (1:1000) overnight at 4 ¡æ. Then the blot was washed and incubated with HRP-coupled second antibody and signals were detected by ECL film.

Statistical analysis All data are expressed as mean±SD and were analyzed with t test.

RESULTS

Effect of SA-B on HSC proliferation SA-B 0.1-100 ¦Ìmol/L inhibited the HSC intracellular uptake of [³H]TdR in a concentration-dependent manner and the inhibitory rates were 94.1 %, 82.4 %, 62.7 %, and 4 % of the control, respectively (Tab 1).

Tab 1. Effect of SA-B on HSC intracellular [³H]TdR uptake. n=4 samples. Mean±SD. ^bP<0.05, ^cP<0.01 vs Control. The data is a representative of 2-3 experiments.

Group	[3H]TdR (Bq/well)
Control	408¡À52
SA-B 0.1  mmol/L	384¡À48b
1      mmol/L	336¡À24b
10      mmol/L	256¡À28c
100     mmol/L	16¡À4c

Effect of SA-B on HSC collagen production Overproduction of extracellular matrix, particularly collagen, is an important feature of liver fibrosis. SA-B 1, 10, and 100 ¦Ìmol/L suppressed soluble type I collagen to 75.3 %, 69.8 %, and 63.5 % of the control, and inhibited HSC total collagen deposition to 86.2 %, 75.4 %, and 73.4 % of the control, respectively (Tab 2). Also RT-PCR analysis showed that SA-B 1 and 10 ¦Ìmol/L down-regulated the expression of type I pro-collagen mRNA to 77.0 % and 51.8 % of the control (Tab 3, Fig 1).

Tab 2. SA-B effect on HSC collagen secretion and deposition. n=4 samples. Mean±SD. ^bP<0.05, ^cP<0.01 vs Control.

Group	Soluble type I collagen/mg¡Ág-1 protein	Matrix collagen (A)
Control	475¡À74	812¡À196
SA-B 0.1  mmol/L	424¡À22	766¡À48
1   mmol/L	358¡À57b	699¡À117b
10   mmol/L	332¡À69b	613¡À194c
100   mmol/L	302¡À56c	596¡À134c

Tab 3. Effect of SA-B on type I procollagen mRNA and active TGF-¦Â1 level. n=4 samples. Mean±SD. ^bP<0.05, ^cP<0.01 vs Control.

Group	Pro-collagen a1(I)mRNA /% of b-actin	Active TGF-b1 /mg¡ÁL-1
Control	54¡À8	3451¡À661
SA-B 0.1  mmol/L	42¡À6b	2185¡À734b
10   mmol/L	28.0¡À1.5c	539¡À282c

Fig 1. Effect of SA-B on the gene expression of type I pro-collagen in HSC. RT-PCR product 5 ¦ÌL was run on 2 % agarose gel stained with EB. Lane 1: PCR markers; Lane 2 and 3: ¦Á₁(I) pro-collagen and ¦Â-actin in SA-B 1 ¦Ìmol/L treated HSC; Lane 4 and 5: ¦Á₁(I) pro-collagen and ¦Â-actin in SA-B 10 ¦Ìmol/L treated HSC; Lane 6 and 7: ¦Á₁(I) pro-collagen and ¦Â-actin in the control respectively.

Effect of SA-B on HSC TGF-¦Â1 secretion TGF-¦Â1 is one of the most potent cytokines for extracellular matrix synthesis and HSC is main effector and producer of TGF-¦Â1 in the liver. On acidification, the active TGF-¦Â1 was released from the complex of TGF-¦Â1 and latency-associated peptide (LAP), while SA-B 1 and 10 ¦Ìmol/L significantly decreased the active TGF-¦Â1 content in the culture medium (Tab 3).

Effect of SA-B on HSC MAPK activity Ras/MAPK is an important pathway of TGF-b signaling in HSC. To elucidate the possible effects of SA-B on TGF-¦Â1 signaling in HSC, we examined MAPK activity. As shown in Fig 2, the TGF-¦Â1 5 mg/L increased MAPK phosphorylation level remarkably, while SA-B 1 ¦Ìmol/L and 10 ¦Ìmol/L reduced TGF-¦Â1 stimulated-MAPK phosphorylation level by 1- and 2-fold respectively.

Fig 2. SA-B effect on phosphorylated MAPK level in HSC. Lane 1: protein markers; Lane 2: normal HSC; Lane 3: control HSC stimulated with TGF-¦Â1 5 ¦Ìg/L; Lane 4: the HSC treated SA-B 1 ¦Ìmol/L and TGF-¦Â1 5 ¦Ìg/L; Lane 5: the HSC stimulated with SA-B 10 ¦Ìmol/L and TGF-b1 5 ¦Ìg/L.

DISCUSSION

It is widely known that HSC is the key cellular basis for liver fibrosis. On liver injury, free radicals, cytokines, etc, would stimulate HSC activation. HSC cultured in uncoated plastic plate can spontaneously activate in vitro, which mimics the activation processes in vivo. As HSC activated, they transformed from the quiescent state full of cytoplasmic lipid droplets to the myofibroblast like cell phenotypically, produced a lot extracellular matrix in liver and led to liver fibrosis^[11]. Activation of stellate cells were mediated by cytokines, which include platelet derived growth factor (PDGF), basic fibroblast growth factor (bFGF), hepatic growth factor (HGF), and transforming growth factor-¦Â1 (TGF-¦Â1), etc. Theses cytokines promoted HSC proliferation and matrix production. Moreover, activated HSC could secrete TGF-¦Â1 itself, which help to perpetuate HSC activation. Therefore, cultured HSC is a good and reliable cell model for hepatic fibrosis research and interference with HSC activation and cytokines effect on HSC was usually regarded as the main strategy for curing liver fibrosis.

SA-B is one of main phenol acid compounds from Radix Salviae Miltiorrhizae^[12]. It had good effect against liver injury and fibrosis^[2,3]. HSC activation had two obvious features, proliferation and ECM production. The deposition and accumulation of ECM, particularly type I collagen, is the main feature of liver fibrosis. In this study, we found that SA-B inhibited the proliferation of HSC in a concentration-dependent manner, decreased excessive type I collagen secretion and ECM deposition, and down-regulated pro-collagen gene expression. It suggests that SA-B can inhibit or reverse the HSC activation and this effect is one of the important mechanisms of SA-B action against hepatic fibrosis.

TGF-¦Â1 is known to be the most potent fibrogenic cytokine^[13]. HSC was the main producer and effector of TGF-¦Â1 during liver fibrogensis^[14]. TGF-¦Â1 was synthesized in HSC as precursor that includes LAP, proteolytic cleavage of the precursor occurs intracellu-larly, but active peptide (M_r 25 000) and noncovalently associated LAP are secreted as a complex that lacks biological activity. With plasmin or acid condition, etc, the active TGF-¦Â1 is secreted^[15]. The active TGF-¦Â1 could bind its cell membrane receptors, then downstream the signal through the signal molecules such as MAPK from plasma to the nucleus, and regulate the target gene transcription, through combinatoril binding of transcription factors (such as AP1 and SP1) to the gene regulatory elements^[16]. Pro-collagen gene is one of main target genes of TGF-¦Â1. In this study, SA-B 1 and 10 ¦Ìmol/L suppressed active TGF-¦Â1 secretion and MAPK phosphorylation in HSC. It indicated that SA-B can inhibit TGF-¦Â1/MAPK signal transduction, and that is important reason for SA-B down-regulation of collagen gene and protein expression.

In summary, SA-B could effectively inhibit HSC proliferation and collagen production and interfere with TGF-¦Â1/MAPK signal transduction. These actions underlie the mechanism of SA-B effect against hepatic fibrosis.

REFERENCES

• 1 Pinzani M, Marra F, Carloni V. Signal transduation in hepatic stellate cells. Liver 1998; 18: 2-13.
• 2 Xu YX, Yang HQ, Zhu L, Cao L, Liu ZL, Zhu LR, et al. Clinic observation on treating hepatocihosis with heavy dose of Radix Salviae Miltiorrhizae. Shanghai J Tradit Chin Med 1994; (2): 8-10.
• 3 Liu P, Mizoguzhi Y, Morisawa S. Effect of magnesium lithospermate B on liver injuty induced by D-galactosamine. Chin J Integr Med 1993; 13 : 352-3.
• 4 Hu YY, Liu P, Liu C, Xu LM, Gu HT, Ji G. Study of Magnesium Lithospermate B action against liver fibrosis. Chin Med Abstr Intern Med 1997; 14 (suppl): 139-42.
• 5 Xu LM, Liu C, Liu C, Liu CH, Gu HT, Li FH, et al. A stable and producible method of isolating fat storing cells. J Cell Biol 1995; 17: 143-5.
• 6 Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods 1983; 65: 55-63.
• 7 Rennard SI, Berg R, Martin GR, Foidart JM, Robey PG. Enzyme-linked immunoassay (ELISA) for connective tissue components. Anal Biochem 1980; 104: 205-14.
• 8 Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162:156-9.
• 9 Iimuro Y, Gallucci RM, Luster MI, Kono H, Thurman RG. Antibodies to tumor necrosis factor alfa attenuate hepatic necrosis and inflammation caused by chronic exposure to ethanol in the rat. Hepatology 1997; 26: 1530-7.
• 10 Li SW, Khillan J, Prockop DJ. The complete cDNA coding sequence for the mouse pro alfa 1(I) chain of type I procollegan. Matrix Biol 1995; 14: 593-5.
• 11 Olaso E, Friedman SL. Molecular regulation of hepatic fibrogenesis. J Hepatol 1998; 29: 836-47.
• 12 Chen ZX, Gu WH, Huang HZ, Yang XM, Sui CJ, Chen WZ, et al. Study on Hydrophilic Components in Radix Salviae Miltiorrhizae. Chin Pharm Bull 1981; 16 : 24-5.
• 13 Bachem MG, Meyer D, Melchior R, Sell KM, Gressner AM. Activation of rat liver perisinusoidal lipocytes by transforming growth factors derived from myofibroblastlike cell: a potential mechanism of self-perpetuation in liver fibrogenesis. J Clin Invest 1992; 89: 19-27.
• 14 Yao J, Li LS. Glomerularmesangial cells produced growth inhibitory factor-transforming growth factor-b in rats. Chin J Pathophysiol 1994; 10: 644-8.
• 15 Roth E, Kuhl K, Gressner AM. Subcellular localization of (latent) transforming growth factor b and the latent TGFb binding protein in rat hepatocytes and hepatic stellate cells. Hepatology 1998; 28: 1588-96.
• 16 Reimann T, Hempel U, Krautwald S, Axmann A, Scheibe R, Seidel D, et al. Transforming growth factor-¦Â1 induces activation of Ras, Raf-1, MEK and MAPK in rat hepatic stellate cells. FEBS Lett 1997; 403: 57-60.