Effects of salvianolic acid-A on NIH/3T3 fibroblast proliferation, collagen synthesis and gene expression

Abstract

AIM: To investigate the mechanisms of salvianolic acid A (SA-A) against liver fibrosis in vitro.

METHODS: NIH/3T3 fibroblasts were cultured routinely, and incubated with 10^-4 mol/L-10^-7 mol/L SA-A for 22 h. The cell viability was assayed by [³H]proline incorporation, cell proliferation by [³H]TdR incorporation, cell collagen synthetic rate was measured with [³H]proline impulse and collagenase digestion method. The total RNA was prepared from the control cells and the drug treated cells respectively, and α (1) I pro-collagen mRNA expression was semi-quantitatively analyzed with RT-PCR.

RESULTS: 10^-4 mol/L SA-A decreased cell viability and exerted some cytotoxiciy, while 10^-5 mol/L-10^-7 mol/L SA-A did not affect cell viability, but inhibited cell proliferation significantly, and 10^-6 mol/L SA-A had the best effect on cell viability among these concentrations of drugs. 10^-5 mol/L-10^-6 mol/L SA-A inhibited intracellular collagen synthetic rate, but no significant influence on extracellular collagen secretion. Both 10^-5 mol/L and 10^-6 mol/L SA-A could decrease α (1) I pro-collagen mRNA expression remarkably.

CONCLUSION: SA-A had potent action against liver fibrosis. It inhibited NIH/3T3 fibroblast proliferation, intracellular collagen synthetic rate and type I pro-collagen gene expression, which may be one of the main mechanisms of the drug.

Key Words: salvianolic acid-A; NIH/3T3 fibroblast; cell viability; cell proliferation; collagen; gene expression

INTRODUCTION

Radix salviae miltiorrhizae, one of the most frequently used Chinese herbs, is regarded to have effects on both blood production and circulation by traditional Chinese medicine, and is widely applied in clinical therapy for liver diseases, such as chronic hepatitis, hepatic cirrhosis, etc. Salvianolic Acid-A is one of the water soluble components from Radix salviae miltiorrhizae. It was reported to have good actions on peroxidation[1]. Lipid peroxidation could stimulate hepatic stellate cell (HSC) transformed into myofibro blast like cell (MFBC) and collagen gene expression in vivo and in vitro, and played an important role in liver fibrogenesis[2]. In our previous work[3], it was found that SA-A could protect hepatic lipid peroxidation, and had marked effects against liver injur y and fibrosis in carbon tetrachloride induced fibrotic rats. In order to investigate the mechanism by which SA-A protects against liver fibrosis, we observed the effects of SA-A on NIH/3T3 fibroblast proliferation, collagen protein production and procollagen gene expression.

MATERIALS AND METHODS

Drug

SA-A, molecular formular as C₂₆H₂₂O₁₀, molecular structure as shown in Figure 1, molecular weight 494, was extracted and identified by Shanghai Institute of Materia Medica, Chinese Academy of Sciences.

Figure 1 SA-A molecular structure.

Main reagents and solutions

PRMI-1640 Medium and Dubocal modified Eagle Medium (DMEM) were purchased from Gibco BRL Co., new brown serum (NBS) from Shanghai Sino-American Co., purified type III collagenase (specific activity, 960 U/mg), N-ethylmaleimide (NEM) and β-aminopropionitrile from Sigma Co. [5³H]proline ([³H]Pro) from Amersham Co. methyl-[³H] thymidine (TdR) from Shanghai Institute of Atomic Energy, guanidium thiocynate from Serva Co. Access RT-PCR Sy stem Kit, PCR marker from Promega Co., Diethypyrocarbonate, saturated phenol/chloroform mix and agarose from Shanghai Sangon Biotech Co. Other reagents all were of analytical grade.

The non-homogeneous scintillation liquid was dimethylbenzene solution containin g 5 g/L 2,5-diphenyloxazol (PPO) and 0.5 g/L 1,4-bis[5-phen yloxazol-2] benzene (POPOP), the homogeneous scintillation liquid was dimethy lbenzene solution containing 7 g/L PPO, 0.5 g/L POPOP, 100 g/L naphthalene and 400 mL/L₂-ethoxy-ethanol.

Cell line

Mouse NIH/3T3 fibroblasts were purchased from Shanghai Institute of Cell Biology, Chinese Academy of Sciences, and cultured with PRMI-1640 medium containing 100 g/L NBS, 100 KU/L penicillin and 100 mg/L streptomycin. After the cell growth became confluent, they were digested with trypsin-EDTA and subcultured.

PCR Primers

The PCR primers for pro-collagen α2(I) and β-actin were designed according to the published sequences and references in Table 1[4], and were synthesized by Gibco BRL Co.

Table 1 PCR primer sequences and expected size of amplified products.

Primers	Sequence	Size
α 2(I) collagen upstream	5'TGTTCGTGGTTCTCAGGGTAG3'
α 2(I) collagen downstream	5'TTGTCGTAGCAGGGTTCTTTC3'	254 bp
β-actin upstream	5'ACATCTGCTGGAAGGTGGAC3'
β-actin downstream	5'GGTACCACCATGTACCCAGG3'	163 bp

Cell proliferation assay

Confluent NIH/3T3 fibroblasts in 24 well plates were incubated with 10^-4 mol/L-10^-7 mol/L SA-A diluted in PRMI-1640 medium containing 100 mL/L NBS for 22 h, and [³H]TdR (55.5 KBq/well) was impulsed in the last 16 h. Then cells were harvested with trypsin digestion an d collected on the filtration membrane, then sample radioactivity (cpm) in the non-homogeneous scintillation liquid was measured by Backman Wallac 1410 Scintillator. All tests were repeated 3 times.

Cell viability assay

According to Mallat's method[5], confluent NIH/3T3 fibroblast s in 24-well plates were incubated with 10^-4 mol/L-10^-7 mol/L SA-A re solved in PRMI-1640 medium without NBS for 22 h, and [³H]Pro (55.5 KB q/well) was impulsed in the last 16 h. Then cells were collected and the cpm was measured as above.

Assay of cell collagen synthetic rate

According to Greets' method[6], confluent NIH/3T3 fibroblasts in 6 well plates were incubated with 10^-5 mol/L-10^-6 mol/L SA-A diluted in PRMI-1640 without NBS for 22 h, during the later 16 h the culture media were changed to DMEM containing 185 KBq/mL [³H]Pro, 100 mg/L-β-aminopropionitrile, 50 mg/L ascorbic acid as well as the same drugs. Then the culture media and cell layer extract were collected respectively, dialyzed thoroughly and reacted with collagenase, etc. The total radioactivity in the samples (cpm_t), radioactivity in the samples treated with colla genase (cpm_c) and not treated with collagenase (cpm_b) were counted in the ho mogeneous scintillation liquid by Backman Wallac 1410 Scintillator. The new collagen that cell produced, i.e. the fraction of collagenous protein express edas percentage of total radiolabeled protein, was calculated using the formula:

% of collagen = 100 ÷ [5.4 × (cpm_t - cpm_c)/(cpm_c - cpm_b) + 1]

RNA extraction and RT-PCR (reverse transcription and polymerase chain reaction)

The total RNA was extracted from the control cells and SA-A incubated cells by the acid guanidium thiocynate-phenol-chloroform method[7]. The RNA quantity was determined by absorption at 260 nm, its purity was confirmed with A₂₆₀/A₂₈₀ specrophoto meter readings that ranged from 1.6 to 1.9, and its integrity was checked by 9 g/L agarose gel electrophoresis with ethidium bromide (EB) staining of 18S and 28S ribosomal RNA (Figure 2). With Access RT-PCR system kit, the cDNA synthesis and amplification was done in one tube following the manufacturer's instructions. In brief, 1 μg RNA, 50 pmol/L primers for α (1) I pro-collagen or β-actin were added to each reaction mixture respectively, which included 10 mmol/L dNTPs 1 μL, 25 mmol/L MgSO₄ 2 μL, AMV reverse transcriptase 5 U, Tfl DNA polymerase 5 U, AMV/Tfl-5 × buffer 10 μL. The reaction final volume was 50 μL and was covered with 20 μL mineral oil. Then with PCR Touchdown thermal cycler (Hybaid, England), RT-PCR reaction was run in the following procedures: ① 48 °C for 45 min, 1 circle. ② 94 °C for 2 min, 1 circle. ③ 94 °C for 30 s, 60 °C for 1 min, 38 °C for 2 min, 30 circles. ④ 68 °C for 7 min, 1 circle. Five μL PCR product was run on 15 g/L agarose gel and observed by EB staining under UV light, the electrophores is photo was transformed into computer, and α 1(I) pro-collagen intensity was analyzed with MPIAS500 image system, while the β-actin band intensity was subtracted as an internal standard.

Figure 2 Total RNA gel electrophoresis photograph. 28S and 18S of total RNA run on 9 g/L agarose gel stained with EB.

Statistical analysis

Data were analyzed by Student's t test.

RESULTS

Effects on cell morphology and viability

10^-5 mol/L-10^-7 mol/L SA-A had no marked effects on cell morphology, but 10^-4 mol/L SA-A led to shrinkage and detachment of some cells, showing cytotoxicity to some degree. 10^-4 mol/L-10^-7 mol/L SA-A did not decrease intercellular [³H]Pro incorporation, while 10^-6 mol/L SA-A could increase [³H]Pro impulse(P < 0.05) and enhance cell viability (Table 2).

Table 2 Effects of SA-A on cell intracellular [³H]TdR and [³H]Pro incorporation (cpm/well, ^-x±s, n = 4).

Group	[3H]TdR	[3H]Pro
Control	1482 ± 486	21018 ± 5473
10-4 mol/L SA-A	675 ± 201b	18659 ± 2363
10-5 mol/L SA-A	969 ± 183a	23761 ± 5430
10-6 mol/L SA-A	868 ± 183a	31408 ± 4981a
10-7 mol/L SA-A	1056 ± 187	26080 ± 4504

^aP < 0.05,

^bP < 0.01, vs control.

Effects on cell proliferation

10^-4 mol/L-10^-6 mol/L SA-A remarkably decreased intercellular [³H]TdR incorporation and inhibited cell proliferation(P < 0.05), 10^-4 mol/L SA-A showed more significant effect (P < 0.01), but it induced some cell death, which may be associated with its cytotoxic action. 10^-7 mol/L SA-A had no obvious effect on cell [³H]TdR incorporation (Table 2).

Effects on cell collagen synthetic rates

10^-5 mol/L-10^-6 mol/L SA-A could inhibit intracellular collagen synthetic rate significantly (P < 0.01), but did not influence extracellular synthetic rate (Table 3).

Table 3 Effects of SA-A on NIH/3T3 fibroblast collagen synthetic rates (%,^-x±s, n = 4).

Group	Intracellular	Extracellular
Control	0.78 ± 0.03	2.57 ± 0.37
10-5 mol/L	0.48 ± 0.24b	2.54 ± 0.91
10-6 mol/L	0.43 ± 0.26b	3.02 ± 0.69

^bP < 0.01, vs control.

Effects on procollagen α2(I) mRNA expression

Both 10^-5 mol/L and 10^-6 mol/L SA-A decreased procollagen α1(I) mRNA expression significantly (P < 0.05), but the re was no difference between the two different concentration groups (Table 4, Figure 3).

Figure 3 RT-PCR product gel electrophoresis photograph. Five μL RT-PCR products of procollagen α2(I) and β-actin run on 1.5% agarose gel stained with EB. Lane 1 as PCR marker, lane 2 and 3 as the control for procollagen α2(I) and β-actin respectively, lane 4 and 5 SA-A 10^-6 mol/L for procollagen α2(I) and β-actin respectively, lane 6 and 7 as SA-A 10^-5 mol/L for procollagen α2(I) and β-actin respectively.

Table 4 The relative expression amount of α2(I) procollagen mRNA (^-x±s,% of β-actin).

Group	n	Col α1(I) mRNA
Control	3	98.71 ± 9.96
10-5 mol/L SA-A	3	76.23 ± 12.02a
10-6 mol/L SA-A	3	68.44 ± 8.06a

^aP < 0.05, vs control.

DISCUSSION

Hepatic fibrosis, a precursor of cirrhosis, is a common and important pathologic al feature of chronic liver diseases, which involves the abnormal accumulation of extracellular matrix (ECM) proteins, particularly collagen[8]. In fibrotic liver, ECM components are mainly produced by HSC and fibroblasts. It is known that during fibrogenesis, HSC undergoes a process of activation, developing a myofibroblast-like phenotype associated with increased proliferation and ECM production, especially type I collagen synthesis. The mouse NIH/3T3 fibroblast also shared the features that active HSC (MFBC) presented, such as remarkable proliferation and substantial production of collagen, and stable cell line. In practice, NIH/3T3 fibroblast is often used as a desirable cell model for investigation of antifibrotic drugs.

In order to rule out the possibility of SA-A cytotoxic influence in vitro, the intracellular [³H] Pro incorporation was measured, and inverted microscopic observation was done. It was found that only 10^-4 mol/L SA-A cause d some cell detachment, decreased [³H] Pro incorporation, and showed cytotoxicity to some extents. 10^-5 mol/L-10^-7 mol/L SA-A did not influence cell morphology or inhibit cell viability. However, 10^-6 mol/L SA-A enhanced cell viability. Both 10^-5 mol/L-10^-6 mol/L SA-A could inhibit intracellular [³H]TdR impulse that NBS stimulated. It is suggested that SA-A had an effective action against NIH/3T3 fibroblast proliferation.

Type I collagen is the predominant component of ECM during liver fibrosis. Its production involve two processes: the first is intracellular synthesis, including gene transcription, translation and modification to form procollagen, then procollagen alpha chains are secreted to the outside of the cell to form helix collagen by sorting and alignment etc. In the study, it was found that SA-A d ownregulated procollagen α2(I) steady-state mRNA expression, and intracellular collagen synthetic rate, but exerted no effect on extracellula r synthetic rate. It is suggested that SA-A influence on collagen production through the intracellular synthetic process. The fibrogenic cells have two predominant features: one is active in cell proliferation, which led to increase in c ell number, another is strong fibrogenic ability per cell, which led to accumulation of ECM. In the study, SA-A not only inhibited NIH/3T3 fibroblast proliferation, but also decreased collagen synthesis, showing a good action against liver fibrosis.

Salvianolic radix is widely used as an important component in Chinese herbal formulas for the treatment of chronic liver diseases. Salvianolic Acid-A, one of water-soluble ingredients from Salvianolic radix, had effective actions on hepatic peroxidation and fibrosis in vivo[3]. In the paper, it is for the first time found that SA-A has the potential action against hepatic fibrosis in vitro, and its main mechanisms of antifibrotic action perhaps was associated with the inhibition of fibrogenic cell proliferation, collagen gene expression and protein synthesis.