Effect of IL-10 on the expression of HSC growth factors in hepatic fibrosis rat

Abstract

AIM: To study the effect of IL-10 on the expression of growth factors - transforming growth factor-β1 (TGF-β1), epidermal growth factor (EGF), hepatocyte growth factor (HGF)and platelet-derived growth factor (PDGF) of hepatic stellate cells (HSCs) of hepatic fibrosis rat and the anti-fibrogenic role of exogenous IL-10.

METHODS: Hepatic fibrosis was induced by CCl4 administration intra-peritoneally. Sixty clean male Sprague-Dawley (SD) rats were randomly divided into three groups: normal control group (GN, 8 rats), hepatic fibrosis model group (GC, 28 rats) and IL-10 treated group (GI, 24 rats). At the beginning of the 7^th and 11^th wk, rats in each group were routinely perfused with pronase E and type IV collag-enase through a portal vein catheter and the suspension obtained from the liver was spun by centrifugation with 11% Nycodenz density gradient to isolate HSCs. Histological examination was used to determine the degree of hepatic fibrosis. RT-PCR was employed to analyze mRNA expression from freshly isolated cells. Immunocytochemistry was performed to detect protein expression in primary cultured HSCs.

RESULTS: Rat hepatic fibrosis was developed with the increase of injection frequency of CCl4, and HSCs were successfully isolated. At the 7^th and 11^th wk, TGF-β1, EGF, and HGF mRNA in GC increased obviously compared with GN (P = 0.001/0.042, 0.001/0.001, 0.001/0.001) and GI (P = 0.001/0.007, 0.002/0.001, 0.001/0.001). For TGF-β1, no difference was observed between GI and GN. For EGF, mRNA level in GI increased compared with GN during the 7^th wk (P = 0.005) and 11^th wk (P = 0.049). For HGF, mRNA level in GI decreased compared with GN at the 7^th wk (P = 0.001) and 11^th wk (P = 0.021). Between these two time points, TGF-β1 expression at the 7^th wk was higher than that of the 11^th wk (P = 0.049), but for EGF, the former was lower than the latter (P = 0.022). As for PDGF mRNA, there was no significant difference between these groups, but difference seemed to exist in protein levels. Results by immunocytochemistry of TGF-β1 and EGF were paralleled with the above findings.

CONCLUSION: The expression of TGF-β1, EGF and HGF increased in HSC of hepatic fibrosis rat and decreased after treatment with IL-10. IL-10 plays an anti-fibrogenic role by suppressing growth factors expression.

Key Words: Hepatic fibrosis; Hepatic stellate cells; Interleukin-10; Transforming growth factor-β1; Epidermal growth factor; Hepatocyte growth factor; Platelet-derived growth factor

INTRODUCTION

Hepatic fibrosis, which represents the wound healing response of the liver is a common sequel to diverse liver injuries, characterized by increased deposition and altered composition of extracellular matrix (ECM). Hepatic stellate cells (HSCs) are the major source of ECM, and believed to be the crucial cell type in the development of hepatic fibrosis[1,2] and growth factors were considered to exert their effects through autocrine or paracrine on HSCs during the process[3]. It has been reported that interleukin-10 (IL-10) could relieve the degree of rat hepatic fibrosis induced by carbon tetrachloride (CCl4). In the present study, based on the established hepatic fibrosis rat model and IL-10 treated model, we isolated HSCs, detected the expression of TGFβ1, EGF, HGF and PDGF of HSC, and tried to explore the relationship between growth factors and hepatic fibrosis and the possible mechanisms of the anti-fibrogenic activities of exogenous IL-10 in vivo.

MATERIALS AND METHODS

Establishment of animal models

Sixty clean male SD rats, weighing 200-300 g, were divided randomly into three groups: normal control group (GN, 8 rats), hepatic fibrosis model group (GC, 28 rats) and IL-10 treated group (GI, 24 rats). All rats were bred under routine conditions (room temperature, 22±2 °C; humidity,55±5%; light, 12 h per day; drinking tap water and food ad libitum. Animal food was provided by BK Company in Shanghai, China. Rats in GN were injected intra-peritoneally with saline at a dose of 2 mL/kg twice a wk; rats in the other two groups received intra-peritoneal injection of 500 ml/L CCl4 dissolved in castor oil 2 mL/kg twice a week as described previously[4]. From the 3^rd wk, rats in the treated group were given intra-peritoneally IL-10 dissolved in saline 4 ug/kg, 20 min before CCl₄ administration, as proposed by Nelson et al[5]. The intervention stated above lasted to the end of the experiment.

Histological examination

At the beginning of the 7^th and 11^th wk, two rats of each group were selected randomly for histological examination. The liver tissues were fixed in 40 g/L formaldehyde and embedded with paraffin. Sections were stained with hematoxylin and eosin (HE) and examined under a light microscope.

Isolation, culture and identification of HSCs

Isolation, culture and identification of HSCs have been described thoroughly in our previous experiments[6]. Briefly, at the beginning of the 7^th and 11^th wk, five rats of each group were selected randomly to perfuse successively with 0.13% pronase E and 0.025% type-IV collagenase through a portal vein catheter. The liver tissue suspension was incubated with 0.02% pronase E and 0.025% type-IV collagenase with agitation. Then the suspension obtained from the digested liver was spun by centrifugation with 11% Nycodenz density gradient to purify HSCs. Thereafter, HSCs were seeded at approximately 1×10⁶ cells/mL of Dulbecco’s modified eagle medium (DMEM) with 20% fetal calf serum in 96-well plates. Then HSCs were kept in culture at 37 °C in a 50 mL/L CO2 atmosphere for 72 h. They were identified by their typical phase-contrast microscopic appearance and by immunocytochemistry using antibody directed against desmin. Cell vitality was checked by trypan blue exclusion.

RNA extraction and RT-PCR

Total RNA was extracted from freshly isolated HSCs, according to the RNA isolation kit instructions (Jingmei Biotechnology Company of Shenzhen). Its quantity and purity were assessed by measuring the optical density at 260 and 280 nm. After measurement of RNA amount, samples were either used immediately for reverse transcription (RT) or stored at -70 °C.

For RT, 1 μg total RNA was reversely transcribed following the instructions of first strand cDNA synthesis kit (Jingmei Biotechnology Company of Shenzhen). Reaction mixtures of 20 μL were transcribed using the following program: at 42 °C for 60 min, 99 °C for 5 min, and stored at -20 °C.

For PCR, primers coding for the house-keeping gene β-actin was added to the reaction mixture spontaneously to standardize the results. Reactive systems contained cDNA 2 μL, 10×buffer 5 μL, 25 mmol/L MgCl2 5 μL, 10 mmol/L d NTP 1 μL, 20 mmol/L target gene sense and anti-sense primers 1 μL, 20 mmol/L β-actin primer pairs 1 μL, Taq DNA polymerase 3 U, with a total volume of 50 μL by adding aqua. Then PCR was performed with cycle parameters of 45 s at 94 °C, 30 s at 55 °C (TGF-β1) or 58 °C (EGF) or 66 °C (HGF) or 62 °C (PDGF) and 60 s at 72 °C after pre-denaturation for 5 min at 94 °C; final elongation time was 7 min at 72 °C. The number of corresponding cycles for TGF-β1, EGF, HGF and PDGF were 30, 32, 34 and 34, respectively. Primers were designed with reference to GenBank (Table 1). PCR products were immediately analyzed by 2% agarose gel electrophoresis and the density of resultant bands was semi-quantified by scanning densitometry with the ratio of TGF-β1/β-actin2, EGF/β-actin1, HGF/β-actin1 and PDGF/β-actin2.

Table 1 Primer sequences used in this study.

mRNA	Sense primers	Antisense primers	Product length (bp)
TGF-β1	CTC TGC AGG CGC AGC TCT G	GGA CTC TCC ACC TGC AAG AC	392
EGF	GAC AAC TCC CCT AAG GCT TA	CATGCA CAC GCC ACC ATT	567
GAG		GCA GTA CCC ATC GTA CGA
HGF	TCT TGA CCC TGA CAC CCC	GTG ATT CAG CCC CAT CCG G	269
PDGF	GAT CCG CTC CTT TGA TGA TC	GTC TCA CAC TTG CAT GCC AG	435
β-actin1	AGC ACT TGC GGT CCA CGA	GAG CTA TGA GCT GCC TGA	410
β-actin2	CCA ACC GTG AAA AGA TGA CC	CAG GAG GAG CAA TGA TCT TG	660

Immunocytochemistry

Most of HSCs were attached to the dishes after 72 h of primary culture. Then the 96-well plates were washed twice with 0.1 mol/L PBS and fixed with poly-formaldehyde at 4 °C overnight. The following procedures were performed according to the instructions of streptavidin-peroxidase (S-P) kit (Beijing Zhongshan Company). Briefly, cells were washed with PBS, incubated with bovine serum albumin in PBS, reacted with primary antibody dissolved in PBS, washed, incubated with peroxidase-conjugated second antibody, washed, and reacted for 20 min with S-P. Color reaction was developed by incubation with DAB. For negative controls, the primary antibody was replaced by PBS. Table 2 summarizes the characteristics of the antibodies used in this study.

Table 2 Antibodies used for immunocytochemistry.

Antibody	Type	Source	Working solution
Desmin	Polyclonal, mouse	Beijing Zhongshan Co.	0.111111111
TGF-β1	Polyclonal, rabbit	BOSTER Co.	1:20
EGF	Polyclonal, rabbit	BOSTER Co.	1:50
PDGF	Polyclonal, rabbit	Beijing Zhongshan Co.	1:20
Antibody to rabbit IgG	Goat	Beijing Zhongshan Co.	Ready to use
Antibody to mouse IgG	Goat	Beijing Zhongshan Co.	Ready to use

Statistical analysis

All data were expressed as mean±SE. The significance for the difference between the groups was analyzed with SPSS10. 0 by one-way ANOVA.

RESULTS

Histological examination

Hepatic fibrosis, as shown histologically, became remarkable with the treatment of CCl4. During the 7th wk, specimens from GC showed steatosis and ballooning degeneration, the collagen fibers increased and began to extend to the parenchyma, a great number of mononuclear cells and unusual neutrophils surrounded the centrilobular veins and fibrotic septa (Figure 1A), while only a few inflammatory cells infiltrated around centrilobular veins without evident changes of lobular structure in GI (Figure 1B). At the 11th wk, the reticular fibers extended into hepatic plate and full delimitation was developed in GC (Figure 1C), while less fibrous and inflammatory infiltrates were seen in GI (Figure 1D). Specimens from GN showed normal lobular structure (Figure 1E). Due to the limit of samples, no statistical data could be obtained to present disparity between the two groups, but it still illustrated the trend that fibrogenesis of GI was much less severe than that of GC.

Figure 1 Effect of IL-10 on liver histology of hepatic fibrosis rats. A: Rat liver in group C at the 7^th wk (HE×100); B: Rat liver in group I at the 7^th wk (HE×100); C: Rat liver in group C at the 11^th wk (HE×100); D: Rat liver in group I at the 11^th wk (HE×100); E: Rat liver in group N (HE×100).

Isolation and identification of HSCs

Totally 2-4.5107 cells per rat were harvested by the present method. HSCs were identified by immunoreaction for desmin (Figure 2), the mean purity of freshly isolated HSCs was (955)%. Cell vitality checked by trypan blue exclusion was higher than 95%.

Figure 2 Desmin expression of HSCs from primary culture for 5 d by immunochemistry (SP×100).

Expression of mRNA

RNA purity was determined by the ratio of A260-A280, which ranged from 1.8 to 2.0. At the 7^th and 11^th wk, TGF-β1, EGF, and HGF mRNA in GC increased obviously compared with GN and GI (P<0.01). For TGF-β1, no difference was seen between GI and GN. For EGF, mRNA level in GI increased compared with GN at the 7^th wk (P<0.01) and 11^th wk (P<0.05). For HGF, mRNA level in GI decreased compared with GN at the 7^th wk (P<0.01) and 11^th wk (P<0.05). Between these two time points, TGF-β1 expression at the 7^th wk was higher than that of the 11^th wk; but for EGF, the former was lower than the latter. As for PDGF mRNA, there was no significant difference between these groups (Table 3, Table 4, Table 5, Figure 3 A-C).

Table 3 Expression of TGF-β1 mRNA in HSCs (mean±SE).

Wk	n	Group N	Group C	Group I
7th wk	5	0.143±0.009	0.267±0.025b	0.140±0.008d
11th wk	5	0.141±0.004	0.207±0.029ab	0.123±0.009d

^aP<0.05 vs 7^th wk;

^bP<0.01 vs group N;

^dP<0.01 vs group C.

Figure 3 RT-PCR analysis of growth factor gene expression of HSC before and after intervention. A: Electrophoretogram of PCR for TGF-β1 gene of HSCs. 1-3: groups N, C, I of 11^th wk; 4-6: groups N, C, I of 7^th wk; M: DNA ladder of 100 bp; B: Electrophoretogram of PCR for EGF gene of HSCs. 1-3: groups N, C, I of 11^th wk; 4-6: groups N, C, I of 7^th wk; M: DNA ladder of 100 bp; C: Electrophoretogram of PCR for HGF gene of HSCs. 1-3 groups N, C, I of 11^th wk; 4-6: groups N, C, I of 7^th wk; M: DNA ladder of 100 bp.

Table 4 Expression of EGF mRNA in HSCs (mean±SE).

Wk	n	Group N	Group C	Group I
7th wk	5	0.181±0.012	0.425±0.013b	0.293±0.024b,d
11th wk	5	0.179±0.004	0.514±0.008a,b	0.258±0.021c,d

^aP<0.05 vs 7^th wk;

^bP<0.01 vs group N;

^cP<0.05 vs group N;

^dP<0.01 vs group C.

Table 5 Expression of HGF mRNA in HSCs (mean±SE).

Wk	n	Group N	Group C	Group I
7th	5	0.364±0.015	0.636±0.009b	0.210±0.010bd
11th	5	0.383±0.004	0.622±0.022b	0.265±0.028ad

^aP<0.05 vs group N;

^bP<0.01 vs group N;

^dP<0.01 vs group C.

Protein expression and immunophenotyping of cells

TGF-β1, EGF and PDGF positive expressions were localized in cytoplasm and nuclei of HSCs in all groups by immunocytochemistry after primary culture for 72 h. By then, most of the cells were attached and spread over the plastic sub-stratum. At the 7^th wk, the size of HSCs in GC and GI was slightly larger than GN, as well as the number and length of pseudopodium. At the 11^th wk, cell phenotype in GI was in the form of a circle or an ellipse and slightly smaller than the 7^th wk, no obvious change of phenotype was seen in GC compared to the 7^th wk. Despite the small number of samples and lack of statistical data, we could still find the tendency that the expression of TGF-β1, EGF and PDGF in GC was higher than that of GN and decreased after treatment with IL-10 (Figure 4A-F).

Figure 4 Expression of growth factor protein of HSCs in different groups. A: Expression of PDGF protein of HSCs in group C at the 7^th wk (SP×100); B: Expression of PDGF protein of HSCs in group I at the 7^th wk (SP×100); C: Expression of TGF-β1 protein of HSCs in group C at the 7^th wk (SP×100); D: Expression of TGF-β1 protein of HSCs in group I at the 7^th wk (SP×100); E: Expression of EGF protein of HSCs in group C at the 11^th wk (SP×100); F: Expression of EGF protein of HSCs in group I at the 11^th wk (SP×100).

DISCUSSION

Although significant progress has been made in understanding the pathogenesis of hepatic fibrosis, a rational therapy that prevents the progression or even reverses established fibrosis remains elusive[7-9]. IL-10 is produced mainly by TH2 cells and inhibits functions of TH1 cells. It downregulates pro-cytokine synthesis and is associated with amelioration of the inflammatory response[10-12] and fibrosis. The present study also found the trend of IL-10 relieving the degree of inflammation. Then, except for anti-inflammation, is there any other mechanism to fulfill its anti-fibrotic role? In this experiment, we detected the mRNA level of TGF-β1, EGF, HGF and PDGF of HSCs of rat in hepatic fibrosis group and IL-10 treated group at the 7^th and 11^th wk.

TGF-β1 is the most important growth factor involved in the fibrotic and cirrhotic liver. It is known to promote the development of liver fibrosis by inducing synthesis of ECM proteins and downregulating the expression of matrix degrading enzymes and stimulating the synthesis of their respective inhibitors[13-16]. The present study showed that with the development of hepatic fibrosis, TGF-β1 mRNA increased in GC compared to GN, and decreased after treatment with IL-10. Specifically, TGF-β1 mRNA level did not parallel with the progression of hepatic fibrosis. These illustrated further that TGF-β1 correlated closely with hepatic fibrosis, and it could not be regarded as a predictable factor of the degree of hepatic fibrosis and HSCs capacity of producing TGF-β1 declined when fibrosis developed to a certain extent and that IL-10 exerted its anti-fibrotic activities by attenuating the expression of TGF-β1 of HSCs.

EGF is a potent mitogen for many cells. EGF up-regulation is a characteristic feature of fibrotic liver disease[17]. Apart from stimulating hepatocyte and bile duct epithelial proliferation, EGF also has a chemotactic mitogen effect on HSCs[18]. A recent study demonstrated that direct stimulation with TGF-β1 and EGF results in an increased migratory response of activated HSCs[19]. In the present study, EGF level increased with the presence of hepatic fibrosis or cirrhosis and dropped after intervention with IL-10. It showed that IL-10 inhibited HSCs from proliferating and migrating by reducing autocrine secretion of EGF, thus it limited the spread of disease, and retarded the progression of hepatic fibrosis.

HGF is secreted as a precursor polypeptide that requires proteolytic cleavage to a disulfide-linked heterodimer and binds to a specific receptor known as c-Met[20] for biological activity. Physiologically, HGF has potent hepatotrophic and cytoprotective functions[21]. Reports have shown treatment with HGF suppressed the increase of TGF-beta1, reduced mRNA levels of procollagen, inhibited fibrogenesis and cell apoptosis, and produced complete resolution of fibrosis in the cirrhotic liver, thereby improving the survival rate of rats with this severe illness[22-26]. In the present experiment, HGF mRNA level of HSCs increased in hepatic fibrosis rats which might reflect intra-hepatic inflammation or the compensation of steatosis of hepatocytes. After intervention with IL-10, HGF mRNA level dropped. So we postulated that IL-10 probably suppressed some cytokines that positively regulated HGF expression by anti-inflammation or immunoreaction or that it indicated normal liver function, or a manifestation of decreased degree of steatosis and degeneration.

PDGF has been identified as the most potent-polype-ptide growth factor able to stimulate the proliferation of HSCs[27-29], which play a critical role in the development of hepatic fibrosis. Although no statistical data were obtained to prove any difference of PDGF mRNA in these groups, which probably resulted from the difference in experimental animals, drugs, reagents and methods, PDGF protein level seemed to show the disparity that it increased in GC and decreased in GI compared to GN.

Our experiment showed IL-10 exerted anti-fibrotic effect not only by anti-inflammatory reaction but by suppressing the expression of growth factors. Whether IL-10 plays an adverse role in liver regeneration by downregulating the expression of HGFmRNA awaits further studies. Based on the biological effect model of HGF, it is necessary to elucidate its activities at protein and receptor levels, which should provide a promising prospect for the therapy of hepatic fibrosis.