Leptin in hepatic fibrosis: evidence for increased collagen production in stellate cells and lean littermates of ob/ob mice

Abstract

Leptin is a 16-kd hormone that mediates a range of metabolic effects by using a transduction pathway from the long form of the leptin receptor, OB-R(L,) through Janus kinase-signal transducer and activator of transcription (Jak-Stat) signaling components. Leptin is produced by hepatic stellate cells (HSCs) but only following their "activation." Because activation of stellate cells is a central event in the fibrotic response to liver injury, we hypothesized that leptin may directly stimulate fibrogenesis in activated stellate cells via OB-R(L). We analyzed leptin receptors and their signaling partners in a stellate cell line (HSC-T6) as well as in primary stellate cell isolates. We also examined the effect of leptin on stellate cell expression of alpha(2)(I) collagen messenger RNA (mRNA) levels by ribonuclease protection analysis (RPA). Finally, we examined the role of leptin in in vivo fibrogenesis by inducing a wounding response in ob/ob mice, which lack functional leptin. HSC-T6 and culture-activated stellate cells expressed OB-R(L). Scatchard analysis verified specific binding of leptin to HSCs, with an association constant (K(d)) equal to 660 +/- 5.8 pmol/L. Exposure of HSCs to leptin resulted in significant increases in alpha(2)(I) collagen mRNA expression. Transient transfection with a promoter reporter construct showed a 3-fold increase in alpha(2)(I) collagen transgene activity. Leptin stimulated activation of Stat3 in activated HSCs. Finally, lean animals, but not ob/ob littermates, had significant fibrosis as assessed by picrosirius red staining and abundant alpha-smooth muscle actin staining. In conclusion, these results indicate that leptin is profibrogenic in activated HSCs and can signal via the Jak-Stat pathway. Up-regulation of leptin signaling in liver injury could contribute to enhanced fibrogenesis, particularly in states in which leptin levels are high.

Figures

Fig. 1

RPA of the α2(I) collagen mRNA from HSC-T6 cells and primary cultured HSCs. β-actin, 304 nucleotides; α2(I)COL (α2(I) collagen), 533 nucleotides. (A) Primary HSCs were isolated from normal rats and allowed to grow in culture in standard serum-containing medium. After 3 days, serum-containing medium was exchanged with serum-free medium; cells were exposed to 100 ng/mL leptin in SF medium, total cellular RNA was isolated at the indicated time points, and RPA was performed as described in Materials and Methods. Results with primary cultured HSCs include a representative histogram of densitometric analysis of autoradiography shown. (B) A representative autoradiogram and histogram results from RPA from HSC-T6 cells. RPA from primary cultured HSCs and HSC-T6 cells was performed separately in triplicate in 3 independent experiments. Statistical analysis of RPA products was performed by Student’s t test. All treatment data were compared with controls (untreated cells), *P < .05. Error bars represent SEM.

Fig. 2

Transient transfection of the α2(I) collagen promoter in HSC-T6 cells exposed to leptin. Transient transfections with HSC-T6 were performed with 2 μg of the α2(I) collagen promoter (pGL3-1009) and 0.5 μg of pSV-βgal. Leptin treatment (100 ng/mL) is as in Fig. 1 and Materials and Methods. Luciferase activity was determined and results shown as relative light units/microgram protein. The pSV-βgal reporter gene was cotransfected to normalize for transfection efficiency. Untreated pGL3-1009 (□) represents transgene activity of the α2(I) collagen promoter in cells not exposed to leptin for each treatment time indicated. The transfected pGL3 empty vector (B) was used as a negative control; the pGL3-SV40 enhancer vector (P, ) served as a positive control. All assays were performed in triplicate. Data shown represent the average of 3 independent experiments ± SEM. Data analysis was performed by Student’s t test, where *P < .05 was significant and compares collagen transgene expression between untreated cells (□) and leptin-treated cells (■) normalized to total protein assessed by Lowry assay for each time point indicated in hours.

Fig. 3

Specific binding of 125I-leptin to HSCs. (A) Time course showing leptin binding to HSCs. Cells were cultured as monolayers and incubated with 62.5 pmol/L 125I-leptin for time indicated in minutes (◆). To determine specific leptin-HSC binding, an identical assay was performed with excess unlabeled leptin (5 μmol/L) (●). Bound radioactive ligand was measured and standardized to microgram protein assayed. (B) Scatchard analysis. Incubation with labeled leptin was performed for 2 hours with cell monolayers at 4°C. Cell-associated radioactivity was determined by scintillation counting. Data for Kd are expressed as mean ± SEM. The plots represent duplicate experiments performed twice. The details of these studies are described in Materials and Methods.

Fig. 4

Detection of OB-RL in HSC-T6 cells and primary, culture-activated rat HSCs by immunoprecipitation and RT-PCR. (A) Immunoprecipitation of OB-RL. Whole-cell lysate was prepared from primary HSCs in culture for 1 day (D1), 3 days (D3), and 7 days (D7). OB-RL in HSC-T6 (T6) cells was detected by immunoprecipitation. An equal quantity of protein was immunoprecipitated with antibody against the cytoplasmic domain of the leptin receptor. Positive control was from mouse brain (+) and was supplied by Santa Cruz Biotechnology. (B) RT-PCR detecting mRNA for OB-RL. Details of reactions and primer pairs are described in Materials and Methods. Total RNA was isolated from primary HSCs and HSC-T6 in culture at times identical to immunoprecipitation analysis as previously described. The expression of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. (C) Immunoprecipitation of OB-R with antibody directed at the extracellular domain of the receptor common to all OB-R isoforms. Molecular weight is on the left in kilodaltons.

Fig. 5

Immunoprecipitation of pStat3 in stellate cells exposed to leptin. One microgram of polyclonal antibodies to pStat3 was added to (A) primary cultured HSCs or (B) HSC-T6 as described in detail in Materials and Methods. Equal protein loading was confirmed based on expression of total, unphosphorylated Stat3 (STAT3 in figures). Lanes 2 to 6 indicate leptin exposure for the times indicated in hours for either primary HSCs (A) or cell culture model (B). The corresponding histograms compare pStat3 from leptin-treated HSCs with untreated HSCs and are from 3 independent experiments performed in triplicate. Data analysis comparing pStat3 in leptin-treated cells for each time point vs. untreated (Lane 1, control [C]) HSCs was by Student’s t test. *P < .05.

Fig. 6

Histopathologic analysis for type I collagen in the liver after CCl4 administration in ob/ob mice and their lean littermates. Representative photomicrographs of (A) CCl4-treated male lean littermates of ob/ob mice and (B) CCl4-treated ob/ob mice. Simultaneous treatment with vehicle (saline and olive oil) is also shown in C for the lean littermates and D for ob/ob mice. These photomicrographs represent 3 independent studies in which 4 male mice were included in each treatment group (picrosirius red stain, original magnification 100×).

Fig. 7

Expression of α-SMA in the liver after administration of CCl4 in ob/ob mice and their lean littermates. The expression and localization of α-SMA were detected by immunohistochemical staining as described in detail in Materials and Methods. Representative micrographs from 3 individual experiments are shown (original magnification 400×). (A) CCl4-treated lean littermates showing α-SMA–positive cells (white arrows), (B) CCl4-treated ob/ob mice, (C) vehicle-treated lean littermates, and (D) vehicle-treated ob/ob mice. Primary antibody dilution was 1:400; the experimental design and resultant liver sections used for α-SMA staining were as in Fig. 6.