Norepinephrine induces calcium spikes and proinflammatory actions ...

Am J Physiol Gastrointest Liver Physiol 291: G877–G884, 2006. First published June 15, 2006; doi:10.1152/ajpgi.00537.2005.

Norepinephrine induces calcium spikes and proinflammatory actions in human hepatic stellate cells Pau Sancho-Bru,1 Ramo´n Bataller,1 Jordi Colmenero,1 Xavier Gasull,2 Montserrat Moreno,1 Vicente Arroyo,1 David A. Brenner,3 and Pere Gine`s1 1

Liver Unit, Institut Clı´nic de Malalties Digestives i Metabo`liques, Hospital Clı´nic, and 2Laboratory of Neurophysiology, University of Barcelona School of Medicine, Institut d’Investigacions Biome`diques August Pi i Sunyer, Barcelona, Spain; and 3Department of Medicine, Columbia University College of Physicians and Surgeons, New York, New York Submitted 28 November 2005; accepted in final form 6 June 2006

CATECHOLAMINES regulate a number of physiological functions in the liver. The normal human liver contains abundant sympathetic innervation, and circulating catecholamines regulate hepatic vascular tone as well as hepatic metabolic functions (10, 13, 43). In patients with cirrhosis, catecholamines may participate in the pathogenesis of portal hypertension. In these patients, elevated serum catecholamines are believed to contribute to the increased intrahepatic vascular resistance to blood

flow (4, 16). Moreover, ␣- and ␤-adrenoceptor blockers markedly decrease portal pressure. The effect of nonselective ␤-blockers is mainly mediated by a reduction in portal venous blood flow, whereas the beneficial effects of ␣1-receptor blockers is basically due to a reduction in intrahepatic resistance (3, 13). In rats with experimentally induced cirrhosis, ␣1-adrenoceptor agonists increase intrahepatic vascular resistance (17). We hypothesized that ␣1-adrenoceptors induce vasoactive effects in contractile (myofibroblastic) hepatic cell types, which might mediate changes in vascular tone in advanced liver diseases. Besides their role in the regulation of vascular tone, catecholamines seem to be involved in the wound healing response to injury (11, 15). Locally released catecholamines regulate key biological features of the tissue repair process, including myofibroblast proliferation and collagen secretion (1, 2). Therefore, catecholamines are currently viewed as a system that participates in the local response to chronic injury. Recent evidence suggests that catecholamines may participate in the pathogenesis of liver fibrogenesis (29). Experimentally induced liver fibrosis is enhanced in spontaneously hypertensive rats, which have elevated serum levels of catecholamines (21). Moreover, both chemical sympathectomy and treatment with prazosin, an ␣1-adrenoceptor blocker, attenuate experimentally induced liver fibrosis in rats (15, 30). Finally, genetic ablation of the sympathetic nervous system (SNS) reduces liver fibrosis in mice (32). Moreover, it has been shown that rodent hepatic stellate cells (HSC) express some adrenoceptor subtypes and that stimulation with norepinephrine (NE) induces proliferation and collagen gene expression (32, 33). The cell types mediating the beneficial effects of ␣1adrenoceptor blockers on portal hypertension and liver fibrosis are unknown. HSC are potential targets for the vasoactive actions of catecholamines in the liver. Contraction of HSC in response to vasoconstrictors (i.e., endothelin-1 and angiotensin II) is believed to contribute to the increased resistance to blood flow in advanced chronic liver diseases (6, 35–37). Moreover, activated HSC secrete large amounts of collagen I as well as inflammatory mediators in chronic liver diseases, regulating fibrogenesis and inflammation (27). We hypothesized that NE, a major catecholamine, induces vasoactive effects on activated HSC. Besides this action, we explored whether NE also regulates fibrogenic and proinflammatory actions in these cells.

Address for reprint requests and other correspondence: R. Bataller, Liver Unit, Hospital Clı´nic, Villarroel 170, Barcelona 08036, Spain (e-mail: [email protected]).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Sancho-Bru, Pau, Ramo´n Bataller, Jordi Colmenero, Xavier Gasull, Montserrat Moreno, Vicente Arroyo, David A. Brenner, and Pere Gine`s. Norepinephrine induces calcium spikes and proinflammatory actions in human hepatic stellate cells. Am J Physiol Gastrointest Liver Physiol 291: G877–G884, 2006. First published June 15, 2006; doi:10.1152/ajpgi.00537.2005.—Catecholamines participate in the pathogenesis of portal hypertension and liver fibrosis through ␣1-adrenoceptors. However, the underlying cellular and molecular mechanisms are largely unknown. Here, we investigated the effects of norepinephrine (NE) on human hepatic stellate cells (HSC), which exert vasoactive, inflammatory, and fibrogenic actions in the injured liver. Adrenoceptor expression was assessed in human HSC by RT-PCR and immunocytochemistry. Intracellular Ca2⫹ concentration ([Ca2⫹]i) was studied in fura-2-loaded cells. Cell contraction was studied by assessing wrinkle formation and myosin light chain II (MLC II) phosphorylation. Cell proliferation and collagen-␣1(I) expression were assessed by [3H]thymidine incorporation and quantitative PCR, respectively. NF-␬B activation was assessed by luciferase reporter gene and p65 nuclear translocation. Chemokine secretion was assessed by ELISA. Normal human livers expressed ␣1A-adrenoceptors, which were markedly upregulated in livers with advanced fibrosis. Activated human HSC expressed ␣1A-adrenoceptors. NE induced multiple rapid [Ca2⫹]i oscillations (Ca2⫹ spikes). Prazosin (␣1blocker) completely prevented NE-induced Ca2⫹ spikes, whereas propranolol (nonspecific ␤-blocker) partially attenuated this effect. NE caused phosphorylation of MLC II and cell contraction. In contrast, NE did not affect cell proliferation or collagen-␣1(I) expression. Importantly, NE stimulated the secretion of inflammatory chemokines (RANTES and interleukin-8) in a dose-dependent manner. Prazosin blocked NE-induced chemokine secretion. NE stimulated NF-␬B activation. BAY 11-7082, a specific NF-␬B inhibitor, blocked NE-induced chemokine secretion. We conclude that NE stimulates NF-␬B and induces cell contraction and proinflammatory effects in human HSC. Catecholamines may participate in the pathogenesis of portal hypertension and liver fibrosis by targeting HSC. portal hypertension; collagen; inflammation; catecholamines; liver fibrosis

http://www.ajpgi.org

0193-1857/06 $8.00 Copyright © 2006 the American Physiological Society

G877

G878

EFFECTS OF NOREPINEPHRINE IN HEPATIC STELLATE CELLS

Table 1. PCR primer pairs for adrenoceptors and housekeeping genes Gene

Sense

Antisense

Product, bp

GADPH ␣1A ␣1B ␣2A ␣2B ␤1 ␤2

5⬘-CCATGGAGAAGGCTGGGG-3⬘ 5⬘-TGGCCGACCTCCTGCTCACCTC-3⬘ 5⬘-CCCCCGACGCCGTGTTCAAGGTG-3⬘ 5⬘-GCCGGCCTGCTCATGCTGCTCAC-3⬘ 5⬘-CAGGCCGATGGAGGTGGTTCAAAT-3⬘ 5⬘-GTGGCCCTACGCGAGCAGAA-3⬘ 5⬘-TCTGGCCGCGTTTCTGTGTTGG-3⬘

5⬘-CAAAGTTGTCATGGATGACC-3⬘ 5⬘-GGCCCCGGCTTTCCCTCTTG-3⬘ 5⬘-CTCAGGCGCGGGCAGGCTCAGGA-3⬘ 5⬘-GCCGCCGCCGCCCTTCTTCTC-3⬘ 5⬘-CCGGAGGGGTGCTGGGTAAGGA-3⬘ 5⬘-GCGGCAGTAGATGATGGGGTTGAA-3⬘ 5⬘-GGCCCCGGCTTTCCCTCTTG-3⬘

198 444 407 420 423 195 295

MATERIALS AND METHODS

Human liver specimens and HSC isolation. Human cirrhotic livers were obtained from liver explants at the time of liver transplantation from patients with end-stage liver disease due to hepatitis C virus infection (n ⫽ 5). Human control livers were obtained from resections of liver metastasis of colon cancer (n ⫽ 5). Human HSC were isolated

Fig. 1. Expression of adrenoceptors in human hepatic stellate cells (HSC). A: ␣1A-, ␣2B-, and ␤2-receptors were detected in HSC and whole liver, as assessed by RT-PCR. ␣1B-, ␣2A-, and ␤1-receptors were not detected in HSC but were amplified in mRNA isolated from human livers. GAPDH was amplified as a housekeeping gene to assess correct amplification (data not shown). B: detection of ␣1A-receptor in HSC freshly isolated from normal human livers (quiescent). Green staining shows ␣1A-receptor expression. C: detection of ␣1A-receptor in HSC activated after prolonged culture by immunocytochemistry. Note the ␣1A-adrenoceptor expression at the cell membrane. A negative control using an isotype-matched primary antibody was performed (D). AJP-Gastrointest Liver Physiol • VOL

from fragments of normal livers as described in detail previously (6). Experiments were performed with HSC activated in culture (after the second serial passage). A subset of experiments was performed in activated nonpassaged HSC. The immunocytochemistry studies were performed also in HSC freshly isolated from normal human livers (quiescent phenotype). In all cell cultures, no staining was found for CD45, factor VIII-related antigens, and cam 5.2 (Dako, Glostrup, Denmark), indicating the absence of mono/macrophagic, endothelial, and epithelial cells. Cells were cultured in standard conditions in Iscove’s modified Dulbecco’s medium (BioWhittaker, Verviers, Belgium) containing 15% FCS. Cells were serum starved for at least 12 h before the experiments. The protocol was approved by the Investigational Review Board of the Hospital Clı´nic of Barcelona. Immunocytochemistry studies. Cultured HSC were fixed in methanol at ⫺20°C for 10 min, blocked in PBS containing 0.1% BSA for 30 min, and incubated with anti-p65 or anti-␣1A-adrenoceptor for 1 h (Santa Cruz Biotechnology, Santa Cruz, CA). Cells were incubated with Cy3- or fluorescein-labeled secondary antibody for 1 h. An isotype-matched antibody was used as a negative control. Cells were analyzed with a Leica TCS SL laser scanning confocal spectral microscope (Leica Microsystems Heidelberg, Mannheim, Germany) at a wavelength of 500 –535 nm. Measurement of intracellular Ca2⫹ concentration. Changes in intracellular Ca2⫹ concentration ([Ca2⫹]i) were measured in fura-2 (Calbiochem, San Diego, CA)-loaded cells with an inverted epifluorescence microscope as described in detail previously (6). A representative experiment was performed in fluo-4-loaded cells in a confocal spectral microscope. Cells were tested with NE, prazosin, and

Fig. 2. Expression of ␣1A adrenoceptors in normal and cirrhotic human livers. A: Western blot analysis of 50 ␮g protein from normal and fibrotic human livers. ␣1A-Adrenoceptors were detected in normal livers and were overexpressed in cirrhotic livers (51.5 kDa). Actin expression was included in Western blotting to ensure equal protein loading. B: analysis of ␣1Areceptor expression by densitometry with respect to actin. **P ⬍ 0.01 vs. normal livers. 291 • NOVEMBER 2006 •

www.ajpgi.org

G879

EFFECTS OF NOREPINEPHRINE IN HEPATIC STELLATE CELLS

Table 2. Effect of norepinephrine on [Ca2⫹]i in human hepatic stellate cells

NE (10 ␮M) Prazosin (1 ␮M) ⫹ NE (10 ␮M) Propranolol (1 ␮M) ⫹ NE (10 ␮M)

n

Responder cells, %*

Peak [Ca2⫹]i, nM†

Oscillations, % of cells

51 27 69

84 0 79

317⫾138

57

260⫾170‡

36‡

Values are given as means ⫾ SE; n, no. of cells. *Responder cells are those showing at least a 1.5-fold increase in intracellular Ca compared with resting values. †Responder cells only. ‡P ⬍ 0.05 vs. cells stimulated with norephinephrine (NE).

2⫹

propranolol (Sigma, St. Louis, MO). To study the Ca2⫹ source, experiments were performed with and without extracellular Ca2⫹ in selected groups. To deplete intracellular Ca2⫹ stores, cells were incubated with thapsigargin (Sigma) for 20 min before NE stimulation. Cells were considered to be responders when [Ca2⫹]i increased ⬎50% above the resting value. Assessment of cellular contraction. The contractile responses of cultured HSC to NE were evaluated by wrinkle formation (19, 24). A thin layer of the polydimethyl siloxane (12.500 cP; Sigma) was spread on microscope coverslips. The silicone fluid layer was briefly heated on a low flame from a Bunsen burner to promote cross-linking of the silicone fluid surface while the overlying non-cross-linked fluid served as a lubricant and allowed the surface to move independently of the coverslip. Cells plated on the silicone rubber membrane showed basal wrinkles after 2–3 days in culture. Cells were photographed every 5 min on an Olympus IX70

concentration ([Ca2⫹]i)

inverted phase-contrast microscope with a digital charge-coupled device camera (Hamamatsu, Shizuoka, Japan) and stimulated with the agonists to assess cellular contraction by wrinkle formation. Cell contraction was considered to occur when there was a significant increase in the number of wrinkles or an increase in the length of the preexisting ones. Gene expression assays. RNA was isolated from activated HSC with TRIzol (Life Technologies, Rockville, MD). PCR primer pairs were designed to amplify GAPDH and ␣1A-, ␣1B-, ␣2A-, ␣2B-, ␤1-, and ␤2-adrenoceptors (Table 1). Quantitative PCR was performed with predesigned Assays-on-Demand TaqMan probes and primer pairs for collagen-␣1(I) and ribosomal subunit 18S. Information on these Assays-on-Demand is available at http:// myscience.appliedbiosystems.com/cdsEntry/Form/gene_expression_ keyword.jsp. TaqMan reactions were carried out in duplicate on an ABI PRISM 7900 machine (Applied Biosystems, Foster City, CA).

Fig. 3. Effects of norepinephrine (NE) on intracellular Ca2⫹ concentration ([Ca2⫹]i) and cell contraction in HSC. A: representative graph showing the effect of NE in a single HSC. NE (10 ␮M) induced an oscillatory [Ca2⫹]i increase (Ca2⫹ spikes). B: preincubation with prazosin (1 ␮M) for 10 min blocked NE-induced [Ca2⫹]i increase. C: incubation for 10 min with propranolol (1 ␮M) decreased both Ca2⫹ peak and the percent of cells with oscillations. D: representative images of [Ca2⫹]i increase induced by NE (10 ␮M) in HSC. Images were obtained with a confocal microscope in fluo-4-loaded HSC. Images were taken every 10 s. Color change to red reveals [Ca2⫹]i increase. E: phosphorylation (p) of myosin light chain II (MLC II) as assessed by Western blotting (18 kDa). Stimulation with NE (10 ␮M) or ANG II induced MLC II phosphorylation at 5 min. F: representative images of wrinkle formation (arrows) formed by HSC stimulated with NE (10 ␮M).

AJP-Gastrointest Liver Physiol • VOL

291 • NOVEMBER 2006 •

www.ajpgi.org

G880

EFFECTS OF NOREPINEPHRINE IN HEPATIC STELLATE CELLS

Cell proliferation assay. DNA synthesis was estimated by [methylH]thymidine (Amersham Biosciences, Little Chalfont, UK) incorporation, as described in detail previously (7). Cells were serum starved for 24 h, stimulated for 18 h with agonists, and then pulsed for 6 h with 1 ␮Ci/ml [methyl-3H]thymidine. Results are expressed as fold stimulation compared with cells incubated with buffer. Determination of chemokine secretion. HSC were cultured in six-well plates at a density of 4 ⫻ 105 cells/well for 24 h. Medium was removed, and cells were incubated in serum-free medium for 24 h in the presence of agonists. Supernatants were collected, and a sandwich ELISA for human IL-8 (BLK Diagnostics, Barcelona, Spain) and RANTES (R&D Systems, Minneapolis, MN) was performed. Cells were preincubated with SP-600125, PD-231445 (Sigma), and BAY 11-7082 (Calbiochem, Darmstadt, Germany) in indicated experiments. Results are expressed as fold increases of chemokine secretion compared with cells treated with buffer. Recombinant adenoviral infection. Recombinant adenoviral vectors expressing a luciferase reporter gene driven by NF-␬B transcriptional activation (Ad5NF-␬BLuc; Ref. 39), a dominant-negative mutant form (S32A/S36A) of I␬B (Ad5I␬B; Ref. 23), or control green fluorescent protein (Ad5GFP; Ref. 28) were used. HSCs were infected with Ad5NF-␬BLuc [multiplicity of infection (MOI) 500] and/or Ad5I␬B (MOI 1,000) and Ad5GFP (MOI 1,000) for 12 h in DMEM containing 0.5% FCS. After infection, the medium was replaced with medium containing 0.5% FCS, and the culture was continued for an additional 8 h before the individual experiments were performed. NF-␬B-responsive luciferase assay. HSC were infected with Ad5NF-␬BLuc for 12 h. Medium was replaced, and cells were stimulated with agonists for 8 h. NF-␬B-mediated transcriptional induction was assessed with the luciferase assay system (BD Pharmingen, San Diego, CA). Luciferase activity (relative light units) was normalized to the protein concentration. Electrophoretic mobility shift assay. Cell nuclear proteins were extracted as described previously (8). Eight micrograms of nuclear proteins were incubated with 100 pg of a 32P-labeled probe containing the activator protein-1 (AP-1) consensus site (5⬘-GTAAAGCATGAGTCAGACACCTC-3⬘) in buffer containing (in mM) 10 HEPES (pH 7.8), 2 MgCl2, 50 KCl, 1 DTT, and 0.1 EDTA with 20% glycerol in the presence of single-stranded oligonucleotide (25 ␮g/ml) and poly(dI-dC) (25 ␮g/ml) for 20 min at room temperature. For the competition assay, one sample was incubated with 10 ng of unlabeled probe. Western blot analysis. Whole cell extracts were obtained in Triton lysis buffer containing protease and phosphatase inhibitors. Twentyfive micrograms were loaded onto 10% or 15% SDS-acrylamide gels and blotted onto nitrocellulose membranes. Membranes were then incubated with antibodies against SAPK/JNK, phospho-SAPK/JNK, p44/42 MAPK (ERK), phospho-ERK, phospho-myosin light chain II (MLC II) (Cell Signaling, Beverly, MA), and ␣1A-adrenoceptor (Santa Cruz Biotechnology). After an extensive wash, membranes were incubated with blocking buffer containing horseradish peroxidase-conjugated secondary antibody. Proteins were detected by enhanced chemiluminescence (ECL, Amersham). Data analysis. Data are representative of at least three experiments. Results are expressed as means ⫹ SD. Statistical analysis was performed by Student’s t-test and ANOVA. Statistical analysis was performed using SPSS software (Chicago, IL). 3

the total human liver. Because ␣1-receptors mediate key biological effects of the SNS in fibrogenic cell types (15, 38), we further investigated their expression in HSC as well as in normal and fibrotic human liver specimens. Both quiescent and culture-activated HSC contained ␣1A-receptors, as assessed by immunocytochemistry (Fig. 1, B–D). Moreover, Western blot analysis of whole liver extracts revealed that normal human livers express ␣1A-receptors, which are markedly upregulated in livers with advanced fibrosis (Fig. 2). These results suggest that ␣1A-receptors are upregulated during liver fibrogenesis and that HSC are potential targets for adrenergic agonists and antagonists. NE induces Ca2⫹ spikes and cell contraction in human HSC. We next stimulated HSC with NE, a major neurotransmitter of the SNS. Stimulation of cells with NE (10 ␮M) increased [Ca2⫹]i in fura-2-loaded HSC (Table 2 and Fig. 3, A and D). The typical pattern was characterized by an abrupt increase of [Ca2⫹]i followed by an oscillatory response in 57% of the cells (Ca2⫹ spikes). NE (10 ␮M) induced a significant increase of [Ca2⫹]i in 84% of the cells and a mean Ca2⫹ peak of 317 ⫹ 138 nM. Cell stimulation with 1 and 0.1 ␮M NE also induced a marked [Ca2⫹]i increase (not shown). Removal of Ca2⫹ from the extracellular solution did not affect the NE-induced Ca2⫹ response. Moreover, cell preincubation with thapsigargin, a cell-permeant sarco(endo)plasmic reticulum Ca2⫹-ATPase pump inhibitor that depletes intracellular Ca2⫹ stores, completely abolished the NE-induced Ca2⫹ response (data not shown). These results indicate that NE mainly releases Ca2⫹

RESULTS

␣1A-Adrenergic receptors are expressed in human HSC and are upregulated in human fibrotic livers. We first assessed the expression of adrenoceptors in human HSC by RT-PCR. Cultured human HSC expressed ␣1A-, ␣2B-, and ␤2-adrenoceptors (Fig. 1A). In contrast, ␣1B-, ␣2A-, and ␤1-adrenoceptors were not expressed in HSC but were detected in RNA obtained from AJP-Gastrointest Liver Physiol • VOL

Fig. 4. Effect of NE on cell growth and collagen gene expression. A: effect of NE (10 ␮M) on cell growth as assessed by [3H]thymidine incorporation. HSC were treated for 24 h with NE PDGF-BB (20 ng/ml), or both. PDGF-BB induced a marked increased in DNA synthesis (*P ⬍ 0.001 vs. untreated cells). NE did not stimulate cell growth or amplify PDGF-induced mitogenic effect. B: effect of NE (10 ␮M) on procollagen-␣1(I) gene expression, as assessed by quantitative PCR. Values are means of 5 independent experiments. 291 • NOVEMBER 2006 •

www.ajpgi.org

EFFECTS OF NOREPINEPHRINE IN HEPATIC STELLATE CELLS

from intracellular stores. Prazosin completely prevented the NE-induced [Ca2⫹]i increase in all cells (Table 2 and Fig. 3B). However, preincubation of cells with propranolol only attenuated the NE-induced [Ca2⫹]i peak and the percentage of cells showing an oscillatory response (Fig. 3C). Similar results were obtained with nonpassaged cells activated in culture (data not shown). We next explored whether NE induces cell contraction. After 5 min, NE induced MLC II phosphorylation, which is a known prerequisite for cell contraction (20, 44) (Fig. 3E). Preincubation with 10 ␮M BAPTA-AM did not significantly modify MLC II phosphorylation, suggesting that it could be Ca2⫹ independent (data not shown). Moreover, we found that stimulation with NE induced cell contraction, as assessed by wrinkle formation (Fig. 3F). Together, these results indicate that NE induces cell contraction and may be involved in the vasoactive effects of HSC. NE does not stimulate proliferation of human HSC. We investigated whether NE regulates human HSC growth. Different concentrations of NE (10 nM–10 ␮M) were tested. Incubation with NE did not affect DNA synthesis, as assessed by [3H]thymidine incorporation (Fig. 4A). Stimulation with phenylephrine or terbutaline (␣- and ␤-receptor agonists, respectively) did not affect cell proliferation (data not shown).

G881

To elucidate whether NE regulates agonist-induced cell proliferation, HSC were pretreated with NE (10 ␮M) and then challenged with PDGF-BB (20 ng/ml), a potent mitogen for these cells. NE pretreatment did not affect PDGF-induced mitogenic effects. These results indicate that NE is not mitogenic for human HSC. Effect of NE on gene expression of collagen-␣1(I) in human HSC. We next investigated whether NE regulates the expression of collagen-␣1(I), the main extracellular protein found in fibrotic human livers. Stimulation of cells with NE (10 ␮M) did not affect collagen expression, as assessed by quantitative PCR (Fig. 4B). These results suggest that NE does not directly regulate collagen synthesis in activated human HSC. NE stimulates chemokine production by human HSC. We next investigated whether NE regulates the inflammatory actions of HSC by stimulating chemokine secretion. NE (10 ␮M) induced a significant increase in secretion of IL-8 and RANTES, which are chemokines potentially involved in liver fibrogenesis (34, 40), to the culture medium (Fig. 5A). Preincubation of cells with prazosin markedly reduced both IL-8 and RANTES secretion. Propranolol also attenuated NE-induced chemokine production, yet only RANTES secretion was significantly reduced. These results suggest that the inflammatory

Fig. 5. Effect of NE on chemokine secretion in human HSC. A: NE stimulated the secretion of IL-8 and RANTES to the culture medium, as assessed by specific ELISA. Incubation with prazosin prevented NE-induced chemokine secretion. Propranolol reduced NE-induced RANTES secretion. B: preincubation with BAY 11-7082 (a NF-␬B inhibitor), but not with SP-600125 (a JNK inhibitor) and PD-231445 (a ERK inhibitor), prevented NE-induced IL-8 secretion. C: Ad5I␬B reduced IL-8 production induced by NE. *P ⬍ 0.05 vs. untreated cells (Ad5GFP was used as a control); #P ⬍ 0.05 vs. cells treated with NE alone (n ⫽ 5). AJP-Gastrointest Liver Physiol • VOL

291 • NOVEMBER 2006 •

www.ajpgi.org

G882

EFFECTS OF NOREPINEPHRINE IN HEPATIC STELLATE CELLS

actions of NE on human HSC could be mediated by both ␣1and ␤-adrenoceptors. Effect of NE on intracellular signaling pathways in human HSC. Finally, we explored the effect of NE on intracellular pathways known to regulate the inflammatory actions in human HSC (8, 41). These pathways include MAPK and the transcription factors NF-␬B and AP-1. NE (10 ␮M) induced a transient increase in ERK and SAPK/JNK phosphorylation, as assessed by Western blot analysis (Fig. 6A). NE activated NF-␬B, as indicated by increased NF-␬B-dependent gene expression as well as nuclear translocation of the heterodimer p50/p65 (Fig. 6, B and C). In addition, NE increased TNF-␣-induced NF-␬B activation, suggesting that it could amplify the inflammatory response evoked by other cytokines (data not shown). Finally, NE stimulated AP-1 DNA binding, as assessed by the electrophoretic mobility shift assay (Fig. 6D). To elucidate the role of these signaling pathways on chemokine production, HSC were preincubated with specific antagonists before NE stimulation. Preincubation of cells with BAY 11-7082 (a NF-␬B inhibitor), but not with SP-600125 (a JNK inhibitor) and PD-231445 (an

ERK inhibitor), attenuated IL-8 secretion (Fig. 5B). Moreover, transfection with Ad5I␬B reduced IL-8 production, confirming NF-␬B involvement (Fig. 5C). These data strongly indicate that NE exerts proinflammatory actions in human HSC through NF-␬B activation. DISCUSSION

In the present study, we investigated the biological effects of NE in human HSC. We provide evidence that HSC express some ␣- and ␤-receptor subtypes. NE stimulates key intracellular signaling pathways in these cells and induces cell contraction and proinflammatory actions. These actions are mainly mediated by ␣1-receptors, although ␤-receptors also modulate some of the effects evoked by NE. Moreover, ␣1-receptors are markedly upregulated in human fibrotic livers. These results suggest that HSC are potential targets for the pathogenic effect of catecholamines in the liver, in keeping with recent data indicating that catecholamines not only regulate vascular ho-

Fig. 6. Effect of NE on intracellular signaling pathways in human HSC. A: effect of NE on ERK and SAPK/JNK phosphorylation, as assessed by Western blotting. Stimulation with NE induced a transient phosphorylation of ERK and SAPK/JNK. Images are representative of 3 independent experiments. B: NF-␬B-dependent gene expression as assessed by a luciferase assay. NE stimulated NF-␬B-dependent gene expression that was completely blunted by prazosin but not by propranolol. *P ⬍ 0.05 vs. untreated cells; #P ⬍ 0.05 vs. cells treated with NE alone. Results are means of 4 independent experiments. C: immunocytochemistry analysis of the p65 subunit of NF-␬B in HSC. Stimulation for 60 min with NE (10 ␮M) resulted in the translocation of the heterodimer p50/p65 to the nucleus, which is indicative of NF-␬B activation. TNF-␣ (10 ng/ml) was used as a positive control. D: activation of the transcription factor activator protein-1 (AP-1) as assessed by electrophoretic mobility shift assay. Stimulation with NE (10 ␮M) for 60 min markedly stimulated AP-1 DNA binding. Preincubation with cold probe was performed to ensure specificity. Image is representative of 3 independent experiments. AJP-Gastrointest Liver Physiol • VOL

291 • NOVEMBER 2006 •

www.ajpgi.org

EFFECTS OF NOREPINEPHRINE IN HEPATIC STELLATE CELLS

meostasis but also contribute to tissue repair and inflammation (11, 15, 30, 31). The hallmark biological action of NE in vascular cell types is to induce Ca2⫹ increase and cell contraction. This effect has been extensively studied in mesangial cells and vascular smooth muscle cells (18, 26). Similarly, a large number of studies have investigated the vasoactive effects of vasoconstrictors in cultured human HSC, including angiotensin II, endothelin-1, thrombin, and leukotriene D4 (36). Currently, studies using activated HSC are widely used to investigate the cellular basis of portal hypertension to identify potential targets for therapy. Surprisingly, little information is available on the actions of adrenergic agonists in these cells. Here, we provide evidence that NE is a potent agonist for human HSC to induce [Ca2⫹]i increase. An intriguing finding of the present study is that the Ca2⫹ response pattern evoked by NE markedly differs from that observed with other vasoconstrictors. Previously, we (6) showed that angiotensin II induces an abrupt [Ca2⫹]i increase followed by a sustained phase. This effect was mainly due to the entrance of Ca2⫹ from the extracellular space through voltage-operated Ca2⫹ channels. Interestingly, the effect of NE on [Ca2⫹]i consists of numerous Ca2⫹ spikes. Baseline Ca2⫹ spikes are characterized by rapidly rising transient increases in [Ca2⫹]i, rising from a baseline that is close to the resting concentration (9). This pattern of response has been described in many cells in response to hormones and neurotransmitters (18, 25). Ca2⫹ oscillations permit cells to respond to agonists without being exposed to sustained levels of [Ca2⫹]i. It has been shown that oscillations and especially their frequency activate cell signaling like calmodulin-dependent protein kinase II and NF-␬B (14, 22). Moreover, Ca2⫹ oscillations have been involved in many cellular processes like contraction, migration, cell secretion, or phagocytosis (9, 12). The biological relevance of NE-induced Ca2⫹ spikes in HSC is unknown. Although NE induced MLC II phosphorylation, a prerequisite for cell contraction, and induced cell contraction, it failed to induce cell proliferation, which is commonly stimulated by vasoconstrictors in HSC. It is possible that sustained [Ca2⫹]i increase, rather than [Ca2⫹]i oscillations, is required to induce cell proliferation. Further studies should investigate the pathophysiological relevance of agonist-induced Ca2⫹ spikes in HSC. A relevant finding of the present study is that NE stimulates key intracellular signaling pathways implicated in the pathogenesis of hepatic inflammation and fibrogenesis. These pathways include NF-␬B, JNK/AP-1, and ERK. Importantly, NE enhances chemokine secretion in a NF-␬B-dependent manner. Besides their role in collagen synthesis, recent data suggest that activated HSC play a supportive role in the inflammatory response to injury (27). These cells proliferate at the areas of liver injury, and cross-talk between HSC and infiltrating leukocytes is likely to occur. HSC can present antigens, express cell adhesion molecules, regulate lymphocyte proliferation, and secrete powerful chemokines such as IL-8 and RANTES (27, 42). On the other hand, mediators derived from lymphocytes (e.g., CD40 ligand, RANTES) can regulate HSC biological functions and stimulate their fibrogenic potential (40, 41). Therefore, by stimulating chemokine secretion, NE may enhance inflammation in chronic liver diseases. Whether NE also exerts proinflammatory actions in other nonparenchymal heAJP-Gastrointest Liver Physiol • VOL

G883

patic cell types (i.e., Kupffer cells and sinusoidal endothelial cells) is unknown and deserves further investigation. In the present study, despite the possible proinflammatory effect of NE, we did not find a direct fibrogenic effect of NE on collagen expression. In rodent early-cultured HSC, NE stimulates collagen secretion, suggesting that it directly induces fibrogenic effects in these cells (32, 33). These discrepant results can be explained by the fact that we used passaged, fully activated human HSC. In this cell phenotype, collagen is highly expressed and fibrogenic agonists commonly fail to further increase its expression. Other biological functions, such as chemokine secretion, can be stimulated by inflammatory mediators (40, 41). However, it is well known that liver inflammation precedes and promotes the progression of liver fibrosis, and there is a positive correlation between the degree of sustained hepatic inflammation and the progression of fibrosis. Moreover, substances inhibiting the inflammatory response of the liver also limit fibrogenesis (5). Therefore, we speculate that the proinflammatory effect of NE on HSC may play a role in promoting hepatic inflammation in patients with chronic liver diseases, thus favoring fibrosis development. The biological effects of NE in HSC may have pathophysiological consequences. First, contraction of activated HSC is believed to increase intrahepatic vascular resistance in the cirrhotic liver and contribute to the pathogenesis of portal hypertension (37). The finding that NE induces contraction in HSC may explain, at least in part, the effects of catecholamines in regulating intrahepatic vascular resistance in cirrhosis. Moreover, the fact that HSC synthesize NE suggests that not only may an endocrine pathways regulate HSC contractility but an autocrine/paracrine system may also be involved (32). Further studies should study this hypothesis. Our finding that NE enhances chemokine secretion in HSC suggests that catecholamines may exert proinflammatory effects in the liver. The proinflammatory effect of NE on HSC has not been previously described in other cell types. The finding that NE may promote proinflammatory actions in HSC expands the traditional view of the SNS as a system that basically regulates systemic and hepatic hemodynamics. In summary, the results of the present study indicate that NE mediates vasoactive and proinflammatory actions in human HSC mainly through ␣1-adrenoceptors. Because HSC are able to synthesize adrenergic agonists and the SNS is activated in patients with advanced liver fibrosis, these findings suggest that catecholamines may increase intrahepatic vascular resistance and promote inflammatory actions by targeting HSC. Our data explain, at least in part, the mechanisms underlying the beneficial effects of ␣1-receptor antagonists in the treatment of portal hypertension and suggest the potential use of these drugs to modulate hepatic inflammation and subsequent fibrogenesis. ACKNOWLEDGMENTS We thank Joan Carles Garcia-Paga´n for helpful discussion, Elena Juez, and Cristina Milla´n for excellent technical support. We thank Maria Calvo and Anna Bosch [Serveis Cientificote`cnics, Universitat de Barcelona-Institut d’Investigacions Biome`diques August Pi i Sunyer (IDIBAPS)] for excellent technical assistance in in vivo confocal microscopy. GRANTS This work was supported by National Institute on Alcohol Abuse and Alcoholism Grant R01-AA-15055, Ministerio de Ciencia y Tecnologı´a, Direccio´n General de Investigacio´n Grant SAF2002-03696, and Instituto de 291 • NOVEMBER 2006 •

www.ajpgi.org

G884

EFFECTS OF NOREPINEPHRINE IN HEPATIC STELLATE CELLS

Salud Carlos III Grant CO3/02. P. Sancho-Bru had a grant from IDIBAPS. M. Moreno had a grant from Instituto Reina Sofı´a de Investigacio´n Nefrolo´gica and iDiBAPS. REFERENCES 1. Akiyama-Uchida Y, Ashizawa N, Ohtsuru A, Seto S, Tsukazaki T, Kikuchi H, Yamashita S, and Yano K. Norepinephrine enhances fibrosis mediated by TGF-beta in cardiac fibroblasts. Hypertension 40: 148–154, 2002. 2. Akiyoshi H and Terada T. Mast cell, myofibroblast and nerve terminal complexes in carbon tetrachloride-induced cirrhotic rat livers. J Hepatol 29: 112–119, 1998. 3. Albillos A, Lledo JL, Rossi I, Perez-Paramo M, Tabuenca MJ, Banares R, Iborra J, Garrido A, Escartin P, and Bosch J. Continuous prazosin administration in cirrhotic patients: effects on portal hemodynamics and on liver and renal function. Gastroenterology 109: 1257–1265, 1995. 4. Ballet F, Chretien Y, Rey C, and Poupon R. Differential response of normal and cirrhotic liver to vasoactive agents. A study in the isolated perfused rat liver. J Pharmacol Exp Ther 244: 283–289, 1988. 5. Bataller R and Brenner DA. Hepatic stellate cells as a target for the treatment of liver fibrosis. Semin Liver Dis 21: 437– 451, 2001. 6. Bataller R, Gines P, Nicolas JM, Gorbig MN, Garcia-Ramallo E, Gasull X, Bosch J, Arroyo V, and Rodes J. Angiotensin II induces contraction and proliferation of human hepatic stellate cells. Gastroenterology 118: 1149 –1156, 2000. 7. Bataller R, Nicolas JM, Gines P, Esteve A, Nieves GM, Garcia-Ramallo E, Pinzani M, Ros J, Jimenez W, Thomas AP, Arroyo V, and Rodes J. Arginine vasopressin induces contraction and stimulates growth of cultured human hepatic stellate cells. Gastroenterology 113: 615– 624, 1997. 8. Bataller R, Schwabe RF, Choi YH, Yang L, Paik YH, Lindquist J, Qian T, Schoonhoven R, Hagedorn CH, Lemasters JJ, and Brenner DA. NADPH oxidase signal transduces angiotensin II in hepatic stellate cells and is critical in hepatic fibrosis. J Clin Invest 112: 1383–1394, 2003. 9. Berridge MJ, Bootman MD, and Roderick HL. Calcium signalling: dynamics, homeostasis and remodelling. Nat Rev Mol Cell Biol 4: 517– 529, 2003. 10. Bioulac-Sage P, Lafon ME, Saric J, and Balabaud C. Nerves and perisinusoidal cells in human liver. J Hepatol 10: 105–112, 1990. 11. Bonnefont-Rousselot D, Mahmoudi A, Mougenot N, Varoquaux O, Le Nahour G, Fouret P, and Lechat P. Catecholamine effects on cardiac remodelling, oxidative stress and fibrosis in experimental heart failure. Redox Rep 7: 145–151, 2002. 12. Bootman MD, Lipp P, and Berridge MJ. The organisation and functions of local Ca2⫹ signals. J Cell Sci 114: 2213–2222, 2001. 13. Burt AD, Tiniakos D, MacSween RN, Griffiths MR, Wisse E, and Polak JM. Localization of adrenergic and neuropeptide tyrosine-containing nerves in the mammalian liver. Hepatology 9: 839 – 845, 1989. 14. De Koninck P and Schulman H. Sensitivity of CaM kinase II to the frequency of Ca2⫹ oscillations. Science 279: 227–230, 1998. 15. Dubuisson L, Desmouliere A, Decourt B, Evade L, Bedin C, Boussarie L, Barrier L, Vidaud M, and Rosenbaum J. Inhibition of rat liver fibrogenesis through noradrenergic antagonism. Hepatology 35: 325–331, 2002. 16. Esler M, Dudley F, Jennings G, Debinski H, Lambert G, Jones P, Crotty B, Colman J, and Willett I. Increased sympathetic nervous activity and the effects of its inhibition with clonidine in alcoholic cirrhosis. Ann Intern Med 116: 446 – 455, 1992. 17. Graupera M, Garcia-Pagan JC, Abraldes JG, Peralta C, Bragulat M, Corominola H, Bosch J, and Rodes J. Cyclooxygenase-derived products modulate the increased intrahepatic resistance of cirrhotic rat livers. Hepatology 37: 172–181, 2003. 18. Hajjar RJ and Bonventre JV. Oscillations of intracellular calcium induced by vasopressin in individual fura-2-loaded mesangial cells. Frequency dependence on basal calcium concentration, agonist concentration, and temperature. J Biol Chem 266: 21589 –21594, 1991. 19. Harris AK, Wild P, and Stopak D. Silicone rubber substrata: a new wrinkle in the study of cell locomotion. Science 208: 177–179, 1980. 20. Hirano K, Hirano M, and Kanaide H. Regulation of myosin phosphorylation and myofilament Ca2⫹ sensitivity in vascular smooth muscle. J Smooth Muscle Res 40: 219 –236, 2004. 21. Hsu CT. The role of the sympathetic nervous system in promoting liver cirrhosis induced by carbon tetrachloride, using the essential hypertensive animal (SHR). J Auton Nerv Syst 37: 163–173, 1992. 22. Hu Q, Deshpande S, Irani K, and Ziegelstein RC. [Ca2⫹]i oscillation frequency regulates agonist-stimulated NF-␬B transcriptional activity. J Biol Chem 274: 33995–33998, 1999. AJP-Gastrointest Liver Physiol • VOL

23. Iimuro Y, Nishiura T, Hellerbrand C, Behrns KE, Schoonhoven R, Grisham JW, and Brenner DA. NFkappaB prevents apoptosis and liver dysfunction during liver regeneration. J Clin Invest 101: 802– 811, 1998. 24. Kawada N, Klein H, and Decker K. Eicosanoid-mediated contractility of hepatic stellate cells. Biochem J 285: 367–371, 1992. 25. Kawanishi T, Blank LM, Harootunian AT, Smith MT, and Tsien RY. Ca2⫹ oscillations induced by hormonal stimulation of individual fura-2loaded hepatocytes. J Biol Chem 264: 12859 –12866, 1989. 26. Koshimizu TA, Tanoue A, Hirasawa A, Yamauchi J, and Tsujimoto G. Recent advances in ␣1-adrenoceptor pharmacology. Pharmacol Ther 98: 235–244, 2003. 27. Marra F. Hepatic stellate cells and the regulation of liver inflammation. J Hepatol 31: 1120 –1130, 1999. 28. Moriyoshi K, Richards LJ, Akazawa C, O’Leary DD, and Nakanishi S. Labeling neural cells using adenoviral gene transfer of membranetargeted GFP. Neuron 16: 255–260, 1996. 29. Oben JA and Diehl AM. Sympathetic nervous system regulation of liver repair. Anat Rec A Discov Mol Cell Evol Biol 280: 874 – 883, 2004. 30. Oben JA, Roskams T, Yang S, Lin H, Sinelli N, Li Z, Torbenson M, Huang J, Guarino P, Kafrouni M, and Diehl AM. Sympathetic nervous system inhibition increases hepatic progenitors and reduces liver injury. Hepatology 38: 664 – 673, 2003. 31. Oben JA, Roskams T, Yang S, Lin H, Sinelli N, Li Z, Torbenson M, Thomas SA, and Diehl AM. Norepinephrine induces hepatic fibrogenesis in leptin deficient ob/ob mice. Biochem Biophys Res Commun 308: 284 –292, 2003. 32. Oben JA, Roskams T, Yang S, Lin H, Sinelli N, Torbenson M, Smedh U, Moran TH, Li Z, Huang J, Thomas SA, and Diehl AM. Hepatic fibrogenesis requires sympathetic neurotransmitters. Gut 53: 438 – 445, 2004. 33. Oben JA, Yang S, Lin H, Ono M, and Diehl AM. Norepinephrine and neuropeptide Y promote proliferation and collagen gene expression of hepatic myofibroblastic stellate cells. Biochem Biophys Res Commun 302: 685– 690, 2003. 34. Pinzani M and Marra F. Cytokine receptors and signaling in hepatic stellate cells. Semin Liver Dis 21: 397– 416, 2001. 35. Pinzani M, Milani S, De Franco R, Grappone C, Caligiuri A, Gentilini A, Tosti-Guerra C, Maggi M, Failli P, Ruocco C, and Gentilini P. Endothelin 1 is overexpressed in human cirrhotic liver and exerts multiple effects on activated hepatic stellate cells. Gastroenterology 110: 534 –548, 1996. 36. Reynaert H, Thompson MG, Thomas T, and Geerts A. Hepatic stellate cells: role in microcirculation and pathophysiology of portal hypertension. Gut 50: 571–581, 2002. 37. Rockey D. The cellular pathogenesis of portal hypertension: stellate cell contractility, endothelin, and nitric oxide. Hepatology 25: 2–5, 1997. 38. Saeed AE, Parmentier JH, and Malik KU. Activation of ␣1A-adrenergic receptor promotes differentiation of rat-1 fibroblasts to a smooth musclelike phenotype. BMC Cell Biol 5: 47, 2004. 39. Sanlioglu S, Williams CM, Samavati L, Butler NS, Wang G, McCray PB Jr, Ritchie TC, Hunninghake GW, Zandi E, and Engelhardt JF. Lipopolysaccharide induces Rac1-dependent reactive oxygen species formation and coordinates tumor necrosis factor-␣ secretion through IKK regulation of NF-␬B. J Biol Chem 276: 30188 –30198, 2001. 40. Schwabe RF, Bataller R, and Brenner DA. Human hepatic stellate cells express CCR5 and RANTES to induce proliferation and migration. Am J Physiol Gastrointest Liver Physiol 285: G949 –G958, 2003. 41. Schwabe RF, Schnabl B, Kweon YO, and Brenner DA. CD40 activates NF-␬B and c-Jun N-terminal kinase and enhances chemokine secretion on activated human hepatic stellate cells. J Immunol 166: 6812– 6819, 2001. 42. Vinas O, Bataller R, Sancho-Bru P, Gines P, Berenguer C, Enrich C, Nicolas JM, Ercilla G, Gallart T, Vives J, Arroyo V, and Rodes J. Human hepatic stellate cells show features of antigen-presenting cells and stimulate lymphocyte proliferation. Hepatology 38: 919 –929, 2003. 43. vom Dahl S, Bode JG, Reinehr RM, Monnighoff I, Kubitz R, and Haussinger D. Release of osmolytes from perfused rat liver on perivascular nerve stimulation: alpha-adrenergic control of osmolyte efflux from parenchymal and nonparenchymal liver cells. Hepatology 29: 195–204, 1999. 44. Yanase M, Ikeda H, Ogata I, Matsui A, Noiri E, Tomiya T, Arai M, Inoue Y, Tejima K, Nagashima K, Nishikawa T, Shibata M, Ikebe M, Rojkind M, and Fujiwara K. Functional diversity between Rho-kinaseand MLCK-mediated cytoskeletal actions in a myofibroblast-like hepatic stellate cell line. Biochem Biophys Res Commun 305: 223–228, 2003. 291 • NOVEMBER 2006 •

www.ajpgi.org