Kidney International, Vol. 68 (2005), pp. 96–109 The antifibrotic effects of relaxin in human renal fibroblasts are mediated in part by inhibition of the Smad2 pathway2 MALTE H.J. HEEG,1 MICHAEL J. KOZIOLEK,1 RADOVAN VASKO, LILIANA SCHAEFER, KUMAR SHARMA, GERHARD A. MU¨LLER, and FRANK STRUTZ Department of Nephrology and Rheumatology, Georg-August-University, Go¨ttingen, Department of Internal Medicine D, University of Muenster, both in Germany, and Department of Nephrology, Thomas Jefferson University, Philadelphia, Pennsylvania The antifibrotic effects of relaxin in human renal fibroblasts are mediated in part by inhibition of the Smad2 pathway. Background. The peptide hormone relaxin has been demon- strated to exert antifibrotic effects in renal and extrarenal tissues. The aims of this study were to identify potential anti- fibrotic effects of relaxin on human renal fibroblasts in vitro and to analyze their mechanisms. Methods. All experiments were performed in established re- nal fibroblast cell lines and in primary cortical fibroblasts. Ef- fects of relaxin were analyzed on cell proliferation, apoptosis, activation of renal fibroblasts, synthesis and secretion of colla- gen type I and fibronectin, as well as on the secretion of matrix metalloproteinases (MMPs). Effects on transforming growth factor-b1 (TGF-b1) receptor binding were analyzed by flow cytometry and on TGF-b1 signal transduction by immunoblot analyses for Smad4 and 7, translocation from cytosol to nucleus for Smad2 and 3 as well as for phosphorylated and unphospho- rylated forms of p38, c-Jun NH2 terminal kinase (JNK) and extracellular-regulated protein kinase (ERK). Finally, specific siRNAs for Smad2 and 3 were applied to assess the signal trans- duction pathway. Results. After stimulation with relaxin, tyrosine phospho- rylation of a 220 kD protein was demonstrated, indicat- ing interaction with the receptor. Relaxin had only modest inhibitory effects on cell proliferation, and no effects on apop- tosis. Conversely, relaxin exerted robust effects on TGF-b1– induced fibroblast to myofibroblast transformation as well as on matrix synthesis and secretion even at the smallest dose tested. The secretion of MMP-2 and MMP-9 was induced no- ticeably by all investigated relaxin concentrations. TGF-b1 receptor binding was not influenced by relaxin; however, it pre- vented Smad2 phosphorylation, translocation to nucleus, and complex formation between Smad2 and 3 indicating a possible 1These two authors contributed equally to this manuscript. 2See editorial by Gaedeke and Peters, p. 405. Key words: fibrosis, TGF-b1 signal transduction, fibroblasts, ECM, MMP, Smad. Received for publication November 4, 2003 and in revised form October 20, 2004, and January 6, 2005 Accepted for publication January 28, 2005 C© 2005 by the International Society of Nephrology interaction with TGF-b1 signaling. These findings were corrob- orated by studies using siRNAs to Smad2 and 3 where siRNA to Smad2 but not to Smad3 inhibited the TGF-b1 induction of fibronectin synthesis. There was no influence of relaxin on intracellular Smad3, Smad4, and Smad7 translocation or phos- phorylation of mitogen-activated protein (MAP) kinases. Conclusion. Relaxin is a potent inhibitor of TGF-b1–induced extracellular matrix (ECM) synthesis and secretion as well as fibroblast activation. Furthermore, it induces ECM degradation by induction of MMP-2 and MMP-9. These effects are mediated, at least in part, by inhibition of TGF-b1 signaling. The typical morphologic correlate of end-stage renal failure is interstitial fibrosis, tubular atrophy, and dilata- tion. Renal fibrosis is mainly characterized by an acti- vation and proliferation of interstitial fibroblasts and by an excessive synthesis and accumulation of extracellu- lar matrix (ECM) components, including fibronectin and collagen types I, III, and IV along with a decreased rate of ECM degradation [1]. This ECM degradation is often inhibited during fibrogenesis due to the up-regulation of tissue inhibitor of matrix metalloproteinase 1 (TIMP-1) and/or a decreased activity of matrix metalloproteinase (MMP). Cytokines such as transforming growth factor- b1 (TGF-b1), connective tissue growth factor (CTGF), platelet derived growth factor (PDGF), fibroblast growth factor 2 (FGF-2), and angiotensin II (Ang II) play impor- tant roles in the process of fibrogenesis [2]. TGF-b1 is cur- rently viewed as the most important fibrosis-promoting cytokine [3]. It is responsible for several profibrotic events during fibrogenesis, including ECM synthesis in fibroblasts and tubular epithelial cells, differentiation of fibroblasts into myofibroblasts, epithelial-mesenchymal transformation of tubular epithelial cells, and inhibition of matrix degradation by inhibiting the expression of MMPs and up-regulating the expression of TIMPs. In- creased expression has been demonstrated in a number of animal models associated with progressive renal disease including anti-Thy1.1 nephritis [4], diabetic nephropa- thy [5], Heyman nephritis [6], and obstructive nephropa- thy [7]. TGF-b signal transduction is mainly regulated 96 Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts 97 via the Smad proteins, the receptor mediated Smad2, Smad3, the common mediator Smad, Smad4, and the in- hibitory Smad7 [8]. Furthermore, mitogen-activated pro- tein (MAP) kinase pathways play an important role for TGF-b signal transduction [9]. Interstitial scarring often progresses despite resolution of the primary disease process. Recently, a number of po- tential factors have emerged with the potential to possibly reverse fibrotic disease in the kidney, including hepato- cyte growth factor (HGF) [10] and bone morphogenetic protein-7 (BMP-7) [11]. However, these substances have not been tested in clinical studies and additional ap- proaches are needed. Another potential candidate for a novel therapeutic approach to progressive renal disease is relaxin. It is a peptide hormone with a molecular weight of 6 kD and a member of the insulin-like growth factor (IGF) family. Like insulin, relaxin is composed by two disulfide-linked chains, the A and B chains [12]. It is pre- dominantly produced by the corpus luteum during preg- nancy whereby serum levels in pregnant women range from 0.2 to 1.2 ng/mL [13]. The primarily biologic func- tions of relaxin are the widening of the birth channel in preparation of birth and the renal vasodilation and hyper- filtration during pregnancy [14]. In humans two genes (H1 and H2) code for functional relaxin. However, H2 relaxin is the principal expression product in vivo. Relaxin has already demonstrated promising antifibrotic potential in vitro and in vivo. In dermal and pulmonary as well as in hepatic fibroblasts recombinant human (rh) relaxin pre- vented TGF-b–induced collagen synthesis [15–17]. Fur- thermore, in pulmonary fibroblasts, relaxin inhibited the fibronectin synthesis and stimulated MMP secretion [16]. Relaxin deficiency in mice results in an age-related pro- gression of lung fibrosis [18]. In vivo relaxin showed antifi- brotic properties in models of pulmonary [15] and renal fibrosis [19, 20]. Furthermore, it reduced skin thickening and improved mobility in a clinical trial in scleroderma patients [21]. However, the mechanisms of its actions are largely unknown. The aims of this study were to investigate the effects of relaxin on cell growth and activation, matrix synthesis, and degradation in human renal fibroblasts and to iden- tify the intracellular signaling mechanism of these pro- cesses. We demonstrate that relaxin influences fibroblast proliferation only modestly but has robust inhibitory ef- fects on fibroblast activation and ECM synthesis while stimulating matrix degradation. These effects are medi- ated by the prevention of Smad2 phosphorylation and complex formation with Smad3, whereas no interference with the MAP kinase pathway was observed. METHODS Materials rhrelaxin was a generous gift from Dr. Unemori (Connectics Corp., Palo Alto, CA, USA), rhTGF-b1 was purchased from R&D Systems (Minneapolis, MN, USA). The following mouse monoclonal antibodies were used for cell characterization: anticollagen type I [also used for enzyme-linked immunosorbent assay (ELISA)], anticollagen type III (both from Southern Biothechnology, Birmingham, AL, USA), anticytok- eratin (Dako, Carpinteria, CA, USA), antivimentin (Boehringer Mannheim, Mannheim, Germany), an- tifactor VIII (Dako), antihuman leukocyte antigen (anti-HLA-DR) (Dako), anti-a-smooth muscle actin (anti-a-SMA) (Paesel and Lorei, Wiesbaden, Germany), anti-CD44 (Pharmingen, San Diego, CA, USA), anti- CD54, and anti-CD68 (Dako). Rabbit polyclonal an- tibody to collagen type I from Biotrend (Cologne, Germany) was used for immunoblot analyses; rabbit polyclonal antibody to fibronectin (Sigma Chemical Co., St. Louis, MO, USA) was used for ELISA as well as for immunoblots. Antirabbit IgG alkaline phosphatase (AP) and antisheep/goat IgG AP, Fab fragments (both from Boehringer Mannheim, Mannheim, Germany) were used as secondary antibodies for ELISA. Mouse polyclonal anti-MMP-1 was purchased from Oncogene (Boston, MA, USA) and rabbit polyclonal anti-MT- MMP-1 from Calbiochem-Novabiochem (Nottingham, UK). The rabbit antibodies to extracellular-regulated protein kinase 1/2 (ERK1/2) and c-Jun NH2 terminal kinase 1/2 (JNK1/2), the mouse monoclonal antibod- ies to phosphorylated JNK1/2, p38 MAP kinase, and phosphorylated p38 MAP kinase as well as the goat polyclonal antibody to phosphorylated ERK1/2 were ob- tained from Santa Cruz Biotechnology (Santa Cruz, CA, USA), as was the donkey antigoat-horseradish peroxi- dase (HRP) secondary antibody for immunoblots. Re- combinant antiphosphotyrosine HRP, goat polyclonal Smad7, and mouse polyclonal Smad4/DPC4 were used for immunoblots and purchased from Transduction Lab- oratories (Lexington, KY, USA). The rabbit polyclonal antibody to phosphorylated Smad2 was a kind gift from Serhiy Souchelnytski (Ludwig Institute for Cancer Research, Uppsala, Sweden) and the rabbit poly- clonal antibody to phosphorylated Smad3 was purchased from Cell Signaling (Beverly, MA, USA). The mouse monoclonal antibody to a-SMA was purchased from Progen (Heidelberg, Germany). The mouse antibody to caspase 3 was a friendly gift from Dr. Schweyer (Department of Pathology, Georg-August-University, Goettingen, Germany). As secondary antibodies for im- munofluorescence stainings served rabbit antimouse flu- orescein isothiocyanate (FITC) and rabbit antimouse Rhodamin from Dianova (Hamburg, Germany). Trypsin ethylenediaminetetraacetic acid (EDTA), Dulbecco’s modified Eagle’s medium (DMEM), Iscove’s modified Eagle’s medium (IMDM), and fetal calf serum (FCS) were obtained from Gibco Ltd. (Paisley, Scotland). Cell culture dishes were from Becton Dickinson (Franklin Lakes, NJ, USA). 98 Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts Cell culture and quantitation of a-SMA–positive cells Human renal fibroblast cell lines Tk 173 (normal kid- ney derived) and Tk 188 (from a kidney with tubu- lointerstitial fibrosis) have been described previously [22]. These cells were cultured in DMEM containing 10% FCS, 1% L-glutamine (both from Gibco) and 5 lg/mL M-plasmocin to avoid mycoplasma contamination (Paesel and Lorei). Primary renal fibroblasts were iso- lated from human kidney biopsies. Isolation and charac- terization were performed as described before [23]. Cells that were positive for vimentin, CD44, CD54, a-SMA, and the collagen types I and III, and negative for cytok- eratin, factor VIII, and major histocompatability com- plex (MHC) class II were considered to be fibroblasts and used for induction assays in passages 2 to 8. Addi- tional staining with the Hoechst dye H33258 (Boehringer Mannheim) was performed to facilitate cell counting. Pri- mary cortical fibroblasts Tk 489 were established from a 61-year-old female with membranous gomerulonephritis without interstitial involvement, Tk 461 fibroblasts were cultured from a biopsy of a 25-year-old female with IgA nephritis and tubulointerstitial fibrosis involving 30% of the interstitium. Quantitation of a-SMA expression was performed as described previously [24]. Briefly, cells were counted and the relative percentage of positive cells determined by a blinded investigator after stimulation with 10 ng/mL TGF-b1 and relaxin in concentrations of 0.1, 1, and 10 ng/mL for 48 hours. Stimulation with TGF-b1 alone served as positive control. Cells with the typical a- SMA structure were considered to be a-SMA–positive, whereas cells with faint green halo were deemed negative. Tyrosine phosphorylation Since the exact nature of the relaxin receptors has not been elucidated, we examined a possible specific cellu- lar response of human renal fibroblasts to relaxin stim- ulation. Fibroblasts were stimulated with relaxin (0.1, 1, and 10 ng/mL) for 15, 30 and 45 minutes and pro- tein was isolated as described before [23]. Briefly, fi- broblasts were lysed with a detergent based buffer and 100 lg of cellular protein were run on a 7.5% sodium do- decyl sulphate-polyacrylamide gel elctrophoresis (SDS- PAGE) gel and transferred to a nitrocellulose membrane (HybondTM ECLTM) (Amersham, Arlington Heights, IL, USA) by wet electroblotting. To prevent degradation of phosphorylated proteins the tyrosine phosphatase in- hibitor sodium vanadate (0.5 mmol/L) was added to the lysis buffer. Membranes were stained with Ponceau red to control for equal loading and transfer. After block- ing [phosphate-buffered saline (PBS) + 0.1% Tween + 5% dry milk] for 6 hours, the membrane was incubated with the primary antibody (antiphosphotyrosine anti- body 1:500) followed by the secondary antibody (antirab- bit HRP antibody) (1:3000). Positive reaction products were identified by chemiluminescence detection with en- hanced chemiluminescence (ECL)+ (Amersham). Anti- bodies were diluted in PBS containing 3% bovine serum albumin (BSA). Effects on proliferation, apoptosis, and necrosis Proliferation studies were performed using cell counts and nonradioactive bromodeoxyuridine (BrdU) incorpo- ration assays (Amersham) as described before [23]. For cell count experiments, Tk 173 and Tk 188 fibroblasts were plated at 1 × 105 cells per well in a 24-well mi- crotiter plate containing DMEM medium with 10% FCS and incubated overnight. Cells were made quiescent by incubation with FCS-free Iscove’s medium for 24 hours. Subsequently, cells were incubated with various additives for additional 72 hours. IMDM medium alone served as negative, IMDM medium containing 10% FCS as pos- itive control. After trypsination, cells were stained with trypane blue and counted using a Neubauer chamber. All experiments were repeated three times. For BrdU incorporation assay, Tk 173 and Tk 188 fi- broblasts were plated at 4 × 103 cells per well in 96-well microtiter plates containing DMEM medium with FCS and incubated overnight. Cells were subsequently made quiescent as described above. Medium containing 10% FCS was used as positive control or FCS+ relaxin. IMDM without additives served as negative control. Relaxin was added in concentrations of 0.1, 1, and 10 ng/mL. Prolif- eration was measured after 24, 48, and 72 hours. Opti- cal densities were determined photometrically at 405 nm (Dynatech MR 4000) (Denkendorf, Germany). All ex- periments were performed in triplicates and repeated five times. Apoptosis and necrosis rates were investigated in Tk 173 and Tk 188 fibroblasts after incubation with either IMDM (control), TGF-b1, or different concentrations of relaxin for 72 hours. Apoptosis was detected by immuno- cytochemical staining for caspase 3. After cytospin, cells were air dried, briefly fixed in acetone/methanol (1:1) and stained for caspase 3 using a modified APAAP method as described elsewhere [25]. Using fast red as peroxidase substrate resulted in a red signal in positively stained cells. The number of positively stained cells was counted and results are shown in percent of total cell number. Necrosis assay was performed by detection of propid- ium iodide by FACS analysis. After trypsination, cells were counted, and 5 × 105 cells per sample were in- cubated with phycoerythrin-conjugated antipropidium iodide antibody for 15 minutes, washed twice before they were counted by a FACS Calibur Flow Cytometer (Becton Dickinson). Results are given in percentage of control IMDM. Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts 99 Matrix secretion and synthesis ELISAs and immunoblots were performed for colla- gen I and fibronectin, the main components of the inter- stitial matrix. ELISAs of supernatants were performed as described before [23]. A total of 8 × 103 cells were plated per well, and cells were again made quiescent by incubation in serum-free IMDM. After 24 hours, IMDM was replaced by medium containing 1 ng/mL TGF-b1 ei- ther alone (as positive control) or in combination with relaxin. IMDM without additives served as negative con- trol. Relaxin was again added in concentrations of 0.1, 1, and 10 ng/mL. Ascorbic acid (50 lg/mL) and propi- onitrile (50 lg/mL) were added in experiments evalu- ating collagen secretion. After 48 hours of incubation, supernatants were transferred to a MicroliteTM plate and incubated overnight at 37◦C. Plates were subsequently dried for 2 hours and blocked with PBS/3% dried milk. Incubation with 50 lL/well of the primary antibody (an- ticollagen type I in a dilution of 1:300 and antifibronectin antibody at 1:5000 in PBS/0.1% BSA) for 2 hours was followed by washing with PBS/0.1% Tween and incuba- tion with the secondary antibody (antirabbit IgG AP and sheep/antigoat IgG AP, both 1:1000 in PBS/BSA). After three final wash steps, 100 lL/well Boehringer Mannheim (BM) chemiluminescence ELISA substrate (Boehringer Mannheim) was added. Chemiluminescence was then measured in a luminometer (Mikrolumat CB 96P) (Berthold, Bad Wildbad, Bermany) using MikroWinTM software (Mikrotek, Overath, Germany). Nonspecific binding was determined by incubation with the secondary antibody only. The value was normally less than 5% of the total chemiluminescence and was subtracted from each assay. All experiments were performed in triplicates and repeated three times. Standardization was obtained for the collagen type I ELISA using human type I collagen (Becton Dickinson). The limit of detection was 1 ng/mL, with a working range of 10 to 1000 ng/mL. Human fi- bronectin (Sigma Chemical Co.) served as standard for the fibronectin ELISA. The working range of hat ELISA was 0.1 to 10 lg/mL, with a limit of detection of 10 ng/mL. The influence of relaxin on the matrix synthesis was evaluated by immunoblot analyses for collagen type I and fibronectin. Cells were treated in flasks under condi- tions as described above and proteins were isolated after 48 hours. Specificity of the antibodies was determined as described [23]. Measurement of TGF-b1 receptor binding To investigate TGF-b1 receptor binding a Fluo- rokineTM Cytokine Flow Cytometry Kit was used. Exper- iments were performed according to the manufacturer’s instructions with some modifications. Cells were culti- vated as described above. After trypsination, cells were incubated in DMEM for 6 hours on a rocker platform to enable receptor regeneration. The rocker platform was used to prevent reattachment to the substrate. Then cells were washed in PBS and resuspended at 4 × 106 cells per mL in PBS or PBS + relaxin (0.1, 1, and 10 ng/mL, re- spectively). Twenty-five microliters of each solution was incubated with 12.5 ng of biotinylated TGF-b1 for 60 minutes at 4◦C. Then 10 lL avidin FITC reagent was added and incubated for 30 minutes at 4◦C. Probes were washed twice, resuspended in 200 lL of RDF1 buffer and measured at 488 nm in a FACS Calibur Flow Cytometer (Becton Dickinson). Biotinylated soy bean trypsin in- hibitor served as negative control. Specificity was con- trolled by incubation of 40 lg antihuman-TGF-b1 with 12.5 ng biotynilated TGF-b1 for 15 minutes at room tem- perature and identical treatment. Zymography Gelatine zymography was used to evaluate the effects of relaxin on gelatinase (MMP-2 and MMP-9) expres- sion. Briefly, cells were plated at a density of 8 × 104/mL and growth-arrested in Iscove’s medium for 24 hours before incubation with relaxin (0.1, 1, and 10, ng/mL) in the presence or absence of 10% FCS or neutralizing TGF-b-antibody, respectively, for 48 hours. Cell super- natants were 10× concentrated and equal amounts of total protein were loaded on SDS gel containing 1.0 mg gelatin per mL and electrophoresis on nonreducing SDS- 10% polyacrylamide gels (Bio-Rad) was performed. Af- ter 30 minutes, incubation in renaturation buffer contain- ing 2.5% Triton X-100, gelatinolytic activity was detected by an 18-hour incubation step in 5 mmol/L CaCl2, 50 mmol/L Tris/HCl, pH 7.5, at 25◦C before Coomassie bril- liant blue staining of the gels (0.5% Coomassie blue in 40% methanol/10% acetic acid for 1 hour). Destaining was performed in the same buffer devoid of Coomassie blue for 15 minutes. The presence of the MMPs was indi- cated by an unstained proteolytic zone of the substrate. Molecular sizes of the bands displaying enzymatic activity were identified by comparison to standard proteins and to purified MMPs (both from Calbiochem). Quantifica- tion was achieved by densitometric analysis as described above. Immunoblot analyses for Smad2, 3, 4, 7, phosphorylated Smad2, phosphorylated Smad3, JNK, ERK, p38, phosphorylated JNK, phosphorylated ERK, and phosphorylated p38 MAP kinase Cells were solubilized and cytosolic proteins were ex- tracted by addition of cytosolic lysis buffer [10 mmol/L Hepes, 1.5 mmol/L MgCl2, 10 mmol/L KCl, 1 mmol/L phenylmethylsulfonyl fluoride (PMSF), 0.1% NP-40, pH 7.9] for 10 minutes on ice. Solubilized proteins were iso- lated by centrifugation with 3500 U/min for 4 minutes and nuclear proteins were extracted out of the pellet by 100 Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts adding of nuclear lysis buffer (20 mmol/L Hepes, 420 mmol/L NaCl, 1.5 mmol/L MgCl2, 0.2 mmol/L EDTA, and 25% glycerol, pH 7.9) for 30 minutes on ice. Total cellular proteins were isolated as described before. Solu- bilized proteins were extracted by additional centrifuga- tion with 10.000 U/min for 10 minutes at 4◦C. The Western blot analyses for Smad2 were performed as described above. For phosphorylated Smad2 analy- ses, cells were stimulated for 60 minutes with 10 ng/mL TGF-b1 ± relaxin (0.1, 1, and 10 ng/mL). Immunoblots were performed as described above. Rabbit antiphos- phorylated Smad2 antibody was diluted 1:600 and rabbit antiphosphorylated Smad3 1:1000, secondary antirabbit- HRP antibody 1:3000. The HRP-conjugated recombinant antiphosphotyrosine antibody was diluted 1:500 in PBS containing 3% BSA. Immunoprecipitation Proteins were isolated as described above and 50 lg protein per sample were incubated with anti-Smad2 or anti-Smad3 antibody, respectively, for 1 hour, followed by immunoprecipitation with agarose A overnight at 4◦C. After five washes, the immunoprecipitates were boiled for 3 minutes in Laemmli sample buffer. The re- sulting precipitated complexes were separated on SDS- polyacrylamide gels, blotted and complex-bound Smad3 (Smad2/3 complex) or phosphorylated Smad3 were de- tected by a specific antibody and quantification was achieved by densitometric analysis as described above. Smad2 and Smad3 siRNA transfection and immunoblot Tk 173 and Tk 188 were plated on a 6-well plate in a density of 4 × 105 cells per well and grown for 48 hours. Then the medium was changed and cells were transfected with Smad2 siRNA SMART Pool (Upstate Cell Sig- nalling Solutions, Waltham, MA, USA) or Smad3 siRNA (Ambion, Austin, TX, USA) using RNAiFectTM Trans- fection Reagent (Qiagen, Valencia, CA, USA) follow- ing manufacturer’s instructions. Cells were left untreated (negative control) or stimulated with 10 ng/mL TGF- b1 in the presence or absence of 10 ng/mL relaxin for 48 hours. Cells were lysed and protein isolated as de- scribed above. Effective transfection was documented by Western blot analysis for Smad2 or Smad3, respectively. Effects of TGF-b1 on synthesis of fibronectin in the pres- ence or absence of 10 ng/mL relaxin in transfected and untransfected cells were analyzed by immunoblots as de- scribed above. Statistical analysis All values are expressed as mean ± SEM. One-way analysis of variance (ANOVA) was used to determine sta- tistical differences between growth factor treated groups C 15’ 30’ 45’ 15’ 30’ 45’ 15’ 30’ 45’ 0.1 ng/mL RLX 1 ng/mL RLX 10 ng/mL RLX 220 kD Fig. 1. Relaxin stimulates the tyrosine phosphorylation of a 220 kD protein. After stimulation of Tk 173 fibroblasts with relaxin (RLX) (0.1, 1, and 10 ng/mL) for 15, 30, and 45 minutes a tyrosine phosphorylation of a 220 kD protein was seen. This protein may represent a relaxin receptor. One representative blot is shown. and controls using Sigma-StatTM-software 2.03 (Jandel Scientific, San Rafael, CA, USA). Bonferroni’s method was used to control for multiple testing. P values < 0.05 were considered significant. RESULTS Stimulation with relaxin results in tyrosine phosphorylation in human renal fibroblasts Relaxin is an insulin-like hormone. Therefore, its re- ceptor may be, like the insulin receptor, a tyrosine kinase. Relaxin did induce tyrosine phosphorylation in all exam- ined cells. In Tk 173 fibroblasts a significant increase in tyrosine-phosphorylation was seen at relaxin dosages of 1 ng/mL (172.6 ± 16.0% of control after 30 minutes) and 10 ng/mL (151.6 ± 12.2% of control after 45 minutes). Figure 1 displays a characteristic blot. Relaxin inhibits proliferation of human fibroblasts, but has no effect on either apoptosis or necrosis The peptide hormone relaxin significantly decreased cell count to 30 ± 13% in Tk 173 or 58 ± 16% in Tk 188 of controls after 72 hours. The effects were reproducible in Tk 188 fibroblasts in the presence of 10% FCS, whereas in Tk 173 they were not (Fig. 2A). Proliferation rate of human fibroblasts were decreased after stimulation with relaxin after 24, 48, and 72 hours at all investigated con- centrations (Fig. 2B). However, these effects were not robust and a significant inhibition was only found in Tk 173 cells after 72 hours of stimulation with 0.1 ng/mL relaxin (−10.8 ± 3.7%). The maximum inhibition in Tk 188 was also seen after 72 hours of stimulation with 10 ng/mL relaxin (−22.0 ± 12.6%). There were no effects on apoptosis or necrosis rates (data not shown). Relaxin inhibits TGF-b1–induced fibroblast activation into myofibroblasts The differentiation of fibroblasts into myofibroblasts is thought to represent one of the key steps during fibro- genesis. Thus, in order to determine if relaxin influences this process in vitro, primary fibroblasts Tk 461 were in- cubated with TGF-b1 ± relaxin as described above and the number of a-SMA–positive cells was determined. Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts 101 350 300 250 200 150 100 50 0 % o f N eg at ive c on tro l Tk 173 Tk 188 NC RL X1 0 RL X1 RL X0 .1ø RL X1 0 RL X1 RL X0 .1 + 10% FCS * * * * * * § § § A Tk 188 Tk 173 B 0 20 40 60 80 100 120 % o f P os itiv e co nt ro l 0.1 RLX 1 RLX 10 RLX * ** ** ** Fig. 2. Effects of human recombinant relaxin (rhRLX) on cell count (A) and proliferation (B) of immortalized normal kidney-derived (Tk 173) and fibrotic kidney-derived (Tk 188) cortical human renal fibrob- lasts. Relaxin was added in concentrations of 0.1, 1 and 10 ng/mL for 72 hours. Cell growth was either determined by cell count (A) or bromod- oxyuridine (BrdU) incorporation (B). Relaxin showed antiproliferative effects which were more prominent in Tk 188 fibroblasts. Values are given in percentage of negative control (NC is Iscove’s medium alone) (A) or positive control [(PC is 10% fetal calf serum (FCS)] (B) and are the mean of five (Tk 173) and six (Tk 188) experiments. All experiments were performed in triplicates. P values were significant against nega- tive control (A) or positive control (B). ∗P < 0.05 vs. negative control; ∗∗P < 0.01 vs. negative control; §P < 0.05 vs. 10% FCS. Under basal conditions (IMDM/1% FCS) the number of a-SMA–positive cells was 11.5 ± 1.7%. Stimulation with 10 ng/mL TGF-b1 raised the number of a-SMA– positive cells to 33.7 ± 1.4%. Relaxin was able to reverse the TGF-b1–induced increase of a-SMA–positive cells in all investigated concentrations. The maximum inhibition was found after stimulation with 1 ng/mL relaxin (13.8 ± 1.6% a-SMA–positive cells). Figure 3 illustrates a typical staining pattern. Relaxin has robust antifibrotic effects by inhibiting the TGF-b1–induced ECM synthesis and secretion To study the interactions between relaxin and the TGF-b1–induced ECM synthesis immunoblot analyses A 0 10 20 30 40 NC PC 0.1 RLX 1 RLX 10 RLX ****** ** B % α – SM A po sit ive c e lls Fig. 3. Expression of a-smooth muscle actin (a-SMA) in primary cor- tical fibroblasts. The number of a-SMA–positive cells was determined after incubation with 10 ng/mL transforming growth factor-b1 (TGF- b1) and three different concentrations of relaxin (RLX) (0.1, 1, and 10 ng/mL) for 48 hours. Single stimulation with TGF-b1 served as posi- tive control (PC), Iscove’s medium as negative control (NC). (A) gives a representative example of the positive control, (B) summarizes the findings graphically. Relaxin reduced a-SMA expression at all investi- gated concentrations significantly. ∗∗P < 0.01 vs. positive control. for type I collagen and fibronectin were performed. The ECM secretion was investigated by ELISAs for type I col- lagen and fibronectin. The TGF-b1–induced ECM syn- thesis was strongly inhibited by relaxin in all investigated cell lines. In Tk 173 the maximum inhibition for collagen I synthesis was found at a dosage of 0.1 ng/mL relaxin (−52.7 ± 6.7%), in Tk 188 at a dosage of 1 ng/mL re- laxin (−25.2 ± 8.4%) (Fig. 4). The fibronectin-synthesis was highly significant inhibited at all investigated relaxin doses as well as in all investigated cell lines [Tk 173 and Tk 188 (Fig. 5A) and in primary kidney fibroblasts Tk 489 and Tk 461 (data not shown)]. The maximum inhibition of fibronectin synthesis was seen at dosages of 1 ng/mL re- laxin for Tk 461(−34.4 ± 5.6%) and 10 ng/mL relaxin for Tk 173 (−55.0 ± 4.6%), Tk 188 (−51.3 ± 2.4%), and Tk 489 (−40.5 ± 5.7%). However, relaxin alone had no ef- fect on constitutive fibronectin synthesis in neither FCS- starved nor in unstarved Tk 173 and Tk 188 fibroblasts (Fig. 5B). The TGF-b1–induced secretion of collagen type I was inhibited by relaxin in both cell lines (Fig. 6A and B). In the supernatants of the positive control (1 ng/mL 102 Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts NC PC 0.1 1 10 RLX, ng/mL 140 kD 97 kD Tk 173 NC PC 0.1 1 10 RLX, ng/mL 140 kD 97 kD Tk 188 0 20 40 60 80 100 120 140 % o f P os itiv e co nt ro l PC NC 0.1 RLX 1 RLX 10 RLX ** ** ** * * Tk 173 Tk 188 Fig. 4. Inhibition of collagen type I synthesis in Tk 173 (dark shad- ing) and Tk 188 cells (white shading). The effect of relaxin (RLX) on collagen type I synthesis (alpha 1 and alpha 2 chain) was examined by immunoblot analyses of cellular protein after incubation with 1 ng/mL transforming growth factor-b1 (TGF-b1) and three different concen- trations of relaxin (0.1, 1, and 10 ng/mL) for 48 hours. Single stimulation with TGF-b1 served as positive control (PC), Iscove’s medium as neg- ative control (NC). Values are given in percentage of positive control and are the mean of four independent experiments. At this figure also one representative blot is shown. ∗P < 0.05; ∗∗P < 0.01. TGF-b1) 649.9 ± 119.2 ng/mL (Tk 173) and 2072.2 ± 627.8 ng/mL (Tk 188) collagen I were measured. The most visible reduction of secreted collagen I was found at a dosage of 1 ng/mL relaxin (298.5 ± 84.3 ng/mL for Tk 173 and 651.3 ± 310.4 ng/mL for Tk 188). Likewise, the secretion of fibronectin was inhibited by relaxin (Fig. 6C and D). In Tk 173 the amount of fibronectin in the super- natants of the positive control was 1039.6 ± 375.7 ng/mL, in Tk 188 it was 659.59 ± 59.0 ng/mL. The strongest reduc- tion of the fibronectin concentration was found at dosages of 0.1 (Tk 173 460.9 ± 128.6 ng/mL) and 10 ng/mL RLX (Tk 188 372.5 ± 39.8 ng/mL). Relaxin stimulates secretion of MMP-2 and MMP-9 in renal fibroblasts The effects of relaxin on matrix-degrading enzymes were studied by gel zymography analyses. Relaxin dis- played robust antifibrotic effects by inducing the secre- tion of MMP-2 and MMP-9 in Tk 173 and Tk 188 fibrob- lasts (Fig. 7). In both cell lines, MMP-2 secretion was induced highly significantly at all investigated relaxin doses. The most robust increase was seen after stimula- tion with 10 ng/mL relaxin (131.1 ± 5.7% of control in Tk 173 and 167.0 ± 15.0% of control in Tk 188). Moreover, in Tk 188 incubation with 1 ng/mL relaxin induced a sig- nificant induction of the MMP-9 secretion 191.7 ± 22.1% NC PC 0.1 1 10 RLX, ng/mL 220 kD Tk 173 NC PC 0.1 1 10 RLX, ng/mL 220 kDTk 188 0 20 40 60 80 100 % o f P o si tiv e c o n tro l PC NC 0.1 RLX 1 RLX 10 RLX ** ** * ** ** ** ** ** Tk 173 Tk 188 NC RL X 0 .1 RL X 1 RL X 1 0 NC a- TG Fβ RL X 0 .1 RL X 1 RL X 1 0 + 10% FCS 220 kD B A Fig. 5. Inhibition of relaxin (RLX) to transforming growth factor-b1 (TGF-b1)–induced fibronectin synthesis in Tk 173 and Tk 188 human renal fibroblasts. The effect of relaxin on fibronectin synthesis was ex- amined by immunoblot analyses of cellular protein after incubation with 10 ng/mL TGF-b1 and three different concentrations of relaxin (0.1, 1, and 10 ng/mL) for 48 hours. Single stimulation with TGF-b1 (10 ng/mL) served as positive control (PC), Iscove’s medium as neg- ative control (NC). Relaxin had robust inhibitory effects on TGF-b1 fibronectin synthesis in both, Tk 173 (dark shading) and Tk 188 (white shading). Values are given in percentage of positive control and are the mean of nine (Tk 173) and seven (Tk 188) independent experiments (A). Relaxin alone had no effects on fibronectin synthesis neither in fetal calf serum (FCS)-starved nor in unstarved Tk 173 or Tk 188 fi- broblasts (B). ∗P < 0.05; ∗∗P < 0.01. of control. In Tk 173 the strongest increase of MMP-9 se- cretion was demonstrated by stimulation with 10 ng/mL relaxin (129.7 ± 20.8% of control); however, this effect was not significant. Additionally, relaxin induced MMP-2 and MMP-9 secretion was determined in unstarved fibroblasts un- der equal conditions. Relaxin exerted a significant up- regulation of both MMP-9 (+68 ± 5.1% vs. control in Tk 173 and +55.5 ± 5.0% vs. control in Tk 188 fibroblasts) (P < 0.05) and MMP-2-activity (+107 ± 8.6 vs. control in Tk 188) (P < 0.05) after stimulation with 10 ng/mL relaxin (data not shown). To prove the expression of MMP-1 immunoblot, casein gel zymography, and ELISA analyses were performed. MT-MMP-1 expression was examined by immunoblots. However, both MMP-1and MT-MMP-1 expression could Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts 103 NC PC 0.1 RL X 1 R LX 10 RL X 0 200 400 600 800 Fi br on ec tin , n g/ m L D NC PC 0.1 RL X 1 R LX 10 RL X 0 200 400 600 800 1600 1000 1200 1400 Fi br on ec tin , n g/ m L C 0 200 400 600 800 1000 NC PC 0.1 RL X 1 R LX 10 RL X Co lla ge n ty pe I, n g/ m L A NC PC 0.1 RL X 1 R LX 10 RL X 0 500 1000 1500 2000 2500 3000 Co lla ge n ty pe I, n g/ m L B ** ** ** ** * Fig. 6. Influence of relaxin (RLX) on colla- gen type I and fibronectin secretion of human renal fibroblasts. The effect of relaxin trans- forming growth factor-b1 (TGF-b1)–induced extracellular matix (ECM) secretion was evaluated by enzyme-linked immunosorbent assay (ELISAs) for collagen type I and fi- bronectin of cell supernatants. Experiments were performed after stimulation with 10 ng/mL TGF-b1 and three different concentra- tions of relaxin (0.1, 1, and 10 ng/mL) for 48 hours. Single stimulation with TGF-b1 served as positive control (PC), Iscove’s medium as negative control (NC). Relaxin had legible in- hibitory effects on collagen type I secretion of Tk 173 (A) and Tk 188 (B) as well as on fi- bronectin secretion of Tk 173 (C) and Tk 188 (D). Values are the mean of six independent experiments. ∗P < 0.05; ∗∗P < 0.01. not be demonstrated in the human renal fibroblast cell lines Tk 173 and Tk 188 (data not shown). Relaxin does not interact with TGF-b1 action by affecting its receptor binding Possible effects of relaxin on TGF-b1 receptor binding were examined by flow cytometric analyses. Relaxin does not affect TGF-b1 receptor binding in any investigated relaxin dose (data not shown). In Tk 173 and Tk 188 fibroblasts, relaxin effects the TGF- b1 signal transduction via Smad-proteins by inhibiting Smad2 but not Smad3 pathway In order to examine the mechanisms of relaxin action, the effects of relaxin on different TGF-b1 signal transduc- tion pathways was investigated. Therefore immunoblot analyses for common Smad4 and inhibitory Smad7 and, moreover, for phosphorylated Smad2, translocation of Smad2 and 3 from cytosol to nucleus, and, for heteromeric formation between Smad2 and 3 after immunoprecipita- tion were performed. In addition, effects of relaxin on MAP kinase pathway were investigated by immunoblots for the unphosphorylated and phosphorylated forms of the MAP kinases JNK1/2, ERK1/2, and p38. Smad2 phosphorylation was inhibited by relaxin at all investigated relaxin doses (Fig. 8A). The strongest inhi- bition was seen at a concentration of 1 ng/mL relaxin (−24.3 ± 10.7% vs. positive control). In contrast, re- laxin had no effects on Smad3 phosphorylation (Fig. 8B). Since the antiphosphorylation Smad3 antibody cross- reacts with pSmad1, 5, and 8, we precipitated proteins prior to immunoblot analysis with a specific antibody to Smad3. Additionally, relaxin significantly inhibited translocation of Smad2 from cytosol to nucleus, with a maximal inhibition at a dosage of 10 ng/mL relaxin (−47.3 ± 7.3% vs. positive control in Tk 173 and −61.0 ± 9.2% vs. positive control in Tk 188 fibroblasts) (Fig. 9A). Since phosphorylated Smad2 forms a heteromeric com- plex with Smad3 and 4, we investigated whether this com- plex formation was inhibited by relaxin. Consecutively, reduction of Smad2 phosphorylation was associated with a reduction of active complex bound Smad3 as it has been detected by immunoblot after immunoprecipitation for Smad2 in Tk 188 cells (Fig. 9B). The strongest inhibition was seen at the dosage of 0.1 ng/mL relaxin (−26.3 ± 14.1% vs. positive control). Conversely, relaxin did have no effect on Smad3 translocation (data not shown) and intracellular Smad7 concentrations (Fig. 9C). To give further evidence that the effects of relaxin are mediated by Smad2, and not Smad3 inhibition, we com- pared its effects with those of inhibitory RNA to Smad2 or Smad3. Successful transfection was demonstrated by a reduction of total Smad2 or Smad3, respectively, of more than 80% compared to controls (Fig. 10A). Fibronectin synthesis was chosen as an easy read-out system. Both, relaxin (10 ng/mL) and siRNA to Smad2, significantly blocked TGF-b1–induced fibronectin up-regulation by −66.33 ± 18.0% using relaxin (P < 0.05 vs. positive con- trol) and –80.67 ± 21.1% using siRNA to Smad2 (P < 0.05 vs. positive control), but not siRNA to Smad3 (Fig. 10B and C). Representative blots and results are summarized in Figure 10. Conversely, unphosphorylated and phosphorylated forms of MAP kinases JNK1/2, 104 Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts 0 50 100 150 200 % o f C on tro l PC 0.1 RLX 1 RLX 10 RLX * ** ** ** ** **Tk 173 Tk 188 A 0 50 100 150 300 200 250 % o f C on tro l PC 0.1 RLX 1 RLX 10 RLX * Tk 173 Tk 188 B Co ntr ol 0.1 1 10 RLX, ng/mL MMP-9 MMP 2 Tk 173 Co ntr ol 0.1 1 10 RLX, ng/mL MMP-9 MMP 2 Tk 188 Fig. 7. Effect of relaxin (RLX) on secretion of matrix metallopro- teinase 2 (MMP-2) and MMP-9. The effect of relaxin on MMP secre- tion was evaluated by gel zymography of cell supernatants. Experiments were performed after incubation with three different concentrations of relaxin (0.1, 1, and 10 ng/mL) for 48 hours. Iscove’s medium served as control. Relaxin had significant stimulatory effects on MMP-2 secretion (A) in both, Tk 173 (dark shading) and 188 (white shading). MMP-9 se- cretion was also stimulated by relaxin (B), where the effect on Tk 188 was more robust than in Tk 173. Values are given in percentage of con- trol and are the mean of six (Tk 173) and three (Tk 188) independent experiments. Also one representative zymography for each cell line is shown. ∗P < 0.05; ∗∗P < 0.01. ERK, and p38 MAP kinase were not affected by stimu- lation with relaxin in the presence or absence of TGF-b1 (Fig. 11). DISCUSSION Tubulointerstitial fibrosis is the final common path- way of almost all forms of chronic progressive renal disease [2]. An effective antifibrotic therapy should in- hibit all parts of renal fibrogenesis, including fibroblast proliferation, differentiation into myofibroblasts, ECM synthesis, and matrix degradation. The peptide hormone relaxin has already shown antifibrotic properties on der- mal, cardiac, and pulmonary fibroblasts [15, 16]. How- 0 20 40 60 80 100 120 % o f P o si tiv e c o n tro l PC NC 0.1 RLX 1 RLX 10 RLX ** * * NC PC 0.1 1 10 RLX, ng/mL p-Smad2 A IP smad3 p-Smad3 N C TG Fβ 10 n g/ m L R LX 1 0 n g/ m L R LX 0 .1 n g/ m L R LX 1 n g/ m L R LX 1 0 n g/ m L +TGFβ 10 ng/mL B Fig. 8. Influence of relaxin (RLX) on transforming growth factor-b1 (TGF-b1)–induced Smad2 (A) and Smad3 phosphorylation (B). Hu- man renal fibroblasts were incubated with 10 ng/mL TGF-b1 and three different concentrations of relaxin (0.1, 1, and 10 ng/mL) for 60 min- utes. Single stimulation with TGF-b1 served as positive control (PC), Iscove’s medium as negative control (NC). In Tk 461 primary human renal fibroblasts, relaxin inhibited Smad2 phosphorylation at all inves- tigated concentrations (A), but there were no effects on Smad3 phos- phorylation. Since the antibody against pSmad3 cross-reacts with phos- phorylated Smad1, 5, and 8, proteins were immunoprecipitated with a Smad3-specific antibody prior to immunoblot analyses (IP Smad3). Values are given in percentage of positive control and are the mean of four independent experiments. Also one representative blot is shown. ∗P < 0.05 vs. negative control; ∗∗P < 0.05 vs. positive control. ever, fibroblasts from different origins may display sub- stantial heterogeneity [26]. Our results show that relaxin may inhibit fibroblast pro- liferation (albeit moderately) and the TGF-b1–induced transformation of these cells to myofibroblasts. More- over, robust inhibitory effects of relaxin on TGF-b1– induced collagen type I and fibronectin synthesis and se- cretion could be demonstrated in human lung fibroblasts [16], rat cardiac [27], and very recently in rat renal cortical fibroblasts [28]. Those results implicate strong antifibrotic properties. It has been demonstrated, that fibronectin is able to activate latent TGF-b [29]. Thus, an inhibition of fibronectin synthesis and secretion does inhibit fibrosis Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts 105 c Smad2 n Smad2 β-actin N C TG Fβ 10 n g/ m L R LX 1 0 n g/ m L R LX 0 .1 n g/ m L R LX 1 n g/ m L R LX 1 0 n g/ m L +TGFβ 10 ng/mL A NC TGFβ 10 ng/mL RLX 10 ng/mL RLX 0.1 ng/mL RLX 1 ng/mL RLX 10 ng/mL +TGF-β 10 ng/mL Tk 173 Tk 188 0 20 40 60 80 100 120 140 160 180 Cy to so lic s m ad 2- ex pr es sio n, % o f N C * * ** ** ** ** NC TGFβ 10 ng/mL RLX 10 ng/mL RLX 0.1 ng/mL RLX 1 ng/mL RLX 10 ng/mL +TGF-β 10 ng/mL Tk 173 Tk 188 * * ** ** ** 0 50 100 150 200 250 N uc le ar S m ad 2 ex pr es sio n, % o f N C Smad2 Smad3 Smad3 N C TG Fβ 10 n g/ m L R LX 1 0 n g/ m L R LX 0 .1 n g/ m L R LX 1 n g/ m L R LX 1 0 n g/ m L +TGFβ 10 ng/mL B IP Smad2 NC TGFβ 10 ng/mL RLX 10 ng/mL RLX 0.1 ng/mL RLX 1 ng/mL RLX 10 ng/mL +TGFβ 10 ng/mL Tk 173 Tk 188 * * ** ** ** ** ** 0 50 100 150 200 250 Sm ad 3 ex pr es sio n, % o f c on tro l IP Smad2 Smad-7 β-actin N C TG Fβ 10 n g/ m L R LX 1 0 n g/ m L R LX 0 .1 n g/ m L R LX 1 n g/ m L R LX 1 0 n g/ m L +TGFβ 10 ng/mL C Fig. 9. Influence of relaxin (RLX) on transforming growth factor-b1 (TGF-b1)–induced Smad2 translocation (A), heteromer formation between Smad2 and 3 (B), and expression of Smad7 (C). Human renal fibroblasts (Tk 173 and 188) were incubated with 10 ng/mL TGF-b1 and three different concentrations of relaxin (0.1, 1 and 10 ng/mL) for 60 minutes. Single stimulation with TGF-b1 served as positive control (PC), Iscove’s medium as negative control (NC). Cytosolic and nuclear proteins were isolated and cytosolic and nuclear Smad2 was detected by immunoblot analysis. In Tk 188 relaxin inhibited shift of Smad2 from cytosol to nucleus (A). In (B) isolated proteins were immunoprecipitated for Smad2 and immune complex bound Smad3 was detected by immunoblot after sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). In Tk 188 human renal fibroblasts, relaxin reduced complex formation between Smad2 and 3 at all concentrations. Total cellular Smad2 and 3 expressions were taken as internal controls. Values are given in percentage of negative control and are the mean of four independent experiments. Also one representative blot is shown. ∗P < 0.05 vs. negative control; ∗∗P < 0.05 vs. positive control. There were no effects of relaxin on inhibitory Smad7 expression detectable (C). by two mechanisms. First, it decreases ECM deposition and second, it results in reduced matrix expression by di- minishing TGF-b activity. However, those effects could mainly prevent the development of renal fibrosis but they are not useful to improve organ function by degrading ECM depositions. Therefore we investigated the influ- ence of relaxin on matrix degrading MMPs. Relaxin was able to raise the secretion of MMP-2 and MMP-9 robustly. 106 Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts Smad3 β-actin NC siRNA Smad3 Smad2 β-actin NC siRNA Smad2 A β-actin FN − + − − + + − − + − + − − TGF-β (10 ng/mL): RLX (10 ng/mL): siRNA Smad2: − − + − + B β-actin FN − + − − + + − − + − + − − TGF-β (10 ng/mL): RLX (10 ng/mL): siRNA Smad3: − − + − + − + − + + + − − − + − − − TGF-β (10 ng/mL): RLX (10 ng/mL): siRNA Smad2: − − + − − − + − − + − −siRNA Smad3: − − − + − − + 0 50 100 150 200 250 Fi br on ec tin e xp re ss io n, % o f n eg at ive c o n tro l * ** ** C Fig. 10. Relaxin (RLX) as well as siRNA to Smad2, but not siRNA to Smad3 exert comparable inhibitory effects to fibronectin synthesis in Tk 188 renal fibroblasts. (A) Effective transfection with inhibitory RNA to Smad2 and Smad3 (siRNA) was demonstrated by a decreased Smad2 or Smad3 expression, respectively, after 48 hours of incubation with siRNA. (B) Effects of relaxin in comparison to siRNA to Smad2 or Smad3, respectively, on transforming growth factor-b1 (TGF-b1)–induced fibronectin synthesis was demonstrated. Values are given in percentage of negative control and are the mean of three independent experiments (C). ∗P < 0.05 vs. negative control; ∗∗P < 0.05 vs. positive control. In murine pulmonary fibroblasts an increase of MMP-1 and MMP-2 has been reported before [16]. Surprisingly, MMP-1 and MT-MMP-1 protein expression could not be demonstrated in any of the investigated human renal fi- broblasts indicating that they do not seem to play a role in these cells. All those results suggest powerful antifibrotic proper- ties of relaxin also in vivo. Relaxin showed robust antifi- brotic effects in in vivo models of pulmonary [16] and renal fibrosis [19] as well as well as in a clinical trial of patients with progressive systemic sclerosis [21]. In the treatment of chronic renal failure relaxin might act due to an additional effect. Danielson, Sherwood, and Con- rad [14] were able to demonstrate that relaxin attenuates Ang II–induced renal vasoconstriction. Consequently, re- cently Garber et al [29] noted beneficial effects of relaxin in two models of renal mass reduction. Interestingly, these effects were in part dependent of its effects on blood pres- sure reduction and in part independent of vasodilation. McDonald et al [20] recently demonstrated that relaxin may increase ubiquitin-dependent fibronectin degrada- tion. In that study, decreased glomerular and interstitial Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts 107 pErk-2 pErk-1 Erk-2 Erk-1 β-actin N C TG Fβ 10 n g/ m L R LX 1 0 n g/ m L R LX 0 .1 n g/ m L R LX 1 n g/ m L R LX 1 0 n g/ m L +TGFβ 10 ng/mL A pJNK-1 JNK-1 β-actin N C TG Fβ 10 n g/ m L R LX 1 0 n g/ m L R LX 0 .1 n g/ m L R LX 1 n g/ m L R LX 1 0 n g/ m L +TGFβ 10 ng/mL C pp38MAPK p38MAPK β-actin N C TG Fβ 10 n g/ m L R LX 1 0 n g/ m L R LX 0 .1 n g/ m L R LX 1 n g/ m L R LX 1 0 n g/ m L +TGFβ 10 ng/mL B Fig. 11. Relaxin (RLX) has no influence on transforming growth factor-b1 (TGF-b1)–induced phosphorylation of mitogen-activated pro- tein (MAP) kinases, extracellular-regulated protein kinase (ERK1/2) (A), p38 MAP kinase (B), and c-Jun NH2 terminal kinase (JNK1/2) (C). Human renal fibroblasts were incubated with 10 ng/mL TGF-b1 and three different concentrations of relaxin (0.1, 1, and 10 ng/mL) for 60 minutes. Single stimulation with TGF-b1 served as positive control, Iscove’s medium as negative control (NC). There were no detectable effects of relaxin on TGF-b1–induced phosphorylation of MAP kinases. scarring was noted in a rat model of antiglomerular base- ment membrane disease when the animals were treated with relaxin. Moreover, Samuel et al [18] were able to demonstrate that relaxin−/− mice displayed progression in age-related fibrosis (albeit only in male animals) and that this process could be reversed by addition of relaxin, providing additional evidence for an antifibrotic effect of the hormone. Interestingly, some of the effects of relaxin we observed were not purely dose-dependent. In most cases the most robust effects were seen at dosages of 0.1 and 1 ng/mL relaxin, the range of physiologic serum concentrations in human pregnancies. These findings correspond to those of Danielson, Sherwood, and Conrad [14] in a clinical trial of relaxin in scleroderma patients and some of the findings by McDonald et al [20]. It is tempting to spec- ulate that they may be explainable by a receptor down- regulation or by a relaxin molecule aggregation at higher relaxin concentrations. As described above, it is not much known about how relaxin exerts its mechanisms since the exact nature of the relaxin-receptor is still unknown. Recently, re- ceptor binding and activation of orphan receptors by relaxin has been demonstrated [30] although the signif- icance of this finding for the effects of relaxin remains to be determined. Moreover, relaxin may bind to gluco- corticoid receptors, as was recently shown by Dschiet- zig and Stangl [31]. In view of these facts, we indirectly demonstrated specific cellular responses of renal fibrob- lasts. Considering that relaxin is an insulin-like hormone and that the insulin receptor is a tyrosine kinase, we in- vestigated cellular tyrosine phosphorylation in depen- dence of relaxin stimulation. In fact, we could demon- strate a relaxin-induced tyrosine phosphorylation of a >220 kD protein. Based on these results, we postulate that the phosphorylated >220 kD protein may repre- sent a relaxin receptor. In a next step we investigated the influence of relaxin on TGF-b1 receptor interaction and signal transduction pathways. In a first set of ex- periments we could demonstrate, that relaxin does not interfere with TGF-b1 receptor binding. The central sig- nal transduction pathway of TGF-b molecules is medi- ated by Smad proteins. There exists a common mediator Smad, Smad4. Besides Smad4, the most important Smad proteins for the TGF-b1 signal transduction are Smad2 and 3 as well as the inhibitory Smad7. The cytosolic re- ceptor Smad2 and 3 are activated by type I TGF-b re- ceptor and become phosphorylated. Subsequently, they form heteromeric complexes either as Smad2/Smad4, Smad3/Smad4, or as Smad2/Smad3/Smad4. These re- ceptor Smad/costimulatory Smad heteromeric complex translocates to the nucleus where they bind to DNA. Furthermore the TGF-b1 signaltransduction is mod- ulated by MAP kinase pathways via p38, JNK, and ERK. Consequently, the influence of relaxin on all those 108 Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts proteins was investigated. Relaxin had no influence on cellular amounts of Smad4 and Smad7 as well as on phos- phorylated or unphosphorylated forms of MAP kinases JNK1/2, ERK, or p38 MAP kinase. However, relaxin had distinct inhibitory effects on TGF-b1–induced Smad2 phosphorylation resulting in a reduction of transloca- tion of Smad2, but neither on Smad3 phosphorylation nor translocation of Smad3 from cytosol to the nucleus and, moreover, a reduction of heteromeric formation be- tween Smad2 and Smad3 as could be demonstrated by the results of the immunoprecipitation studies. Those re- sults suggest that relaxin affects TGF-b1 signal transduc- tion through an inhibition of receptor regulated Smad2, whereas Smad3, Smad4, or inhibitory Smads were un- affected. Further proof for the specificity of this effect is provided by the fact that inhibition of Smad2 with siRNA resulted in similar results, whereas siRNA to Smad3 did have no noticeable effect on TGF-b1–induced fibronectin synthesis. There is some controversy in the literature regarding the mediation of TGF-b1–mediated fibronectin synthesis and secretion. Whereas our results seem to confirm findings by Isono et al [32] in mesangial cells, others have described that fibronectin synthesis may be independent of the Smad pathway in mouse embryo fibroblasts [33] and a rat fibroblast cell line [34]. Possibly, species differences may explain the differences observed. CONCLUSION We conclude that relaxin may represent a potent ther- apeutic agent to reverse renal fibrosis and remodel renal function. Those properties are at least in part mediated by an attenuation of the profibrotic TGF-b1 effects. Fur- ther studies are necessary to delineate the exact nature of the interaction between relaxin and TGF-b1 signal trans- duction pathways. ACKNOWLEDGMENTS M. Heeg is the awardee of a graduate grant from the Deutsche Forschungsgemeinschaft (DFG). This work was supported in part by the EU project “Common molecular pathways in progression of kidney disease” (QLG1-2002–01215) and by the “Forschungsfo¨rderungsprogramm” from the Georg-August-University to M. Koziolek. The authors wish to acknowledge E. Unemori/ Connectics Corp. for making available rh relaxin. Parts of this work were presented in abstract form at the National Meeting of the Amer- ican Society of Nephrology in November 1999 in Miami, FL, USA (J Am Soc Nephrol 10:581A). The authors wish to thank Mrs. S. Woock for excellent technical assistance. Reprint requests to Frank Strutz, M.D., Department of Nephrology and Rheumatology, Georg-August University Go¨ttingen, Robert-Koch- Str., 40 37075 Go¨ttingen, Germany. E-mail: fstrutz@gwdg.de REFERENCES 1. EDDY AA: Molecular basis of renal fibrosis. Pediatr Nephrol 15:290– 301, 2000 2. NATH KA: The tubulointerstitium in progressive renal disease. Kid- ney Int 54:992–994, 1998 3. BORDER WA, NOBLE NA: TGF-beta in kidney fibrosis: A target for gene therapy. Kidney Int 51:1388–1396, 1997 4. YAMAMOTO T, NOBLE NA, MILLER DE, et al: Sustained expression of TGF-b1 underlies development of progressive kidney fibrosis. Kidney Int 45:916–927, 1994 5. ZIYADEH FN, HOFFMAN BB, HAN DC, et al: Long-term prevention of renal insufficiency, excess matrix gene expression, and glomerular mesangial matrix expansion by treatment with monoclonal anti- transforming growth factor-beta antibody in db/db diabetic mice. Proc Natl Acad Sci USA 97:8015–8020, 2000 6. EDDY A: Protein restriction reduces transforming growth factor b and interstitial fibrosis in chronic purine aminonucleoside nephro- sis. Am J Physiol 266:F884–F893, 1994 7. KANETO H, MORRISSEY J, KLAHR S: Increased expression of TGF- b1 mRNA in the obstructed kidney of rats with unilateral ureteral ligation. Kidney Int 44:313–321, 1993 8. MASSAGUE J: TGF-beta signal transduction. Annu Rev Biochem 67:753–791, 1998 9. INOKI K, HANEDA M, ISHIDA T, et al: Role of mitogen-activated pro- tein kinases as downstream effectors of transforming growth factor- beta in mesangial cells. Kidney Int (Suppl 77):S76–S80, 2000 10. LIU Y: Hepatocyte growth factor and the kidney. Curr Opin Nephrol Hypertens 11:23–30, 2002 11. ZEISBERG M, HANAI J, SUGIMOTO H, et al: BMP-7 counteracts TGF- beta1-induced epithelial-to-mesenchymal transition and reverses chronic renal injury. Nat Med 9:964–968, 2003 12. BANI D: Relaxin: A pleiotropic hormone. Gen Pharmacol 28:13–22, 1997 13. EDDIE LW, BELL RJ, LESTER A, et al: Radioimmunoassay of relaxin in pregnancy with an analogue of human relaxin. Lancet 1:1344– 1346, 1986 14. DANIELSON LA, SHERWOOD OD, CONRAD KP: Relaxin is a potent renal vasodilator in conscious rats. J Clin Invest 103:525–533, 1999 15. UNEMORI EN, BECK LS, LEE WP, et al: Human relaxin decreases collagen accumulation in vivo in two rodent models of fibrosis. J Invest Dermatol 101:280–285, 1993 16. UNEMORI EN, PICKFORD LB, SALLES AL, et al: Relaxin induces an extracellular matrix-degrading phenotype in human lung fibroblasts in vitro and inhibits lung fibrosis in a murine model in vivo. J Clin Invest 98:2739–2745, 1996 17. BENNETT RG, KHARBANDA KK, TUMA DJ: Inhibition of markers of hepatic stellate cell activation by the hormone relaxin. Biochem Pharmacol 66:867–874, 2003 18. SAMUEL CS, ZHAO C, BOND CP, et al: Relaxin-1-deficient mice develop an age-related progression of renal fibrosis. Kidney Int 65:2054–2064, 2004 19. GARBER SL, MIROCHNIK Y, BRECKLIN CS, et al: Relaxin decreases renal interstitial fibrosis and slows progression of renal disease. Kid- ney Int 59:876–882, 2001 20. MCDONALD GA, SARKAR P, RENNKE H, et al: Relaxin increases ubiquitin-dependent degradation of fibronectin in vitro and amelio- rates renal fibrosis in vivo. Am J Physiol Renal Physiol 285:F59–F67, 2003 21. SEIBOLD JR, KORN JH, SIMMS R, et al: Recombinant human re- laxin in the treatment of scleroderma. A randomized, double- blind, placebo-controlled trial. Ann Intern Med 132:871–879, 2000 22. MU¨LLER GA, FRANK J, RODEMANN HP, et al: Human renal fibroblast cell lines (tFKIF and tNKF) are new tools to investigate patho- physiologic mechanisms of renal interstitial fibrosis. Exp Nephrol 3:127–133, 1995 23. STRUTZ F, ZEISBERG M, HEMMERLEIN B, et al: Basic fibroblast growth factor (FGF-2) expression is increased in human renal fibrogene- sis and may mediate autocrine fibroblast proliferation. Kidney Int 57:1521–1538, 2000 24. STRUTZ F, ZEISBERG M, RENZIEHAUSEN A, et al: Transforming growth factor (TGF)-b1 induces proliferation in human renal fibroblasts via induction of basic fibroblast growth factor (FGF-2). Kidney Int 59:579–592, 2001 25. THELEN P, BURFEIND P, GRZMIL M, et al: cDNA microarray analy- sis with amplified RNA after isolation of intact cellular RNA from Heeg and Koziolek et al: Antifibrotic effects of relaxin in human renal fibroblasts 109 neoplastic and non-neoplastic prostate tissue separated by laser mi- crodissections. Int J Oncol 24:1085–1092, 2004 26. MU¨LLER GA, STRUTZ F: Renal fibroblast heterogeneity. Kidney Int 48 (Suppl 50):S33–S36, 1995 27. SAMUEL CS, UNEMORI EN, MOOKERJEE I, et al: Relaxin modu- lates cardiac fibroblast proliferation, differentiation, and collagen production and reverses cardiac fibrosis in vivo. Endocrinology 145:4125–4133, 2004 28. MASTERSON R, HEWITSON TD, KELYNACK K, et al: Relaxin down- regulates renal fibroblast function and promotes matrix remodelling in vitro. Nephrol Dial Transplant 19:544–552, 2004 29. GARBER SL, MIROCHNIK Y, BRECKLIN C, et al: Effect of relaxin in two models of renal mass reduction. Am J Nephrol 23:8–12, 2003 30. HSU SY, NAKABAYASHI K, NISHI S, et al: Activation of orphan recep- tors by the hormone relaxin. Science 295:671–674, 2002 31. DSCHIETZIG T, STANGL K: Relaxin: A pregnancy hormone as central player of body fluid and circulation homeostasis. Cell Mol Life Sci 60:688–700, 2003 32. ISONO M, CHEN S, HONG SW, et al: Smad pathway is activated in the diabetic mouse kidney and Smad3 mediates TGF-beta-induced fibronectin in mesangial cells. Biochem Biophys Res Commun 296:1356–1365, 2002 33. PIEK E, JU WJ, HEYER J, et al: Functional characteriza- tion of transforming growth factor beta signaling in Smad2- and Smad3-deficient fibroblasts. J Biol Chem 276:19945–19953, 2001 34. SUZUKI H, UCHIDA K, NITTA K, et al: Role of mitogen-activated pro- tein kinase in the regulation of transforming growth factor-beta- induced fibronectin accumulation in cultured renal interstitial fi- broblasts. Clin Exp Nephrol 8:188–195, 2004

Original Article

Abstract

Extracted data

Original Article

Abstract

Extracted data

Related items

Related items