Signalling and regulation of collagen I synthesis by ET-1 and TGF-b1 Angelika Horstmeyer 1 , Christoph Licht 2 , Gabriele Scherr 1 , Beate Eckes 1 and Thomas Krieg 1 1 Department of Dermatology and 2 Children’s Hospital, University of Cologne, Cologne, Germany Connective tissue homeostasis requires the tightly coordinated regulation of synthesis and degradation of extracellular matrix constituents (ECM). This involves a number of different factors including signalling of cytokines in an auto- and paracrine manner, e.g. trans- forming growth factor-b1 (TGF-b1), platelet-derived growth factor (PDGF), basic fibroblast growth factor (bFGF) and tumour necrosis factor-a (TNF-a). Collagen I is the major fibrillar collagen in the der- mal connective tissue and comprises approximately 80% of the total amount of the collagens synthesized by fibroblasts. Dysregulated collagen I synthesis at any step leads to diseases such as scurvy or fibrosis of skin and internal organs. TGF-b1 transmits its biological effects via serine- threonine receptors and Smad-proteins [1]. Some of Keywords Connective tissue growth factor; dermal fibroblast; endothelin receptor subtypes; fibrosis; phospholipase Correspondence A. Horstmeyer, Department of Dermatology, University of Cologne, Joseph-Stelzmann- Str. 9, D-50931 Cologne, Germany Fax: (+)49 221 478 5949 Tel: (+)49 221 478 6152 E-mail: ahorstmey@gmx.de (Received 29 May 2005, revised 7 October 2005, accepted 13 October 2005) doi:10.1111/j.1742-4658.2005.05016.x Endothelin-1 (ET-1) plays an important role in tissue remodelling and fibrogenesis by inducing synthesis of collagen I via protein kinase C (PKC). ET-1 signals are transduced by two receptor subtypes, the ETA- and ETB-receptors which activate different Ga proteins. Here, we investigated the expression of both ET-receptor subtypes in human primary dermal fibroblasts and demonstrated that the ETA-receptor is the major ET-receptor subtype expressed. To determine further signalling intermedi- ates, we inhibited Gai and three phospholipases. Pharmacologic inhibition of Gai, phosphatidylcholine-phospholipase C (PC-PLC) and phospholipase D (PLD), but not of phospholipase Cb, abolished the increase in collagen I by ET-1. Inhibition of all phospholipases revealed similar effects on TGF-b1 induced collagen I synthesis, demonstrating involvement of PC-PLC and PLD in the signalling pathways elicited by ET-1 and TGF-b1. ET-1 and TGF-b1 each stimulated collagen I production and in an additive manner. ET-1 further induced connective tissue growth factor (CTGF), as did TGF-b1, however, to lower levels. While rapid and sustained CTGF induction was seen following TGF-b1 treatment, ET-1 increased CTGF in a biphasic manner with lower induction at 3 h and a delayed and higher induction after 5 days of permanent ET-1 treatment. Coincidentally at 5 days of permanent ET-1 stimulation, a switch in ET-receptor subtype expression to the ETB-receptor was observed. We conclude that the signal- ling pathways induced by ET-1 and TGF-b1 leading to augmented collagen I production by fibroblasts converge on a similar signalling pathway. Thereby, long-time stimulation by ET-1 resulted in a changed ET-receptor subtype ratio and in a biphasic CTGF induction. Abbreviations bFGF, basic fibroblast growth factor; CTGF, connective tissue growth factor; DAG, 1,2-diacylglycerol; ECM, extracellular matrix; ET-1, Endothelin-1; ET-receptor, endothelin receptor; FBS, fetal bovine serum; IP 3 , inositol triphosphate; Ga,Ga protein; MMP, matrix metalloproteinase; PC, phosphatidylcholine; PDGF, platelet-derived growth factor; PKC, protein kinase C; PLC, phospholipase; PTX, Pertussis toxin; TGF-b1, transforming growth factor-b1; TIMP-1, tissue inhibitory matrix metalloproteinase-1; TNF-a, tumour necrosis factor-a. FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6297 the actions of TGF-b1 are mediated by the induction of connective tissue growth factor (CTGF) [2], which stimulates collagen production by fibroblasts [3]. CTGF is coexpressed with TGF-b1 at sites of tissue fibrosis in numerous tissues [4]. Besides TGF-b1 and CTGF, the peptide hormone endothelin-1 (ET-1) was also reported to be involved in the pathophysiology of cardiac [5,6], renal [7], pul- monary [8–10], and dermal fibrosis such as systemic sclerosis ⁄ scleroderma [11–13]. Serum concentrations of ET-1 as well as TGF-b1 were increased in scleroderma patients [8,14,15]. These patients showed changes in total ET-receptor expression with an increased percent- age of the ETB-receptor, one of the two ET-receptor subtypes [11,12,15]. An antagonist against both ET-receptor subtypes abolished ET-1 induced synthesis of matrix proteins such as fibronectin and collagens I and III, suggesting that their synthesis is mediated via both ET-receptor subtypes [11]. In contrast, other authors reported that these effects were solely mediated via ETA-receptor stimulation [5]. Besides these inducing effects, ETA- receptor stimulation also contributes to ECM degra- dation by reducing expression of tissue inhibitor of matrix metalloproteinase-1 (TIMP-1) and inducing that of matrix metalloproteinase-2 (MMP-2) [7,16], which in turn degrades MMP-1 cleaved collagen I [17]. Regulation of MMP-1 expression by ET-1 has been discussed controversially demonstrating repression as well as induction [11,18]. However, the opposing effects of ET-1 on collagen I synthesis and breakdown clearly indicate an important role for ET-1 in the dynamic regulation of ECM and its dysregulation in disease [19]. ET-1 and its isoforms ET-2 and )3 belong to a peptide hormone family, which was originally charac- terized as a vasoconstrictor in the systemic and pulmonary circulation [15]. Since then, apart from regulating ECM deposition and turnover [7,11,18,19], ET-1 was further shown to induce differentiation [13], mitogenesis [11] or release of pro-inflammatory cyto- kines such as the ECM regulatory cytokines TGF-b1 [20,21], PDGF, bFGF [22] and TNF-a [23]. The endothelins exert their biological effects via two specific cell surface receptor subtypes, the ETA- and the ETB- receptor, which are members of the G protein- coupled receptor family. The ETA-receptor couples to heterotrimeric Ga proteins such as Gaq, Gas, Ga12 ⁄ 13 [24–26] and in some cell types to Gai [27–29], whereas the ETB- receptor couples to Gai and Gaq [25]. Signal transduction triggered by binding of one of the three endothelins leads to changed concentrations of various second messengers such as Ca 2+ , 1,2-diacylglycerol (DAG), inositol triphosphate (IP 3 ), cyclic adenosine 3¢,5¢-monophosphate (cAMP) and arachidonic acid [24,27]. The induction of collagen I synthesis by ET-1 was described to involve protein kinase C-d (PKC-d), p44 ⁄ 42 MAPK (pERK1 ⁄ 2) and SP-1 [30,31]. PKC-d is a mem- ber of the novel PKCs. Their activation depends exclu- sively on DAG, in contrast to the conventional PKCs, which require DAG and Ca 2+ in addition to phospha- tidylserine for activation [32]. ET-1 increases cellular DAG concentrations derived from 4,5-bis-phosphate (PIP2) by PIP2-PLC-b [24] and phosphatidylcholine (PC) by PLD [33–35] or by PC-PLC activation [34–36]. To better understand the molecular mechanisms underlying the role of ET-1 in regulating ECM depos- ition and its involvement in fibrotic processes, we analysed the signalling pathway inducing collagen I synthesis between the cell surface receptor and PKC-d by pharmacological intervention using specific inhibi- tors. We also investigated the effects of combined ET-1 and TGF-b1 stimulation and of prolonged stimu- lation with respect to ET-receptor subtype expression and the induction of CTGF and TGF-b1 transcripts. Results Expression of ET-receptor subtypes by human primary dermal fibroblasts Cell surface binding of [ 125 I]ET-1 revealed specific ET- 1 binding sites on human primary dermal fibroblasts. Binding was confirmed by displacing labelled ligand by specific ET-receptor antagonists including the nonsub- type specific antagonist PD156252, the ETA-receptor selective antagonist BQ123 and the ETB-receptor selective antagonist BQ788. The IC 50 values were 5.4 · 10 )10 m for BQ123, and 3.8 · 10 )7 m for BQ788, clearly demonstrating high affinity binding to the ETA-receptor (Fig. 1A). These data are in agreement with previously published IC 50 values [37,38]. Results from surface binding analysis were further confirmed by immunoblotting (Fig. 2A). In total cell lysates, the ETA-receptor was clearly detected as well as in membrane fractions. The ETA-receptor was enriched in membrane fractions as revealed by two specific bands, a predominant band of approximately 55 kDa and a less prominent one of approximately 45 kDa as described [24,39,40]. Using immunodetec- tion, we were unable to show ETB-receptor expression in fibroblasts. However, applying RT-PCR, both, the ETA- as well as the ETB-receptor were detectable as single bands at 135 bp (ETA-receptor) and 150 bp (ETB-receptor; Fig. 2B). ET-1 and TGF-b1 signalling A. Horstmeyer et al. 6298 FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS We therefore conclude that the ETA-receptor is the receptor predominantly expressed by human primary dermal fibroblasts, while ETB-receptor expression lev- els are much lower as reported elsewhere [11]. Additive induction of collagen I levels by ET-1 and TGF-b1 ET-1 stimulation (100 nm) induced a twofold increase in the synthesis of collagen I compared to unstimu- lated controls (Fig. 3A), in concordance with published data [11,13]. By immunoblotting, collagen I was detec- ted by two different antibodies against human collagen I, a polyclonal and a monoclonal one (data not shown). Both antibodies recognized collagen I derived from fibroblast culture supernatants as well as purified collagen I. As a positive control, cultures stimulated by 180 pm TGF-b1 [41,42] were included in all subse- quent analyses. Induction of collagen I by TGF-b1 and by ET-1 was perceptible by coomassie staining as well as by immunoblotting (Fig. 3B). The immunoblot thus confirms the specifity of the coomassie staining (see Experimental procedures). Comparing collagen I levels induced by either ET-1 or TGF-b1 or a combination of both revealed that either hormone alone enhanced collagen I levels in fibroblasts compard to unstimulated cells (control) and that the combination induced an additive effect. Fibroblasts responded to stimulation by increasing collagen I synthesis in the order: unstimulated AB Fig. 2. Expression of ET-receptors. (A) Immunoblotting: human primary dermal fibroblasts were cultured for two days and cells were lysed or fractionated. Samples were separated by 10% SDS ⁄ PAGE and electrophoretically transferred to nitrocellulose. ET-receptors were visual- ized with polyclonal rabbit anti-ETA- or anti-ETB-receptor antibody. Lanes: membrane fraction (membrane), total cell lysate (lysate). The immunoblot shown represents three independent experiments, each performed with cells from different donors and cell isolations. (B) RT-PCR: human primary dermal fibroblasts were cultured for one day, then starved for one day. RNA was isolated, reverse-transcribed and subjected to PCR in presence of specific primer pairs for the ETA- or ETB-receptor. Samples were separated by 3% agarose-ethidium bro- mide gel electrophoresis. Results from two different donors (lane 1, 2) and negative controls (H 2 O) are shown. The gel shown represents four independent experiments, performed with cells from different donors and cell isolations. Fig. 1. Displacement of [ 125 I]ET-1 by specific ET-receptor antago- nists. Human primary dermal fibroblasts were cultured on 96 well plates for two days. Fibroblasts were incubated with [ 125 I]ET-1 (25 p M) with antagonists as indicated, lysed, and bound ligand was determined using a c-counter. (A) Results are summarized as IC 50 values and means ± SEM of six independent experiments each performed with cells from different donors and cell isolations. (B) Representative displacement curves of the three antagonists: PD156252 (n), BQ123 (m), BQ788 (d). Results are means ± SEM. A. Horstmeyer et al. ET-1 and TGF-b1 signalling FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6299 control < ET-1 < TGF-b1 < ET-1 + TGF-b1 (Fig. 3B). Analysis of ET-1 and TGFb1 signalling mechanisms resulting in collagen I synthesis To determine the signalling pathway between the ET- receptor and its cognate PKC-isoform, that mediates the ET-1 induced collagen I synthesis in fibroblasts, specific inhibitors were applied. Cadwallader et al. [29] described Gai-dependent ET-1 stimulation of PKC via the ETA-receptor in rat fibroblasts. We therefore inves- tigated the involvement of Gai in human primary dermal fibroblasts. Pertussis toxin (PTX), a specific Gai inhibitor [26], strongly decreased ET-1-induced collagen I synthesis. But PTX did not alter unstimulated, basal collagen I synthesis (Fig. 4) as previously described [43]. Further, collagen I levels induced by TGF-b1 stimulation were also not altered by PTX, as described [43]. This indicated involvement of Gai leading to enhanced collagen I levels after ET-1 stimulation. To determine the phospholipases responsible for PKC activation [32], we examined the possible DAG donors PLC-b, PC-PLC and PLD. PLCb, a down- stream effector of the Gaq [44] as well as of the bc subunit of Gai [45] was inhibited specifically by U73122 [46]. Inhibiting PLC-b by U73122 did not alter ET-1-induced collagen I synthesis, nor did it influence basal or TGF-b1-stimulated collagen I levels (Fig. 5A). Next, we analysed the involvement of the PC-phospho- lipases. We used D609 to specifically inhibit PC-PLC [46], and neomycin [47–49] and N-butanol [50–52] as specific inhibitors of PLD. The inhibition of PC-PLC by D609 (Fig. 5B) and inhibition of PLD by neomycin (Fig. 5C) as well as by N-butanol (not shown) abol- ished the increase in collagen I levels after ET-1 A B Fig. 3. ET-1 induces collagen I synthesis. Human primary dermal fibroblasts were cultured without or with hormone (100 nM ET-1, 180 pM TGF-b1) for 2 days. The entire cell culture media were incubated with pepsin, TCA-precipitated and separated by 6% SDS ⁄ PAGE. (A) Pro- teins were electrophoretically transferred to nitrocellulose and collagen I was visualized with polyclonal rabbit anticollagen I antibody. Lanes: unstimulated fibroblasts (control), fibroblasts stimulated with ET-1 (ET-1), fibroblasts stimulated with TGF-b1 (TGFb1), one lg of purified colla- gen I (positive control). The histogram represents densitometric ratios of (stimulated: unstimulated) collagen I signals from three independent experiments, performed with cells from different donors and cell isolations. Unstimulated values were set as 1.0. Results are means ± SEM. Asterisks represent statistically significant differences, P < 0.05. (B) Two specific collagen I bands representing the a1anda2 chains of collagen I were stained by Coomassie Blue. Lanes: unstimulated fibroblasts (control), fibroblasts stimulated with ET-1 (ET-1), fibroblasts stimulated with TGF-b1 (TGFb1), fibroblasts stimulated with ET-1 and TGF-b1 (ET-1 TGFb1). The histogram represents densitometric ratios (stimulated: unstimulated) of collagen I signals from four independent experiments, performed with cells from different donors and cell iso- lations. Unstimulated values were set as 1.0. Results are means ± SEM. ET-1 and TGF-b1 signalling A. Horstmeyer et al. 6300 FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS stimulation. Of note, simultaneous PLD inhibition and ET-1 stimulation significantly reduced collagen I levels to below basal values. TGF-b1 signalling was affected similarly, with no effect by inhibiting PLC-b (Fig. 5A), strong down-regulation of collagen I levels after inhi- bition of PC-PLC (Fig. 5B), and similar collagen I down-regulation after PLD inhibition (Fig. 5C). These results demonstrate a crucial role for the PC-phospho- lipases in collagen I induction by both ET-1 and TGF- b1. Our results are in agreement with previous reports that demonstrated the involvement of PC-PLC in TGF-b1-mediated regulation of other genes not inclu- ding collagen I in further cell types [53,54]. Long-term stimulation by ET-1 and TGF-b1 Increased serum concentrations of ET-1 and TGF-b1 were shown to be present in sera of patients with fibro- tic diseases [8,14,15]. In addition, it has been described that TGF-b1 causes a persistent increase in collagen I [41] partly via CTGF gene activation [55]. It is thereby reported for both hormones that an induced CTGF gene expression correlates with an increased amount of the CTGF hormone [18,56]. To clarify whether ET-1 also mediates its effects via CTGF, we stimulated fibroblasts with ET-1 and TGF-b1 for five days and examined CTGF and TGF-b1 gene activation (Fig. 6A). ET-1 stimulation resulted in a biphasic induction of CTGF. We observed a weak and short initial increase of CTGF transcripts three hours after ET-1 stimulation, which thereafter decreased to basal levels. However, a stimulation for five days resulted in a second, stronger increase in CTGF RNA levels. In contrast to ET-1, stimulation with TGF-b1 led to per- sistent induction of CTGF transcripts during the entire stimulation period. As ET-1 induced TGF-b1 RNA expression in var- ious cell types [20,21] and both, ET-1 and TGF-b1 act additionally to induce collagen I synthesis (Fig. 3B) we examined whether ET-1 induces TGF-b1 gene activa- tion in human primary dermal fibroblasts. In contrast to TGF-b1, that induced its own gene activation, ET-1 stimulation failed to induce TGF-b1 RNA levels (Fig. 6B). This indicates a TGF-b1 independent action of ET-1. Previous reports described altered ratios of ETA- to ETB-receptors in fibroblasts cultured from the skin of patients with systemic sclerosis [11,12]. We were inter- ested to see whether long-term in vitro stimulation of fibroblasts from healthy donors with ET-1 also leads to similar modulations. Indeed after stimulation with ET-1 over five days we observed induced levels of the ETB-receptor, while ETA-receptor levels were strongly diminished (Fig. 7C). Discussion Using human primary dermal fibroblasts, the ETA- receptor was clearly detectable by immunoblotting in contrast to the ETB-receptor, which required more sensitive investigation by RNA analysis for detection. Furthermore, ET-1 displacement studies with subtype- specific antagonists confirmed binding only to the ETA-receptor. Thus, the ETA-receptor is the main Fig. 4. Analysis of signalling mechanism: Gai. After pretreatment with PTX (100 ngÆmL )1 , overnight) human primary dermal fibroblasts were cultured without or with hormone (100 n M ET-1, 180 pM TGF-b1) for 2 days. Entire cell culture media were incubated with pepsin, TCA-preci- pitated and separated by 6% SDS ⁄ PAGE. Two specific collagen I bands representing the a1anda2 chains of collagen I were stained by Coomassie Blue. Lanes: unstimulated fibroblasts (control), fibroblasts stimulated with (ET-1), fibroblasts stimulated with TGFb1. The histo- gram represents densitometric ratios (stimulated: unstimulated) of collagen I signals from three independent experiments, performed with cells from different donors and cell isolations. Unstimulated values were set as 1.0. Results are means ± SEM. Asterisks represent statisti- cally significant differences, P < 0.05. A. Horstmeyer et al. ET-1 and TGF-b1 signalling FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6301 ET-receptor subtype expressed in human primary der- mal fibroblasts, consistent with findings by Shi-wen et al. [11]. Interestingly, increased ET-1 serum concentrations and a change in the ratio of the ETA- to the ETB- receptor with an increase of the ETB-receptor on fibro- blasts from patients with systemic scleroderma have been described [11,12]. This is in agreement with a change in ET-receptor subtype expression pattern towards the ETB-receptor after long-term stimulation of five days by ET-1. The receptor subtype shift towards the ETB-receptor described here is different from cardiac and pulmonary fibrosis, where increased numbers of ETA-receptor binding sites were observed [9,57]. A similar receptor subtype shift towards the ETB-receptor is described in smooth muscle cells under the development of arteriosclerosis [58] and for endo- thelial cells under the condition of hypertension [59]. A changed ratio of ET-receptor subtypes results in a modified signal transduction, because ETA- and ETB-receptors couple in part to different Ga proteins. For example in smooth muscle cells, a shift towards the ETB-receptor causes a deformation of these cells [60], whereas in human myocardial fibroblasts a shift towards the ETA-receptor profoundly interferes with cellular Ca 2+ regulation [57]. Moreover, it is described that the ETA-receptor regulates proliferation by pro- and antimitogenic actions [24,61]. In addition, the ETA-receptor is a regulator of collagen I homeostasis by inducing both, its synthesis and its degradation [7,16] in contrast to the ETB-receptor that only indu- ces collagen I synthesis [11,62]. A changed ET-receptor subtype relation towards ETB-receptor thus may dis- turb the balance between synthesis and degradation of collagen I by ET-1. In this context, an increased num- ber of ETB-receptors may in turn be responsible for a change of signalling properties via different Ga coup- ling induced by ET-1 [63]. This may be potentiated by a changed internalization of ETB-receptors. Homo- dimers and monomers of the ETB-receptor are A B C Fig. 5. Analysis of signalling mechanism: DAG mediators PLC-b, PC-PLC and PLD. After pretreatment with specific phospho- lipase inhibitors (15 min) human primary der- mal fibroblasts were cultured without or with hormone (100 n M ET-1, 180 pM TGF- b1) for two days. Entire cell culture media were incubated with pepsin, TCA-precipita- ted and separated by 6% SDS ⁄ PAGE. (A) PLCb inhibition by U73122 (10 l M). (B) PC-PLC inhibition by D609 (185 l M). (C) PLD inhibition by neomycin (500 l M). Two specific collagen I bands representing the a1 and a2 chains of collagen I were stained by Coomassie Blue. Lanes: unstimu- lated fibroblasts (control), fibroblasts stimu- lated with ET-1 (ET-1), fibroblasts stimulated with TGF-b1 (TGFb1). The histograms repre- sent densitometric ratios (stimulated: unstimulated) of collagen I signals from five independent experiments, performed with cells from different donors and cell isola- tions. Unstimulated values were set as 1.0. Results are means ± SEM. Asterisks repre- sent statistically significant differences, P < 0.05. ET-1 and TGF-b1 signalling A. Horstmeyer et al. 6302 FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS lysosomally degradated after ET-1 binding [64,65]. Elevated numbers of ETB-receptors increase the prob- ability of heterodimerization between ETA- and ETB- receptor proteins. These heterodimers are recycled after ligand binding as is the case for the monomeric ETA-receptor [66]. This additionally increases amount of ETB-receptors present at the cell membrane. It is described [67] that the internalization and subsequent lysosomal degradation of the ETB-receptor is respon- sible for its short-term signalling period after ligand binding. Therefore, this changed ETB-receptor inter- nalization may lead to a prolonged signal transduction period via the ETB-receptor. The augmented autoinduction of ET-1 synthesis and secretion via increased ETB-receptor numbers [68] may also augment the intracellular signalling effects of ETB-receptors. This may result into an overshooting collagen I synthesis by ET-1. This is supported by our results demonstrating CTGF gene activation after ET-1 stimulation: ET-1 led to an activation of the early response gene CTGF at two time-points. The first peak in CTGF gene activation was observed only three hours after ET-1 stimulation, whereas a second one was detectable after continuous stimulation for five days by ET-1. Reports from several groups indica- ted that induction of CTGF mRNA by ET-1 is regu- lated in a cell type-specific manner. In human pulmonary fibroblasts, CTGF RNA was strongly induced by ET-1 for more than 12 hours [18], while in cardiac myocytes, CTGF was transiently induced for barely four hours, while in cardiac fibroblasts, no CTGF induction was observed [69]. A B Fig. 6. Long-term effects of CTGF and TGFb1 gene activation by ET-1 and TGF-b1. Human primary dermal fibroblasts were cul- tured without or with hormone (100 n M ET-1, 180 pM TGF-b1) for the indicated times. Specific CTGF RNA or TGF-b1 RNA were detected by nothern hybridization with 32 P labelled probes and visualized on X-ray films. (A) CTGF gene activation: samples are unstimulated fibroblasts (C), fibroblasts stimulated with ET-1 (E), fibroblasts stimula- ted with TGF-b1 (T). Top panel: CTGF detection, bottom panel: 18S detection (loading control). The histograms represent densitometric ratios of CTGF to 18S signals from three to seven independent experi- ments, performed with cells from different donors and cell isolations. Unstimulated values were set as 1.0. Results are means ± SEM. Asterisks represent statisti- cally significant differences, P < 0.05. (B) TGF-b1 gene activation: samples are unstimulated fibroblasts (C), fibroblasts stimulated with ET-1 (E), fibroblasts stimula- ted with TGF-b1 (T). Top panel: TGF-b1 detection, bottom panel: 18S detection (loading control). The histograms represent densitometric ratios of TGF-b1 signals normalized to 18S signals from three to seven independent experiments, performed with cells from different donors and cell isolations. Unstimulated values were set as 1.0. Results are means ± SEM. Asterisks represent statistically significant differences, P < 0.05. A. Horstmeyer et al. ET-1 and TGF-b1 signalling FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6303 Our results indicate that in human primary dermal fibroblasts ET-1 may mediate induced synthesis of col- lagen I by activation of CTGF, which is not detectable in unstimulated adult dermal fibroblasts [14]. Upon activation by TGF-b1, CTGF acts in an autocrine fashion as a strong inducer of collagen I synthesis [3,14]. Further investigations should reveal if ET-1 eli- cits induced collagen I synthesis solely via induction of CTGF or via additional mechanisms as it is the case for TGF-b1 stimulation [2,55]. In contrast to ET-1, TGF-b1 induced persistent CTGF gene activation dur- ing the stimulation period resulting in prolonged colla- gen I synthesis, as described [41]. It seems that ET-1 and TGF- b1 show different properties in regulating the induction of collagen I synthesis by CTGF. Further examination of the intracellular signalling mechanisms revealed an essential involvement of Gai, PLD and PC-PLC. Their inhibition abolished ET-1 induced collagen I expression. The PC-phospholipases are also important signalling intermediates for TGF-b1 induced collagen I synthesis. Furthermore, it has been reported that ET-1 and TGF-b1 require PKC-d for induction of collagen I [30,31,70], that may be activa- ted by DAG derived from PLD and PC-PLC. Our results demonstrate that the ET-1 and TGF-b1 signal- ling pathways terminating in induced collagen I syn- thesis overlap partially. This is in contrast to the early postreceptor pathways, which differ between the ET-receptors belonging to the family of G protein-cou- pled receptors and the TGF-b1 receptors as members of the serine-threonine kinase receptor group [1,24,25]. Here, we demonstrated an additive induction of col- lagen I production by ET-1 and TGF-b1. We observed an additive CTGF gene activation by both hormones (data not shown). A further example of additive action between ET-1 and TGF-b1 is induction of myofibro- blast differentiation [71]. Some hormones acting in auto- and paracrine man- ner are capable of inducing each other’s expression [72]. In human primary dermal fibroblasts, however, TGF-b1 neither induced the synthesis of ET-1 [71] nor did ET-1 induce the synthesis of TGF-b1. The inter- play between TGF-b1 and ET-1 therefore occurs within their partly common intracellular signalling pathway. However, their roles may diverge because of differences in CTGF gene activation and the different effects of PLD inhibition on collagen I levels. Experimental procedures Materials Experimental material was purchased from the following suppliers: ET-1 (human), BQ123, BQ788, TGF-b1 (human), D609, U73122, Pertussis toxin (PTX), pepsin: Calbiochem- Novabiochem, Bad Soden, Germany. PD156252, neomycin trisulphate, magnesium salt of l-ascorbic acid 2- or 3-phos- phate, Ponceau Red, Coomassie blue, N-butanol: Sigma- Aldrich, Deisenhofen, Germany. [ 125 I]ET-1: Amersham ⁄ Pharmacia Biotech, Freiburg, Germany. Human collagen I: Sircoll, Cologne, Germany. Cell culture Primary human dermal fibroblasts were obtained by explant culture from several biopsies of healthy skin of age- and sex-matched adult donors. Fibroblasts were maintained in Dulbecco’s modified Eagle’s medium (DMEM), supple- mented with ascorbic acid (71 lg per mL), penicillin (100 UÆmL )1 ), streptomycin (100 lgÆmL )1 ), 0.2 m glutamine (all from Biochrom, Berlin, Germany) and fetal bovine serum (10%, FBS; PAA, Linz, Austria) and cultured in the humidified atmosphere of 5% CO 2 . Fibroblasts were sub- cultured to confluence and used between passages 2–5. Northern hybridization analysis Fibroblasts were seeded at 5 · 10 4 cells per cm 2 into 100 mm dishes and cultured for 24 h followed by starvation with FBS-free DMEM for 24 h. Thereafter, cells were cul- tured in FBS-free DMEM in the absence or with either Fig. 7. Long-term effects of ET-receptor gene activations by ET-1. Human primary dermal fibroblasts were cultured without or with 100 n M ET-1 for indicated times. RNA was isolated and 2 lg RNA was reverse-transcribed and subjected to Q-PCR in presence of specific primer pairs for the ETA- and the ETB-receptor. Signals from three independent experiments were determined in triplicate and normalized to b-actin, performed with cells from different donors and cell isolations. Unstimulated values were set as 1.0. Results are means ± SEM. Asterisks represent statistically signifi- cant differences, P < 0.05. ET-1 and TGF-b1 signalling A. Horstmeyer et al. 6304 FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS 100 nm ET-1 or 180 pm TGF-b1 (daily addition) for the indicated times. Total RNA was isolated using Qiagen RNA columns (Qiagen, Hilden, Germany). Purified RNA (2–5 lg per lane) was electrophoresed on formaldehyde agarose gels running in Mops buffer (40 mm Mops, 10 mm sodium acetate, 1 mm EDTA, pH 7.0) and transferred to nylon membranes (Amersham ⁄ Pharmacia Biotech, Frei- burg, Germany). Membranes were hybridized to an a- 32 P- dCTP labelled 0.4 kbp human TGF-b1 or a 1.1 kbp human CTFG cDNA probe using Hybrimax solution (Ambion, Huntingdon, UK) according to the manufacturer’s instruc- tions. RNA loading and transfer was evaluated by re-pro- bing the membranes with an 18S ribosomal RNA cDNA probe. Real Time PCR (RT-PCR) Two lg of purified RNA were reverse-transcribed by reverse transcriptase core kit (Eurogentec, Cologne, Germany) and subjected to PCR in the presence of specific primer pairs (ETA-receptor: forward primer: 5¢-TGCCCTCAGTGAACATCTTAAGC-3¢, reverse pri- mer: 5¢-TCCATTTCGTTATACACAGTTTTCTTC-3¢; ETB- receptor: forward primer: 5¢-GCTGCACATCGTCATTGA CAT-3¢, reverse primer: 5 ¢-GCACATAGACTCAGCA CAGTGATT-3¢). Primer pairs yielded products of 135 bp for the ETA-receptor and of 150 bp for the ETB-receptor. The annealing temperature used was 60 °C and 40 cycles were performed. PCR products were separated using agarose gels containing ethidium bromide and signals visualized under UV light. For semiquantitative PCR (Q-PCR), RNA was reverse- transcribed by qPCR core kit (Eurogentec Deutschland GmbH, Cologne, Germany) and subjected to Q-PCR in the presence of the specific ET-receptor subtype primer pairs in addition to corresponding FAM-labelled probes (ETA- receptor: 5¢-FAM-TGCTGGTTCCCTCTTCACTTAAG CCG-3¢-TAMRA, ETB-RECEPTOR-receptor: 5¢-FAM- CAGTCCTCTGCCAGCAGCTTGTAGACA-3¢-TAMRA) using the Taq Man 7700 Sequence Detection System (ABI Applied Biosystems, Darmstadt, Germany). Assays were run in triplicates and normalized to b-actin (forward pri- mer: 5¢-CTAGACTTCGAGCAAGAGATGG-3¢, reverse primer: 5¢-ATAGAGGTCTTTACGGATGTCAAC-3¢,FAM probe: 5¢-FAM-CAGCCTTCCTTCCTGGGTATGGAAT CC-3¢-TAMRA). Binding studies Receptor binding and displacement studies were performed as described [24]. Fibroblasts were seeded (5 · 10 4 cells per cm 2 ) into 96 well plates and cultured for 2 days. Cells were washed 3 times with PBS at room temperature and incu- bated with binding buffer (137 mm NaCl, 2.7 mm KCl, 6.5 mm Na 2 HPO 4 , 1.5 mm KH 2 PO 4 , pH 6.8) containing [ 125 I]ET-1 (25 pm) for 4 h at 4 °C. Non-specific binding was determined in the presence of unlabelled ET-1 (200 nm). For displacement studies, binding buffer contained either the nonsubtype specific antagonist (PD156252), the ETA- receptor selective antagonist (BQ123) or the ETB-receptor selective antagonist (BQ788) over a concentration range of 0.1 pm)1 lm. After an incubation period of 4 h, cells were rinsed with PBS (4 °C), treated with cell lysis buffer (1% SDS, 0.1 N NaOH) for 10 min and the bound activity was determined using a c-counter. Collagen I synthesis Fibroblasts were seeded (3 · 10 4 cells per cm 2 ) into 35 mm dishes and cultured for 24 h in full medium, which was replaced by FBS-free DMEM (0.2 mL per cm 2 ) with or without hormones or inhibitors as indicated, and cells were cultured for further 2 days ([73], modified protocol). Cul- ture media were collected and 1800 lL of each medium sample were incubated with pepsin (40 lgÆmL )1 , at pH 2.5, 4 °C) overnight under rotation to cleave all proteins present in the sample with the exception of collagen I. After TCA precipitation (10% TCA, 0.1% Triton X-100), modification of [74], samples of one experimental set were resuspended with nonreducing SDS loading buffer and applied to one gel for electrophoresis (6% polyacrylamide gel). Gels were fixed and stained with Coomassie Brilliant Blue R B-0149 (0.25%, 10% acetic acid, 25% isopropanol), followed by destaining with 10% acetic acid and 12.5% isopropanol. As an indirect loading control, cells were counted before and after stimulation to exclude that increased collagen I amounts may result from different cell numbers instead of the induction of collagen I synthesis. Immunoblot analysis To determine ET-receptor protein expression, fibroblasts were seeded (5 · 10 4 cells per cm 2 ) into 100 mm dishes and cultured for 2 days. Then the cells were washed three times with PBS and lysed with RIPA buffer (150 mm NaCl, 0.1% SDS, 0.5% desoxycholate, 1% Nonidet P 40, 50 mm Tris- HCl pH 8.0) to obtain total cell lysate or with membrane buffer (10 mm KCl, 1.5 mm Mg 2 Cl, 1 mm EDTA, 1 mm EGTA, 1 mm DTT, 20 mm Hepes, pH 7.5) to obtain mem- brane fractions [75]. Samples were diluted in reducing SDS loading buffer and subjected to 10% SDS ⁄ PAGE. Separ- ated proteins were transferred to nitrocellulose membranes (Amersham ⁄ Pharmacia Biotech, Freiburg, Germany). Transfer efficiency was determined by Ponceau Red stain- ing. Nitrocellulose membranes were blocked by incubation with 5% nonfat dried milk (BioRAD, Mu ¨ nchen, Germany) in PBS for 1 h at room temperature. Antigens were detec- ted using polyclonal antibodies diluted 1 : 200 in PBS for one h at room temperature [rabbit anti-(ETA receptor) IgG A. Horstmeyer et al. ET-1 and TGF-b1 signalling FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS 6305 or rabbit anti-(ETB receptor) IgG, Chemicon, Hofheim, Germany]. The specificities of both antibodies were demon- strated [76–78]. To determine secreted collagen I, TCA-precipitated pepsin-resistent proteins derived from cell culture media were resuspended in nonreducing SDS-buffer and subjected to precast 4–12% gradient Tris-glycine polyacrylamide gel electrophoresis (anamed Elektrophoreses GmbH, Darms- tadt, Germany). Separated proteins were transferred and further processed for immunoblotting as described above. Antigens were detected using antibodies diluted in PBS for 1 h at room temperature [1 : 100, rabbit anti-(collagen I) IgG, Quartett, Berlin, Germany or 5 lgÆmL )1 mouse anti- (collagen I) IgG, clone CP17L, Calbiochem-Novabiochem, Bad Soden, Germany]. Primary antibodies were visualized using species-specific secondary horseradish peroxidase-conjugated antibodies (IgG) diluted 1 : 3000 in PBS [mouse anti-(rabbit IgG) Ig or swine anti-(mouse IgG) Ig, Dako, Hamburg, Germany]. Resulting antigen-antibody complexes were detected by chemiluminescence using ECL kit (Amersham ⁄ Pharmacia Biotech, Freiburg, Germany). Chemiluminescence signals were recorded on X-ray films (Amersham ⁄ Pharmacia Bio- tech, Freiburg, Germany). Densitometrical analysis Gels and X-ray films were analysed by scanning densitome- try [Imager Fluor-S TM Multilmager (BioRAD, Heidelberg)]. The resulting data were evaluated using quantify one 4.0.3. Densitometric values obtained from samples of unstimu- lated and untreated human primary dermal fibroblasts were statistically not different from values obtained from sam- ples of pretreated and unstimulated human primary dermal fibroblasts. However, the mean values of densitometric data derived from unstimulated and untreated human primary dermal fibroblasts of each experiment were set to a relative value of ‘1’, serving as baseline value for the other data, thereby representing the status of both groups of unstimu- lated human primary dermal fibroblasts. Statistical analysis Statistical analysis was performed using GraphPad Prism version 3.02. All data were expressed as means ± SEM and statistically analysed using an unpaired t-test. Acknowledgements This study was supported by the Deutsche Fors- chungsgemeinschaft (KR558 ⁄ 13 to TK) and by the Koeln Fortune programme ⁄ Faculty of Medicine, Uni- versity of Cologne (to TK and CL). References 1 Border WA & Noble NA (1994) Transforming growth factor b in tissue fibrosis. New Engl J Med 331, 1286– 1292. 2 Grotendorst GR (1997) Connective tissue growth factor: a mediator of TGF-b action in fibroblasts. Cytokine Growth Factor Rev 8, 171–179. 3 Frazier K, Williams S, Kothapalli D, Klapper H & Grotendorst GR (1996) Stimulation of fibroblast growth, matrix production, granulation tissue formation by connective tissue growth factor. J Invest Dermatol 107, 404–411. 4 Igarashi A, Nashiro K, Kikuchi K, Sato S, Ihn H, Fujimoto M, Grotendorst GR & Takehara K (1996) Connective tissue growth factor gene expression in tissue sections from localized scleroderma, keloid, and other fibrotic skin disorders. J Invest Dermatol 106, 729–733. 5 Fraccarollo D, Galuppo P, Bauersachs J & Ertl G (2002) Collagen accumulation after myocardial infarc- tion: effects of ETA receptor blockade and implications for early remodelling. Cardiovasc Res 54, 599–567. 6 Hafizi S, Wharton J, Chester AH & Yacoub MH (2004) Profibrotic effects of Endothelin-1 via the ETA receptor in cultured human cardiac fibroblasts. Cell Physiol Biochem 14, 285–292. 7 Tostes RCA, Touyz RM, He G, Ammarguellat F & Schiffrin EL (2002) Endothelin A receptor blockade decreases expression of growth factors and collagen and improves matrix metalloproteinase-2 activity in kidneys from stroke-prone spontaneously hypertensive rats. J Cardiovasc Pharmacol 39, 892–900. 8 Cambrey AD, McAnulty RJ, Harrison NK, Dawes JS, Campa RW, du Bois DM, Black CM, Laurent GJ & McAnulty RJ (1992) Endothelin is present in the lungs of patients with systemic scleorsis and stimulates lung fibroblast proliferation in vitro. Am Rev Respir Dis 145, A15. 9 Wendel M, Petzold A, Koslowski R, Kasper M, Augu- stein A, Knels L, Bleyl JU & Koch T (2004) Localiza- tion of endothelin receptors in bleomycin-induced pulmonary fibrosis in the rat. Histochem Cell Biol 122, 507–517. 10 Giad A, Michel RP, Sheppard SM, Corrin B & Hamid O (1993) Expression of endothelin in lungs of patients with cytogenic fibrosing alveolitis. Lancet 341, 1550–1554. 11 Shi-wen X, Denton CP, Dashwood MR, Abraham DJ & Black CM (1998) Endothelins: Effect on matrix biosynth- esis and proliferation in normal and scleroderma fibro- blasts. J Cardiovasc Pharmacol 31 (Suppl. 1), S360–S363. 12 Shi-wen X, Denton CP, Dashwood MR, Holmes AM, Bou-Gharios G, Pearson JD, Black CM & Abraham DJ (2001) Fibroblast matrix gene expression and connective ET-1 and TGF-b1 signalling A. Horstmeyer et al. 6306 FEBS Journal 272 (2005) 6297–6309 ª 2005 The Authors Journal Compilation ª 2005 FEBS [...]... A, Offermanns S, Wilkie TM & Schulz G (1999) Differential involvement of Ga12 and Ga13 in receptor mediated stress fiber formation J Biol Chem 274, 17901–17907 27 Jones LG (1996) Inhibition of cyclic AMP accumulation by endothelin is pertussis toxin sensitive and calcium independent in isolated adult feline myocytes Life Sci 58, 115–123 28 Hilal-Dandan R, Merck DT, Lujan JP & Brunton LL (1994) Coupling... Whiteside CI (2001) High glucose-enhanced mesangial cell extracellular signal-regulated protein kinase activation and alpha1 (IV) collagen expression in response to enothelin-1: role of specific protein kinase C isoenzymes Diabetes 50, 2376– 2383 31 Chintalgattu V & Katwa LC (2004) Role of PKCd in endothelin induced type I collagen expression in cardiac myofibroblasts isolated from the site of myocardial infarction... Transforming growth factor-beta stimulates the expression of fibronectin and collagen and their incorporation into extracellular matrix J Biol Chem 261, 4337–4345 43 McAnulty RJ, Chambers RC & Laurent GJ (1995) Regulation of fibroblast procollagen production Biochem J 307, 63–68 44 Smrcka AV & Sternweis PC (1993) Regulation of purified subtypes of phosphatidylinositol-specific phospholipase Cb by G-protein and. .. N-terminal kinase (JNK)-interacting protein, Sab (SH3BP5), associates with mitochondria Biochem J 367, 577–583 76 Kang YJ, Li Y, Zhou Z, Roberts AM, Cai L, Myers SR, Wang L & Schuschke DA (2002) Elevation of serum endothelins and cardio toxicity induced by particulate matter (PM2.5) in rats witha cute myocardial infarction cardiovascular Toxicology 2, 253–262 77 Ramzy D, Rao V, Tumiati L, Xu N, Miriuka... collagen accumulation, cell poliferation and formation of a three-dimensional tissue like substance by skin fibroblasts J Cell Physiol 138, 8–16 74 Koyano Y, Hammerle H & Mollenhauer J (1997) Ana¨ lysis of 3H-prolin-labeled protein by rapid filtration in multiwell plates for the study of collagen metabolism Bio Techniques 22, 706–716 75 Wiltshire C, Matsushita M, Tsukada S, Gillespie DAF & May GHW (2002)... factor-betainduced expression of connective tissue growth factor in human fibroblasts J Invest Dermatol 124, 1162–1169 Drimal J, Mislovicova M, Ismail A & Moncek F (1999) Detrimental subtype-specific endothelin-1 (ET-1) signaling in myocardial cells: the ETA mediated proliferation and ETB receptor down -regulation Physiol Res 48, 9–19 Iwasa S, Fan J, Shimokama T, Nagata M & Watanabe T (1999) Increased immunoactivity... Compilation ª 2005 FEBS A Horstmeyer et al 62 Jeyabalam A, Novak J, Danielson LA, Kerchner LJ, Opett SL & Conrad KP (2003) Essentail role for vascular gelatinase activity in relaxin-induced renal vasodilation, hyperfiltration, and reduced myogenic reactivity of small arteries Circ Res 93, 2149–1257 63 Takagi Y, Ninomiya H, Sakamoto A, Miwa S, Masaki T (1995) Structural basis of G protein specificity of. .. Tetrahydrobiopterin prevents cyclosporine induced vasomotor dysfunction Transplantation 78, 876–881 78 Uematsu F, Takahashi M, Yoshida M, Igarashi M, Watanabe N, Suzuki N, Abe M, Rusyn I, Floyd RA & Nakae D (2005) Distinct patterns of gene expression in hepatocellular carcinomas and adjacent non-cancerous cirrhotic liver tissue in rats fed a choline-deficient, I- amino acid-defined diet Cancer Sci 96, 414–424... Cell Mol Biol 17, 10–16 47 Huang Y, Quershi JA & Chen H (1999) Effects of phosphatidylinositol 4,5-bisphosphate and neomycin on phospholipase D: kinetic studies Mol Cell Biochem 197, 195–201 48 Kristiansen S, Nielsen JN, Bourgoin S, Klip A, Franco M & Richter EA (2001) GLUT-4 translocation in skeletal muscle studied with a cell-free assay: involvement of phospholipase D Am J Physiol Endocrinol Metab... Oksche A (2004) Ligand-dependent differences in the internalization of endothelin A and endothelin B receptor heterodimers J Biol Chem 279, 27679–27687 41 Varga J, Rosenbloom J & Jimenez S (1987) Transforming growth factor beta causes a persistent increase in steady state amounts of type I and type III collagen and fibronectin mRNAs in normal human dermal fibroblasts Biochem J 247, 597–604 42 Ignotz RA & . involvement of PC-PLC and PLD in the signalling pathways elicited by ET-1 and TGF-b1. ET-1 and TGF-b1 each stimulated collagen I production and in an additive manner [11]. Additive induction of collagen I levels by ET-1 and TGF-b1 ET-1 stimulation (100 nm) induced a twofold increase in the synthesis of collagen I compared