Relaxin, a Pregnancy Hormone, Is a Functional Endothelin-1 Antagonist
Attenuation of Endothelin-1–Mediated Vasoconstriction by Stimulation of Endothelin Type-B Receptor Expression via ERK-1/2 and Nuclear Factor-κB
We have recently demonstrated that relaxin (RLX) acts as compensatory mediator in human heart failure. RLX inhibits the stimulation of endothelin-1, the most potent vasoconstrictor in heart failure. Upregulation of the endothelin type-B receptor (ETB), which mediates endothelin-1 clearance and endothelial release of NO, represents a pivotal mode of RLX action. However, signal transduction and abundance of this phenomenon are unknown. Therefore, we investigated RLX-induced regulation of ETB in human umbilical vein endothelial, epithelial (HeLa), and vascular smooth muscle cells. In human umbilical vein endothelial cells and HeLa cells, but not in human vascular smooth muscle cells, RLX upregulated ETB expression and activated extracellular signal–regulated kinase-1/2 (ERK-1/2) and nuclear factor-κB (NF-κB), a transcription factor. PD-98059, a selective inhibitor of the mitogen-activated protein kinase kinase-1 (MEK-1)–ERK-1/2 pathway, abolished ERK-1/2 and NF-κB activation and ETB upregulation. NF-κB inhibition also prevented RLX-mediated ETB stimulation. In NF-κB-luciferase reporter assays, we demonstrated complete inhibition of RLX-induced NF-κB activation in cells transfected with dominant-negative Raf-1, MEK-1, or ERK-1/2 constructs, whereas dominant-negative Ras had no effect. In rat aorta and mesenteric artery, RLX pretreatment, in an ETB-dependent fashion, mitigated the maximum contractile response to endothelin-1, by 38±4% and 43±6%, and the endothelin-1 sensitivity (−log[EC50]: aorta, 8.2±0.2 for vehicle versus 7.2±0.2 for RLX; mesenteric artery, 8.0±0.2 for vehicle versus 7.1±0.1 for RLX). RLX pretreatment augmented the dilator effect of the ETB agonist endothelin-3 by 100±8% and 133±13%. In conclusion, RLX stimulates endothelial and epithelial ETB via a Ras-independent Raf-1–MEK-1–ERK-1/2 pathway that activates NF-κB. On vascular smooth muscle cells, ETB, a contributor to endothelin-mediated vasoconstriction, remains unaffected. This renders RLX a functional endothelin-1 antagonist.
The insulin-related peptide relaxin (RLX) has long been regarded as a central hormone of human pregnancy that contributes to changes in connective tissue composition and to the regulation of implantation, myometrial activity, and labor.1 On the other hand, we have recently found that RLX is constitutively expressed in human cardiovascular tissues and that myocardial gene expression of RLX is remarkably upregulated in human congestive heart failure.2
What renders RLX particularly intriguing as a neurohumoral player and as a potential new therapeutic tool is its antagonism vis-à-vis endothelin-1 (ET-1), one of the most potent vasoconstrictors and a powerful mediator of salt retention, fibroproliferation, and myocardial remodeling.3 We have shown that RLX induces upregulation of the endothelial endothelin type-B receptor (ETB).2 This receptor subtype stimulates endothelium-dependent vasodilation via NO and prostacyclin as well as clearance of ET-1.3 Moreover, RLX, via this ETB mechanism, inhibits stimulated ET-1 release in endothelial cells and may cause the inverse correlation between circulating RLX and ET-1 in patients with end-stage heart failure.2
Given the obvious importance of the RLX-induced ETB upregulation, it is crucial to know whether this phenomenon is restricted to endothelial cells only, because ETB is also found on epithelial and vascular smooth muscle cells (VSMCs).3
Additionally, it would be of particular interest to investigate the ensuing signaling cascades. Because other members of the insulin family represent strong activators of extracellular signal–regulated kinases-1/2 (ERK-1/2)4,5 and of nuclear factor-κB (NF-κB, a transcription factor),6 we hypothesized that ERK-1/2 and NF-κB were involved in RLX actions.
In the present study, we investigated mechanisms, abundance, and functional consequences of RLX-dependent ETB upregulation in human endothelial, epithelial, and VSMCs.
Materials and Methods
Primary human umbilical vein endothelial cells (HUVECs) were prepared according to standard procedures7 and grown in MCDB-131 medium (GIBCO) supplemented with 1% l-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin, 15% FCS, 1 ng/mL basic fibroblast growth factor, 0.1 ng/mL epidermal growth factor, 5 U/mL heparin, 1 μg/mL hydrocortisone, and 0.4% endothelial cell growth supplement.
For primary human VSMCs (hVSMCs) prepared from mammary arteries,8 we used DMEM (GIBCO) with 15% FCS, 100 U/mL penicillin, and 100 μg/mL streptomycin.
HeLa cells and bovine aortic endothelial cells (BAECs) were obtained from American Type Culture Collection and grown in RPMI (GIBCO) supplemented with 10% FCS, 100 U/mL penicillin, and 100 μg/mL streptomycin.
In all cells, FCS was lowered to 2% at 3 hours before the experiments. All cells were grown to subconfluence in a humidified 5% CO2 atmosphere and used at passages 3 to 4. The primary cultures showed a purity of at least 95% as determined by immunostaining with anti–von Willebrand factor (HUVEC) and anti–smooth muscle actin (hVSMC) antibodies (Dako).
Western Blot Analysis
This analysis has been described previously.2 For primary antibodies (dilution 1:2500), we used phospho-p44/42 (activated ERK-1/2), p44/42 (ERK-1/2), phospho-p38, and p38 mitogen-activated protein (MAP) kinase antibodies (Cell Signaling), ETB, and actin antibodies (Research Diagnostics).
Electrophoretic Mobility Shift Assays
Preparation of nuclear extracts and electrophoretic mobility shift assays were conducted as explained previously.2
We used an H-2K oligo probe containing the NF-κB recognition site (5′-GATCCCTGTGGAGATGGGGAATCCCCAGCCCTC-3′). We also performed supershift assays using monoclonal antibodies (Santa Cruz) against p65, c-Rel, p50, Rel-B, and p52.
RNA analysis has been described in detail.2 Polymerase chain reaction (PCR) amplification of single-stranded cDNA was performed using primer pairs specific for human GAPDH (housekeeping gene) (5′ primer, 5′-TGAAGGTCGGAGTCAACGGATTTGGT-3′; 3′ primer, 5′-CATGTGGGCCATGAGGTCCACCAC-3′; Clontech), human ETB (5′ primer, 5′-TTGGAGCTGAGATGTGTAAGC-3′; 3′ primer, 5′-CAGTGAAGCCATGTTGATACC-3′), and human ETA (5′ primer, 5′-CTGGCACTGGTTGGATGTGTA-3′; 3′ primer, 5′-GCTTCAG-GAATGGCCAGGAT-3′). Subsequently, Southern blot hybridization was conducted for quantification of the amplified sequences using radioactively labeled oligonucleotides specific for human GAPDH, human ETB, and human ETA, as described.2 Autoradiographs were quantified using ImageMaster 1D Prime software (Pharmacia Biotech).
Transient Transfections and Reporter Assays
HeLa cells and BAECs were seeded in tissue culture plates and grown to 70% confluence. Using cationic liposomes (Qiagen), we cotransfected cells with the NF-κB-luciferase reporter gene construct (pGL2 vector) containing the NF-κB recognition motif and the following dominant-negative (DN) expression constructs: DNRasN179 (pCMV5 vector), DNRaf-110 (pRSV), DNMEK-111 (pCEP4L; MEK-1 indicates mitogen-activated protein kinase kinase-1), DNERK-112 (pCEP4L), and DNERK-212 (pCEP4L). Transfections of 0.5 μg cDNA each were carried out according to the manufacturer’s instructions (Qiagen). After 6 hours of incubation at 37°C, cells were maintained in RPMI for 24 hours. Thereafter, experiments were performed, cells were harvested for measurement of transactivation activity, and luciferase activity was assayed in a β-counter, as recommended in the manufacturer’s instructions (Promega). Data were normalized to transfection efficiency determined by cotransfection with the pSV-β-galactosidase vector (Promega) using a spectrophotometric assay. The HeLa cells and BAECs exhibited no relevant endogenous β-galactosidase activity. The number of experiments was four in all cases.
Membrane preparation and 125I-ET-1 saturation binding were performed as previously described.13 Nonspecific binding determined by coincubation with 100 nmol/L unlabeled ET-1 was <15% of total binding at all concentrations used. In endothelial cells, binding of 150 pmol/L 125I-ET-1 was displaced by the ETB-selective antagonist A-192621 (−log[IC50]=9.0±0.04, n=3), whereas the ETA-selective antagonist ABT-627 showed no displacement up to 10 μmol/L. This confirmed the commonly accepted notion that no relevant ETA population resided on these endothelial cells.3,14
Male Wistar rats (200 to 250 g body weight, Charles River Laboratories, Sulzfeld, Germany) were used for these experiments, which were approved by the institutional animal care and use committee. Thoracic aortas and renal and mesenteric arteries were cleaned, cut into strips, and mounted under 10-mN resting tension in an organ bath containing oxygenated Krebs-Henseleit buffer (37°C, pH 7.35 to 7.45). Isometric tension was recorded by means of a force transducer (Scientific Instruments). After a 120-minute equilibration period, preparations were exposed to a KCl (60 mmol/L) salt solution to assess maximum contractility. The presence of an intact endothelium was then verified by the occurrence of significant relaxations in response to acetylcholine (0.01 to 1 μmol/L) after precontraction with phenylephrine (1 μmol/L). We chose the following protocol (n=4 for each group): groups 1, 3, 5, 7, and 9 were treated with 1 nmol/L human RLX for 6 hours; groups 2, 4, 6, 8, and 10 received vehicle alone. Thereafter, the bathing medium was completely exchanged in all groups. In groups 1 and 2, we elicited cumulative dose-response curves to ET-1 (10−11 to 10−6 mol/L). In groups 3 and 4, these experiments were performed in the presence of the ETB-selective antagonist A-19262115 (500 nmol/L). During the 6-hour incubation period, groups 5 and 6 and groups 7 and 8 additionally received 2-[2′-amino-3′-methoxyphenyl]-oxanaphtalen-4-one (PD-98059, 50 μmol/L), a selective inhibitor of the MEK-1/2–ERK-1/2 pathway,16 and cycloheximide (CHX, 10 μg/mL), a protein synthesis inhibitor, respectively. Finally, groups 9 and 10 were mechanically endothelium-denuded before the experiments.
In distinct vessel preparations, an identical experimental protocol was performed to record cumulative dose-response curves to ET-3 (10−12 to 10−8 mol/L) after precontraction with phenylephrine (1 μmol/L).
Tension data were normalized to the KCl or phenylephrine responses.
Recombinant human RLX H2 was from Immundiagnostik. Angiotensin II, ET-1, ET-3, phorbol myristate acetate (PMA), PD-98059, tumor necrosis factor-α (TNF-α), CHX, and N-α-tosyl-l-phenylalanyl chloromethyl ketone (TPCK) were purchased from Sigma Chemical Co. A-192621 and ABT-627 were gifts from Abbott Laboratories (Abbott Park, Ill).
Data are presented as mean±SEM. A value of P<0.05 was regarded as significant. Densitometric and transfection data were analyzed using Kruskal-Wallis ANOVA on ranks. Dose-response curves were compared using a nonparametric ANOVA for repeated measures. A multiple-comparison procedure with Bonferroni-Holm adjustment of the P value was performed after global testing.
ETB Expression, ERK-1/2, and NF-κB Activation
We investigated the effect of RLX (0.1, 1, and 10 nmol/L) on ETB gene expression in the different cell types. RLX, in a concentration-dependent fashion reaching a plateau at 5 nmol/L, induced marked increases in ETB mRNA levels in HeLa cells and HUVECs, which were detectable as early as 2 hours after RLX and reached a plateau at 4 and 8 hours of exposure (data shown for 4 hours and 1 nmol/L RLX in Figure 1 and the Table). In contrast, no change of ETB gene expression in response to RLX was found in hVSMCs. Control experiments showed ETB-stimulating effects of PMA (50 nmol/L) and angiotensin II (100 nmol/L) in HeLa cells and HUVECs and a decrease in ETB mRNA after exposure to angiotensin II in hVSMCs.
Binding studies with 125I-ET-1 using membranes from endothelial cells (Figure 2) revealed a marked increase in the ETB receptor density Bmax after a 6-hour exposure of the cells to 1 nmol/L RLX (Bmax, 200±36 fmol/mg membrane protein for vehicle versus 611±56 fmol/mg membrane protein for RLX; n=3), whereas ETB binding affinity remained unchanged (KD, 126±18 pmol/L for vehicle versus 139±23 pmol/L for RLX).
In HeLa cells and HUVECs, RLX, like PMA and angiotensin II, was found to evoke activation of ERK-1/2 and enhanced DNA binding of the transcription factor NF-κB at 45, 60, and 90 minutes after RLX exposure (data shown for 60 minutes in Figure 1 and in the Table). In hVSMCs, PMA and angiotensin II, but not RLX, evoked substantial ERK-1/2 activation, whereas all three substances increased nuclear binding of NF-κB.
In hVSMCs, but not in HeLa cells or HUVECs, RLX induced p38 activation at 45, 60, and 90 minutes after exposure (data shown for 45 minutes in Figure 1). RLX affected ETA gene expression in none of the studied cell types (not shown).
Supershift experiments in HeLa cells (Figure 3) revealed substantial contribution of the p50 homodimer to the NF-κB activation evoked by RLX and PMA.
Inhibition of ERK-1/2 and NF-κB
Causal implication of the ERK-1/2 cascade in the RLX effect was confirmed by the finding that PD-98059 (50 μmol/L), a selective inhibitor of the MEK-1/2–ERK-1/2 pathway,16 completely suppressed translocation of NF-κB and ETB upregulation in HeLa cells and HUVECs in response to RLX and angiotensin II (Figure 4 and Table). Effects of PMA on ERK and NF-κB activation and on ETB expression were partially antagonized by PD-98059, with the NF-κB response in HeLa cells being unaffected. Furthermore, TPCK (25 μmol/L), which prevents NF-κB activation by inhibiting degradation of different IκB proteins,17 abrogated not only the enhanced DNA binding of NF-κB but also the increase in ETB gene expression in response to RLX. Thus, NF-κB activation was critical to this RLX action. Similarly, the effects of PMA and angiotensin II on NF-κB activation and ETB regulation were also abolished.
We first ensured that BAECs also exhibited RLX-induced upregulation of ETB protein, which was sensitive to ERK-1/2 inhibition by PD-98059 and NF-κB inhibition by TPCK (Figure 5). As shown in Figure 5 for HeLa cells and for BAECs, treatment with 1 nmol/L RLX for 45 minutes induced a markedly increased transactivation activity in control cells cotransfected with NF-κB-luciferase reporter construct and empty vector. This RLX-related stimulation of NF-κB activity was not affected in cells transfected with DNRas but was completely prevented in cells transiently expressing DNRaf-1, DNMEK-1, or DNERK-1 plus DNERK-2. We used TNF-α (10 ng/mL) as a control substance because its mode of NF-κB activation is independent of the MEK-1–ERK-1/2 pathway.18 Specificity of the effects evoked by the DN kinase constructs was confirmed by the fact that these transfections did not affect the TNF-α–induced NF-κB activation.
Functional Relevance of RLX-Induced ETB Stimulation
In all preparations (rat renal artery, thoracic aorta, and mesenteric artery), maximum tensions to KCl (mean obtained was 16±1.2×103 N/m2 in aortas, 15±1.3×103 N/m2 in mesenteric arteries, and 15±1.0×103 N/m2 in renal arteries) and acetylcholine-induced dilator responses in phenylephrine-precontracted preparations (−72±7% [aorta], −68±7% [mesenteric artery], and −72±6% [renal artery] of the phenylephrine response, determined at 1 μmol/L acetylcholine) were comparable among groups 1 to 10. Likewise, responses to phenylephrine did not differ between groups 1 to 10.
Figure 6 shows data obtained in aortic preparations that are representative of all three anatomic sites. Pretreatment with RLX (1 nmol/L) significantly mitigated the maximum contractile response to ET-1 by 38±4% (aorta), 43±6% (mesenteric artery), and 33±7% (renal artery) as well as the ET-1 sensitivity (−log [EC50]: aorta, 8.2±0.2 for vehicle pretreatment versus 7.2±0.2 for RLX; mesenteric artery, 8.0±0.2 versus 7.1±0.1, respectively; and renal artery, 7.8±0.2 versus 7.2±0.1, respectively). Moreover, the effect of the ETB-selective agonist ET-3 was significantly augmented: by 100±8% (aorta), 133±13% (mesenteric artery), and 73±8% (renal artery) at maximum ET-3 concentration. In presence of the ETB-selective antagonist A-192621 (500 nmol/L), we found no difference between the RLX or vehicle-pretreated groups with respect to ET-1, and the ET-3 response was completely abolished. In endothelium-denuded strips, the ET responses of RLX- and vehicle-pretreated preparations were also indistinguishable. Furthermore, the effects of RLX could be abolished by treatment with the MEK-1/2–ERK-1/2 inhibitor PD-59098 (50 μmol/L) or with the protein synthesis inhibitor CHX (10 μg/mL).
Known for many years as a hormone of human reproduction,1 RLX has attracted more attention since the 1990s, when new studies revealed the remarkable pleiotropy of this mediator.19 Meanwhile, implication of RLX in the neurohumoral response taking place in human congestive heart failure has been demonstrated.2 In the present study, we offer first evidence that one crucial effect of RLX, ie, the upregulation of ETB receptors, (1) is obviously limited to endothelial and epithelial cells but is not present in VSMCs and (2) is mediated via Ras-independent activation of the Raf-1–MEK-1–ERK-1/2 pathway and the transcription factor NF-κB(p50).
The latter conclusion is based on the facts that (1) both ERK-1/2 and NF-κB are activated on RLX exposure in HeLa cells and HUVECs, (2) functional abrogation of the ERK-1/2 and the NF-κB pathways by PD-98059 and TPCK, respectively, completely prevents ETB upregulation, and (3) the RLX-induced stimulation of NF-κB transactivation activity is completely abolished in cells transiently transfected with DNRaf-1, DNMEK-1, or DNERK-1/2 constructs.
The finding that a cell type–dependent restriction of the RLX effect exists is of great relevance because ETB receptors fulfill distinct functions in endothelial and epithelial as opposed to VSMCs3,14: whereas endothelial/epithelial ETB expedites the release of vasodilatory and anti-fibroproliferative mediators (NO and prostacyclin) as well as clearance of ET-1 (thereby providing a physiological negative feedback within the ET-1 system), ETB on VSMCs contributes to the vasoconstrictor profile of ET-1. In human congestive heart failure, the ETB contribution to the overall vasoconstrictor effect of ET-1 is even increased.20 In other words, the selective stimulation of endothelial/epithelial ETB expression renders RLX a functional ET-1 antagonist. Accordingly, we demonstrated, in three different vessels, an RLX-dependent augmentation of the ETB-mediated vasodilation along with a mitigation of the vasopressor effect of ET-1, which was clearly abolished after ETB blockade. The present results confirm the functional relevance of RLX-induced ETB upregulation, which is also supported by reports on a RLX-induced functionally ETB-dependent renal vasodilation in pregnant rats.21 This functional evidence, ie, the fact that the net RLX effect is vasodilation, is particularly important because NF-κB has also been reported to upregulate gene expression of the vasoconstrictors ET-1 and angiotensin II in certain cell types.22 In a recent study, we have found that RLX is capable of inhibiting angiotensin II–mediated as well as pressure-induced ET-1 stimulation in pulmonary artery endothelial cells by an ETB-dependent mechanism2; in detail, this RLX-related mitigation of ET-1 stimulation was attributed to massive ETB upregulation, which more than outweighed, by increased clearance, the slight to moderate elevation by RLX of ET-1 gene expression.
The question arises as to which mechanism(s) might underlie the different regulation by RLX of the ETB gene in endothelial/epithelial compared with smooth muscle cells. Our results favor the view that the activation by RLX of distinct kinase cascades, ie, ERK-1/2 in HeLa cells and HUVECs, but p38 in hVSMCs, accounts for this difference. Both kinase pathways may target NF-κB,23 but our data indicate the necessity of concerted NF-κB and ERK-1/2 activation to stimulate ETB expression. This, in turn, suggests critical involvement of distinct kinases downstream from ERK-1/2, ie, p90 ribosomal S6 kinases, a point that must be addressed in future studies.
Our observation that the ERK-1/2–NF-κB pathway is involved in RLX signaling represents one of the first investigations at this subreceptor level of RLX signal transduction. Until now, it was known that RLX links tyrosine phosphorylation to phosphodiesterase and adenylyl cyclase activity.24 The present findings are in good accordance with a report by Zhang et al,25 who, in human endometrial stromal cells and in the human monocytic cell line THP-1, also demonstrated RLX-induced ERK-1/2 activation. Most likely, all these effects are mediated via the recently identified RLX receptors26 LGR 7 and LGR 8, which are G protein–coupled 7-transmembrane-domain receptors.
As to the functional relevance of RLX-induced vasodilation in humans, it has recently been shown that exogenous RLX represents a highly potent dilator in the human systemic circulation.27,28 Thus, the importance of RLX as a potential therapeutic tool in humans seems a matter of fact. Moreover, the pivotal role of RLX in the circulatory changes occurring during pregnancy has already been evidenced, at least in rats21: endogenous RLX mediates both the decrease in plasma osmolality and the increase in renal plasma flow. Whether endogenous RLX affects basal vessel tone in men and nonpregnant women remains to be investigated. It may be misleading to judge from the comparably low circulating RLX levels (<1 pmol/L in men and postmenopausal women,2 ≈15 to 20 pmol/L in menstruating women,29 and ≈150 pmol/L during pregnancy30) because plasma levels are often poor indicators of locally effective concentrations.
In control experiments with PMA and angiotensin II, we reproduced several well-established findings: PMA as well as angiotensin II have repeatedly been reported to activate both ERK-1/2 and NF-κB pathways.23,31,32 On the other hand, PMA has also been found to stimulate p38 and c-Jun N-terminal kinase cascades and activator protein-1, a transcription factor.23 Stimulation of p38 kinases by PMA and subsequent p38-mediated activation of NF-κB may account for the fact that the ERK-1/2 inhibitor PD-98059 did not prevent (in HeLa cells) or only partially prevented (in HUVECs) the PMA-induced NF-κB activation. The fact that PD-98059 inhibited PMA-related ERK-1/2 stimulation only partially in HeLa cells and HUVECs is most likely a dose-related problem because PMA is a very strong kinase inductor. Regarding the regulation of ETB, there are some reports in accordance with our finding of angiotensin II–induced receptor downregulation in hVSMCs,33 whereas the precise role of angiotensin II in the regulation of endothelial ET-1 receptors appears unknown from the literature. In the experiments shown in the present study, angiotensin II stimulated both epithelial and endothelial ETB via ERK-1/2 and NF-κB.
As to the limitations of the present study, we mostly used tumorigenic epithelial cells (HeLa cells) and HUVECs, both of which are not necessarily representative of normal epithelial and endothelial cells. However, the results obtained in HUVECs could be confirmed in BAECs. In our vessel preparations, we further demonstrated complete endothelium dependence of the vasodilatory RLX effect. In addition, the existence of a RLX-ERK axis has been confirmed by others for human endometrial cells25 and human uterine fibroblasts,34 ie, for nontumorigenic cells.
We conclude that RLX promotes endothelial and epithelial, but not vascular smooth muscle, ETB via a Ras-independent Raf-1–MEK-1–ERK-1/2 kinase cascade that activates NF-κB(p50). One of the consequences is significant mitigation of the vasopressor effect of ET-1. This finding may have an impact on future therapeutic strategies exploiting RLX in human heart failure or other states of disease.
This study was supported by an institutional grant to Dr Dschietzig (Charité, 2001-10743).
Original received July 19, 2002; revision received November 27, 2002; accepted November 27, 2002.
Dschietzig T, Richter C, Bartsch C, Laule M, Armbruster FP, Baumann G, Stangl K. The pregnancy hormone relaxin is a player in human heart failure. FASEB J. 2001; 15: 2187–2195.
Bertrand F, Atfi A, Cadoret A, L’Allemain G, Robin H, Lascols O, Capeau J, Cherqui G. A role for nuclear factor κB in the antiapoptotic function of insulin. J Biol Chem. 1998; 271: 2931–2938.
Vaccaro PS, Joseph LB, Titterington L, Stephens RE. Methods for the initiation and maintenance of human endothelial cell culture. Vasc Surg. 1987; 21: 391–400.
Chamley-Campbell J, Campbell GR, Ross R. The smooth muscle cell in culture. Physiol Rev. 1979; 59: 1–61.
Feig LA, Cooper GM. Inhibition of NIH 3T3 cell proliferation by a mutant ras protein with preferential affinity for GDP. Mol Cell Biol. 1988; 8: 3235–3243.
Cai H, Erhardt P, Troppmair J, Diaz-Meco MT, Sithanandam G, Rapp UR, Moscat J, Cooper GM. Hydrolysis of phosphatidylcholine couples Ras to activation of Raf protein kinase during mitogenic signal transduction. Mol Cell Biol. 1993; 13: 7645–7651.
Robbins DJ, Zhen E, Owaki H, Vanderbilt CA, Ebert D, Geppert TD, Cobb MH. Regulation and properties of extracellular signal-regulated protein kinases 1 and 2 in vitro. J Biol Chem. 1993; 268: 5097–5106.
Elshourbagy NA, Korman DR, Wu HL, Sylvester DR, Lee JA, Nuthalaganti P, Bergsma DJ, Kumar CS, Nambi P. Molecular characterization and regulation of the human endothelin receptors. J Biol Chem. 1993; 268: 3873–3879.
Von Geldern TW, Tasker AS, Sorensen BK, Winn M, Szczepankiewicz BG, Dixon DB, Chiou WJ, Wang L, Wessale JL, Adler A, Marsh KC, Nguyen B, Opgenorth TJ. Pyrrolidine-3-carboxylic acids as endothelin antagonists, 4: side chain conformational restriction leads to ETB selectivity. J Med Chem. 1999; 42: 3668–3678.
Dudley DT, Pang L, Decker SJ, Bridges AJ, Saltiel AR. A synthetic inhibitor of the mitogen-activated protein kinase cascade. Proc Natl Acad Sci U S A. 1995; 92: 7686–7689.
Kim H, Lee HS, Chang KT, Ko TH, Baek KJ, Kwon NS. Chloromethyl ketones block induction of nitric oxide synthase in murine macrophages by preventing activation of nuclear factor-κB. J Immunol. 1995; 154: 4741–4748.
Baud V, Liu ZG, Bennett B, Suzuki N, Xia Y, Karin M. Signaling by proinflammatory cytokines: oligomerization of TRAF2 and TRAF6 is sufficient for JNK and IKK activation and target gene induction via an amino-terminal effector domain. Genes Dev. 1999; 13: 1297–1308.
Cowburn PJ, Cleland JG, McArthur JD, MacLean MR, McMurray JJ, Dargie HJ, Morton JJ. Endothelin B receptors are functionally important in mediating vasoconstriction in the systemic circulation in patients with left ventricular systolic dysfunction. J Am Coll Cardiol. 1999; 33: 932–938.
Kyriakis JM, Avruch J. Mammalian mitogen-activated protein kinase signal transduction pathways activated by stress and inflammation. Physiol Rev. 2001; 81: 545–605.
Bartsch O, Bartlick B, Ivell R. Relaxin signalling links tyrosine phosphorylation to phosphodiesterase and adenylyl cyclase activity. Mol Hum Reprod. 2001; 9: 799–809.
Hsu SY, Nakabayashi K, Nishi S, Kumagai J, Kudo M, Sherwood OD, Hsueh AJ. Activation of orphan receptors by the hormone relaxin. Science. 2002; 295: 671–674.
Zsebo KM. Development of recombinant human relaxin: clinical experience in scleroderma . . . and beyond. In: Tregear GW, Ivell R, Bathgate RA, Wade JD, eds. Proceedings of the 3rd International Conference on Relaxin and Related Peptides, Broome, Australia. Dordrecht, the Netherlands: Kluwer Academic Publishers; 2000: 273–281.
Fisher C, MacLean M, Morecroft I, Seed A, Johnston F, Hillier C, McMurray J. Is the pregnancy hormone relaxin also a vasodilator peptide secreted by the heart? Circulation. 2002; 106: 292–295.
Purcell NH, Tang G, Yu C, Mercurio F, DiDonato JA, Lin A. Activation of NF-κB is required for hypertrophic growth of primary rat neonatal ventricular cardiomyocytes. Proc Natl Acad Sci U S A. 2001; 98: 6668–6673.
Kim KW, Kim SH, Lee EY, Kim ND, Chung BS, Kang CD. Extracellular signal-regulated kinase/90-KDA ribosomal S6 kinase/nuclear factor-κB pathway mediates phorbol 12-myristate 13-acetate-induced megakaryocytic differentiation of K562 cells. J Biol Chem. 2001; 276: 13186–13191.