Activation of multiple mitogen-activated protein kinase signal transduction pathways by the endothelin B receptor requires the cytoplasmic tail.

Endothelin is a 21-amino acid peptide with remarkably diverse biological properties, including potent vasoconstriction, induction of mitogenesis, and a role in the development of blood vessels. In the present study, stimulation of the endothelin B receptor was found to activate three distinct mitogen-activated protein kinase signal transduction pathways, the extracellular regulated kinase (ERK) 2, c-Jun N-terminal kinase 1 (JNK), and p38 kinase. These mitogen-activated protein kinase isozymes are thought to mediate very different biological outcomes, suggesting that the observed pattern of kinases activation may be important for the diverse biological properties of endothelin. The cytoplasmic tail of the endothelin B receptor was found to be required for activation of all three mitogen-activated protein kinases and stimulation of intracellular calcium levels. An endothelin B receptor truncated at the C-terminal tail was not able to stimulate the mitogen-activated protein kinases or increase cytosolic free calcium. Furthermore, ectopic expression of the cytoplasmic tail attenuated signaling through the wild type receptor. The observed ERK activation appeared to be mediated by heterotrimeric G proteins, since ectopic expression of a transducin α-subunit inhibited endothelin-stimulated ERK activation. The data suggest that the cytosolic tail of the endothelin B receptor is involved in calcium mobilization and mitogen-activated protein kinase activation via a G protein-dependent mechanism.

The putative ET receptor topology includes three extracellular domains, three intracellular cytosolic loops, and a cytoplasmic C-terminal tail, separated by seven highly hydrophobic regions thought to span the lipid bilayer. Ligand binding and specificity is conferred by the conformation of the extracellular portions of the receptor, whereas the intracellular domains physically interact with heterotrimeric G proteins to initiate signal transduction events. Receptor-G protein complexes are thought to form the high affinity ligand binding state of the receptor (8). For example, constitutively activated adrenergic and muscarinic receptor mutants have increased affinity for agonist with little effect on antagonist binding, supporting the view that the receptor-G protein complex is the high affinity binding complex (9,10).
Endothelin receptor mutagenesis studies suggest that residues within extracellular loop two are important for ligand binding. The intracellular domains of the receptor, thought to play a role in G protein coupling, are located within cytosolic loop three and the C-terminal tail. Mutations of the ETA receptor tail decrease the high affinity binding state of the receptor, suggesting that the C-terminal tail stabilizes the receptor-G protein complex that forms the high affinity ligand binding state (11)(12)(13)(14)(15).
The role of endothelins in disease is highlighted by the finding that a mutated ETB receptor has been mapped to a recessive susceptibility locus associated with Hirschsprung's disease (16 -18). Mice lacking an ET-1 gene display severe malformations in large blood vessels, further underscoring the importance of ET during development (19). Endothelin has also been shown to be a mitogenic agonist in different cell types, suggesting that the role of endothelin in development of blood vessels may involve developmentally regulated mitogenesis (20,21). The signal transduction mechanisms that mediate the mitogenic properties of endothelin include the activation of phosphorylation events that ultimately direct gene transcription. The signaling pathway by which ET-1 promotes cell proliferation includes activation of intracellular kinase cascades and transcription factor stimulation. For example, ET-1 stimulation results in activation of c-Src and the mitogen-activated protein (MAP) kinases (20 -22). In addition, ET-1 stimulates phosphoinositide turnover resulting in increased cytosolic free calcium (3)(4)(5).
The mechanism of seven transmembrane receptor activation of the ERK MAP kinases is not precisely defined, but it appears that the G protein ␤/␥-subunits are able to stimulate MAP kinase activation. In COS cell transient transfection experiments, ␤/␥-subunits activate ERK2 in a Ras-dependent manner in the absence of agonist stimulation, and JNK is also stimulated by ␤/␥ (23)(24)(25)(26). Furthermore, both the ␣2 adrenergic receptor and ␤/␥-subunits are able to induce tyrosine phosphorylation of Shc by an unidentified tyrosine kinase (27). The PYK2 tyrosine kinase is activated by G protein-coupled receptors, such as the bradykinin and muscarinic receptors, and results in Shc tyrosine phosphorylation and MAP kinase activation in neuronal cells (28). GPCR activation of the ERK MAP kinases is Ras-dependent, since dominant negative Ras mutants block activation of agonist-stimulated ERK activation, and activated mutants stimulate this cascade. In contrast, activation of the JNK MAP kinase pathway appears to be mediated by members of the Rho subfamily of small GTPbinding proteins. In addition to activating the PAK65 kinase, members of the Rho subfamily, Rac1 and Cdc42Hs, appear to activate the JNK MAP kinase cascade. Whereas dominant negative Rac and Cdc42 alleles block agonist-stimulated GPCR activation of the JNK cascade, and activated mutants stimulate this pathway, neither Rac1 nor Cdc42 appear to interact with the ERK arm of the MAP kinase cascade (29 -32).
The putative structure of G protein-coupled receptors includes seven hydrophobic membrane spanning regions that form three extracellular loops involved in ligand binding, and three intracellular loops that are important for activation of intracellular signaling pathways. In many GPCRs the third intracellular loop and the cytoplasmic tail have been demonstrated to be important for receptor signaling functions. For example, the third intracellular loop and C-terminal tail of the adrenergic and muscarinic receptors are required for effective receptor-G protein interactions (33)(34)(35). Mutagenesis studies of the endothelin A receptor indicate that portions of the third loop and cytoplasmic tail are critical for increases in cytosolic calcium concentration, suggesting that these regions of the receptor are involved in G protein coupling (36).
In order to learn more about the signal transduction mechanisms employed by the endothelin B receptor (ETB), we have examined receptor stimulation of MAP kinases and calcium mobilization. We report here the ability of the ETB receptor to stimulate the ERK, JNK, and p38 MAP kinase pathways and have identified a region of the ETB receptor required for activation of these kinase pathways. Furthermore, ETB receptorstimulated calcium mobilization and MAP kinase activation are both dependent on the cytoplasmic tail of the ETB receptor, suggesting a common G protein-dependent mechanism for both signaling functions.

MATERIALS AND METHODS
Cell Culture and Transfection-COS cells were grown in Dulbecco's modified Eagle's medium, 10% fetal bovine serum. Cells were transfected at 25% confluence using lipofectamine (20 g) (Life Technologies, Inc.) and 1 g of DNA. The media were replaced with Dulbecco's modified Eagle's medium, 10% fetal bovine serum after 16 -18 h, and the cells were incubated for 36 h. The transfected cells were serumstarved overnight in Dulbecco's modified Eagle's medium, 0.1% fetal bovine serum prior to stimulation and immunoprecipitation. The rat ETB receptor cDNA was generously provided by Dr. Shigetada Nakanishi and has been described previously (4). The ETB DNA was subcloned into the pCDNA 3 vector and epitope tagged by fusing oligonucleotides encoding the flag epitope (Kodak) (recognized by the M2 antibody) in-frame at the 3Ј-terminus of the coding sequence, with a stop codon inserted after the last flag codon to form ETB-flag. The C-terminal deletion mutant was engineered by deleting the ETB receptor at the internal EcoRI site which resulted in the loss of the Cterminal 64 amino acids from Asn 377 . The deletion mutant was similarly tagged at the C terminus with the flag epitope to form ETB-⌬N377-flag. The ERK2 MAP kinase was tagged with the Myc epitope at the N terminus as described previously (37). The JNK1 and p38 MAP kinases were flag-tagged at the N terminus as described previously (38). To construct the ETB-tail fragment, polymerase chain reaction primers were designed to encompass codons 108 -153 of the rat ETB receptor with BamHI sites incorporated into both primers. The resulting 0.45-kilobase pair polymerase chain reaction fragment was cloned into the pKH3 vector, and the plasmid was sequenced. The polymerase chain reaction fragment was fused in-frame with DNA sequences encoding a triple repeat of the HA1 epitope on the 5Ј end of the construct.
The resulting plasmid, pKH3-ETB-tail, expressed amino acids Thr 363 -Ser 509 of the rat ETB receptor with a triple HA1 tag fused onto the N terminus. A 1.2-kilobase pair HindIII fragment containing the entire open reading frame of the bovine retina transducin ␣-subunit was cloned into pCDNA3, and the resulting plasmid, pCDNA3-␣ t , was used in transfection experiments. Expression of the ␣-subunit of transducin was monitored by Western blot using a G␣ t1 antibody (K-20, Santa Cruz).
Immunoprecipitation and Kinase Assay-COS cells ectopically expressing the indicated plasmids were lysed in Nonidet P-40 IPB (1% Nonidet P-40, 10 mM Hepes, 2 mM EDTA, 50 mN NaF, 0.1% ␤-mercaptoethanol, 1% aprotinin) and immunoprecipitated with 3 g of the 9E10 (ERK2) or M2 (JNK1, p38) antibodies. The washed immunoprecipitates were incubated in the presence of [ 32 P]ATP and bacterially expressed substrate fusion proteins as indicated. The phosphorylated substrates were separated on SDS gels that were subsequently dried and exposed to film.
Immunofluorescence-The wild type and ⌬N377 ETB receptors were transfected into COS cells as described above. The cells were fixed with paraformaldehyde and permeabilized with methanol prior to incubation with the M2 antibody. The receptor-antibody complexes were incubated with a CY3 goat anti-mouse secondary antibody (Jackson Laboratories), and the receptors were visualized under fluorescence microscopy.
Cytosolic Free Calcium Measurements-Measurements of cytosolic free calcium were performed by loading the cells with Fura-2 (50 M).
The cells were examined under a Nikon microscope which is part of an IMAGE-1/FL quantitative fluorescence PC imaging system (Universal Imaging Inc., West Chester, PA) using a Hammamatsu C2400 SIT camera to acquire images. After acquiring background images for realtime background subtraction, ratio images of the 510-nm emissions resulting from alternate illumination with 340/380 nm light were acquired at a rate of 0.2 Hz using a 4-frame average from each wavelength. The fluorescence signal was monitored until a steady-state level was reached for at least 2 min prior to addition of endothelin. Following endothelin treatment, calcium ionophore (10 M ionomycin) was added to the bath, and the maximal fluorescence was recorded to compare and ensure equal dye loading among different cell populations. There were no significant differences in apparent dye loading, judging from the maximal fluorescent ratio values (R max ) obtained in this manner. Furthermore, there were no significant differences in resting levels of calcium (F o ) between the three groups. In total 23 dishes with cells from four independent transfections were studied. Data were analyzed with Origin 4.0 using its' pClamp module to read the axotape binary data files and are expressed as stimulation over base-line values prior to endothelin addition.
Endothelin Binding Studies-COS cells were transfected with either the wild type ETB or ⌬N377-ETB receptors. Crude membranes were prepared from the cells 48 h post-transfection. Briefly, cells were scraped into hypotonic lysis buffer (10 mM Hepes, 7.5, 5 mM MgCl 2 , 2 mM KCl, 1% aprotinin, 1 mM phenylmethylsulfonyl fluoride). The postnuclear supernatant was centrifuged at 100,000 ϫ g, and the pellet was resuspended in hypotonic lysis buffer and stored at Ϫ70°C. The cell membranes (50 g) were incubated with varying concentrations of 125 I-endothelin 1 (DuPont NEN) for 3 h at 25°C in a total volume of 0.1 ml. The binding reaction was terminated by addition of 1 ml of ice-cold phosphate-buffered saline, and the membranes were trapped on glass filters (Whatman, GF/B) and counted in a gamma counter. Nonspecific binding was estimated by addition of the given dose of 125 I-endothelin in the presence of excess unlabeled endothelin (0.1 M).

Reconstitution of the Endothelin Signaling Pathway in COS
Cells-The signaling pathways utilized by the endothelin receptor are known to include phospholipase activation, inositol phosphate turnover, increases in cytosolic free calcium, and activation of protein kinases, including mitogen-activated protein (MAP) kinase (3)(4)(5)20). To further study the signaling mechanisms of the ETB receptor, transient transfection studies were undertaken in a cell system that lacked endogenous ETB receptors so that wild type and mutant ETB receptors could be systematically examined. COS cells were transfected with 1 g of ETB receptor and ERK2 or JNK1 MAP kinases or pCDNA3 vector, treated with or without endothelin (40 nM) for 10 min, and immune complex kinase assays were performed to examine the catalytic activity of each kinase. The ERK2 or JNK1 ki-nases were immunoprecipitated by means of their epitope tags (Myc for ERK2, flag for JNK1) which are recognized by the 9E10 and M2 antibodies, respectively. Endothelin-stimulated cells expressing the ETB receptor, but not a MAP kinase, had background levels of substrate phosphorylation indicating that the observed phosphorylation was attributable to the ectopically expressed MAP kinases. In contrast, cells expressing ETB receptor and ERK2 or JNK1 had a sharp increase in substrate phosphorylation in cells treated with endothelin ( Fig. 1). Cells expressing only the ERK2 or JNK1 kinase showed background levels of substrate phosphorylation following endothelin treatment confirming that COS cells do not have endogenous endothelin receptors (Fig. 1).
The C-terminal Tail of the ETB Receptor Is Required for Kinase Activation-The intracellular loops and C terminus of the ETB receptor are presumed to form the domain of the receptor that interacts physically with heterotrimeric G proteins, although the exact intracellular portion of the receptor that interacts with the G protein has not been clearly identified. In contrast to other GPCRs, the intracellular loops of the ETB receptor are significantly smaller. For example, loop 1 is approximately 11 amino acids, loop 2 is 21 amino acids, and loop three is 28 amino acids. In comparison, loop three of the muscarinic receptor, which is thought to participate in G protein binding, is approximately five times as large (39). The C-terminal tail of the ETB receptor is 53 amino acids, the largest intracellular domain of the receptor. We examined the requirements of the C-terminal tail in activation of the ERK2, JNK1, and p38 MAP kinases by truncating the receptor at Asn 377 , which results in a deletion of the C-terminal portion of the receptor. COS cells were transiently transfected with either the wild type or ⌬N377 ETB receptors and either ERK2, JNK1, or p38 expression vectors. Following endothelin treatment for 10 min, the kinases were immunoprecipitated by means of their respective epitope tags, and immune complex kinase assays were performed to examine their catalytic activity (Fig. 2). While endothelin treatment stimulated the kinase activity of all three MAP kinases in cells expressing wild type receptor, the ⌬N377 ETB receptor was not able to similarly activate these kinases ( Fig. 2A). This lack of kinase activation could not be attributed to varying levels of expression since the expression of the three kinases was essentially equal (Fig. 2B).
If the cytoplasmic tail of the ETB receptor is critical for coupling to MAP kinase activation, then expression of the tail fragment may be expected to interfere with coupling of the wild type ETB receptor to MAP kinase activation. Ectopic expression of minigenes encoding various intracellular domains of the ␣ adrenergic receptor has proven helpful in defining regions of the receptor involved in coupling to downstream signal trans- The cells were treated with or without endothelin-1 (40 nM) for 10 min prior to lysis and immunoprecipitation. The ERK2 or JNK1 kinases were immunoprecipitated with 9E10 (MT-ERK2) or M2 (flag-JNK1) antibodies, and catalytic activity was assayed by immune complex kinase assay using myelin basic protein (MBP) or glutathione S-transferase-Jun-  as in vitro substrates. In the first lane, cells were transfected with ETB and pCDNA3 but not ERK or JNK to verify that the observed kinase activation was attributed to ERK or JNK and not an endogenous kinase activity. Cells transfected with ETB and ERK2 or JNK1 and subsequently treated with endothelin resulted in catalytic activation of the kinases as demonstrated by phosphorylation of an exogenous substrate (myelin basic protein (MBP) or glutathione S-transferase (GST)-Jun). This result confirmed that the signaling molecules between the receptor and the MAP kinases were intact in COS cells. Furthermore, activation of the kinases was not observed in cells treated with ET that did not express the ETB receptor indicating the absence of endogenous ET receptors (last two lanes).
FIG . 2. Truncation of the endothelin B receptor C terminus inhibits MAP kinase activation. COS cells were transiently transfected with the indicated expression plasmids and treated with or without ET-1 prior to immunoprecipitation of the ERK2, JNK1, or p38 kinases as described in Fig. 1. A, the substrate phosphorylation is indicative of the catalytic activity of each kinase. The wild type ETB receptor activated the ERK2, JNK1, and p38 kinases as determined by an endothelin-dependent activation of kinase activity. The truncated ETB receptor was not able to activate any of the MAP kinases. The expression levels of the kinase were approximately equal as determined by Western blot in B. duction pathways (35). Therefore, a construct encoding the C-terminal tail of the ETB receptor, with an HA-1 epitope tag fused to the N terminus, was co-expressed in cells with the wild type ETB receptor to determine if this fragment was able to interfere with signaling through the receptor. Expression of the ETB tail fragment inhibited the ability of the wild type ETB receptor to activate ERK2 (Fig. 3A). The decrease in ERK activation by the receptor was correlated with increasing levels of expression of the ETB-tail fragment, as determined by Western blot with the 12CA5 antibody (Fig. 3B). Taken together the data suggest that the C-terminal portion of the ETB receptor is critically involved in effector coupling.
G Protein-dependent Coupling to ERK Activation-The current paradigm of GPCR activation of signal transduction pathways involves receptor activation of heterotrimeric G proteins. Evidence from several laboratories has confirmed the central role of the G ␤␥-subunits in GPCR activation of the ERK subgroup of MAP kinase (23)(24)(25)(26). Therefore, we initiated experiments to further define the role of heterotrimeric G proteindependent activation of ERK by the ETB receptor. Free G␤␥subunits are able to activate signaling pathways, and binding to their respective ␣ GDP -subunit is thought to quench the signaling capability of the ␤␥-subunits. Therefore, overexpression of ␣-subunits may be expected to inhibit G␤␥-mediated signal transduction events. Indeed, ectopic expression of the ␣-transducin subunit has been demonstrated to inhibit the signaling functions of the muscarinic receptor (41). To explore the role of heterotrimeric G proteins in the observed ETB receptor activation of MAP kinase, the ␣-subunit of transducin was co-expressed in cells with the wild type ETB receptor. Activation of ERK2 by the ETB receptor was significantly decreased in cells expressing ␣ transducin (Fig. 4A). Western blotting with a transducin antibody revealed increasing levels of the protein correlated with inhibition of ERK2 activation (Fig. 4B). Since expression of the G protein ␣-subunit significantly attenuated the ability of the ETB receptor to activate the ERK pathway, it is likely that this is a G␤␥-mediated process.
Expression and Localization of the Wild Type and ⌬N377 ETB Receptors Are Similar-A possible explanation for the lack of kinase activation by the ⌬N377 ETB receptor may be lack of expression or proper localization of the mutant receptor.
Therefore to examine the expression levels and localization of the wild type and ⌬N377 ETB receptors, COS cells expressing these constructs were examined by immunofluorescence using the M2 anti-FLAG epitope antibody that recognizes the epitope-tagged receptors. Transfection of the cells with the empty expression vector pCDNA3, followed by incubation with the M2 antibody and a CY3 secondary antibody, revealed a lack of red fluorescence; whereas DAPI staining of the nuclei revealed the presence of cells (Fig. 5A). In contrast, cells transfected with either the wild type or ⌬N377 ETB receptor plasmids exhibited marked fluorescence with the CY3 secondary antibody indicating the presence of receptor-primary antibody complexes (Fig. 5, B and C). To ensure that the flag epitope on the C terminus of the ⌬N377 ETB receptor was properly expressed on the cytoplasmic side of the plasma membrane, cells were fixed in paraformaldehyde but not permeabilized, followed by incubation with the primary and secondary antibodies. In cells that were not permeabilized, no specific staining was observed (Fig. 5D) confirming the intracellular location of the flag epitope. The intensity of fluorescence was similar between the wild type and mutant receptors suggesting comparable levels of expression of both receptors. In addition, fluorescence was generally excluded from the nucleus and appeared on the surface of the cell as well as areas where two or more cells were in contact. The results reveal that both the wild type and truncated ETB receptor were expressed and localized similarly.
The Cytoplasmic Tail Is Required for Calcium Mobilization-Stimulation of endothelin receptors is known to result in cytosolic calcium release, presumably through G protein-mediated stimulation of inositol phosphate production. To examine the role of C-terminal tail of the receptor in calcium mobilization, COS cells transiently expressing either the wild type or ⌬N377 ETB receptors were loaded with Fura-2 prior to stimulation with endothelin. Cells expressing wild type receptors showed a characteristic pattern of calcium increase, whereas the ⌬N377 ETB receptor expressing cells did not (Fig. 6A). Wild type ETB-transfected COS cells responded to endothelin with a 1.5 Ϯ 0.2-fold (n ϭ 10) increase in intracellular calcium levels, whereas calcium levels in cells transfected with the ⌬N377 ETB receptor or empty expression vector did not respond to endothelin. The calculated stimulation was 1.0 Ϯ 0.0 (n ϭ 5) for the mock-transfected cells and 1.0 Ϯ 0.1 (n ϭ 8) for ⌬N377 ETB-transfected cells (Fig. 6B). The lack of a calcium response in the mutant receptor expressing cells suggests the C-terminal tail may be involved in stabilizing the receptor-G protein complex. Furthermore, the data reveal that the C terminus is required for multiple signaling functions of the ETB receptor, that is MAP kinase activation and cytosolic free calcium increases.
Ligand Binding Properties of the Wild Type and ⌬N377 ETB Receptors-The ternary complex model of GPCR activation suggests that the high affinity binding state of a given G protein-coupled receptor consists of a complex of receptor, G protein, and ligand (8). This model has been modified lately with the discovery of single point mutations that result in agonist-independent, constitutively activated forms of GPCRs. The constitutively activated receptors bind G protein independent of agonist binding and have increased affinity for the agonist with little or no effect on antagonist binding. The data suggest that receptor molecules in contact with G protein have higher affinity ligand binding properties than receptors not bound to G protein. In the case of the ETB receptor, the G protein interacting domain is not clearly defined; however, mutagenesis studies of the C-terminal tail of the ETA receptor reveal that ligand binding is decreased in C-terminal deletion mutants. This decrease in binding has been ascribed to a loss of interaction with the G protein (12,14). The ligand binding properties of the ⌬N377 ETB receptor were compared with the wild type receptor to estimate the ability of the mutant receptor to couple to G protein. The ⌬N377 ETB receptor had reduced binding of 125 I-endothelin as compared with the wild type ETB receptor. At 5 nM of labeled ET the wild type ETB receptor specifically bound 349 fmol of labeled ET, whereas the ⌬N377 ETB receptor bound 9.3 fmol of labeled ET (Table I). The decrease in binding by the ⌬N377 ETB receptor further suggests that the cytoplasmic tail is required for stabilizing the G protein-receptor interactions.

DISCUSSION
Endothelin is a peptide with a broad array of critical biological properties, as evidenced by its role in the development of the vasculature, Hirschsprung's disease, vasoconstriction, and mitogenesis (17,18,20,21). The intracellular signal transduction pathways that mediate the varied properties of endothelin are therefore of interest in understanding the biology of the endothelins. Our findings extend the understanding in this area in several ways. The ETB receptor was demonstrated to activate three MAP kinase signal transduction pathways that have very different biological consequences with regard to cell proliferation and cell death. In addition, the cytoplasmic tail of the receptor was determined to be critical for coupling the receptor to intracellular kinase pathways and calcium flux, by two criteria. Deletion of C-terminal tail portion of the receptor inhibited the ability of the receptor to activate the MAP kinases. While this result is suggestive of the role of the tail in signaling, the data must be interpreted with caution owing to the possibility of gross structural changes induced by the truncation. Further evidence for the role of the cytoplasmic tail in ERK activation came from ectopic expression of the C-terminal tail of the ETB receptor. Expression of the cytoplasmic tail inhibited ERK activation by the wild type ETB receptor, confirming the importance of this portion of the receptor in MAP kinase activation. The 125 I-endothelin binding studies revealed less binding for the truncated receptor, consistent with a loss of stabilization of the high affinity binding state (Table I). However, guanine nucleotide sensitivity, which is a measure of the high affinity state, was not observed in this heterologous expression system. 2 The observed differences in ligand binding are unlikely to be accounted for by differences in receptor expression levels since the immunofluorescence data suggest similar levels of expression for the wild type and truncated endothelin B receptors. 2 J. Posada, unpublished data.
FIG. 5. Immunofluorescence of wild type and ⌬N377 ETB receptors. COS cells transiently transfected with either the wild type or ⌬N377 ETB receptors (both flag-tagged) were fixed and permeabilized prior to incubation with the M2 antibody and a Cy3 secondary antibody. A, cells were transfected with an empty expression vector, the only staining observed was DAPI staining of the nuclei. B and C, the cells transfected with the wild type or ⌬N377 ETB receptors, respectively, show staining at the cell periphery and point of cell-to-cell contact. D, ⌬N377 ETB-expressing cells were fixed but not permeabilized and incubated with primary and secondary antibodies as in B and C to ensure the mutant receptors expressed the flag epitope on the intracellular surface of the plasma membrane. In each sample a portion of the cells are untransfected, and only the nuclei are visible by DAPI staining.
While there is ample evidence to infer a role for heterotrimeric G proteins in the calcium and phospholipase signaling functions of the ETB receptor, less is known regarding the mechanism of MAP kinase activation by this receptor. We examined the role of G␤␥-subunits in the ETB receptor signal transduction pathway indirectly by ectopic expression of the transducin ␣-subunit. Since expression of this ␣-subunit significantly attenuated the ability of the wild type ETB receptor to activate MAP kinase, presumably by binding free G␤␥-subunits, it is likely that the ETB receptor employs G␤␥-subunits in the ERK activation mechanism, although direct evidence of this will be difficult to obtain.
Given the diverse effects of endothelin in the development of the vasculature, contraction of blood vessels, and mitogenesis, it is interesting to note that three different MAP kinase isozymes are activated by ligand binding to the ETB receptor. Activation of the ERK2 MAP kinase is required for mitogenesis and is sufficient to induce cell transformation, pointing to the central role of ERK2 in cell growth (42,43). In contrast, the JNK1 MAP kinase is activated by agents that induce cell stress, such as UV light, tumor necrosis factor, and anisomycin (38,44). The p38 MAP kinase is activated by cytokines such as interleukin-1, as well as UV light. Activation of JNK1 and p38 is the result of phosphorylation of the kinases by two distinct dual specificity kinases, MKK4 and MKK3, respectively, suggesting the existence of distinct signaling pathways involving JNK1 and p38 activation (45,46). Activation of the JNK MAP kinase pathway has been discovered to be causally associated with apoptosis resulting from nerve growth factor withdrawal from PC12 cells. The association of ERK activation with mitogenesis and JNK activation with apoptosis has fostered the hypothesis that ERK and JNK MAP kinases mediate stimulatory and inhibitory effects on cell growth, respectively (47). Extending the hypothesis is the notion that the decision to proliferate or not is reflected in the balance of ERK and JNK activities. The ETB receptor activates all three MAP kinase isozymes, and in airway smooth muscle cells the ERK activity appears to predominate since endothelin is a mitogenic agonist (20). However, the balance of ERK versus JNK or p38 kinase activities may be altered by endothelin in other contexts such as in vascular constriction or during embryogenesis where endothelin plays a critical role in the developing vasculature. The broad repertoire of MAP kinase activation by endothelin may be an important aspect of its diverse biological functions. Activation of individual MAP kinase isoforms at specific times during development and cell-type specific differential activation of MAP kinases may explain the diverse biological effects of endothelin.
How G protein-coupled receptors activate MAP kinases is an area of intense interest. Considerable evidence indicates that ␤/␥ G protein subunits are able to activate the ERK2 and JNK MAP kinases (23)(24)(25)(26). Muscarinic receptor-mediated ␤/␥ activation of the ERK pathway appears to be Ras-dependent, since dominant negative Asn 17 Ras interrupts this activation (48). Similarly, JNK activation induced by the muscarinic receptor FIG. 6. Endothelin stimulates an increase in cytosolic free calcium in COS cells transfected with wild type but not ⌬N377 ETB receptors. Transiently transfected COS cells expressing either the wild type or ⌬N377 ETB receptors or transfected with an empty expression plasmid were loaded with Fura-2 prior to treatment with ET-1 as described under "Materials and Methods." A, endothelin (100 nM) caused a fast transient rise in cytosolic calcium as reflected by an increase in the fluorescence ratio calculated as fluorescence at maximal response (F) over pre-stimulus levels (F o ). Cells transfected with either the truncated N377-ETB receptor or the empty pCDNA3 expression plasmid showed no increase in calcium levels following endothelin. B, summary of cytosolic calcium levels stimulated by endothelin in COS cells transiently transfected with empty expression plasmid (pCDNA3), wild type ETB (ETB), and truncated ETB (⌬N377-ETB) receptors, respectively. Increases in endothelin-stimulated calcium levels over resting calcium levels were calculated as the average of maximal response (F) divided by pre-stimulus (F o ) values.

FIG. 7. Model of endothelin signal transduction pathways.
The wild type ETB receptor on the left couples to G proteins and results in increased cytosolic free calcium, presumably through activation of phospholipase C. Ligand binding also results in activation of the ERK, JNK, and p38 MAP kinases. The truncated ETB receptor, shown on the right, which is lacking the cytosolic tail, is not able to activate the MAP kinases or increase cytosolic free calcium, possibly due to an inability to couple to G proteins. The data presented in this report indicate that both MAP kinase activation and calcium mobilization are mediated by the receptor tail, via a G protein mechanism. Previous work has demonstrated the catalytic activation of the Src cytosolic tyrosine kinase by endothelin; it is unknown if Src is involved in MAP kinase activation stimulated by endothelin.  (29,30). In the case of the endothelin receptor, the requirements for GTP-binding proteins such as Ras or Rac1 in kinase activation are unknown. Tyrosine phosphorylation events appear to play a critical role in transcriptional activation induced by endothelin, since kinaseinactive Src mutants are able to block endothelin-induced fos transcription (22). The Src kinase is catalytically activated by endothelin in mesangial cells and is involved in the induction of transcriptional events (49); however, it is currently unknown if Src activation is involved in endothelin-induced MAP kinase activation. The current model of ETB signal transduction mechanisms include both kinase activation and intracellular calcium mobilization mediated by the C-terminal tail of the receptor by a G protein-mediated mechanism (Fig. 7). Further work will be required to determine the relationship, if any, between Src and MAP kinase in the endothelin signaling nexus.
Another aspect of our findings relates to the intracellular domains of the ETB receptor required for signal transduction. Chimera and mutagenesis studies with the ETA receptor have identified the third intracellular loop and residues in the cytoplasmic tail as important for coupling to G protein and ligand binding (reviewed in Ref. 2). In particular, Cys 385 in the tail is important for G protein activation and calcium mobilization. This residue is conserved in a large number of G proteincoupled receptors and is palmitoylated in the ␤ 2 -adrenergic receptor, rhodopsin, and the ␣2 adrenergic receptors, and thought to allow insertion into the lipid bilayer. This cysteine has been determined to be important for G protein coupling in some receptors (50 -52). Although the ETA and ETB receptors are highly homologous, especially in the transmembrane regions, they are functionally different with respect to tissue distribution, vasoconstrictor response, and coupling to G proteins. ETA is able to stimulate cAMP through G s , whereas ETB inhibits forskolin-induced cAMP by interaction with G i (11,53). In the present studies we have determined that truncation of the cytoplasmic tail of the rat ETB receptor resulted in a loss of activation of the ERK2, JNK1, and p38 MAP kinases, and calcium mobilization. While the conserved Cys 401 residue was eliminated in the truncation of the tail of the ETB receptor, our studies do not directly address the significance of this residue in mediating the observed kinase activations. It is likely that the ET receptors interact with G protein effectors by means of multiple intracellular domains including the cytoplasmic tail. Indeed, mutagenesis studies to date implicate several intracellular regions in mediating the receptor-G protein interaction (11)(12)(13)(14)(15). The exact structure of the G protein-binding domain of the receptor will be difficult to determine by mutagenesis alone. The recently published structure of the G protein ␣and ␤␥-subunits suggests a large number of contact points are likely to exist between the relatively small intracellular loops of the ET receptors and the comparatively larger structure of the G protein subunits (40, 54 -56). Future mutagenesis experiments combined with molecular modeling may enhance our understanding of the molecular nature of the receptor-G protein interaction.