Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Endothelin-1 (ET-1) has a wide variety of biological effects on various tissues and cell types (Yanagisawa et al., 1988; Masaki, 1993) that are mediated by specific heterotrimeric guanine nucleotide-binding protein (G protein)-coupled receptor subtypes, the endothelin_A receptor (ET_AR) and endothelin_B receptor (ET_BR) (Arai et al., 1990; Sakurai et al., 1990). The two receptors activate multiple subtypes of G proteins and can be distinguished by their selective coupling with specific G protein subtypes. When expressed in Chinese hamster ovary (CHO) cells, ET_AR couples with members of the G_q and G_s families and stimulates phospholipase C (PLC) and adenylyl cyclase. ET_BR couples with members of the G_q and G_i families, stimulates PLC, and inhibits adenylyl cyclase (Aramori and Nakanishi, 1992; Takagi et al., 1995).

ET_AR and ET_BR were shown to be palmitoylated at a cluster of cysteine residues located in the cytoplasmic tail (Horstmeyer et al., 1996; Okamoto et al., 1997). The functional role of palmitoylation and the cytoplasmic tail domain downstream of the palmitoylation site in coupling with G proteins has been studied for ET_AR and ET_BR (Horstmeyer et al., 1996; Okamoto et al., 1997). We found that in the case of ET_BR, palmitoylation is necessary for coupling with both G_q and G_i, whereas the cytoplasmic tail downstream of the palmitoylation sites is also required for coupling with G_i (Okamoto et al., 1997). On the other hand, with ET_AR, palmitoylation is reported to be essential for coupling with G_q but not with G_s, based solely on an experiment using an unpalmitoylated mutant ET_AR (Horstmeyer et al., 1996). Thus, which domain of ET_AR is necessary for coupling with G_s and which of the potential palmitoylation sites is necessary for coupling with G_q remains unknown. In this context, we first attempted to determine the structural basis essential for coupling ET_AR with G_q and G_s by focusing on several potential palmitoylation sites and the cytoplasmic tail downstream of the palmitoylation sites. For this purpose, we constructed CHO cells that stably expressed an unpalmitoylated mutant (Cys³⁸³Cys^385-388Ser³⁸³Ser^385-388) ET_AR (CHO-SerET_AR) and a series of truncated ET_ARs that lacked the cytoplasmic tail downstream of any of the five cysteine residues (Cys³⁸³Cys^385-388).

ET receptors were demonstrated to couple with the G₁₂ subfamily, consisting of G₁₂ and G₁₃, in NIH 3T3 cells (Mao et al., 1998). The G₁₂ subfamily has been shown to mediate important signaling pathways such as for Rho/Rho-associated coiled-coil-forming protein kinase (ROCK)-dependent formation of actin stress fibers (Buhl et al., 1995) and vascular smooth muscle cell contraction (Gohla et al., 2000). These reports suggest that the G₁₂ subfamily may play important roles in several ET-1-induced vascular disorders, such as stroke or vasospasm. Thus, the control of G₁₂ subfamily activation may become a new treatment strategy for these conditions. Recently, it was shown that activation of ET_AR induces actin stress-fiber formation via G₁₂ but not G₁₃ (Gohla et al., 1999). However, the domains in the ET_AR that are necessary for coupling with G₁₂ have not yet been elucidated. The second purpose of the present study is to reveal a functional coupling between ET_AR and G₁₂/G₁₃ in CHO-ET_AR and the functional roles of the palmitoylation site and cytoplasmic tail downstream of the palmitoylation site of ET_AR in coupling of the receptor with G₁₂ using mutated ET_ARs. Furthermore, the conclusion with regard to coupling of ET_AR with G₁₂ is based on an experiment in which actin stress-fiber formation is lost after expression of a dominant negative mutant of G₁₂ in fibroblast cell lines derived from G_q/G₁₁-double deficient mice (Gohla et al., 1999). It remains unknown whether ET-1-induced actin stress-fiber formation requires other G proteins such as G_q and G_s in addition to G₁₂. We have attempted to address this point using CHO cells expressing mutated ET_ARs. Previous reports demonstrated that CHO cells express both G₁₂ and G₁₃ (van de Westerlo et al., 1995; Malek et al., 2001).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Mutagenesis. The entire coding sequence of human ET_AR was subcloned into pGEM-T. The truncated ET_AR cDNAs shown in Fig. 1 were created by polymerase chain reaction. The sequence of the oligonucleotide 5'-primers for all mutants was 5'-CTCGAGGTCGACGGTATCGATAAGCTTGATAT-3'. The sequences of the oligonucleotide 3'-primers for 388, 385, 383, and 382 were 5'-GCGGCCGCTCAACAGCAGCAGCAGAGGCAT-3', 5'-GCGGCCGCTCA-GCAGAGGCATGACTGGAAA-3', 5'-GCGGCCGCTCAGAGGCATGACTGGAAACAA-3', and 5'-GCGGCCGCTCATGACTGGAAACAATTTTTA-3', respectively. Each 3'-primer contained one nucleotide substitution to introduce a termination stop codon with a NotI restriction site, whereas the 5'-primer contained an XhoI restriction site. Fragments were amplified by the 5'-primer and each 3'-primer from ET_AR cDNA as a template. The polymerase chain reaction amplification profiles were denaturation at 94°C for 1 min, primer annealing at 55°C for 30 s, and extension at 72°C for 1 min for 30 cycles. The mutations were confirmed by sequencing, and cDNA fragments were subcloned into a XhoI/NotI restriction site of a mammalian expression vector pME18Sf predigested by XhoI and NotI.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Nomenclature of human ETAR mutants. Aligned are the amino acid sequences of the carboxyl-terminal tail of the wild-type human ETAR, unpalmitoylated mutant and four deletional mutants. The amino acid numbers of the three cysteine residues are indicated. TM VII, seventh transmembrane domain.

Cell Culture and Transfection. CHO cells were maintained in Ham's F-12 medium supplemented with 10% fetal calf serum (FCS) under a humidified 5% CO₂/95% air atmosphere. For stable expression, CHO cells were transfected with expression plasmids together with pSVbsr^r using LipofectAMINE (Invitrogen, Tokyo, Japan). Cell populations expressing the bsr^r gene product were selected in Ham's F-12 supplemented with 10% FCS containing blasticidine (10 µg/ml), and clonal cell lines were isolated by colony lifting and maintained in the same medium.

¹²⁵I-ET-1 Binding Assay. Assays using intact cells or membrane preparations were performed exactly as described previously (Sakamoto et al., 1993).

Cyclic AMP Formation and Inositol Phosphates Formation. Cyclic AMP formation and inositol phosphate (IP) formation were determined as described previously (Okamoto et al., 1997).

Microinjection. Microinjection was performed as described previously (Okazawa et al., 1998). Briefly, cells were seeded onto glass coverslips coated with fibronectin (Iwaki Glass, Chiba, Japan), which were marked with a cross to facilitate the localization of injected cells and incubated overnight in Ham's F-12 medium containing 1% FCS. Plasmids (100 ng/µl) encoding for G₁₂G228A and G₁₃G225A were microinjected into cell nuclei. As a control, expression plasmids without inserts were microinjected in an adjacent field on the same coverslip. Microinjection was performed using a manual microinjection system (Eppendorf-5 Prime, Inc., Hamburg, Germany) equipped with an Axiovert 100 inverted microscope (Carl-Zeiss GmbH, Frankfurt, Germany).

Stress-Fiber Formation. After incubation of cells with serum-free Ham's F-12 medium for 24 h, ET-1 was added at 37°C for 5 min. Cells were washed three times with phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde in PBS at room temperature for 15 min. After being washed five times with PBS containing 0.1% Triton X-100 (PBS-Tx), the cells were incubated with fluorescein rhodamine-phalloidin (Molecular Probes, Eugene, OR) in PBS-Tx (1:200) at room temperature for 10 min. After being washed five times with PBS-Tx, the labeled cells were mounted on glass slides and examined with an MRC 1024 laser-scanning confocal microscope (Bio-Rad, Hercules, CA) equipped with an Axiovert 135 M inverted microscope (Carl-Zeiss GmbH).

Drugs. Y-27632 was kindly provided by Welfide Corporation (Osaka, Japan). Chemicals were obtained from the following sources: ET-1 from the Peptide Institute (Osaka, Japan), ¹²⁵I-ET-1 and myo-[³H]inositol from Amersham Biosciences UK, Ltd. (Little Chalfont, Buckinghamshire, UK), rhodamine-phalloidin from Molecular Probes, U73122 from Funakoshi (Tokyo, Japan), and dibutyryl cAMP from Sigma (St. Louis, MO). All other chemicals were of reagent grade and were obtained commercially.

Statistical Analysis. All results were expressed as mean ± S.E.M. The data were subjected to a two-way analysis of variance, and when a significant F value was encountered, the Newman-Keuls multiple range test was used to test for significant differences between treatment groups. A probability level of P < 0.05 was considered statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Stable Expression of Truncated or Unpalmitoylated Mutant ET_ARs in CHO Cells. By cotransfecting CHO cells with each expression plasmid and pSVbsr^r and then selecting for resistance against blasticidine, we obtained more than five individual clonal cell lines that stably expressed each receptor construct. In CHO cells expressing truncated mutant ET_AR, ¹²⁵I-ET-1 binding assays on membrane preparations from various clones gave K_d values of 30 to 120 pM and B_max values of 0.7 to 1.4 pmol/mg of protein. On the other hand, in CHO-SerET_AR, ¹²⁵I-ET-1 binding assays on membrane preparations from various clones gave K_d values of 50 to 140 pM and B_max values of 0.8 to 1.7 pmol/mg of protein. Cell clones showing similar levels of receptor densities were used in the subsequent study. The K_d and B_max values for the receptors expressed on each clone adopted are listed in Table 1.

Formation of IPs and cAMP in CHO Cells Expressing Truncated or Unpalmitoylated Mutant ET_ARs after Stimulation with ET-1. To reveal the functional significance of the palmitoylation site and the cytoplasmic tail downstream of the palmitoylation site in coupling with G_q and G_s, we tested the abilities of the mutant receptors to stimulate accumulation of [³H]IPs and cAMP, respectively. [³H]Palmitic acid was metabolically incorporated into CHO-ET_AR388 and CHO-ET_AR385 but not into CHO-ET_AR383, CHO-ET_AR382, or CHO-SerET_AR (data not shown).

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Effects of ET-1 on IP (A) or cAMP (B) accumulation in CHO cells expressing wild-type or mutant ETARs. Formation of total IPs or cAMP after stimulation with varying concentrations of ET-1 in CHO-ETAR (), CHO-ETAR388 (closed triangle), CHO-ETAR385 (), CHO-ETAR383 (), CHO-ETAR382 (), and CHO-SerETAR (). A, cells that had been incubated with myo-[3H]inositol for 18 h were stimulated by various concentrations of ET-1 for 30 min. B, cells that were stimulated by various concentrations of ET-1 for 10 min in the presence of 3-isobutyl-1-methylxanthine. Total IPs and cAMP in the cell extract were measured as described under Materials and Methods. Values are expressed as -fold increases above basal values. Each data point represents mean ± S.E.M. of five experiments.

ET-1-Induced Actin Stress-Fiber Formation in CHO-ET_AR. We attempted to determine the structural basis for coupling of ET_AR with G₁₂/G₁₃ and subtypes of G proteins involved in ET-1-induced stress-fiber formation. For these purposes, we examined the effects of inhibition of either one of the G protein-mediated signaling cascades by blockers and dominant negative mutants of G₁₂ or G₁₃ (G₁₂G228A or G₁₃G225A, respectively) on ET-1-induced actin stress-fiber formation in CHO-ET_AR385, CHO-SerET_AR, and CHO-ET_AR. Subsequently, we deduced the domains of ET_AR that were critical for coupling with G₁₂, based on the structure of the mutant ET_ARs that did not have the ability to couple to G₁₂.

View larger version (52K): [in this window] [in a new window]   Fig. 3.   A-D, effects of Y-27632, U73122, and G12G228A on the ET-1-induced actin stress-fiber formation in CHO-ETAR. Cells were stimulated with (B) or without (A) 10 nM ET-1. The effects of preincubation with 10 µM Y27632 (C) and combination treatment of G12G228A microinjection and 5 µM U73122 preincubation (D) on ET-1-induced stress-fiber formation are shown. Y-27632 and U73122 were added 15 min before stimulation with ET-1. Expression plasmids encoding for G12G228A and G13G225A were microinjected into the cell nuclei 24 h before stimulation with ET-1. E, effects of dibutyryl cAMP on actin stress-fiber formation in CHO-ETAR. Cells were incubated for 5 min with 10 µM dibutyryl cAMP alone. Actin stress fibers were visualized as described under Materials and Methods. Representative examples of stress fibers in individual cells are shown. F, pixel intensity of images was quantified using NIH Image software. Values are expressed as -fold increases above the values in the absence of ET-1. Each data point represents mean ± S.E.M. of at least 20 cells.

ET-1-Induced Actin Stress-Fiber Formation in CHO-SerET_AR and CHO-ET_AR385. ET-1 induced stress-fiber formation in CHO-SerET_AR, in which coupling of the receptor with G_s but not G_q was retained (Fig. 4B). Like CHO-ET_AR, ET-1-induced stress-fiber formation was inhibited by preincubation of CHO-SerET_AR with Y-27632 (Fig. 4C) but was not affected by preincubation with U73122 or microinjection of G₁₃G225A (Fig. 4, E and F). Notably, unlike CHO-ET_AR, it was inhibited by microinjection of G₁₂G228A (Fig. 4D).

View larger version (58K): [in this window] [in a new window]   Fig. 4.   Effects of Y27632, U73122, G12G228A, and G13G225A on ET-1-induced actin stress-fiber formation in CHO-SerETAR. Cells were stimulated with (B) or without (A) 10 nM ET-1. C, Y-27632 at 10 µM was added 15 min before stimulation with ET-1. Expression plasmids encoding for G12G228A (D) and G13G225A (E) were microinjected into cell nuclei 24 h before stimulation with ET-1. Actin stress fibers were visualized as described under Materials and Methods. Representative examples of stress fibers in individual cells are shown. F, pixel intensity of images was quantified using NIH Image software. Values are expressed as -fold increases above the values in the absence of ET-1. Each data point represents mean ± S.E.M. of at least 20 cells.

View larger version (38K): [in this window] [in a new window]   Fig. 5.   Effects of Y27632, U73122, G12G228A, and G13G225A on ET-1-induced actin stress-fiber formation in CHO-ETAR385. Cells were stimulated with (B) or without (A) 10 nM ET-1. Y-27632 at 10 µM (C) and U73122 at 5 µM (D) were added 15 min before stimulation with ET-1. Expression plasmids encoding G12G228A (E) and G13G225A (F) were microinjected into the cell nuclei 24 h before stimulation with ET-1. Actin stress fibers were visualized with fluorescein rhodamine-phalloidin as described under Materials and Methods. Representative examples of stress fibers in individual cells are shown. G, pixel intensity of images was quantified using NIH Image software. Values are expressed as -fold increases above the values in the absence of ET-1. Each data point represents mean ± S.E.M. of at least 20 cells.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

¹²⁵I-ET-1 binding assays on intact CHO cells expressing the wild-type or truncated ET_ARs yielded K_d and B_max values within similar ranges (Table 1). These results were consistent with previous data (Hashido et al., 1993) and suggest that truncation of the receptor is not essential for cell surface expression and ligand binding of ET_AR. High affinity binding of ET-1 by the mutant receptors is a good indication that the overall structure of the receptor is unchanged by truncation as described earlier (Hashido et al., 1993).

As reported previously (Horstmeyer et al., 1996), with SerET_AR in which a cluster of five cysteine residues in the cytoplasmic tail as potential palmitoylation sites were substituted with serine, ET-1 failed to stimulate formation of IPs (Fig. 2A). In the present study, we extended this finding using truncated ET_ARs. The truncated ET_ARs holding Cys³⁸⁵ (CHO-ET_AR385 and CHO-ET_AR388) retained the ability to stimulate IP formation, whereas those lacking Cys³⁸⁵ (CHO-ET_AR383 and CHO-ET_AR382) lost such ability (Fig. 2A). These results taken together strongly indicate that Cys³⁸⁵ in ET_AR is critical for coupling of ET_AR with G_q and that the cytoplasmic tail downstream of the palmitoylation site is not necessary for this coupling (Fig. 6).

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Schematic representation of signaling pathways for actin stress-fiber formation activated by ET-1 in CHO-ETAR (A), CHO-SerETAR (B), and CHO-ETAR385 (C). Wild-type ETAR couple to Gq, Gs, and G12, whereas SerETAR or ETAR385 couple to Gs/G12 or Gq, respectively. Actin stress-fiber formation is stimulated by ET-1 via Gq- and G12-dependent pathways in CHO-ETAR, whereas via G12- or Gq-dependent pathway in CHO-SerETAR or CHO-ETAR385, respectively. See Results and Discussion for details.

In the present study, ET-1 stimulated adenylyl cyclase in CHO-SerET_AR, which lacked potential palmitoylation sites but retained the cytoplasmic tail (Fig. 2B). These results are consistent with a previous report (Horstmeyer et al., 1996). In contrast, ET-1 failed to stimulate adenylyl cyclase in all truncated ET_ARs lacking the cytoplasmic tail, regardless of the absence or presence of palmitoylation sites of ET_AR (Fig. 2B). These results, taken together, strongly demonstrate that the cytoplasmic tail of ET_AR is critical for coupling with G_s, although it is not necessary for coupling with G_q (Fig. 6). Moreover, it was previously demonstrated that the second and third intracellular loops of ET_AR were major determinants of the selective coupling of ET_AR with G_s (Takagi et al., 1995). Therefore, we conclude that both the cytoplasmic tail and the second and third intracellular loops of ET_AR are necessary for coupling of ET_AR with G_s.

Next, we attempted to identify the subtypes of G proteins that are involved in ET-1-induced stress-fiber formation using CHO-ET_AR385, CHO-SerET_AR, and CHO-ET_AR. Based on sensitivity to Y-27632, the Rho/ROCK pathway plays important roles in ET-1-induced stress-fiber formation in CHO-ET_AR (Fig. 5C) as in NIH 3T3 cells and fibroblasts (Mao et al., 1998; Gohla et al., 1999). ET-1-induced stress-fiber formation in CHO-ET_AR was affected by neither pretreatment with U73122 nor microinjection of G₁₂G228A or G₁₃G225A (Fig. 5F) but was inhibited by combined treatment with U73122 and G₁₂G228A microinjection (Fig. 3D). These results indicate that ET-1-induced stress-fiber formation is mediated via two signaling pathways (i.e., the G_q/PLC- and G₁₂-dependent pathways in CHO-ET_AR) (Fig. 6) and also that only one of the two is sufficient for actin stress-fiber formation. Moreover, the present study indicates that G_s is not involved in ET-1-induced stress-fiber formation, because dibutyryl cAMP failed to induce actin stress-fiber formation in CHO-ET_AR (Fig. 3E).

These conclusions are supported by findings obtained with SerET_AR. That is, because SerET_AR does not couple with G_q, which is one of the two signaling pathways necessary for ET-1-induced stress-fiber formation, blockade of another signaling pathway with G₁₂G228A leads to inhibition of actin stress-fiber formation. Furthermore, these results indicate that SerET_AR retains the ability to couple with G₁₂.

In CHO-ET_AR385, in which coupling of the receptor with G_q but not G_s is retained, ET-1-induced stress-fiber formation was inhibited by U73122 but not G₁₂G228A. Based on the conclusion obtained from wild-type ET_AR, these data can be interpreted to mean that because ET_AR385 lacks coupling with G₁₂, which is one of the two signaling pathways necessary for ET-1-induced actin stress-fiber formation, blockade of another signaling pathway with U73122 leads to inhibition of actin stress-fiber formation. Therefore, these results indicate that ET_AR385 has lost the ability to couple with G₁₂, although it can still induce stress-fiber formation via the G_q-dependent pathway.

Finally, we deduced the structural determinant for coupling of ET_AR with G₁₂ based on data from experiments using mutated ET_ARs. That is, loss of coupling of ET_AR385 with G₁₂ and retention of coupling of SerET_AR with G₁₂ clearly show that the cytoplasmic tail downstream of Cys³⁸⁵ but not the palmitoylation site of ET_AR is essential for coupling with G₁₂.

In conclusion, the present study showed that 1) the cytoplasmic tail downstream of the palmitoylation site of ET_AR is essential for coupling with G_s and G₁₂, 2) Cys³⁸⁵ of ET_AR is critical for coupling with G_q, and 3) the signal for ET-1-induced stress-fiber formation is mediated via the G_q/PLC- and G₁₂-dependent pathway to Rho/ROCK system in CHO-ET_AR. Thus, the presence of one of the two pathways is sufficient for stress-fiber formation.