Abstract
We demonstrated recently that in Chinese hamster ovary cells stably expressing human recombinant endothelinA receptors (CHO-ETAR), endothelin-1 (ET-1) activates two types of Ca2+-permeable nonselective cation channels (designated NSCC-1 and NSCC-2) and a store-operated Ca2+ channel (SOCC), which can be distinguished by Ca2+ channel blockers such as 1-{β-[3-(4-methoxyphenyl)propoxy]-4-methoxyphenylethyl}-1H-imidazole hydrochloride (SK&F 96365) and (R,S)-(3,4-dihydro-6,7-dimethoxy-isochinolin-1-yl)-2-phenyl-N,N-di[2-(2,3,4-trimethoxyphenyl)ethyl]acetamid mesylate (LOE 908). We also reported that CHO-ETAR couples with G12 in addition to Gq and Gs. The purpose of the present study was to identify the G proteins involved in the activation of these Ca2+ channels by ET-1, using mutated ETARs with coupling to either Gqor Gs/G12 (designated ETARΔ385 and SerETAR, respectively) and a dominant-negative mutant of G12 (G12G228A). ETARΔ385 is truncated immediately downstream of Cys385 in the C terminus as palmitoylation sites, whereas SerETAR is unpalmitoylated because of substitution of all the cysteine residues to serine (Cys383Cys385–388 → Ser383Ser385–388). In CHO-ETARΔ385, stimulation with ET-1 activated only SOCC. In CHO-SerETAR or CHO-ETAR pretreated withU73122, an inhibitor of phospholipase C (PLC), ET-1 activated only NSCC-1. Dibutyryl cAMP alone did not activate any Ca2+channels in the resting and ET-1–stimulated CHO-SerETAR. Microinjection of G12G228A abolished the activation of NSCC-1 and NSCC-2 in CHO-ETAR and that of NSCC-1 in CHO-SerETAR. These results indicate that ETAR activates three types of Ca2+ channels via different G protein-related pathways. NSCC-1 is activated via a G12-dependent pathway, NSCC-2 via Gq/PLC- and G12-dependent pathways, and SOCC via a Gq/PLC-dependent pathway.
Endothelin-1 (ET-1), a 21–amino acid peptide, is one of the most potent endogenous vasoconstricting agents (Yanagisawa et al., 1988). Subsequent studies have described its multiple and wide-ranging biological activities, including modulation of neurotransmission (Koseki et al., 1989) and stimulation of cell proliferation (Komuro et al., 1988; Shichiri et al., 1991). Recent reports have demonstrated that extracellular Ca2+ influx through voltage-independent Ca2+ channels (VICCs) plays a critical role for ET-1–induced contraction of rat aorta (Zhang et al., 1999) and ET-1–induced proliferation of vascular smooth muscle cells (VSMCs) (Kawanabe et al., 2002a). Thus, it is important to elucidate activation mechanisms of VICCs by ET-1. Biological actions of ET-1 are mediated by two distinct receptor subtypes: endothelinA and endothelinB receptors (ETARs and ETBRs, respectively), which belong to a family of G protein–coupled receptors (Arai et al., 1990; Sakurai et al., 1990). Therefore, in the present study, we focused on investigating which G protein subtypes were involved in the activation of each Ca2+ channel by ET-1.
Transfection and functional expression of wild-type or mutant ETAR cDNAs into the same cell type provides a model system for study of the precise characteristics of signal transduction by a single receptor subtype. We used Chinese hamster ovary (CHO) cells stably expressing wild-type or mutant ETARs in this study. We have recently shown that a sustained increase in intracellular free Ca2+concentration ([Ca2+]i) caused by ET-1 results from Ca2+ entry through three types of VICCs into CHO cells stably expressing wild-type ETAR (CHO-ETAR): two types of Ca2+-permeable nonselective cation channels (designated NSCC-1 and NSCC-2) and a store-operated Ca2+ channel (SOCC) (Kawanabe et al., 2001). In particular, these channels can be distinguished using Ca2+ channel blockers such as SK&F 96365 and LOE 908. Thus, NSCC-1 is sensitive to LOE 908 and resistant to SK&F 96365, NSCC-2 is sensitive to both LOE 908 and SK&F 96365, and the SOCC is resistant to LOE 908 and sensitive to SK&F 96365 (Kawanabe et al., 2001). The VICCs activated by ET-1 in CHO-ETAR are pharmacologically identical to those in VSMCs (Kawanabe et al., 2002a). Therefore, CHO-ETAR may be a good model for studying the mechanism of activation of VICCs. ETARs are functionally coupled with Gq and Gs in CHO cells (Aramori and Nakanishi, 1992). Activation of Gqand Gs causes stimulation of phospholipase C (PLC) and adenylyl cyclase, respectively (Aramori and Nakanishi, 1992). In addition, ETARs also couple with G12 via its C terminus to induce actin stress fiber formation in CHO cells (Kawanabe et al., 2002b). In the present study, we used a dominant-negative mutant of G12and two types of mutated ETARs designated ETARΔ385 and SerETAR to clarify the involvement of Gq, Gs, and G12 for Ca2+ channel activation by ET-1. ETARΔ385 lacks a C terminus downstream of Cys385 and couples only with Gq in CHO cells (Kawanabe et al., 2002b). SerETAR is unpalmitoylated because of substitution of all the cysteine residues to serine (Cys383Cys385–388→Ser383Ser385–388) and couples with Gs and G12in CHO cells (Kawanabe et al., 2002b).
Experimental Procedures
Cell Culture.
We used CHO-ETAR, CHO-ETARΔ385, and CHO-SerETAR, which were constructed as described previously (Kawanabe et al., 2002b). TheK d (picomolar) andB max (picomoles per milligram of protein) values for CHO-ETAR, CHO-ETARΔ385, and CHO-SerETAR were 52.8 ± 2.4 and 1.08 ± 0.16, 49.5 ± 4.3 and 1.12 ± 0.08, and 70.2 ± 4.4 and 1.04 ± 0.14, respectively. CHO cells were maintained in Ham's F12 medium supplemented with 10% fetal calf serum under a humidified 5% CO2/95% air atmosphere.
Measurement of Intracellular Free Ca2+Concentration.
[Ca2+]iwas measured using the fluorescent probe fluo-3. The measurement of fluorescence by a CAF 110 spectrophotometer (Jasco, Tokyo, Japan) was performed as described previously (Kawanabe et al., 2001).
Microfluorimetry of fluo-3 was performed as described previously (Zhang et al., 1999). Briefly, the cells seeded on 35-mm glass-bottomed plastic dishes (MatTek, Ashland, MA), which were marked with a cross to facilitate the localization of injected cells, were loaded with fluo-3 by incubating them with Ca2+-free Krebs-HEPES solution containing 10 μM fluo-3/AM for 30 min at 37°C under a reduced light. Ca2+-free Krebs-HEPES solution contained 140 mM NaCl, 3 mM KCl, 1 mM MgCl2, 11 mM glucose, and 10 mM HEPES, pH 7.4, adjusted with NaOH. After washing with Krebs-HEPES solution (2.2 mM CaCl2 was added to Ca2+-free Krebs-HEPES solution), they were kept in fresh Krebs-HEPES solution at 37°C for at least 30 min. Fluo-3 microfluorimetry was performed at 25°C by use of an Attofluor Ratio-Vision real-time digital fluorescence analyzer (Atto Instruments, Rockville, MD) equipped with an Axiovert-100 inverted epifluorescent microscope (Carl Zeiss Inc., Thornwood, NY). A 100-W mercury lamp served as the source of excitation. For measurement of [Ca2+]i, fluo-3 was excited at 450 to 490 nm, and fluorescence was detected at 515 to 565 nm. At the end of the experiment, ionomycin and subsequently EGTA were added at final concentrations of 10 μM and 10 mM, respectively, to obtain the fluorescence intensity maximum (Fmax) and the fluorescence intensity minimum (Fmin). [Ca2+]i was determined from the equilibrium equation [Ca2+]i =K d(F − Fmin) / (Fmax − F), where F was the experimental value of fluorescence andK d was defined as 0.4 μM (Minta et al., 1989).
Microinjection.
For microinjection, 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 were incubated overnight in Ham's F12 medium containing 1% fetal calf serum. Microinjection of G12G228A, constructed as described previously (Kawanabe et al., 2002b), was performed using a Zeiss microinjection system (Carl Zeiss). Plasmid (100 ng/μl) was used for microinjection into the cell nuclei.
Materials.
Boehringer Ingelheim GmbH (Ingelheim, Germany) kindly provided LOE 908. Other chemicals were obtained commercially from the following sources: ET-1 from Peptide Institute (Osaka, Japan); SK&F 96365 from Biomol Research Laboratories (Plymouth Meeting, PA); fluo-3/AM from Dojin Laboratories (Kumamoto, Japan); and U73122 from Funakoshi (Tokyo, Japan).
Statistical Analysis.
All results were expressed as mean ± S.E.M.
Results
Basic Properties of the ET-1–Induced Increase in [Ca2+]i in CHO-ETAR, CHO-ETARΔ385, and CHO-SerETAR.
ET-1 induced a biphasic increase in [Ca2+]i in CHO-ETAR, consisting of an initial transient phase and a subsequent sustained phase (Fig.1A). Both the transient and sustained increase in [Ca2+]i were dependent on the concentrations of ET-1 with EC50values of approximately 1 nM, and they reached the maximal value at concentrations ≥10 nM (Fig. 1, D and E).
In CHO-ETARΔ385, which is coupled with Gq alone, ET-1 also induced a biphasic increase in [Ca2+]i (Fig. 1B). ET-1 caused a transient peak and subsequent sustained increase in [Ca2+]i (Fig.2B). The magnitude of the transient increase in [Ca2+]i in CHO-ETARΔ385 was essentially similar to that in CHO-ETAR (Fig. 1D). On the other hand, the magnitude of the sustained increase in [Ca2+]i in CHO-ETARΔ385 was lower than that in CHO-ETAR (Fig. 1E). Moreover, an ET-1 concentration ≥ 10 nM induced a sustained increase in [Ca2+]i in CHO-ETARΔ385, whereas ET-1 induced an increase at only 0.1 nM in CHO-ETAR (Fig. 1E).
In CHO-SerETAR, which is coupled with Gs and G12, the pattern of the ET-1–induced increase in [Ca2+]i was different from that in CHO-ETAR and CHO-ETARΔ385. That is, ET-1 failed to induce an initial transient increase in [Ca2+]i, and it induced only a sustained increase in [Ca2+]i (Fig. 1C). The magnitude of the sustained increase in [Ca2+]i in CHO-SerETAR was lower than that in CHO-ETAR (Fig. 1E).
Pharmacological Identification of Ca2+ Channels Activated by ET-1 in CHO-ETAR, CHO-ETARΔ385, and CHO-SerETAR.
As described previously (Kawanabe et al., 2001), in CHO-ETAR, the ET-1–induced sustained increase in [Ca2+]i was partially suppressed by the maximally effective concentration (10 μM) of either SK&F 96365 or LOE 908, and it was abolished by combined treatment with both blockers (Fig. 2, A and D). In CHO-ETARΔ385, the ET-1–induced sustained increase in [Ca2+]i was completely inhibited by 10 μM SK&F 96365, whereas LOE 908 at concentrations up to 10 μM had no effects (Fig. 2, B and E). In CHO-SerETAR, the ET-1–induced sustained increase in [Ca2+]i was completely inhibited by 10 μM LOE 908, whereas SK&F 96365 at concentrations up to 10 μM had no effects (Fig. 2, C and F).
Effects of Inhibition of PLC on the Species of ET-1–Activated Ca2+ Channels in CHO-ETAR.
In CHO-SerETAR, coupling between the receptor and Gq is missing, and hence, PLC as an effector of Gq cannot be activated upon stimulation of the receptor. To mimic the stimulation in CHO-SerETAR and confirm that PLC actually acts as an effector for activation of Ca2+ channels, we used U73122, a PLC blocker. Previous reports demonstrated that 5 to 10 μM U73122 inhibits PLC activation completely (Okamoto et al., 1995; Kanki et al., 2001). ET-1 stimulated [3H]inositol phosphates (IPs) formation in CHO-ETAR (Kawanabe et al., 2001). On the other hand, ET-1 failed to induce [3H]IPs formation in CHO-ETAR treated with 5 μM U73122(data not shown). ET-1 at 10 nM induced only the sustained increase in [Ca2+]i in CHO-ETAR treated with 5 μM U73122 (Fig.3, A and B). The magnitude of the sustained increase in [Ca2+]i was approximately 20% of that in the absence of U73122. This sustained increase in [Ca2+]i was completely inhibited by 10 μM LOE 908, whereas SK&F 96365 at concentrations up to 10 μM had no effects (Fig. 3, A and B).
Effects of Dibutyryl cAMP on the Resting [Ca2+]i and the ET-1–Induced Increase in [Ca2+]i in CHO-SerETAR.
Because CHO-SerETAR is coupled with Gs and G12 (Kawanabe et al., 2002b), it is unknown which of the G proteins is involved in the activation of Ca2+ channels. To clarify whether Gs was involved in Ca2+channel activation, we examined the effects of dibutyryl cAMP on the resting [Ca2+]i and the ET-1–induced increase in [Ca2+]i. A previous report demonstrated that 1 mM dibutyryl cAMP activates protein kinase A in CHO cells (Lee and Fraser, 1993). Dibutyryl cAMP at 1 mM failed to evoke an increase in [Ca2+]i in CHO-SerETAR (Fig. 3C). Moreover, the ET-1–induced sustained increase in [Ca2+]i was not affected by 1 mM dibutyryl cAMP in CHO-SerETAR (Fig. 3D). Dibutyryl cAMP also failed to affect the resting [Ca2+]i and the ET-1–induced increase in [Ca2+]i in CHO-ETAR (data not shown).
Effects of G12 on the ET-1–Induced Sustained Increase in [Ca2+]i in CHO-ETAR or CHO-SerETAR.
To confirm that G12is involved in the activation of Ca2+ channels, we investigated the effects of G12G228A on the ET-1–induced increase in [Ca2+]i in CHO-ETAR and CHO- SerETAR. In this experiment, G12G228A was microinjected into CHO-ETAR and CHO-SerETAR, and the ET-1–induced increase in [Ca2+]i in these cells was analyzed using microfluorimetry.
In CHO-ETAR microinjected with G12G228A, ET-1 evoked transient and subsequently sustained increase in [Ca2+]i. The magnitude of the sustained increase in [Ca2+]i was approximately 40% of that in CHO-ETAR microinjected with an expression vector alone (data not shown). The sustained increase in [Ca2+]i was inhibited by 10 μM SK&F 96365, whereas it remained unaffected by 10 μM LOE 908 (Fig. 4A). ET-1 failed to induce a sustained increase in [Ca2+]i in CHO-SerETAR microinjected with G12G228A (Fig. 4B).
Discussion
As reported previously (Kawanabe et al., 2001), the ET-1–induced sustained increase in [Ca2+]i in CHO-ETAR results from extracellular Ca2+ influx through three types of VICCs: NSCC-1, NSCC-2, and SOCC. Pharmacological identification of these Ca2+ channels and calculation for contribution of Ca2+ influx through each channel are explained schematically in Fig. 5. We have pharmacologically defined these channels in CHO cells expressing human recombinant ETAR and VSMCs expressing endogenous ETAR and have found that the same Ca2+ channels are activated in these cells (Kawanabe et al., 2001, 2002a).
In CHO-ETARΔ385, the only Ca2+ channels activated by ET-1 are SOCCs, judging from the sensitivity of the ET-1–induced sustained increase in [Ca2+]i to SK&F 96365 and LOE 908 (Fig. 2, B and E). Because CHO-ETARΔ385 is coupled with Gq but not with Gs or G12 (Kawanabe et al., 2002b), this result indicates that Gq is required (sufficient) for activation of SOCC. Furthermore, activation of SOCC in CHO-ETAR is lost after treatment with U73122(Fig. 3, A and B), indicating that PLC acts as an effector downstream of Gq. These results, taken together, show that SOCC is activated via a Gq/PLC-dependent pathway.
Conversely, the results obtained from CHO-ETARΔ385 suggest that the activation of NSCC-1 and NSCC-2 requires G proteins other than Gq. Because CHO-ETAR is coupled with Gs and G12 in addition to Gq (Aramori and Nakanishi, 1992;Kawanabe et al., 2002b), activation of NSCC-1 and NSCC-2 might be mediated by either Gs or G12. To address this point, we used CHO-SerETAR, which couples with Gs and G12 but not with Gq (Kawanabe et al., 2002b). In CHO-SerETAR, ET-1 activated NSCC-1 but not NSCC-2, because of the pharmacology of the sustained increase in [Ca2+]i (sensitive to LOE 908 and resistant to SK&F 96365) (Fig. 3, C and D). These results indicate that either Gs or G12 is required for activation of NSCC-1, whereas either Gs, G12, or both are not sufficient for activation of NSCC-2. Regarding a mechanism for activation of NSCC-1, dibutyryl cAMP alone was without effect in the resting [Ca2+]i and the ET-1–induced sustained increase in [Ca2+]i in CHO-SerETAR (Fig. 3), excluding the possibility that activation of NSCC-1 is mediated by cAMP, a product of Gs/adenylate cyclase-dependent pathway. Moreover, disruption of signaling through endogenous G12 by its dominant-negative mutant (G12G228A) abrogated activation of NSCC-1 in CHO-SerETAR as well as CHO-ETAR (Fig. 4), indicating that activation of NSCC-1 is mediated by G12. Taken together, these results strongly demonstrate that NSCC-1 is activated via a G12-dependent pathway.
As for a mechanism for activation of NSCC-2, the channel was not activated in CHO-ETAR pretreated with U73122 or in CHO-ETAR microinjected with G12G228A (Figs. 3 and 4). These results indicate that both a Gq/PLC-dependent pathway and a G12-dependent pathway are required for activation of NSCC-2. In accordance with this conclusion, NSCC-2 was not activated after the stimulation of ETAR lacking coupling with either Gq or G12, i.e., SerETAR and ETARΔ385.
Signaling mechanisms downstream of G proteins are presently unknown. In the case of Gq, it is very likely that PLC is activated downstream of Gq, considering that CHO-ETAR treated with U73122 (an inhibitor of PLC) mimicked CHO-SerETAR (which lost the ability to couple with Gq) in terms of temporal pattern and magnitude of the ET-1–induced changes in [Ca2+]i and species of activated channels. Stimulation of PLC leads to increased formation of inositol-1,4,5-trisphosphate and diacylglycerol (DAG). Inositol-1,4,5-trisphosphate acts on its receptor on sarcoplasmic reticulum as an intracellular Ca2+ store to release Ca2+ and subsequently deplete the store (Berridge, 1993). Thus, the store depletion could be a trigger for activation of the Ca2+ channels on the plasma membrane called capacitative Ca2+ channel (Thasreup et al., 1990). In fact, SOCC is activated in CHO-ETAR and VSMCs, after depletion of the Ca2+ store by treatment with thapsigargin (an inhibitor of Ca2+-pump ATPase on the membrane of sarcoplasmic reticulum) (Kawanabe et al., 2001, 2002a). The concentrations of ET-1 that failed to induce a transient increase in [Ca2+]i and IP accumulation did not induce SOCC activation (Kawanabe et al., 2001). Judging from these data, we concluded that the transient increase in [Ca2+]i (depletion of Ca2+ from intracellular Ca2+ store) essentially follows the characteristics of SOCC. On the other hand, another second messenger, DAG, can directly activate Ca2+ channels such as transient receptor potential (Hofmann et al., 1999). Thus, DAG might be an effector downstream of Gq. In the case of G12, several kinds of proteins are reported to be its effectors (Seasholtz et al., 1999). We have recently shown that stimulation of CHO-ETAR with ET-1 induces actin stress fiber formation through G12, and that downstream of G12, a small GTP-binding protein Rho and ρ-associated coiled-coil–forming protein kinase is activated (Kawanabe et al., 2002b). Thus Rho/Rho-associated coiled-coil-forming protein kinase might be a signal downstream of G12. Additional study is needed to identify the effectors downstream of Gq and G12 for activation of NSCC-1, NSCC-2, and SOCC.
In summary, stimulation of ETAR with ET-1 activates NSCC-1, NSCC-2, and SOCC in CHO-ETAR. NSCC-1 is activated via a G12-dependent pathway, NSCC-2 is activated via Gq- and G12-dependent pathways, and SOCC is activated via a Gq-dependent pathway.
Acknowledgments
We thank Boehringer Ingelheim K.G. for the kind donation of LOE 908.
Footnotes
- Received December 13, 2001.
- Accepted March 28, 2002.
-
This work was supported by a Grant-in-Aid from the Ministry of Education, Science, Sports and Culture of Japan, by Special Coordination Funds for Science and Technology from the Science and Technology Agency, by a research grant for cardiovascular disease (11C-1) from the Ministry of Health and Welfare, and by a grant from the Smoking Research Foundation, Japan.
Abbreviations
- ET-1
- endothelin-1
- VICC
- voltage-independent Ca2+ channel
- VSMC
- vascular smooth muscle cell
- ETAR
- endothelinA receptor
- ETBR
- endothelinB receptor
- CHO
- Chinese hamster ovary
- NSCC
- nonselective cation channel
- SOCC
- store-operated Ca2+ channel
- PLC
- phospholipase C
- SK&F 96365
- 1-{β-[3-(4-methoxyphenyl)propoxy]-4-methoxyphenylethyl}-1H-imidazole hydrochloride
- LOE 908
- (R,S)-(3,4-dihydro-6,7-dimethoxy-isochinolin-1-yl)-2-phenyl-N,N-di[2-(2,3,4-trimethoxyphenyl)ethyl]acetamid mesylate
- AM
- acetoxymethyl ester
- U73122
- 1-(6-{[17β-3-methoxyoestra-1,3,5(10)-trien-17-yl]amino}hexyl)-1H-pyrrole-2,5-dione
- IP
- inositol phosphate
- DAG
- diacylglyceride
- G12G228A
- dominant-negative mutant of G12
- CHO-SerETAR
- Chinese hamster ovary cells that express an unpalmitoylated (Cys383Cys385–388→Ser383Ser385–388) human endothelinA receptor
- CHO-ETARΔ385
- Chinese hamster ovary cells that express human endothelinAreceptor truncated at the carboxyl-terminal downstream of Cys385
- CHO-ETAR
- Chinese hamster ovary cells stably expressing human endothelinA receptor
- The American Society for Pharmacology and Experimental Therapeutics