Desensitization of inositol 1,4,5-trisphosphate/Ca2+-induced Cl- currents by prolonged activation of G proteins in Xenopus oocytes.

Expression of G protein α subunits of the Gq family with various G protein-coupled receptors induces activation of an inositol 1,4,5-trisphosphate (IP3)/Ca2+-mediated Cl− conductance in Xenopus oocytes. Our present data show that two members of this family, the human Gα16 subunit and the murine homologue Gα15, can induce both activation and inhibition of these agonist-induced currents. Although extremely low amounts (10-50 pg) of injected Gα16 subunit cRNA cause modest (∼2-fold) enhancement of ligand-induced Cl− currents in oocytes co-injected with thyrotropin-releasing hormone (TRH) receptor cRNA 48 h postinjection, larger Gα16 and Gα15 cRNA injections cause >10-fold inhibition of TRH or 5HT2c receptor responses. The inhibition is analyzed in this study. The inhibited currents are recovered if various Gβγ subunit combinations are also expressed with the Gα subunits. The constitutively active mutant, Gα16Q212L, also causes a strong attenuation of the ligand-induced Cl− currents, but this inhibition is not recovered by co-expression of Gβγ subunits. These results indicate that the free Gα subunit is responsible for the inhibitory signal. Although expression of TRH receptor alone produces maximum responses approximately 48 h after injection, co-expression of TRH receptor with Gα16 results in enhanced responses 6-12 h postinjection, followed by complete attenuation at 36 h. Furthermore, injection of Gα16 cRNA alone at comparable levels gives rise to spontaneous Cl− currents within 6-12 h postinjection, suggesting that the early spontaneous activation underlies the later suppression. Expression of other G protein α subunits of the Gq family, at cRNA levels considerably higher than effective for Gα16, produces both analogous spontaneous Cl− currents and, later, inhibition of ligand-induced Cl− currents. Experiments with direct injection of IP3 and of Ca2+ suggest that this inhibition is consistent with the down-regulation of IP3 receptors. These data indicate that both enhancement and inhibition of signaling through G protein-coupled receptors can be mediated by the expression level and/or activity of an individual G protein.

Many hormones, neurotransmitters, and growth factors act via G protein-coupled receptors (GPCRs) 1 in a wide range of transduction processes. The phosphoinositide cascade, which controls calcium-stimulated processes such as secretion, chemotaxis, and fertilization, is regulated by G proteins. Phosphoinositide phospholipase C (PLC) catalyzes the hydrolysis of phosphatidyl 4,5-bisphosphate to generate two second messengers, inositol 1,4,5-trisphosphate (IP 3 ) and diacylglycerol. Diacylglycerol in turn activates protein kinase C; IP 3 binds to an intracellular IP 3 receptor/channel to release Ca 2ϩ stores (Ref. 1 and references therein). In mammals, several families of PLC isoforms have been described (2,3). The PLC ␤ isoform is regulated by G proteins. All four ␤ isoforms of PLC are activated via the pertussis toxin-insensitive G␣ q family of G proteins: G␣ q , G␣ 11 , G␣ 14 , G␣ 15 , and G␣ 16 , although isoforms appear to vary with respect to potency for activation (4 -6). Two isoforms, PLC ␤2 and PLC ␤3, can also be activated by G␤␥ (6 -9). It also has been shown that the G␣ and G␤␥ subunits bind to different regions of the PLC ␤ enzyme (10,11).
The complexity of the molecular components and the functional interaction among them enable many variations in the control and timing of particular signaling pathways. Such specificity of signaling may be important in determining the cell's response to an extracellular signal. The pattern of expression of receptors, G proteins, and effectors may also play an important role in this selectivity process. For instance, the G␣ 16 and G␣ 15 subunits show a restricted pattern of expression (12,13); conversely, they have a promiscuous behavior with respect to receptor coupling (14,15). Regulatory processes such as short and long term desensitization may also determine the efficiency of signal transduction in a cell.
In Xenopus oocytes, activation of PLC by ligand-activated GPCRs produces the opening of a Cl Ϫ channel via Ca 2ϩ release from intracellular stores; this system has been extensively used for the study of GPCRs. Many receptors couple to an endogenous pertussis toxin-sensitive G protein when expressed in oocytes (16 -18); this pathway seems to involve G o . It is still unclear whether the ␤␥-dependent activation of PLC ␤ accounts for the pertussis toxin-sensitive regulation of PLC. Recent data suggest that the G␤␥ dimer regulates the activity of PLC in oocytes (19). Other receptors, e.g. the thyrotropin-releasing hormone (TRH) receptor, activate PLC via pertussis toxin-insensitive G proteins in oocytes (20,21). Antisense experiments show that these G proteins are members of the G q family. Additionally, the expression level of G proteins influences the selectivity of receptor/G protein coupling (21). Promiscuity of receptor/G protein interactions appears with increas-ing levels of G protein expression, indicating that the level of expression of G proteins and receptors are important for modulating transmembrane signaling.
Regulation of the PLC cascade in oocytes is an important step in oocyte maturation and fertilization. For example, it has been shown that G proteins of the G q family increase in expression level as the oocyte matures and acquires the capacity to respond to IP 3 (22). Although the IP 3 receptors are localized mostly in the cytoplasm of the oocyte, they are found mostly in the cortical region of the mature egg (23). In the egg, an increase in Ca 2ϩ concentration occurs at fertilization and propagates Ca 2ϩ waves that are required for the prevention of polyspermy and for the induction of cell cycle changes (24). In this study, we present further evidence for regulation of the pathway leading from the activation of G proteins to the activation of Cl Ϫ channels. We use this signaling cascade to examine the role of G proteins of the G q family to GPCR stimulation. We show that overstimulation of PLC, by means of expression of G␣ subunits of the G q family, produces signal inhibition and that this effect is the result of a decreased physiological response to IP 3 activation. We suggest that this effect is part of a desensitization process controlling the stimulation of the PLC pathway and that both enhancement and inhibition of signaling through GPCRs can be mediated by the level of G protein expression and/or activity.

EXPERIMENTAL PROCEDURES
In Vitro Synthesis of RNA-In vitro transcription of sense RNA was carried out as described previously (25) with a few modifications. Recombinant plasmids containing cDNA inserts were linearized by digestion with appropriate restriction enzymes. The transcription of linearized templates was performed in 7.6 mM Tris-HCl, pH 7.6, 6 mM MgCl 2 , 0.6 mM NaCl, and 10 mM dithiothreitol containing 0.5 mM each ATP, CTP, and UTP, 0.1 mM GTP, 0.5 mM 5Ј-(7-methyl)-GTP, and 180 units of the respective polymerase in a total volume of 250 l. The reaction mixture was incubated for 150 min at 37°C. The DNA template was subsequently removed by treatment with 5 units of RNase-free DNase I for 15 min at 37°C. Free nucleotides were removed using a Sephadex G50 column. The mRNA was phenol-chloroform-extracted and recovered by ethanol precipitation. The RNA was dissolved in RNase-free water at the corresponding concentration (see figure legends), divided into aliquots, and stored at Ϫ70°C until used.
Oocyte Expression and Electrophysiology-These procedures are described in detail elsewhere (26). Briefly, oocytes were defolliculated and maintained at 18°C in incubation medium containing ND96 (96 mM NaCl, 2 mM KCl, 1 mM MgCl 2 , 5 mM HEPES, pH 7.4), 1.8 mM CaCl 2 , 50 g/ml gentamycin, and 5% horse serum (27). Whole cell currents were measured at room temperature using either a Dagan 8500 amplifier or an Axon Instruments GeneClamp in a standard two-microelectrode voltage clamp configuration. Current was measured on-line by oscilloscope and chart recorder. Electrodes were filled with 3 M KCl and had a resistance of 1-2 megaohms. During experiments, the oocytes were clamped at Ϫ80 mV and superfused continuously in ND96 medium; all drugs were applied in this solution. Injections of IP 3 and Ca 2ϩ were performed using a pressure injector and a small bore (ϳ2-m) injection pipette.
Materials-TRH was from Peninsula Laboratories (Belmont, CA), and 5HT was from Research Biochemicals Inc. (Natick, MA). All other drugs and reagents were from Sigma.

Enhancement and Inhibition of Responses by G Protein
Subunits-Superfusion of TRH onto oocytes voltage-clamped at negative holding potentials (Ϫ80 mV) and expressing TRH receptor produced an inward current characteristic of the PLCand IP 3 -mediated Ca 2ϩ -activated Cl Ϫ conductance (Fig. 1A). We have shown previously (21) that the response increases monotonically with the receptor cRNA concentration under the conditions of these experiments. In addition, co-injection of cRNA for some G protein ␣ subunits (G␣ q , G␣ 11 , G␣ 14 , G␣ o a, and G␣ o b in the range of 1-5 ng of cRNA) as well as for receptor enhances agonist-induced currents (maximum 4 -6-fold in-crease) over the values for receptor injections alone (21).
Surprisingly, when cRNA for another member of the G q family, G␣ 16 , was co-injected with the TRH receptor cRNA, a strong inhibition of the TRH-induced peak inward current was observed 48 h postinjection (Fig. 1A). The inhibition depended on the amount of cRNA injected (Fig. 1B), reaching Ͼ90% at 1 ng of G␣ 16 cRNA, and occurred at cRNA amounts 10 times lower than the injection amounts at which other G protein ␣ subunits produced detectable increases in ligand-induced Cl Ϫ currents (e.g. see Fig. 1C). This suggested that G␣ 16 was more potent than other members of the G␣ q subfamily. High potency is also suggested by the fact that even smaller quantities of G␣ 16 cRNA (0.001-0.05 ng) actually increased TRH-induced currents (Fig. 1B). However, this paper focuses on the novel inhibition produced by G␣ 16 cRNA levels at injected levels Ն0.5 ng. Each of several cRNA preparations yielded such inhibition. As was seen with oocytes expressing G␣ 16 and TRH receptor, G␣ 16 also inhibited peak inward currents in oocytes injected with the serotonin 5HT2c receptor and perfused with 5HT ( Fig.  1B, inset). G␣ 16 is the human homologue of murine G␣ 15 , another member of the G q family of G protein ␣ subunits. The murine homologue G␣ 15 inhibited TRH or 5HT responses similarly to G␣ 16 (Fig. 1B, inset).
We have examined other aspects of the inhibition of the signaling pathway by these two G protein ␣ subunits. We studied first the effect of co-injecting G␣ q cRNA (which leads to enhancement of TRH-induced Cl Ϫ currents) with G␣ 16 cRNA. In oocytes injected with 8 times less G␣ 16 cRNA than G␣ q (at G␣ q levels that by themselves increased the TRH-induced Cl Ϫ current 2-3-fold), the Ͼ5-fold inhibition of Cl Ϫ currents by G␣ 16 was the dominant effect ( Fig. 1C).
G␣ 16 and G␣ 15 are expressed in hematopoietic lineages of human and mouse, respectively (12,13). They activate the PLC ␤2 expressed in these lineages and also other PLC ␤ isoforms (10, 28, 29). We considered the possibility that the inhibition observed in oocytes could be due to the inability of G␣ 16 or G␣ 15 to activate the corresponding PLC ␤ isoform, termed PLC X␤, present in Xenopus oocytes. However, the enhancement observed at low injected G␣ 16 concentrations argues against this possibility. Also, we have recently shown that G␣ 16 and G␣ 15 can activate PLC X␤ 2 by co-expressing Xenopus phospholipase PLC X␤ with various G␣ subunits in transiently transfected COS-7 cells. Therefore, it seems unlikely that the inhibitory effect we observed when expressing G␣ 15 or G␣ 16 in oocytes was due to the inability of these subunits to activate downstream effectors.
These results prompted us to seek other evidence for the hypothesis that G␣ 16 or G␣ 15 subunits in oocytes activate the PLC X␤ and that the prolonged activation by free G␣ subunits induces Cl Ϫ current inhibition. We co-injected G␤␥ subunits with G␣ 16 in presence of the TRH receptor (Fig. 2). Functional activity of the G␤␥ dimer was verified by co-expressing these subunits with the inward rectifying atrial potassium channel (GIRK1), which is activated by free G␤␥ subunits (31)(32)(33). Co-injection of G␤␥ with TRH receptor and G␣ 16 prevented most of the inhibition produced by co-injection of TRH receptor and G␣ 16 alone. Subunit combinations consisting of G␤ 1 ␥ 2 and G␤ 2 ␥ 2 were each capable of recovering these currents. We believe that this effect occurs because G␤␥ sequesters free G␣ and therefore prevents the action of free G␣ subunits on downstream effectors. We also considered the possibility that G␤␥ alone had an effect on the activity of PLC X␤ independent of the G␣ subunit, but we observed no significant change in ligandinduced Cl Ϫ currents when TRH receptor was co-injected with G (in the absence of injected G␣ subunit cRNA) (Fig. 2). This result argued against a direct effect of G␤␥ on ligand-induced PLC activation under our conditions. We further addressed this question by injecting the constitutively active mutant of G␣ 16 , G␣ 16 Q212L, which does not bind to G␤␥. The G␣ 16 Q212L mutation produced the same inhibitory effect on TRH-induced Cl Ϫ currents as did wild-type G␣ 16 , but this effect could not be reversed by the presence of G␤␥ (Fig. 3).
The data presented thus far are consistent with the hypothesis that free G␣ 16 subunits produce the inhibitory effect. More specifically, G␣ 16  evidence for an activity of free G␣ was obtained from examining the effect of expressing G␣ 16 alone or with TRH receptor at various times following cRNA injection. The results of these time course experiments are shown in Fig. 4A. Injection of cRNA for TRH receptor produces a monotonic increase in Cl Ϫ currents from 6 to 50 h postinjection. This response is most likely due to increasing amounts of receptor with time after injection. At 6 h postinjection, co-injection of TRH receptor and G␣ 16 cRNA produced a ligand-induced peak inward current 4 times larger than in oocytes expressing TRH receptor alone. Contrary to the responses of oocytes expressing TRH receptor alone, oocytes with both TRH receptor and G␣ 16  . Measurement of Cl Ϫ currents for each group of oocytes was performed at various times following cRNA injection (abscissa). For the two oocyte groups expressing TRH receptor, currents were elicited with at a holding potential of Ϫ80 mV. The holding current prior to TRH application was less than Ϫ200 nA in all oocytes. For the oocytes injected with G␣ 16 cRNA alone, the oocyte was clamped at the Cl Ϫ equilibrium potential (approximately Ϫ25 mV) and then jumped to Ϫ80 mV for 5 s. Data are the mean Ϯ S.E. for 5-7 oocytes/condition/time point; some oocytes in each group were measured at more than one time point. B, recordings from individual oocytes injected with 10 ng of G␣ 16 cRNA at various time points postinjection. The solid bars above each trace represent a voltage jump from the Cl Ϫ equilibrium potential to Ϫ80 mV. For comparison, the response of a representative uninjected oocyte is included. C, increasing the concentration of injected Gaq cRNA inhibits agonist-induced Cl Ϫ currents. Amounts of injected cRNA (ng) for each condition are shown below the abscissa. Data are the mean Ϯ S.E. for 4 -6 oocytes/condition recorded 48 h postinjection. decrease in peak inward currents after 12 h and complete attenuation after 36 h.
In oocytes injected with G␣ 16 alone, we also observed an important new phenomenon: spontaneous oscillatory Cl Ϫ currents. These spontaneous responses with G␣ 16 also provide further evidence that the G␣ 16 subunit alone can activate PLC X␤. In general, injection of larger cRNA amounts resulted in larger peak spontaneous currents, larger numbers of spontaneous events, and a longer response duration at earlier time points as compared with oocytes injected with smaller amounts of cRNA. For 10-ng injections, the spontaneous signals were maximal (ϳ600 nA) at 12 h and decreased afterwards (Fig. 4, A  and B). For 0.5-ng injections, the spontaneous currents also peaked in amplitude at 12 h and vanished by 48 h but had considerably smaller amplitudes (Table I). For 0.05-ng injections, the spontaneous currents occurred only after Ն24 h, persisted until at least 48 h, and were only a few nA in amplitude (Table I). Fig. 4A shows that the peak and subsequent decline in the spontaneous responses for oocytes expressing G␣ 16 alone has a time course that matches the peak and decline of TRH responses for oocytes expressing both TRH receptor and G␣ 16 , as if the spontaneous responses are first activated and later inhibited by the same mechanism that first enhances and later inhibits the receptor responses. In the most straightforward hypothesis, inhibition of the signaling pathway is the eventual result of spontaneous activation induced by G␣ 16 .
Because we hypothesize that G␣ 16 differs from other G proteins, for instance G␣ q , primarily in the quantitative potency of coupling, we have sought and found spontaneous oscillations induced by the latter G protein as well (Table I). These responses occur at injection levels (5 and 25 ng) greater than those previously studied (21).
Inhibition at High G␣ q Levels-The hypothesis that G␣ 16 differs only quantitatively from other G proteins, because of intrinsic activity, also led us to seek an inhibitory effect of other G proteins. An inhibitory effect analogous to that produced by G␣ 16 could indeed be observed with other G protein ␣ subunits that activate PLC, again when injected at higher cRNA concentrations than previously tested (21). In Fig. 4C we present the results for co-injection of increasing quantities of G␣ q cRNA with TRH receptor cRNA. Similar results were observed with G␣ 11 co-expression and with other receptors (data not shown).
Localization of the Inhibitory Effect at IP 3 Receptors-We next sought to determine where along the pathway the inhibitory signal was produced. Mammalian PLC is the target of phosphorylation by various kinases (34,35). Some of these modifications inactivate the enzyme. The IP 3 receptor can also be phosphorylated by various kinases, although the effects on the activity are not clear (see Ref. 36). We tested the hypothesis that activation of PLC, which leads to generation of diacylglycerol and the concomitant activation of protein kinase C, would in turn lead to the inactivation of a key messenger of this signaling cascade. Oocytes injected with 1 ng of G␣ 16 were treated either chronically (from 1 h postinjection to the time of assay) or acutely (1 h prior to assay) with protein kinase C inhibitors (either staurosporine or bisindolylmaleimide). Little change in spontaneous Cl Ϫ currents induced by G␣ 16 expression was observed with any of the conditions tested (data not shown). We also co-injected excess PLC X␤ cRNA to determine whether it is the rate-limiting step in G␣ 16 inhibition. Coinjection of PLC X␤ produced a 2-fold increase in the ligandinduced Cl Ϫ currents but did not prevent the inhibition produced by the presence of G␣ 16 (data not shown).
IP 3 binds to a specific receptor on intracellular membranes and is a key control point in the regulation of phospholipaseactivating pathways. To study the role of the IP 3 receptor in G␣ 16 inhibition, we injected IP 3 directly into the oocyte, bypassing G protein induction of PLC. These experiments were performed with oocytes expressing either G␣ 16 alone or G␣ 16 plus TRH receptor. A rapid induction of Cl Ϫ currents was observed upon injection of 100 pmol of IP 3 in all oocytes. Fig. 5A summarizes the data and shows representative traces. However, oocytes injected with G␣ 16 with or without receptor showed a marked reduction in IP 3 -induced Cl Ϫ currents (to ϳ30% of control).
We next studied the effect of activating the Cl Ϫ channel directly by injecting Ca 2ϩ into the oocyte. Injection of 50 pmol of CaCl 2 produced comparable responses in control oocytes and in oocytes injected with G␣ 16 (Fig. 5B). These results indicate that the Cl Ϫ channels themselves were not affected by the prolonged stimulation of the PLC ␤ pathway and suggest that the reduction of the peak currents observed in oocytes expressing G␣ 16 , G␣ 15 , or other G␣ subunits in excess is a consequence of physiological alterations in IP 3 receptor responses. DISCUSSION Activation of GPCRs expressed in Xenopus oocytes generates a well characterized second messenger cascade that leads to opening of Ca 2ϩ -activated Cl Ϫ channels. However, regulation of the sensitivity in this cascade is only partially understood. In the present study, we show that the expression of the human G␣ 16 , or the murine homologue G␣ 15 , produces inhibition of peak inward Cl Ϫ currents activated via GPCRs when measured more than 24 h postinjection. This inhibitory effect is likely the result of a desensitization process that blocks the release of Ca 2ϩ from internal stores, due to the prolonged activation of PLC ␤ by free G␣ 16 subunits. As evidence for such a mechanism, we show that (i) expression of G␣ 16 or G␣ 15 inhibits the currents induced by ligands for each of two co-expressed receptors; (ii) co-expression of G␤␥ with G␣ 16 subunits restores this current; (iii) the presence of the activated mutant of G␣ 16 , G␣ 16 Q212L, results in a similar block of the ligand-induced Cl Ϫ conductance, but co-injection of G␤␥ does not restore the signal; (iv) oocytes injected with G␣ 16 alone display a spontaneous Cl Ϫ conductance at 6 -12 h postinjection, after which peak currents decrease; and (v) in oocytes injected with G␣ 16 and TRH receptor cRNA, agonist produces larger than normal currents at early times postinjection (6 -12 h) but a reduced response at later times. These effects suggest that the signaling cascade initiated through a single GPCR can be both enhanced and inhibited depending upon the level of G protein expression and/or activity.
The prolonged activation of PLC ␤ by free G␣ 15 or G␣ 16 that results in signal inhibition is likely not a peculiarity of G␣ 16 or G␣ 15 subunits. Low quantities of other G␣ subunit cRNAs enhance transmitter responses (21), but when higher quantities of other G␣ subunit cRNAs are injected there is an inhibitory effect analogous to that seen with smaller quantities of G␣ 16 , in agreement with previous observations injecting large amounts of wild-type G␣ q (37). We suggest that any prolonged stimulation of the PLC pathway in oocytes will result in signal down-regulation.
It is at first surprising that the rather small spontaneous currents with 0.5 ng of G␣ 16 cRNA (60 -70-nA peak, at 12 h after injection) represent activation sufficient to desensitize most of the release mechanism. We point out, however, that the spontaneous activation is present for many hours, whereas most experiments on oocytes utilize a single application of agonist, lasting just seconds. Furthermore, the IP 3 receptors that mediate the spontaneous currents have presumably been desensitized by the same process that desensitizes the responses to exogenously applied agonists. We do not know whether Ca 2ϩ itself is the desensitizing factor, but previous studies show that neither activation nor desensitization of the Cl Ϫ channel response is simply related to the concentration of intracellular Ca 2ϩ (38,39). The barely detectable spontaneous responses to 0.05 ng of G␣ 16 (Table I) were evidently not associated with activation sufficient to desensitize the signaling pathway (Fig. 1B).
One important question concerns the mechanism by which the wild-type G␣ 16 or G␣ 15 subunit produces this dramatic effect even at relatively low quantities of cRNA injected, while other subunits need to be constitutively active or injected at higher quantities. These two subunits have the peculiarity that they couple to a broad range of GPCRs (14). One possibility could be that these G proteins in the oocyte are being activated by endogenous receptors. On the other hand, a lower affinity for endogenous G␤␥ subunits or a higher efficiency in activating endogenous PLC ␤ isoforms may account for the unusual effects of G␣ 16 or G␣ 15 on the Cl Ϫ current response. It is interesting to note that injection of 1-5 ng of the other G␣ subunit cRNAs produces an increase in peak current when co-injected with TRH receptor (21). Above this amount, all G␣ subunits of the G q family induce desensitization. Presumably, above a threshold at which excess G␣ subunits can no longer interact with G␤␥ and couple to the receptor, the free G␣ subunits induce activation of PLC and induce signal desensitization. Because we have no direct measurements of G␣ subunit levels, it remains possible that the unusual effects of G␣ 16 or G␣ 15 occur simply because unusually large amounts of these G␣ subunits are produced in the oocyte.
This study shows that G␣ 16 and G␣ 15 subunits join the list of pertussis toxin-insensitive G protein subunits that can activate an effector in oocytes, leading to Ca 2ϩ mobilization (21). cDNAs Ca 2ϩ eliminates the G␣ 16 -mediated inhibition of Cl Ϫ currents. Voltageclamp currents were measured at a holding potential of Ϫ80 mV during injection of 25 nl of H 2 O containing 100 pmol of IP 3 . Amounts of injected cRNA (ng) for each condition are shown below the abscissa. Data are the mean Ϯ S.E. for 4 oocytes/condition recorded 48 h postinjection. Representative traces below each graph are for oocytes expressing TRH receptor alone or in combination with G␣ 16 . The initial inward deflection prior to IP 3 or Ca 2ϩ injection is a mechanical artifact due to insertion of the injection pipette. for other pertussis toxin-insensitive G proteins, including G␣ q and G␣ 11 , have been cloned from Xenopus (37,40). Using antibodies against the C-terminal sequence of mouse G q it has been shown that oocytes express several forms of the G q family of ␣ subunits (22). There has, however, been some controversy concerning the ability of G proteins of the G␣ q family to activate the PLC cascade in the oocytes. Our earlier study showed that responses are enhanced by injection of cRNA for G␣ q or G␣ 11 (21), but in other reports G␣ q or G␣ 11 injected into the oocyte has reduced Cl Ϫ currents (19,41). The results presented here support the idea that the G␣ subunit of all members of the G␣ q family can couple to GPCRs and activate the oocyte's IP 3 /Ca 2ϩ -induced Cl Ϫ conductance. Under certain conditions, this coupling can eventually produce an inhibitory signal that we suggest may be due to the overexpression of G protein ␣ subunits and, therefore, to overstimulation of phospholipase C␤. IP 3 is a key second messenger for the mobilization of Ca 2ϩ from internal stores. An IP 3 receptor isoform has been cloned from oocytes (23). Previous experiments had shown that injection of IP 3 into oocytes renders them unresponsive to subsequent stimulation by either receptor activation or IP 3 injection (42). Recently, Honda et al. (43) showed that the loss of IP 3 sensitivity due to down-regulation of the IP 3 receptor is the main reason for the G␣ q -induced desensitization of the plateletactivating factor receptor response. The results presented here demonstrate that the presence of wild type G␣ 16 and G␣ 15 is enough to produce a similar desensitization effect and extend the observation to the other members of the G q family. Taken together, these results suggest that the regulation of the IP 3 receptor plays a pivotal role in the propagation of agonistmediated Ca 2ϩ transients. This study provides a model to study the mechanism of a more physiological effect like heterologous desensitization, where the activation of several receptors may trigger similar down-regulation responses. The precise mechanism of down-regulation of the IP 3 receptor remains unclear.
In mammals three different isoforms of IP 3 receptors have been identified. There is accumulating evidence that individual isoforms have distinct tissue and subcellular localization, pharmacological properties, and regulation by calcium and phosphorylation (see Ref. 36). For example, the type I IP 3 receptor is down-regulated in carbachol-stimulated human neuroblastoma cells (44,45). It will be interesting to ascertain how persistent activation of PLC by different G protein ␣ subunits or ␤␥ subunits affects the regulation of agonist-mediated Ca 2ϩ transients in other cellular environments.