GTP analogues stimulate inositol trisphosphate formation transiently in Dictyostelium

Transcription

1 GTP analogues stimulate inositol trisphosphate formation transiently in Dictyostelium G. NICHOLAS EUROPE-FINNER and PETER C. NEWELL Department of Biochemistry, University of Oxford, South Parks Road, Oxford 0X1 3QU, UK Summary Permeabilization of amoebae of Dictyostelium discoideum with saponin was found not to uncouple the chemotactic cell surface cyclic AMP receptors from inositol trisphosphate (EP3) formation, and stimulation of permeabilized amoebae with 50nM-cyclic AMP produced peaks of IP3 at 5, 15 and 30 s in a manner comparable to that seen previously in non-permeabilized cells. The possible involvement of a GTP-binding protein (Gprotein) in this IP3 signal transduction pathway was investigated by studying the effects on such permeabilized amoebae of added GTP and nonhydrolysable GTP analogues. While GDP produced only very minor effects, stimulation of the amoebae (in the absence of added cyclic AMP) with GTP or the non-hydrolysable GTP analogues GTPyS (guanosine S'-O-(3-thio-triphosphate)) and Gpp(NH)p (5-guanylylimidodiphosphate) induced transient formation of IP3 in an oscillatory manner, with peaks similar in magnitude and timing to those elicited by cyclic AMP. A dose-response curve for GTPyS indicated a concentration for half-maximal stimulation of approximately 8^M. When tested at 300 s after addition of GTPyS, the basal level of IP3 was found to be twofold elevated with shallow (presumably asynchronous) oscillations still just discernible. The significance of the EP 3 oscillations elicited by GTP and its analogues is discussed in relation to the mechanism of signal adaptation and the presumed role of G-proteins. Key words: inositol trisphosphate, IP 3, GTP, GTPyS, Dictyostelium, G-protein Introduction Recent data from our laboratory have indicated that the mechanism of signal transduction used during amoebal chemotaxis of Dictyostelium discoideum involves formation of inositol phosphates. When added to saponinpermeabilized amoebae, both IP 3 * (the 1,4,5 isomer) and Ca 2+ have been found to mimic cyclic AMP in its chemotactic actions in forming cyclic GMP and polymerizing cytoskeletal actin (Europe-Finner & Newell, 1985, 19866; Small et al. 1986), and is able to release Ca 2+ from intracellular (non-mitochondrial) stores (Europe-Finner & Newell, 1986a). More recently, it IP3, inositol trisphosphate. (The chromatographic technique employed in this study does not distinguish between the isomeric forms 1,3,4- and 1,4,5-IP3 and the term IP3 also includes any of the intermediate 1,3,4,5-inositol tetrakisphosphate (IP 4 ) present); GTPyS, guanosine 5'-O-(3-thio-triphosphate); Gpp(NH)p, 5'-guanylylimido-diphosphate. Journal of Cell Science 87, (1987) Printed in Great Britain The Company of Biologists Limited 1987 was demonstrated that rapid and transient formation of IP 3 was detectable at or before 5 s and at 15 s after stimulation of amoebae with 50nM-cyclic AMP (Europe-Finner & Newell, 1987). From these studies we have deduced that stimulation of the cell surface cyclic AMP receptors induces IP3 formation, which is able to liberate Ca 2+ from internal stores that (directly or indirectly) triggers chemotactic cell movement (Newell et al. 1987). In several other signal transduction systems that operate ma IP3, an involvement of GTP-binding proteins (G-proteins) has been inferred from the ability of GTP and non-hydrolysable analogues to stimulate IP3 formation. These studies have employed permeabilized blood platelets (Haslam & Davidson, 1984), permeabilized rat GH 3 pituitary cells (Martin et al. 1986) and also cell-free membrane preparations such as neutrophil plasma membranes (Cockcroft & Gomperts, 1985), rat hepatocyte plasma membranes (Wallace & Fain, 1985), rat GH 3 pituitary cell membranes 513

2 (Straub & Gershengorn, 1986), membranes of insect salivary glands (Litosch et al. 1985) and platelet membranes (Baldassare & Fisher, 1986). In the Dictyostelium system, evidence for G-proteins has come from the work of Leichtling et al. (1981) who reported a 42xlO 3 A/ r, GTP-binding membrane protein that could be ADP-ribosylated, and more recently from studies of the effects of guanine nucleotides on the kinetics of binding of cyclic AMP to isolated membranes (De Wit & Snaar-Jagalska, 1985; Janssensei al. 1985, 1986; Van Haastert et al. 1986) but no evidence has been presented for G-proteins being involved in the inositol phosphate forming transduction pathway. In order to determine whether or not G-proteins regulate IP 3 -mediated signal transduction in Dictyostelium we conducted experiments that monitored the formation of IP3 in saponin-permeabilized amoebae and determined the ability of GTP and non-hydrolysable analogues to stimulate this formation in the presence and absence of the chemotactic stimulant cyclic AMP. Materials and methods Materials L-Myo-[l,2-3 H]inositol (sp. act. 61-2Cimmol~') was obtained from New England Nuclear and passed through a Dowex 1-X8 column (formate form) prior to use to remove trace impurities. [ 3 H]inositol 1,4,5-trisphosphate (potassium salt: sp. act. 1 Cimmol" 1 ) was obtained from Amersham Int. PLC. Hepes (iv-2-hydroxyethyl-piperazine-a f '-2- ethane-sulphonic acid) and EDTA were from Sigma. Dowex 1-X8 ( mesh chloride form) anion exchange resin was obtained from Bio-Rad. GTPyS and Gpp(NH)p were obtained from Boehringer-Mannheim. GTP and GDP were obtained from Sigma. Harvesting of amoebae D. discoideum strain NC4 was grown in association with Websiella aerogenes (strain OXF1) on SM nutrient agar (Sussman, 1966). Amoebae were prepared by growth as lawns on SM agar under conditions permitting uniform clearing of the bacteria by the feeding amoebae. Amoebae were harvested from the bacterial plates in P buffer (17mM- KH2/Na2H phosphate, ph61) and washed free of bacteria by centrifugation at 190 g for 2min. After three such washes the cells were resuspended in P buffer at 2xl0 7 ml~ 1. Labelling of amoebae with [ 3 H]inositol Amoebae suspended in P buffer at 2xlO 7 cells ml" 1 were incubated in the presence of 163 nm-[l,2-3 H]inositol at a specific activity of 10fiCiml~'. Dihydro-streptomycin sulphate and CaClj were added to give a final concentration of 0-25 mgml" 1 and 1 mm, respectively. Cells were incubated at 22 C with aeration in a rotary incubator at 170 revs min" 1. After 4h cells were pulsed with cyclic AMP (50 nm final concentration) at 15-min intervals so as to synchronize the amoebae. After 5 h the amoebae were washed four times with 20ml of P buffer so as to wash out [ 3 H]inositol, and resuspended at 5x 10 7 cells ml"' in P buffer. Permeabilization of amoebae To permeabilize amoebae they were incubated at 5X 10 7 cells ml" 1 in the presence of lmgml" 1 saponin and 5 mm-atp and shaken at 22 C in a rotary incubator at 170 revs min~' for 30 min. The concentration of saponin used was chosen as that showing no significant cell lysis and giving reproducible permeability as judged by uptake of ethidium bromide (100^gml" 1 ) seen under ultraviolet light with an epifluorescence microscope (Europe-Finner & Newell, 1985) or, for recent experiments, by uptake of Giemsa stain (Europe-Finner, Ludfirus & Newell, unpublished). ATP was added to maintain polyphosphoinositide concentration (Bradford & Rubin, 1986). After treatment with saponin, amoebae were washed three times with P buffer and resuspended at 10 s cells ml" 1. Stimulation of labelled permeabilized amoebae with guanine nucleotides Samples (1 ml) of labelled permeabilized amoebae (10 8 cells ml"') were added to plastic vials and the free [Ca 2+ ] was set at 180 nm using Ca z+ /EGTA buffers (Europe-Finner 6 Newell, 19866). To maintain polyphosphoinositide concentrations and thus optimize IP3 formation in permeabilized amoebae, they were pretreated for 5 min at 22 C on an IKA- Vibrax platform shaker at 1400 revs min" 1 with lofim-antimycin A (to prevent endogenous substrate oxidation), 5 mm- ATP and an ATP-regenerating system consisting of 5 mmcreatine phosphate and 5 units ml" 1 creatine phosphokinase. The mitochondrial inhibitors oligomycin (10/iM) and dinitrophenol (0-5 mm) were also included to ensure that the observed responses were entirely those of permeabilized cells and not of any unpermeabilized cells that may have been present. Mg 2+ was excluded from the additions to the incubation medium, since Cockroft & Gomperts (1985) found that this ion increased the activity of inositol phosphate phosphatases in rabbit neutrophils and might inadvertently have affected the kinetics of inositol phosphate accumulation. For time-course studies, amoebae were stimulated with 20^1 of GTPyS, Gpp(NH)p, GTP or GDP to give a final concentration of 68/IM, 70 [1M, 62^M or 82/AM, respectively, and stopped at the appropriate times with 100/il of 10% (v/v) HCIO4. The acid extracts were then left on ice for at least 30 min and neutralized with 135 f.i\ of 1-53M-KOH and then buffered to ph7-2 with 40^1 of 75 mm-hepes (N-2- hydroxyethyl-piperazine-a''-2-ethanesulphonic acid). KCIO4 was precipitated at 0 C for 30 min on ice and removed by brief centrifugation. Measurement of [ 3 H]inositol phosphate Soluble [ 3 H]inositol phosphates were separated by anion exchange chromatography from neutralized extracts on Dowex-1X8 formate form resin (Bonee< al. 1984; Berridgeet al. 1983; Europe-Finner & Newell, 1987). The identification of IP3 was validated using standard 3 H-labelled 1,4,5-IP3. Determination of the radioactivity in the [ 3 H]inositol phosphates peaks was as described by Europe-Finner & Newell (1987). 514 G. N. Europe-Finner and P. C. Newell

3 Results Because previous studies concerned with cyclic GMP formation in D. discoideum had shown that permeabilization of the amoebae with saponin did not uncouple the receptors from the intracellular cyclic GMP response (Small, Europe-Finner & Newell, unpublished), it seemed worthwhile to test the effect of permeabilization on the coupling of the cyclic AMP receptors to the formation of IP3. When saponinpermeabilized amoebae were stimulated at 22 C with 50 nm-cyclic AMP, they rapidly responded by accumulating IP3 in a transient oscillatory manner (Fig. 1). Such an oscillatory response is similar to that reported previously for non-permeabilized amoebae (Europe- Finner & Newell, 1987). The response differs from that with non-permeabilized amoebae in the precise timing of the peaks: the first peak is seen 5 s after stimulation in permeabilized amoebae incubated at 22 C, whereas in non-permeabilized amoebae this first peak often occurred too rapidly at 22 C for the peak to be detected at 5 s and could be routinely measured at this time point only when cellular events had been slowed down by cooling the amoebae to 4 C. A transient response, as seen in Fig. 1, may be due to a negative feed-back system or to some limitation of the duration of the response due to cyclical inactivation of a regulator such as a G-protein. The presence of G- proteins is generally assessed by the stimulatory (or in some cases inhibitory) effects of GTP or non-hydrolysable analogues of GTP on the system. As previous studies had shown that intact D. discoideum amoebae were unaffected in their chemotactic responses by the presence of GTP or analogues (Small, Europe-Finner & Newell, unpublished data), saponin-permeabilized amoebae were used to study the effect of GTP on formation of IP3. When such permeabilized amoebae were treated with 62^M-GTP in the absence of any added cyclic AMP and the cells analysed for IP 3 formed, it was found that IP3 accumulated rapidly in an oscillatory manner (Fig. 2). The time-scale was similar to that for stimulation by cyclic AMP except that the third peak seen at 30 s with cyclic AMP was generally missing. When GDP was used in place of GTP, only a small accumulation of IP3 was observed (Fig. 2). This very minor stimulation can probably be accounted for by phosphorylation of the GDP to GTP since ATP, creatine phosphate and creatine phosphokinase were routinely added to maintain the cellular ATP concentration. When the non-hydrolysable analogue GTPyS was used in place of GTP, a similar pattern of IP3 accumulation was observed (Fig. 3). This result was surprising, as it had been anticipated that the inability to hydrolyse the GTP analogue would have switched on the formation of IP3 in a more permanent fashion, 5000 Fig. 1. Time course of IP3 accumulation in saponinpermeabilized amoebae of D. discoideum after stimulation with 50 nm cyclic AMP ( ) or water (O O). Amoebae were incubated for 4h with lo^ciml" 1 [1,2,- 3 H]inositol (sp. act. 61-2Cimmol~') then permeabilized with saponin and stimulated at 22 C with 50 nm-cyclic AMP or water (controls) and extracts made at the times shown. Although the permeabilization procedure appeared, by ethidium bromide tests, to affect all cells, the possibility of effects due to residual non-permeabilized amoebae was avoided by the routine inclusion of the mitochondrial inhibitors oligomycin (10fiM) and dinitrophenol (0-5mM) in the incubation mixtures. The neutralized extracts were applied to anion exchange columns and IP 3 was determined for each point from the radioactivity in the IP3 peak (peak V, see Europe-Finner & Newell, 1987) eluted from the columns. Results are means of three independent experiments using samples counted in triplicate in each experiment. Error bars represent S.E.M. Fig. 2. Time course of IP3 accumulation in saponinpermeabilized amoebae of D. discoideum after stimulation with GTP (62m) ( ) or GDP (82fm) (A A). Conditions were as described in Fig. 1. Results are the means of samples counted in triplicate. Stimulation with GDP was repeated in an independent experiment with similar results. Inositol trisphosphate in Dictyostelium 515

4 rather than in the observed transient manner. In case this effect was due to some peculiarity of degradation of GTPyS by D. discoideum a similar experiment was carried out using the analogue Gpp(NH)p (5'-guanylylimido-diphosphate). The result (not shown) was similar to that shown in Fig. 3, except that the second peak at 15 s was of a similar magnitude to the 5 s peak, rather than smaller as observed with GTPyS. When different concentrations of GTPyS were used to stimulate permeabilized amoebae, the IP 3 -forming response was found to show saturation, the concentration for half-maximal stimulation being approximately 8JUM (Fig. 4). As GTPyS caused an oscillation in the formation of IP3, it was of interest to see if this were maintained over an extended period and whether this treatment affected the ability of cyclic AMP to stimulate the cells. When permeabilized amoebae were stimulated with GTPyS and then incubated for 300 s before sampling for IP 3 assay, it was found (Fig. 5) that the basal IP3 concentration was considerably elevated over the controls without GTPyS. Very shallow oscillations could also just be detected, which appeared to be non-significant when several experiments were averaged as in Fig. 5. The simplest interpretation of such results is that the oscillations were continuing but had become asynchronous over the 300-s incubation period. When, after 300s incubation with GTPyS, the amoebae were stimulated with cyclic AMP they responded by producing an increment at 5 s that was roughly similar to that produced by cyclic AMP in cells without GTPyS (Fig. 1). Second and third peaks were also seen, although they tended to be shallower (particularly the second peak) and less clearly defined. _ 5000 Discussion The data presented provide evidence for the involvement of a G-protein in the chemotactic signal transduction pathway of D. discoideum. We have found that GTP and non-hydrolysable GTP analogues stimulate 3000 Concentration of GTPyS (/un) Fig. 4. Dose-response curve for IP3 formation by permeabilized D. discoideum amoebae stimulated with various concentrations of GTPyS. The accumulation of IP3 was measured 5 s after stimulation using the method described for Fig. 1. Results are means of three independent experiments using samples counted in triplicate in each experiment. Error bars show S.E.M. 00 = 5000 b ints min / - / i +camp (4-GTPyS) \ + Water ( +GTPyS) n its min ' , T / \ +GTPyS VwIieT^ ^B= ^ i ' Fig. 3. Time course of IP3 accumulation in saponinpermeabilized amoebae of D. discoideum after stimulation with GTPyS (68 fm) ( ) or water (O O). Conditions were as described for Fig. 1. Results are means of four (GTPyS) or three (water) independent experiments using samples counted in triplicate in each experiment. Error bars represent S.E.M. ^2000 x : 1000 i Fig. 5. Time course of IP3 accumulation in saponinpermeabilized amoebae of D. discoideum, which had been preincubated with GTPyS. Amoebae were preincubated for 300 s with GTPyS (68 fim) and then stimulated with either water (O O) or cyclic AMP (50 nm) ( ) at time zero. Data for control amoebae that had not been preincubated with GTPyS but were stimulated with water at time zero are also shown to indicate basal values ( ). Other conditions were as described for Fig. 1. Results are means of three independent experiments using samples counted in triplicate in each experiment. Error bars show S.E.M. 516 G.N. Eumpe-Finner and P. C. Newell

5 IP 3 formation in a manner similar to the natural chemoattractant cyclic AMP, and the concentration of GTPyS needed for half-maximal stimulation is similar to that in other systems (such as permeabilized blood platelets; Haslam & Davidson, 1984) that are thought to involve such GTP-binding proteins. Studies with mammalian systems often incorporate lithium salts into the incubation as this increases accumulation of inositol phosphates (including 1,3,4- IP3) by inhibition of inositol monophosphate phosphatase (Berridge et al. 1982; Bone et al. 1984; Burgess et al. 1985). In D. discoideum, lithium has only modest effects on inositol phosphate metabolism (Europe- Finner & Newell, 1987) and consequently cannot be used to increase their accumulation. This forces us to monitor transient changes rather than steady accumulation of breakdown products such as 1,3,4-IP 3 (which is not separated from the primary product 1,4,5-IP3 using anion exchange chromatography) but has the resulting advantage that we see events that reflect more closely the natural pattern of changes in vivo in the absence of inhibitors. The rapid oscillations that we observe in the accumulation of IP3, which are induced both by cyclic AMP and GTP (and analogues), may reflect the special needs of the amoebae for fast cyclical changes to control the rapid events of chemotaxis. It is conceivable, however, that such IP3 oscillations are more widely occurring in mammalian systems than is at present obvious. The use of lithium without separation of the 1,3,4- and 1,4,5-IP 3 by high-pressure liquid chromatography (HPLC) in some systems and the lack of synchronization of the cell populations would have the effect of integrating transient peaks into a steady accumulation. It is noteworthy that in our studies the effects on IP 3 accumulation of non-hydrolysable analogues of GTP are similar to those of GTP. From these results we deduce that the hydrolysis of GTP to GDP bound to the postulated G-protein does not regulate the duration of the stimulation. Some other regulatory molecule must be responsible for the transient nature of the formation of IP3. This brings into question the role of the postulated G-protein if hydrolysis of its bound GTP does not affect the response. However, the fact that GTP can induce IP3 formation argues strongly that it is a necessary part of the system, but that some other feed-back regulator is responsible for the oscillations observed. The nature of a feed-back regulator that could cause the IP 3 -forming enzyme phospholipase C (or just possibly the degradative phosphatase) to fluctuate in activity is unknown. One possible candidate is Ca 2+. A requirement for Ca 2+ by phospholipase C has previously been reported in mammalian systems (Cockroft et al. 1984; Bradford & Rubin, 1986; Taylor et al. 1986) but the physiological significance of the Ca 2+ concentration required to activate this enzyme is controversial. Oscillations in Ca 2+ have, however, been observed in single Chaos amoebae using aequorin fluorescence (Taylor et al. 1980), in hepatocytes (Woods et al. 1986), and in mouse lymphocytes using the new technique of fluorescence ratio imaging of single cells (Tsien & Poenie, 1986). In this regard it is of interest that IP 4, which is formed from IP 3 (and has been detected in D. discoideum; Europe-Finner & Newell, 1987), has recently been postulated to regulate Ca 2+ uptake by cells (Irvine & Moor, 1986). Whether such regulation of Ca 2+ uptake in permeabilized cells could induce Ca 2+ oscillations is uncertain. It is conceivable, however, that an internal oscillating Ca 2+ feed-back system operates in D. discoideum and experiments to test the effects of Ca 2+ on the formation of IP3 are now in progress. We thank Mr Douglas Ramsay for drawing thefigures,and the Science and Engineering Research Council and the Smith-Kline (1982) Foundation forfinancialsupport. References BALDASSARE, J. J. & FISHER, G. J. (1986). GTP and cytosol stimulate phosphoinositide hydrolysis in isolated platelet membranes. Biochem. biophys. Res. Commun. 137, BERRIDGE, M. J., DAWSON, R. M. C, DOWNES, C. P., HESLOP, J. P. & IRVINE, R. F. (1983). Changes in the levels of inositol phosphates after agonist-dependent hydrolysis of membrane phosphoinositides. Biochem. J. 212, BERRIDGE, M. J., DOWNES, C. P. & HANLEY, M. P. (1982). Lithium amplifies agonist-dependent phosphatidyl inositol responses in brain and salivary glands. Biochem. J. 206, BONE, E. A., FRETTEN, P., PALMER, S., KIRK, C. J. & MlCHELL, R. H. (1984). Rapid accumulation of inositol phosphates in isolated rat superior cervical synpathetic ganglia exposed to Vpvasopressin and muscarinic cholinergic stimuli. Biochem. J. 221, BRADFORD, P. G. & RUBIN, R. P. (1986). Guanine nucleotide regulation of phospholipase C activity in permeabilized rabbit neutrophils. Biochem. J. 239, BURGESS, G. M., MCKINNEY, J. S., IRVINE, R. F. & PUTNEY, J. W. (1985). Inositol 1,4,5-trisphosphate and inositol 1,3,4-trisphosphate formation in Ca z+ - mobilizing-hormone-activated cells. Biochem. J. 232, COCKCROFT, S., BALDWIN, J. M. & ALLAN, D. (1984). The Ca z+ -activated polyphosphoinositide phosphodiesterase of human and rabbit neutrophil membranes. Biochem. J. 221, COCKCROFT, S. & GOMPERTS, B. D. (1985). Role of guanine nucleotide binding protein in the activation of polyphosphoinositide phosphodiesterase. Nature, Land. 314, Inositol trisphosphate in Dictyostelium 517

6 DE WIT, R. J. W. & SNAAR-JAGALSKA, B. E. (1985). Folate and camp modulate GTP binding to isolated membranes of Dictyostelium discoideum. Functional coupling between cell surface receptors and G-proteins. Biochem. biophys. Res. Conrmun. 129, EUROPE-FINNER, G. N. & NEWELL, P. C. (1985). Inositol 1,4,5-trisphosphate induces cyclic GMP formation in Dictyostelium discoideum. Biochem. biophys. Res. Commun. 130, EUROPE-FINNER, G. N. & NEWELL, P. C. (1986a). Inositol 1,4,5-trisphosphate and calcium stimulate actin polymerization in Dictyostelium discoideum. J. Cell Set. 82, EUROPE-FINNER, G. N. & NEWELL, P. C. (19866). Inositol 1,4,5-trisphosphate induces calcium release from a nonmitochondrial pool in amoebae of Dictyostelium. Biochim. biophys. Ada 887, EUROPE-FINNER, G. N. & NEWELL, P. C. (1987). Cyclic AMP stimulates accumulation of inositol trisphosphate in Dictyostelium. J. Cell Set. 87, HASLAM, R. J. & DAVIDSON, M. M. L. (1984). Guanine nucleotides decrease the free [Ca 2+ ] required for secretion of serotonin from permeabilized blood platelets. Evidence of a role for a GTP-binding protein in platelet activation. FEBS Lett. 174, IRVINE, R. F. & MOOR, R. M. (1986). Micro-injection of inositol 1,3,4,5-tetrakisphosphate activates sea urchin eggs by a mechanism dependent on external Ca 2+. Biochem. J. 240, JANSSENS, P. M. W., ARENTS, J. C, VAN HAASTERT, P. J. M. & VAN DRIEL, R. (1986). Forms of the chemotactic adenosine 3',5'-cyclic phosphate receptor in isolated Dictyostelium discoideum membranes and interconversions induced by guanine nucleotides. Biochemistry 25, JANSSENS, P. M. W., VAN DER GEER, P. L. J., AGENTS, J. C. & VAN DRIEL, R. (1985). Guanine nucleotides modulate the function of chemotactic cyclic AMP receptors in Dictyostelium discoideum. Molec. cell. Biochem. 67, LEICHTLING, B. H., COFFMAN, D. S., YAEGER, E. S. & RlCKENBERG, H. V. (1981). Occurrence of the adenylate cyclase "G protein" in membranes of Dictyostelium discoideum. Biochem. biophys. Res. Commun. 102, LITOSCH, I., WALLIS, C. & FAIN, J. N. (1985). 5- Hydroxytryptamine stimulates inositol phosphate production in a cell-free system from blowfly salivary glands. J. biol. Chem. 2, NEWELL, P. C, EUROPE-FINNER & SMALL, N. V. (1987). Signal transduction during amoebal chemotaxis of Dictyostelium discoideum. Microbiol. Sci. 4, MARTIN, T. F. J., LUCAS, D. O., BAJJALIEH, S. M. & KOWALCHYK, J. A. (1986). Thyrotropin-releasing hormone activates a Ca 2+ -dependent polyphosphoinositide phosphodiesterase in permeable GH 3 cells. J. biol. Chem. 261, SMALL, N. V., EUROPE-FINNER, G. N. & NEWELL, P. C. (1986). Calcium induces cyclic GMP formation in Dictyostelium. FEBS Lett. 203, STRAUB, R. E. & GERSHENGORN, M. C. (1986). Thyrotropin-releasing hormone and GTP activates inositol trisphosphate formation in membranes isolated from rat pituitary cells. J. biol. Chem. 261, SUSSMAN, M. (1966). Biochemical and genetic methods in the study of cellular slime mold development. Meth. Cell Physiol. 2, TAYLOR, D. L., BLINKS, J. R. & REYNOLDS, G. (1980). Contractile basis of ameboid movement. VIII. Aequorin luminescence during ameboid movement, endocytosis and capping. J. Cell Biol. 86, TAYLOR, C. W., MERJUTT, J. E., PUTNEY, J. W. & RUBIN, R. P. (1986). Effects of Ca 2+ on phosphoinositide breakdown in exocrine pancreas. Biochem. J. 238, TSIEN, R. Y. & POENIE, M. (1986). Fluorescence ratio imaging: a new window into intracellular ionic signalling. Trends Biochem. Sci. 11, VAN HAASTERT, P. J. M., DE WIT, R. J. W., JANSSENS, P. M. W., KESBEKE, F. & DEGOEDE, J. (1986). G-proteinmediated interconversions of cell-surface receptors and their involvement in excitation and desensitization of guanylate cyclase in Dictyostelium discoideum. J. biol. Chem. 261, WALLACE, M. A. & FAIN, J. N. (1985). Guanosine 5'-Othiophosphate stimulates phospholipase C activity in plasma membranes of rat hepatocytes. J. biol. Chem. 2, WOODS, N. M., CUTHBERTSON, K. S. R. & COBBOLD, P. H. (1986). Repetitive transient rises in cytoplasmic free calcium in hormone-stimulated hepatocytes. Nature, Land. 319, 0-2. {Received 4 February Accepted 26 February 1987) 518 G. N. Europe-Finner and P. C. Newell