J. Cell Sri. 65, 111-121 (1984) \ \ \Printed in Great Britain © The Company of Biologists Limited 1984

TRANSITIONS IN DICTYOSTELIUM DISCOIDEUM

BEHAVIOUR: INFLUENCE OF CALCIUM AND

FLUORIDE ON SLUG PHOTOTAXIS AND

THERMOTAXIS

ULRIKE DOHRMANN, PAUL R. FISHER, MARTINABRUDERLE1N AND KEITH L. WILLIAMSMax-Planck-Institut fur Biochemie, D-8033 Martinsried bet Munchen, FederalRepublic of Germany

SUMMARY

Phototaxis and thermotaxis by slugs of Dictyostelium discoideum show transitions that result inbimodality in phototaxis and temperature-dependent orientation up or down temperature gradients.New steps in the sensory transduction chain for these kinds of behaviour are elucidated from studiesusing inorganic salts (Ca2+, EGTA, KF) and several mutants. KF enhances bimodality ofphototaxis, improves the accuracy of thermotaxis, and affects the transition temperatures frompositive to negative thermotaxis. Changing the Ca2+ concentration has effects on both phototaxisand thermotaxis. At low Ca2+ concentrations phototaxis and thermotaxis are enhanced. In thepresence of EGTA or high concentrations of Ca2+ phototaxis becomes bimodal, thermotaxis isimpaired and spontaneous turning is suppressed. These results obtained by changing Ca2+ con-centrations are analogous to those obtained previously with sensory transduction mutations, whichcoordinately affect phototaxis and thermotaxis.

INTRODUCTION

The behaviour of multicellular Dictyostelium discoideum slugs was once thoughtto be simple — slugs moved directly towards light (Bonner, Clarke, Neely & Slifkin,1950; Poff & Loomis, 1973) and 'up' heat gradients (Bonner, Clarke, Neely & Slifkin,1950; Poff & Skokut, 1977). However, recent results are more complex. Phototaxisis bidirectional, i.e. slugs move at certain angles at either side of the light source, withthe angle of orientation (a) affected by mutation and STF (Slug Turning Factor)levels (Fisher & Williams, 1981). In wild-type strains or is so small that orientationseems to be unidirectional. Thermotaxis exhibits two transitions. At low tem-peratures slugs move towards the cold (Whitaker & Poff, 1980; Fisher & Williams,1982). At intermediate temperatures slugs move up the heat gradient and becomeoriented most accurately, close to the growth temperature (Whitaker & Poff, 1980;Fisher & Williams, 1982). At high temperature a second transition towards the coldis observed (Fisher & Williams, 1982; and this paper).

From studies on behavioural mutants it is clear that phototactic and thermotacticpathways converge (Fisher, Smith & Williams, 1981; Fisher & Williams, 1982;mutants of Schneider, Fontana & Poff, 1982; discussed by Fisher, Dohrmann &Williams, 19836). In this paper we expand the sensory transduction chain by

112 U. Dohrmann and others

demonstrating that inorganic ions, notably calcium and fluoride have specific effectson phototaxis and thermotaxis.

.MATERIALS AND METHODS

Strains and culture o/D. discoideum

Strain X22 and phototaxis mutants HU120 and HU409 have been described elsewhere (Fisher& Williams, 1982). Stocks of D. discoideum were stored as spores on silica gel. Amoebae were grownin association with Klebsiella aerogenes on nutrient SM-agar at 21 ± 1 deg. C in the dark (Williams& Newell, 1976).

Phototaxis and thermotaxis experiments

Previously described methods were used to study slug phototaxis and thermotaxis (Fisher et al.1981; Fisher & Williams, 1982). Amoebae were washed free of bacteria and resuspended in Bonner'ssalt solution (0-6 g/1 NaCl, 0-75 g/1 KG, 0-4 g/1 CaCl2-2H2O), at lfZ/ml or 2-5 X lfZ/ml; 20 jtl ofamoebal suspension was placed in an 1 cm2 origin in the centre of an agar plate containing 35 mlwater agar, and incubated for either 48 h at 21 ± 1 deg. C (phototaxis) or for 72 h (thermotaxis). Forthermotaxis experiments, eight plates were placed in a temperature gradient of 0'2deg. C/cm withtemperatures at the centres of the plates ranging from 14°C to 28 °C. The heat bar was sufficientlylarge to hold 11 such series of plates, so replicates and appropriate controls could be run in eachexperiment. All experiments were performed on water agar rather than on charcoal/water agar (seeFisher et al. 1981). In most experiments low-ash noble agar (Difco) instead of Difco Bacto agar wasused in order to minimize extraneous metal ions (see Materials, below). At the end of each experi-ment slime trails were transferred to PVC discs and stained with Coomassie Blue (Fisher et al.1981). We found that the visibility of the trails was greatly improved after the discs were destainedwith hot water or 10 % acetic acid. Destaining removed the dye from the disc itself but did not affectstaining of the trails.

Spontaneous orientationExperiments to measure spontaneous turning by slugs migrating in the dark in the absence of a

temperature gradient were conducted as described elsewhere (Fisher, Grant, Dohrmann &Williams, 1983a).

Analysis of behaviourThe principles of the data analysis have been explained in detail in earlier reports (Fisher et al.

1981, 1983a). Trails were digitized using a digitizing tablet (Summagraphics) connected to a VAX11 computer. Estimates of the accuracy of phototaxis and thermotaxis (k) and of spontaneous 'noise'in slug steering (rad2/h) were obtained using directional statistics. The figures displayed in thispaper were prepared by digitizing trails and plotting them from a common origin using a VAX 11plot program. The trails were plotted so that the direction towards the point light source or warmth(in thermotaxis experiments) was towards the top of the figure. In some cases, particularly at highcell densities, not all trails were plotted but those trails plotted were representative of the behaviour(i.e. k was the same as that calculated for the whole population).

Materials

Ethyleneglycolbis(/J-aminoethylether)-iV,A''-tetraaceticacid (EGTA), CaCh , NaFand KF wereanalytical grade reagents from Sigma.

Sterile stock solutions and sterile agar were made up with deionized water (18-5 MQ resistance,10^M-Ca2+). Appropriate volumes of stock solutions were added to water agar immediately beforepouring the plates. Our standard water agar was made with Bacto agar (Difco). We observed (seeResults) that this had some effects on behaviour different from those seen when low-ash noble agar(Difco) was used. At present we cannot explain the differences that are almost certainly not due to

Behaviour in D. discoideum slugs 113

Ca2+. The Ca2+ concentrations in the water agar were 0 6 mM (Bacto) and 0'4mM (noble), respec-tively. However, all experiments involving external Ca2+ were done with noble agar. Ca2+ wasdetermined by inductively coupled plasma atomic emission spectrometry (Jarrell-Ash, ICAP 9000)and the measurements were performed by Dr R. Rohl.

RESULTS

Fluoride accentuates bimodal phototaxis

Fluoride had a pronounced effect on the phototaxis of the wild type (X22) and most(e.g. HU120) but not all (e.g. HU409) mutants ofD. discoideum examined (see Fig.1 for 7-5mM-KF). (In all experiments reported here, potassium fluoride was used,

X22

Control 7-5mM-KF

HU120

Control 7-5mM-KF

HU4O91 cm

Control 7-5mM-KF

Fig. 1. Phototaxis by slugs of strain X22 and phototactic mutants HU120 and HU409 inthe presence and absence of KF; 2 X 10* amoebae were placed on an 1 cm2 origin in thecentre of a water/noble agar plate ±7-5mM-KF. The light source was at the top of thepage, and the trails were plotted from a common origin (see Materials and Methods).

114 U. Dohrmann and others

but sodium fluoride produced the same effects.) Strain X22 was detectably bimodalat KF concentrations above 5 mM and exceedingly bimodal at 10 mM and higherconcentrations, while aggregation and slug speed were scarcely affected. At 20 mM-KF not all X22 cells aggregated and aggregates that were formed did not migrate. Atlow concentrations (approx. 1 mM) KF actually improved phototaxis although thiswas not a marked effect because under standard conditions wild-type strains (e.g.X22) are not noticeably bimodal.

Fluoride enhances thermotaxis and alters the transitions to negative thermotaxis

Concentrations of 5 mM to 15 mM-KF, which produced marked bimodal phototaxis(see Fig. 1 for 7-5 mM-KF), improved the accuracy of thermotaxis in most strains andshifted the transitions away from the growth temperature (see Figs 2 and 3 for X22slugs that migrated on normal Bacto/water agar ± KF). Fig. 2 shows the effect of KF(5 mM) over the whole range of temperatures at which slugs can migrate. In Fig. 3,individual trails of slugs migrating at 16 °C and at 20 °C in the presence or absence of7-5 mM-KF were plotted. These particular temperatures were chosen for illustration,since at 16 °C control slugs still migrated towards the cold, whereas slugs on KF/agarhad already undergone the transition from negative to positive thermotaxis. At 20 °Cboth control slugs and slugs on KF were moving towards the warmth, and here it isclearly seen that orientation was improved in the presence of KF. The accuracy of

100-1

8 0 -

ok_

o£ 40

2 20

I0 -

14 18 22Temperature (°C)

26

Fig. 2. Influence of KF on the accuracy of thermotaxis of X22 slugs. The experimentswere performed on normal water agar ( ) or on normal water agar containing 5 mM-KF ( ); 4 X 106 amoebae were placed on a 1 cm2 origin. The temperature gradientwas 0-2 deg. C/cm, and for each test eight plates were assayed at temperatures rangingfrom 14 °C to 28 °C. Trails were plotted as in Fig. 1.

Behaviour in D. discoideum slugs 115

orientation on KF/agar was even greater at 18 °C than at 16 aC, while at this tem-perature the control slugs were close to the transition from negative to positive ther-motaxis (as can be seen in Fig. 2).

Co2* and slugphototaxis

Because of the significant role of Ca2+ in photosensory transduction of other organ-isms (Hildebrand, 1980) as well as in chemotaxis/aggregation of D. discoideumamoebae (Gerisch & Malchow, 1976; Devreotes, 1982), we examined slug behaviourin the presence of Ca2+ and EGTA, as the most specific Ca2+ chelator. Furthermore,EGTA was used to see if the effect of KF on slug behaviour could be explained interms of Ca2+ sequestration: if it did mimic KF, then the effect of KF would mostprobably be due to lowered Ca2+ concentrations (by precipitation of highly insolublecalcium fluoride). Like KF, EGTA caused bimodal phototaxis in essentially allstrains tested (e.g. X22, ±3 mM-EGTA, Fig. 4A, B). At concentrations above 5 mM-EGTA, aggregation was severely impaired in most strains. However, unlike KF, lowconcentrations of EGTA did not lead to improved orientation.

16-5

160-

£ 15-5

Control 7.5mM-KF

210

20-5

2001 cm

Control 7-5mM-KF

Fig. 3. Migration patterns of X22 slugs in a thermogradient in the absence and presenceof KF; 2 X 106 amoebae were placed in the origins of normal water agar plates with orwithout 7-5 mM-KF. Temperatures at the origins were 16°C or 20cC. The temperaturegradient was 0-2 deg. C/cm.

116 U. Dohrmann and others

Control 3mM-EGTA

Control 2-5mM-Ca'.2+ 7-5mM-Ca2+

Fig. 4. Phototaxis of X22 slugs in the absence and presence of EGTA and2 X 106 amoebae were placed on a 1 cm2 origin on noble agar. The agar contained no salts(A, C) or 3mM-EGTA (B), 2-5 mM-CaCl2 (D) or 125 mM-CaCl2 (E), respectively. Otherdetails were as in Fig. 1.

Conversely, adding low amounts of Ca2+ (1-2-5 miw) led to improved (decreasedbimodality) phototaxis (Fig. 4D), which was more pronounced if noble agar was used(see below). At higher concentrations of Ca2+ (7-5-15 miu) the accuracy of phototaxiswas slightly decreased (Fig. 4E). Concentrations of Ca2+ higher than 20 mM oftenreduced the number of aggregates and resulted in failure of slugs of strain X22 (andmost strains tested) to migrate. Thus, the results with Ca2+ showed that there was anoptimal Ca2+ concentration (approx. 1 miu) for slug phototaxis.

Ca2 + and slug thermotaxisAlthough EGTA and high Ca2+ produced effects similar to KF in phototaxis (Fig.

1 versus Fig. 4B, E), their effects on thermotaxis were different (see Fig. 3 versus Fig.5). The accuracy of thermotaxis was decreased by EGTA and high Ca2+ concentra-tions (Fig. 6), and the transitions were shifted towards the growth temperature, i.e.21 °C (and not away from it as with KF). Migration of slugs at high temperature wasgreatly affected in the presence of EGTA (Fig. 5). The accuracy in thermotaxis wasimproved slightly at low Ca2+ concentrations (0-5—1 mM, data not shown). Thesefindings reinforce the observations on phototaxis that slug behaviour was optimal ata Ca2+ concentration in the range from 0-5 to 2mM.

Ca2+ effects on unstimulated behaviour

In a recent report we showed that KF markedly suppressed spontaneous 'noise' in

Behaviour in D. discoideum slugs 117

16-5

160

15-5

aE

200

26-5

260

25-5 Control 12-5mM-Ca2+

1 cm

1 mM-EGTA

Fig. 5. Thermotaxis of X22 slugs in the absence and presence of Ca2+ and EGTA. Theexperiment was performed as in Fig. 3 except that noble agar, without or with 12-5 mM-Ca2+ or 1 mM-EGTA, was used. Temperatures at the origins were 16°C, 20°C and 26°C.Trails were plotted as in Fig. 1.

slug orientation in the dark in the absence of a temperature gradient without affectingthe slug speed (Fisher et al. 1983a). Spontaneous turning on noble agar in thepresence of 1 mM-EGTA was reduced by up to 50 % in three experiments (steeringnoise (rad2/h) in controls: 0-35 ± 0-08, 0-52 ± 0-13, 0-27 ± 0-03; with 1 mM-EGTA:0-22±0-04, 0-26±0-07, 0-13 ±0-03). We observed a smaller decrease in spon-taneous turning in the presence of 15 mM-Ca2+. Steering noise in the presence of 1 ITIM

and 2-5 mM-Ca2+ was similar to control values. We conclude that some spontaneoussignals arise 'upstream' from Ca2+ in the sensory transduction chain in D. discoideumslugs.

Choice of agar

A surprising finding was that the type of agar used was critical, especially in ther-motaxis experiments. When slugs migrated on normal water agar, all slugs wereoriented towards the cold at 16CC (see control slugs in Fig. 3), whereas on noble agar

118 U. Dohrmann and others

60-

40-x

Io 20

0-

-20 J

14 18 22Temperature (°C)

26

Fig. 6. Effect of Ca2+ on thermotaxis of X22 slugs. Experimental conditions were as inFig. 2, except that 2 X 10* amoebae and noble agar were used. Plates contained no Ca2+

( ), 2-5mM-Ca2+ ( ) or 12-5mM-Ca2+ ( ).

about half of the slugs were migrating in each direction at this temperature (see controlslugs in Fig. 5). Furthermore, with noble agar a second transition in orientation(down the heat gradient) was observed at high temperature (Figs 5, 6). This transitionwas not seen reproducibly when X22 thermotaxis was performed on normal Difcoagar. The effects were not due to a difference in the Ca2+ content in the agar (seeMaterials and Methods).

DISCUSSION

The experiments reported here on the behaviour of D. discoideum slugs in thepresence of Ca2+ and EGTA establish a role for Ca2+ in the sensory transductionchain. Like several mutations studied previously (Fisher & Williams, 1982), Ca2+

affected both accuracy of orientation and transitions in phototaxis and thermotaxis.As we have discussed previously for sensory transduction mutations (Fisher &Williams, 1982) these effects can be economically explained via a single mechanism,rather than multiple mechanisms affecting phototaxis and thermotaxis separately.The site(s) of action of Ca2+ must be before or in the region of the sensory transduc-tion chain at which the angle of phototactic orientation and changes in the directionof thermotaxis are set (Fisher, 1981; Williams, 1982; Fisher et al. 19836), since wehave shown here that Ca2+ affected these processes (Figs 4, 5).

Ca2+ has been established as an important effector molecule for (photo)sensory

Behaviour in D. discoideum slugs 119

transduction processes in many lower and higher organisms (Rasmussen et al. 1975;Hildebrand, 1980). In most of the systems studied an optimal extracellular Ca2+

concentration has been observed (e.g. in leucocyte chemotaxis; Sha'afi & Naccache,1981), and our observations inD. discoideum are analogous. It is clear that Ca2+ alsoplays an essential role in the early stages of D. discoideum development, i.e. inaggregation of amoebae, which involves chemotactic responses to a self-generatedcAMP gradient (Mason, Rasmussen & Dibella, 1971; Gerisch & Malchow, 1976;Devreotes, 1982). For example, it has been found that chemotaxis and aggregationwere severely impaired at very low (in the presence of EGTA) and very high(>150mM) Ca2+ concentrations (Loomis, Klein & Brachet, 1978; Maeda, 1970). Ina recent study more precise techniques were used to investigate the influence of Ca2+

in amoebal chemotaxis (Malchow, Bohme & Gras, 1982). A concentration of 1 min-Ca2+ was found to enhance chemotaxis by amoebae to cAMP maximally, althoughhigher concentrations were not tested. We have shown here that slug phototaxis andthermotaxis were optimal at a Ca2+ concentration of about 0-5-2 mM, varying slightlybetween single experiments and depending on the particular strain tested.

Ca2+ is clearly involved in morphogenesis as high Ca2+ (up to 120 mM) increasesthe proportion of stalk cells in D. discoideum fruiting bodies (Maeda, 1970) whileEGTA (3 mM) causes an increase in the percentage of prespore cells in the migratingslug (Krefft, Voet, Mairhofer & Williams, 1983). Since it has been reported thatfluoride increases the proportions of spore cells in mature fruiting bodies (Maeda,1970) and of prespore cells in slugs (Durston & Vork, 1977), we began studies on theinfluence of KF on slug behaviour. KF clearly had a profound effect. Unlike allmutations and other effectors examined (e.g. STF, Ca2+), KF had opposite effectson slug phototaxis and thermotaxis. The fact that cells aggregated normally andthermotaxis was enhanced argues strongly against a non-specific poison effect of KF.The fluoride effect was not due to Ca2+ chelation (studied here using EGTA).Although both EGTA (up to 5 mM) and KF (>5 mM) led to bimodal phototaxis (Figs1, 4) and suppression of spontaneous turning, thermotaxis was enhanced by KF (Figs2, 3) but impaired by EGTA (Figs 5, 6).

The action of fluoride on multiple sites of the sensory pathway cannot of coursebe excluded, but if signals from the photoreceptor and thermoreceptor are bio-chemical opposites (Poff & Loomis, 1973; Fisher et al. 1981), KF could also act ona single element of the sensory transduction chain after convergence of photo-/thermo-transduction. A possible site of action would be activation of adenylatecyclase. It is noteworthy that several groups have proposed a role for cAMP as anextracellular chemical signal in slug behaviour (Maeda, 1977; Durston, Vork &Weinberger, 1979; Matsukuma& Durston, 1979; Kessin, 1982; Fishere* al. 19836).However, while fluoride is a well-established activator of hormone-dependentadenylate cyclases (Ross & Gilman, 1980; Limbird, 1981), stimulation of adenylatecyclase activity by fluoride has not been observed using membranes from aggregatingamoebae of D. discoideum (Klein, 1976). On the other hand, we have measuredelevated cAMP levels in slugs that migrated on KF/agar (Dohrmann, Briiderlein &Williams, unpublished data).

120 U. Dohrmann and others

This work was supported by a grant from the Deutsche Forschungsgemeinschaft (Wi668/1-1).We thank Heidi Dahne und Gabriele Ruger for technical assistance, and Dr R. Rohl (BayrischeLandesanstalt fur Wasserforschung, Miinchen) who did the Ca2+ measurements.

REFERENCES

BONNER, J. T., CLARKE, W. W., NEELY, C. L. & SLIFKIN, M. K. (1950). The orientation to lightand the extremely sensitive orientation to temperature gradients in the slime mold Dictyosteliumdiscoideum. J. cell. comp. Physiol. 36, 144—158.

DEVREOTES, P. N. (1982). Chemotaxis. In The Development of Dictyostelium discoideum (ed. W.F. Loomis), pp. 117-168. New York: Academic Press.

DURSTON, A. J. & VORK, F. (1977). The control of morphogenesis and pattern in the Dictyosteliumdiscoideum slug. In Development and Differentiation in the Cellular Slime Molds (ed. P. Cappuc-cinelli & J. M. Ashworth), pp. 17-26. Amsterdam: Elsevier/North-Holland Biomedical Press.

DURSTON, A. J., VORK, F. & WEINBERGER, C. (1979). The control of later morphogenesis bychemotactic signals in Dictyostelium discoideum. In Biophysical and Biochemical InformationTransfer in Recognition (ed. J. H. G. Vassileva-Popova & E. V. Jensen), pp. 693-708. New York,London: Plenum.

FISHER, P. R. (1981). Orientation behaviour by Dictyostelium discoideum slugs. PhD thesis,Australian National University, Canberra, Australia.

FISHER, P. R., DOHRMANN, U. & WILLIAMS, K. L. (19836). Signal processing in Dictyosteliumdiscoideum slugs. Modern Cell Biology (ed. B. H. Satir). New York: Alan R. Liss, Inc. (in press).

FISHER, P. R., GRANT, W. N., DOHRMANN, U. & WILLIAMS, K. L. (1983a). Spontaneous turningbehaviour by Dictyostelium discoideum slugs. J'. Cell Set. 62, 161-170.

FISHER, P. R., SMITH, E. & WILLIAMS, K. L. (1981). An extracellular chemical signal controllingphototactic behaviour by D. discoideum slugs. Cell 23, 799-807.

FISHER, P. R. & WILLIAMS, K. L. (1981). Bidirectional phototaxis by Dictyostelium discoideumslugs. FEMS Microbiol. Lett. Yl, 87-89.

FISHER, P. R. & WILLIAMS, K. L. (1982). Thermotactic behaviour of Dictyostelium discoideumslug phototaxis mutants. J. gen. Microbiol. 128, 965-971.

GERISCH, G. & MALCHOW, D. (1976). Cyclic AMP receptors and the control of cell aggregationin Dictyostelium. Adv. eye. Nucl. Res. 7, 49-68.

HILDEBRAND, E. (1980). Comparative discussion of photoreception in lower and higher organisms.Structural and functional aspects. In Photoreception and Sensory Transduction in AneuralOrganisms (ed. F. Lenci & G. Colombetti), pp. 319-340. New York, London: Plenum Press.

KESSIN, R. H. (1982). Conservatism in slime mold development. Cell 27, 241-243.KLEIN, C. (1976). Adenylate cyclase activity in Dictyostelium discoideum amoebae and its change

during differentiation. FEBS Lett. 68, 125-128.KREFFT, M., VOET, L., MAIRHOFER, H. & WILLIAMS, K. L. (1983). Analysis of proportion

regulation in slugs of Dictyostelium discoideum using a monoclonal antibody and a FACSIV.Expl Cell Res. (in press).

LIMBIRD, L. E. (1981). Activation and attenuation of adenylate cyclase. Biochem.J. 195, 1-13.LOOMIS, W. F., KLEIN, C. & BRACHET, P. (1978). The effect of divalent cations on aggregation

of Dictyostelium discoideum. Differentiation 12, 83-89.MAEDA, Y. (1970). Influence of ionic conditions on cell differentiation and morphogenesis of the

cellular slime molds. Dev. Growth Differ. 12, 217-227.MAEDA, Y. (1977). Role of cyclic AMP in the polarized movement of the migrating pseudoplas-

modium of Dictyostelium discoideum. Dev. Growth Differ. 19, 201-205.MALCHOW, D., BOHME, R. & GRAS, U. (1982). On the role of calcium in chemotaxis and oscilla-

tions of Dictyostelium cells. Biophys. Struct. Mech. 9, 131-136.MASON, J. W., RASMUSSEN, H. & DIBELLA, F. (1971). 3', 5' AMP and Ca2+ in slime mold

aggregation. Expl Cell Res. 67, 156-160.MATSUKUMA, S. & DURSTON, A. J. (1979). Chemotactic cell sorting in Dictyostelium discoideum.

y. Embryol. exp. Morph. 50, 243-251.POFF, K. L. & LOOMIS, W. F. (1973). Control of phototactic migration in Dictyostelium

discoideum. Expl Cell Res. 82, 236-240.

Behaviour in D. discoideum slugs 121

POFF, K. L. & SKOKUT, M. (1977). Thermotaxis by pseudoplasmodia of Dictyostelium discoideum.Proc. natn. Acad. Set. U.SA. 74, 2007-2010.

RASMUSSEN, H., JENSEN, P., LAKE, W., FRIEDMANN, N. & GOODMAN, D. B. P. (1975). Cyclic

nucleotides and cellular calcium metabolism. Adv. eye. Nucl. Res. 5, 375-394.Ross, E. M. & GILMAN, A. G. (1980). Biochemical properties of hormone-sensitive adenylate

cyclase. A. Rev. Biochem. 49, 533-564.SCHNEIDER, M. J., FONTANA, D. R. & POFF, K. L. (1982). Mutants of thermotaxis in Dictyo-

stelium discoideum. Expl Cell Res. 140, 411-416.SHA'AFI, R. I. & NACCACHE, P. H. (1981). Ionic events in neutrophil chemotaxis and secretion.

Adv. Inflamm. Res. 2, 115-148.WHITAKER, B. D. & POFF, K. L. (1980). Thermal adaptation of thermosensing and negative

thermotaxis in Dictyostelium. Expl Cell Res. 128, 87-93.WILLIAMS, K. L. (1982). Molecules involved in morphogenesis in the multicellular stage of

Dictyostelium discoideum. In Biochemistry of Differentiation and Morphogenesis, vol. 33Colloquium Mosbach (ed. L. Jaenicke), pp. 231-246. Berlin: Springer-Verlag.

WILLIAMS, K. L. & NEWELL, P. C. (1976). A genetic study of aggregation in the cellular slimemould Dictyostelium discoideum using complementation analysis. Genetics 82, 287-307.

(Received 28 May 1983-Accepted 14 July 1983)

CEL65