Selective induction of gene expression and second-messenger accumulation in Dictyostelium discoideum by the partial chemotactic antagonist 8-p-chlorophenylthioadenosine 3',5'-cyclic monophosphate

Proc.Nall.Acad. Sci. USA Vol. 88,pp.9219-9223, October1991 DevelopmentalBiology

Selective induction of

gene

expression

and

second-messenger

accumulation

in

Dictyostelium

discoideum

by

the

partial

chemotactic antagonist

8-p-chlorophenylthioadenosine

3',5'-cyclic

monophosphate

(cAMPderivatives/inositolphospholipidslignalng/GTP-bindingprotein/generegulatlon/transmembranesignaltransduction)

DORIENJ. M. PETERS*, ANTHONYA.

BOMINAARt,

B. EWA SNAAR-JAGALSKA*, RAYMOND BRANDT*, PETERJ. M. VAN

HAASTERTt,

ADRIANO

CECCARELLIt,

JEFFREY G. WILLIAMSI, ANDPAULINE SCHAAP*

*Celi

BiologyandGenetics,Department ofBiology, University of Leiden,Kaiserstraat63,2311GPLeiden,TheNetherlands;tDepartmentofBiochemistry, UniversityofGroningen, Nijenborgh16, 9747AGGroningen,TheNetherlands;andtImperialCancer ResearchFund, ClareHallLaboratories,SouthMimms,

Hertfordshire,EN63L,England

CommunicatedbyJ. T.Bonner,July 24, 1991

ABSTRACT During development of the cellular slime moldDictyosteliumdiscoideum,cAMP induces chemotaxis and

expressionofdifferentclassesof genesbymeansof interaction with surface cAMP receptors. WedescribeacAMPderivative,

8-p-chlqrophenylthioadenosine 3',5'-cyclic monophosphate (8-CPT-cAMP),which inhibitscAMP-inducedchemotaxisatlow concentrations butinduceschemotaxis atsupersaturating con-centrations. This compound, moreover, selectively activates

expression of aggregative genes but not of p reptive

genes. 8-CPT-cAMPinduces normalcGMPand cAMP accu-mulation but in contrast to cAMP, which increases inositol

1,4,5-trisphosphate levels, 8-CPT-cAMP decreases inositol 1,4,5-trisphosphate levels. The derivative induces reduced activation of guanine nucleotide regulatory proteins, which may cause its defective activation ofinositol

1,4,5-trisphos-phate production. Ourdata suggest that disruption of

inosi-tolphospholipidsignaling impairs chemotaxisandexpression of

asubclassofcAMP-regulated genes.

Inthe social amoebae Dictyostelium discoideum, extracel-lular cAMPfunctionsas ahormone-like signal; itinduces the expression of several classes ofgenesandregulates morpho-geneticmovementbyactingas achemoattractant (see ref.1). cAMPsignal processingisverysimilartothatof mammalian hydrophilic hormones, suchasadrenaline,vasopressin,

ace-tylcholine,luteinizing hormone, andmanyothers; itseffects

onchemotaxis andgeneexpressionaremediated by surface

receptors (2-6), which belong to the ubiquitous class of seven-trans-membrane receptors, interacting with guanine nucleotideregulatory protein(G) proteins (7, 8). This inter-action results in activation of target enzymes, suchas ade-nylate cyclase, guaade-nylate cyclase, and phospholipase C(see

ref. 9). Similar to the adrenergic receptor, for example,

cAMPreceptors areencodedbyafamily of differentgenes,

of which three members have been cloned (10). Also, the Dictyostelium G proteins belong to amultigenefamily(11).

Elucidationof signal-transduction cascades involved ingene

regulation and chemotaxis is ofcrucial importancefor our

general understanding of theseprocesses. By using mutants

andmolecular genetic approaches,considerableprogress has

beenmade inunderstandingsomeof the functional relations between cAMP receptors, G proteins, second-messengers

systems, and the ultimate responses that they control (7,

11-17). Wedescribehere amodified cAMP receptor ligand,

8-p-chlorophenylthioadenosine 3',5'-cyclic monophosphate

(8-CPT-cAMP), which may be apowerful pharmacological toolfordissectingcAMPtransduction cascades. This cAMP derivative selectively activates somesecond-messenger sys-tems, asubpopulation of G proteins, and asubpopulation of cAMP-regulated genes.

MATERIALS AND METHODS

Materials. Luciferin andguanosine5'-[y-thio]triphosphate GTP[yS] were from Boehringer Mannheim. 8-CPT-cAMP wassupplied by B.Jastorff(University of Bremen, Bremen, F.R.G.) or purchased from Boehringer Mannheim. 8-Bro-moadenosine 3',5'-cyclic monophosphate (8-Br-cAMP), 6-chloropurineriboside 3',5'-cyclic monophosphate (6-Cl-cPUMP), o-nitrophenyl 3-D-galactoside, Geneticin (G418),

and phenylmethylsulfonyl fluoride were obtained from Sigma, [2,8-3H]cAMP, [a-32P~dATP, and cGMP RIA kits were from Amersham, and

GTP[y35S]

was from New En-gland Nuclear.

Dictyostelium

Strains and Culture Conditions. D. discoi-deum strainNC4 and mutantsynag 7(18) were grown on

glucose/peptone agar in association with Escherichia coli 281. Twotransformedaxenic (AX2) cell lines (D19-lacZ and CP2-luciferase) were grown in HL5 medium (19) in the

presence ofG418 at 10 ug/ml. D19-lacZ cells contain the

vectorpA6PTlac.1, which bearsagene fusion of D19 pro-moterandlacZ(20);CP2-luciferasecells contain thevector

PB10.act.15.BKH.LUC.BAM, which carriesafusion of the

fireflyluciferasegeneand CP2promoter(21).

Growing cellswerefreed fromnutrients byrepeated

wash-ingwith 10mMNa/Kphosphate, pH 6.5 [phosphate buffer

(PB)]. Aggregationcompetence was induced by incubating cellsonPBagar at2.5 x 106cellspercm2for16 hrat6°Cor

bystimulatingcells for4hr with 30nMcAMPpulsesat6-min intervals.

BindingandPhosphodiesterase(PDE) Assays.The affinity of8-CPT-cAMP forcAMP-dependentprotein kinase and its

apparentKmand

Vm.

for cAMP PDEweredetermined by described methods (22, 23). The effects of cAMP,

8-Br-cAMP, and 8-CPT-cAMP on

GTP[y35S]-binding

to

mem-branes of aggregation-competent cells were measured, as

describedbySnaar-Jagalskaetal. (24).

Abbreviations: G protein, guanine nucleotide regulatory protein; PDE, cAMP phosphodiesterase; cAK, cAMP-dependent protein kinase;cAR, cAMP receptor; csA, contact sites A; InsP3, inositol 1,4,5-trisphosphate; GTP[yS], guanosine 5'-[y-thio]triphosphate; 8-CPT-cAMP, 8-p-chlorophenylthioadenosine 3',5'-cyclic mono-phosphate;8-Br-cAMP, 8-bromoadenosine 3',5'-cyclic monophos-phate;6-Cl-cPUMP, 6-chloropurineriboside3',5'-cyclic monophos-phate;PB, phosphate buffer.

9220 Developmental Biology: Peters etal.

Table 1. Binding characteristics of cAMP derivatives

K'dof surface cAMP-binding sites K'dof PDE

AH AL B C cAR K'm VI cAMP (60*) (450*) (15*) (300t) (2.2t) (1000§) Derivative 8-CPT-cAMP 280t 250t 230t 40t 0.24 2.9 0.7 8-Br-cAMP 220t 160t 130t 70t 0.32t 9.2§ 2.2§ 6-Cl-cPUMP 2500* 2200* 1400* 1400t 2.21* 2.7§ 0.7§

Kd,Kdofderivative/Kdof cAMP; V',V..)ofderivative/V,, of cAMP;K', Kmofderivative/Km of cAMP. Kd values (in nM) of cAMP binding to the different receptors are indicated in parentheses. *Data arederived from ref. 35.

tDataarederivedfromref. 34.

*Data

arederived fromref. 22. §Dataarederived fromref. 23.

Analysis of mRNA Levels, lacZ, and Luciferase Gene Expression. Totalcellular RNAwasisolated from 2.5 x 107

cells, purified, size-fractionated on 1.5% agarose gels

con-taining2.2Mformaldehyde, andtransferredtonylon

mem-branes (25). RNA transfers were hybridized to 32P-labeled cDNAs, according to standardprocedures (26).

f3-Galacto-sidase activity in cell linestransformed with D19-lacZ con-structs was measured essentially as described by

Dinger-mann etal. (20). To measure luciferase activity, cellswere

lysed with 100 ,ul of lysis buffer A [8 mM MgCl2/1 mM

EDTA/1mMdithiothreitol/1% TritonX-100/15% (vol/vol) glycerol/0.5 mM phenylmethylsulfonyl fluoride in 100 mM

potassiumphosphate, pH 7.5]. Subsequently 100 ,ul of2% bovine serum albumin in lysis buffer A was added to the lysate. Reactions were started by adding 15 ,lof 0.86 mM luciferin in 0.14 mM ATP to 25 ,ul of cell lysate (27). Chemoluminescencewasmeasuredbyusingthesingle

pho-ton-counting facility of an LKB model 1218

liquid-scintillationcounter.

cGMP,cAMP, and Inositol 1,4,5-trisphosphate (InsP3) Re-sponses. Tomeasure cGMPresponses, aliquotsof108cells

perml werestimulated with cAMP or8-CPT-cAMP in the

presenceof 2 mM dithiothreitol. After 0or10s, incubation

wasterminated byaddinganequal volume of3.5%(vol/vol) perchloric acid,andcGMP levelsweremeasured in neutral-izedextracts by RIA.

cAMP responses were induced by stimulating

27-Al

ali-quots of 2 X 108 cells perml at0°C with 3 ,ul ofcAMPor

8-CPT-cAMP in 5 mM dithiothreitol. After 0 or 4 min of

stimulation, 1.5 ml of ice-cold PB was added, cells were

centrifugedfor 5s at10,000 x g, supernatantwasremoved,

andpelletswerelysedin 30,ulof3.5%perchloricacid. cAMP levels were measured by competition with

[3H]cAMP

for

bindingtoaggregation-competentD.discoideumcells,using

theammonium sulfate stabilizationassay(28).

To measure agonist-induced InsP3 accumulation, cells

wereresuspended in 40mM Hepes, pH 6.5,to5 x

i07

cells

perml and stimulated with cAMP or 8-CPT-cAMP. At 4-s intervals, 30-,ulaliquotswereaddedtoequal volumes of 3.5%

perchloric acid. InsP3 levels were determined by

isotope-dilutionassay (29).

RESULTS

Bindingof8-CPT-cAMPtocAMPReceptorsandInduction of Chemotaxis.Dictyosteliumcells exhibit severalkinetically

distinct classes of surface cAMP-binding sites (30, 31), an

intracellularcAMP-dependentproteinkinase(cAK) (32) and

a cAMP-specific PDE (33). Surface-binding sites can be distinguishedasrapidlydissociating

AH

and

AL

sites, slowly dissociating B sites(30,31), and aputative third class-the

low-affinityCsites, which, incontrast toAand Bsites,are

resistant to downregulation by cAMP (34). The relative

affinity of8-CPT-cAMP, 8-Br-cAMP, and 6-Cl-cPUMPfor

all these binding sites is summarized in Table 1. Both

8-CPT-cAMP and 8-Br-cAMParegood cAKagonists; these

agents bindto Aand B sites with -200-fold lower affinity than does cAMP andtoCsites with -50-fold lower affinity.

Degradationby PDE is similar compared withcAMP. 6Cl-cPUMP binds well tocAK but binds toall surface cAMP-binding sites with >1000-fold lower affinity than doescAMP.

ChemotaxisofaggregativeD. discoideum cellsto 8-CPT-cAMPand cAMPwascompared byusing the small

popula-tion assay (36). Fig. 1 shows that cAMP induces a half-maximal chemotacticresponse at -3 nM,and 8-CPT-cAMP induces thesamelevelresponse at 50 ,uM.Thisconcentration

is -80-fold higherthanexpected from the relative affinity of

8-CPT-cAMP for surface receptors. Most surprisingly, at

lowerconcentrations (0.1-10 uM), 8-CPT-cAMP antagonizes chemotaxis inducedbycAMP. Apparently, atlow

concen-trations8-CPT-cAMPacts as anantagonist of cAMP, andat

high concentrationsitacts as anagonist.

Inductionof GeneExpressionby8-CPT-cAMP.The

expres-sionof aggregativegenescoding for cAMPreceptors(cAR), for example, and contact sites A (csA) can be effectively inducedbynanomolar cAMPpulses (7, 37-39).Fig.2shows

that inwild-type NC4 cells, 8-CPT-cAMP pulsesarealmost

aseffectiveascAMPpulsesatinducing cAR and csAgene

expression. However, because 8-CPT-cAMP can induce cAMPrelay (see Fig. 6), this resultmay be due to 8-CPT-cAMP-induced cAMPproduction.Inmutantsynag7,which isdefective inadenylate cyclaseactivation(15, 18),induction of csA and cARgene expression by 8-CPT-cAMPrequires

100-8 100-80T

00

60-40

,/

7

0 PL 20-0 10-1010-9 10-8 10-7 10-6 10-1 1o-4 Concentration[M]

FIG.1. Inductionof chemotaxis. Aggregation-competent NC-4 cells were deposited as 0.1-,ul droplets of 107 cells per ml on

hydrophobicagar.Dropletsof thesamevolumeof different concen-trationsof cAMP(e),8-CPT-cAMP(A),or acombination of10-8M cAMP with different concentrations of 8-CPT-cAMP (-) were

placed closetothe celldroplets. Eachconcentrationwastestedon 20smallpopulations.Every 15 minthe numberofdropletsshowing apositiveresponsewasscored. Means andSEMs derived from four experimentsarepresented.

Proc. Natl. Acad. Sci. USA 88 (1991) 9221 cAMP 8-CPT-cAMP ( 10 0 100 10 30 100 nM/6 min d. ,-4L 40 .4w .. -, csA NC4 cAR 10-3 0 10-7 Concentration

[Ml

* csA sng7 * .00 * cAR 0 10 30 100 10 30 100 nM/6min cAMP 8-CPT-cAMP

FIG. 2. Induction ofaggregative gene expression. Vegetative NC4 andsynag(sng) 7 cellswereincubatedat5 x 106 cells perml inPB andstimulated withthe indicatedconcentrations of cAMPor

8-CPT-cAMP at6-min intervals. mRNA wasisolated after 3 hr of incubation. Northern (RNA) blotswereprobed with csA and cAR1

cDNAs.

10- to 100-fold higher concentrations than induction by cAMP. This dosedependencyagreeswith the relativeaffinity of 8-CPT-cAMP for surfacecAMP-binding sites and indicates that 8-CPT-cAMP is a full agonist for aggregative gene

expression.

Postaggregativegenesareexpressed inresponseto micro-molar cAMPconcentrations (4, 5). Fig. 3 shows the effect of cAMP, 8-CPT-cAMP, 8-Br-cAMP, and 6-Cl-cPUMP on

expression of thepresporegeneD19(40) and the postaggre-gativegeneCP2, which is preferentially expressedinprestalk cells (41). cAMP induces half-maximal expression of both

genes at -30

AM;

expression is also induced by 1 mM of 6-Cl-cPUMP and8-Br-cAMP, butnoincreaseof D19orCP2 mRNA levelswasdetectableat8-CPT-cAMP concentrations

upto 1 mM.

Dose-response relationships measured during prolonged incubation of cells with cAMP derivatives donotreflecttrue affinitiesof thecAMP-binding proteins because considerable degradation by PDEoccursduring the incubation. Degrada-tion of cAMPand cAMP derivatives by PDEcanbe reduced by incubating cells atlow cell density. With transformants carrying promoter-reporter gene constructs, cell density could be reduced 10-to20-fold(Fig. 4).Atlow celldensity, half-maximal activation of the CP2 and D19 promoter is

FIG. 4. Effectsof cAMP derivativesonD19andCP2promoter

activity. Aggregation-competent cells, transformed with D19-lacZ

constructsorCP2-luciferase constructs,wereincubated as 100-.tl

aliquots of, respectively, 106or5 x 105 cellsperml in microtiter plates and shaken intermittentlyonanEppendorf shaker. cAMP (o), 8-CPT-cAMP(A),or6-Cl-cPUMP (-)wereaddedat60-minintervals. P-Galactosidase(f3-Gal)and luciferaseactivitiesweremeasured after 6 hr of incubation. Data are expressed as percentage of values obtainedafterincubation with10-5McAMP;meansand SEMs of threeexperiments done in quadruplicate arepresented.

induced by 2 u.M cAMP or50 ,uM 6-Cl-cPUMP, whereas

8-CPT-cAMP inducesverylowlevelsofexpressionat1 mM. Activation of SecondMessengers by 8-CPT-cAMP. Stimu-lation of cells withcAMP induces transient accumulation of the intracellularmessengerscGMP, cAMP, and InsP3, which

peak at, respectively, 10 s, 3 min, and 5 s. Fig. 5 shows

dose-response relationships of cGMP accumulation induced by cAMP and 8-CPT-cAMP, measured 10safter stimulation.

About 50 times higher concentrations of 8-CPT-cAMP than cAMParerequiredtoinduceahalf-maximal cGMPresponse.

Induction of cAMP accumulation by 8-CPT-cAMP and cAMPwasalsodetermined.8-CPT-cAMP hasahigh affinity

for bovine cAK, which is generally used in the isotope dilutionassay to measure cAMPaccumulation. Toprevent interference of the stimulus, cells werestimulated at0°C, a

temperature atwhich cAMPsynthesis is normal, but cAMP secretion is strongly retarded (42). Cells could then be washed to remove 8-CPT-cAMP, and accumulated cAMP

levels were determined by competition with [3HJcAMPfor binding to surface cAMP-binding sites, which have a rela-tively low affinity for 8-CPT-cAMP (Table 1). The cAMP relay inhibitor caffeine (43)wasusedasacontroltoshowthat the8-CPT-cAMP stimulus doesnotcontributetomeasured cAMP levels. Fig. 6 shows that 8-CPT-cAMP can induce cAMPsynthesistothesamelevelsascAMPand, therefore,

alsoacts as anagonistonthis second-messenger system. Fig. 7 shows effects of cAMP and 8-CPT-cAMPon InsP3

accumulation. Dictyostelium cells have rather high basal InsP3 levels (44), which showasmallbutsignificantincrease after cAMPstimulation. Stimulation with8-CPT-cAMPdoes notincrease, but rather decreases, InsP3 levels. Possiblythis

CP2

4

cAMP 8-CPT-cAMP II. 8-Br-cAM a i. ;-_P - _ IP 6-CI-cPUMI

FIG. 3. Induction of prespore and prestalk gene

expression. Aggregation-competent cells were

resus-D19 pendedto107 cells perml in PB and stimulated with the indicated concentrations of cAMP, 8-Br-cAMP,

6-Cl-cPUMP,or8-CPT-cAMPat60-minintervals.mRNAwas

isolated after 3or5hr ofincubation, and Northern blots

wereprobed with, respectively,CP2 and D19 cDNAs. The low-intensity D19 mRNA band at 3 x 10-5 M

8-CPT-cAMP is duetoleakage during sample loadingof the lane tothe left.

el

9222 Developmental Biology: Peters etal. 100 -I 80-._

X~

60-C 40-0 20-o10-9 10-7

lo-,

lo-3 Concentration[M]

Fio. 5. Induction of cGMP accumulation. Aggregation-competentNC-4 cells were stimulated withthe indicated concen-trationsofcAMP(e) or8-CPT-cAMP (A) in the presence of 2 mM dithiothreitol. After 10 s, incubation was terminated, and cGMP levels were measured. Data are presented as percentage ofcGMP levelsobtainedafterstimulationwith 10-3 M 8-CPT-cAMP. Means andSEMsof threeexperiments done in triplicate are presented.

derivativeactivates aninhibitory, ratherthanastimulatory, pathway.

Activation ofphospholipase C is mediated by at least one

Gprotein. We measured whether8-CPT-cAMP can increase

GTP[yS]

bindingtomembranes, whichcharacterizes activa-tion of G proteins (24, 45). Fig. 8 showsthat cAMP induces a>2-fold increaseof

GTP[yS]

binding. Half-maximal induc-tion is achieved by 100 nM. 8-Br-cAMP induces the same increase as cAMP at50-fold higherconcentrations. 8-CPT-cAMPstarts to increase GTP[yS] bindingat the same

con-centrationsas8-Br-cAMP,buteven atsaturating

concenltra-tions, the 8-CPT-cAMP-induced increase is only half that inducedby cAMP and 8-Br-cAMP.Apparently8-CPT-cAMP

cannotactivateasubpopulation ofGproteins.

DISCUSSION

WedescribeacAMPderivative,8-CPT-cAMP,which inhib-its cAMP-induced chemotaxisat lowconcentrations, while

inducing chemotaxis at supersaturating concentrations.

8-CPT-cAMP induces virtually normal accumulation of the second-messengers cAMP and cGMP but is defective in inositolphospholipidsignaling andinducesadecrease,rather thananincrease,ofInsP3 levels.This effect of 8-CPT-cAMP

'. 7 z: -I! to. T e--- I1 C~100-"

80-60-

T 40- j 20- A. 0 10 20 30 Time,s

FIG. 7. Inductionof InsP3 accumulation. Vegetative cells starved for 4 hr in PB were resuspended in 40 mM Hepes, pH 6.5. Three batches of cells weresimultaneously stimulated either with10-6M cAMP(e),10-4M8-CPT-cAMP (v), orwith water. At the indicated timeintervals, aliquots were transferred to perchloric acid, and InsP3 levels were measured. Data are expressed as percentageof basal level(stimulation with water). Means and SEMs of three experiments done in triplicate arepresented.**, Data aresignificantbelowbasal level;*,data aresignificantabove basal level(Student'sttest,P< 0.05).

can beexplained by putative control of phospholipase C by both a stimulatory and an inhibitory G protein, with 8-CPT-cAMPonly activating the inhibitory G protein. Compared

withcAMPand 8-Br-cAMP, 8-CPT-cAMP shows astrongly reducedabilitytoincrease the bindingof

GTP[yS]

to Dictyo-stelium membranes(Fig. 8). This result indicates that

8-CPT-cAMPcannotactivate asubpopulation of Gproteins,

pre-sumablythose responsibleforphospholipase C activation. The aberrant 8-CPT-cAMP-induced InsP3 response may

explain its behavior as a partial chemotactic antagonist.

Studiesusingchemotactic mutantsand introduction of

sec-ond messengersintopermeabilizedcells havesuggestedthat cGMP andInsP3signalingmayrespectivelycontrol myosin andactin polymerization(12, 17, 47, 48); 8-CPT-cAMP may

antagonize chemotaxisby counteractingthecAMP-induced

increase of InsP3 levels. However, because 8-CPT-cAMP induces a normal cGMP response, this may, at saturating concentrations, sufficetoinducesomechemotaxis, perhaps duetoenhancedcytokinesis.

240 220 e 200 to 180 n 160 ; 140 EH : . 1LU~ 100 ._ ,_. ,, CAN-11, SUF11 ,q.1).1'

FIG.6. Induction of cAMP accumulation.CompetentNC4 cells

werestimulatedat00Cwiththeindicatedconcentrations of cAMPor

8-CPT-cAMP(8CPT)in 5 mM dithiothreitol in the absence (light bars) or presence (dark bars) of 5 mM caffeine. After 4 min, incubationwasterminated,and cAMP levelsweremeasured. Means

and SEMs of threeexperiments areshown.

0 10~910-8 10-7 10-6 10-5 10-4 10-3 Concentration

[Ml

FIG.8. Activation ofGTP[yS]binding. Equilibriumbindingof

GTP[y"5S]

to membranes from aggregation-competent cells was

measured in thepresenceofcAMP(-), 8-Br-cAMP(*),or 8-CPT-cAMP(A).Data areexpressedaspercentageof

GTP[IyS]

bindingin the absenceofcyclicnucleotides (7000 cpm per107cells). Means and SEMs of threeexperimentsarepresented.

Proc. Natl. Acad. Sci. USA 88

(1991)

Proc. Natl. Acad. Sci. USA 88 (1991) 9223

Theobservation that 8-CPT-cAMP induces accumulation of cGMP butnotof InsP3 contradictsanearlierhypothesis that guanylate cyclase is activated by means of the

InsP3/

Ca2+ pathway (49, 50). Remarkably, the cGMP response

induced by 8-CPT-cAMP reaches much higher levels than that induced by cAMP (Fig. 5), which suggests that the cAMP-induced InsP3response mayhaveanegative effecton

cGMPaccumulation. Thishypothesisis supported by obser-vationsthatboth InsP3 and Ca2+ stronglyinhibit guanylate cyclase activity in vitro(51).

Theambiguous behavior of 8-CPT-cAMPonchemotaxisis

alsoreflectedin its effectson geneexpression.8-CPT-cAMP induces normal aggregative gene expression (Fig. 2) but is

virtuallyineffective in inducingpostaggregativegene

expres-sion. cAMP-induced gene expression maybe mediated by cAMP, cGMP, InsP3/Ca2+,oryet-unknowncAMP-induced

responses. Earlier studies made involvement of cAMP in

generegulation unlikely because both aggregativeand

post-aggregative gene expression occur under conditions that

prevent adenylate cyclaseactivation (5, 15, 16). FgdA mu-tants that are defective in the G protein, G2, mediating

phospholipase C activation (11, 13, 14), show no

cAMP-induced expression of aggregative genes (16, 52) and no

cAMP or cGMP responses (14, 53). It was suggested that G2-mediated inositolphospholipid signaling mediates all

cAMP-induced responses, including aggregative gene

expression(11, 14, 53). The observation that 8-CPT-cAMP reduces InsP3 levels but induces normal aggregative gene

expression, as well as cAMP and cGMP accumulation,

contradicts thissuggestion. Thedefective G proteinis

pos-sibly linkedtoothertargetproteinsorcould berequiredfor

an eventearly indevelopment, which is requiredfor

subse-quent differentiation.

Several dataimplicateInsP3ininduction ofprespore gene

expression. Presporegeneexpressioncannotbeinducedby 8-CPT-cAMP, is effectively inhibited by Ca2+ antagonists (15, 54) and by LiCI which inhibits cAMP-induced InsP3 accumulation (44), andcanbe induced under special condi-tions by InsP3/diacylglycerol pulses (46). Expression of prestalk-relatedgenes,suchasCP2, is probablynotmediated by

InsP3/Ca2+

because thisresponseisnotinhibitedbyCa2+

antagonists (54)orlithium (44) and iscounteractedby

InsP3/

diacylglycerol pulses (46). Why 8-CPT-cAMPcannotinduce CP2 gene expression is unclear. This response may be mediatedby presently unknownintracellularmessenger sys-tems, which cannot be activated by 8-CPT-cAMP. The effects of8-CPT-cAMP andlithium on thecGMPresponse

correlate well with effects on aggregative gene expression.

Bothresponsesareeffectively induced by8-CPT-cAMPand stimulatedby lithium

(unpublished

work),whichsuggestthat cGMP may mediate induction ofaggregative gene

expres-sion.

Thepresentstudy shows that8-CPT-cAMP isaveryuseful tool to unravel involvement of specific cAMP signal-transduction pathways in thegreatvariety of cAMP-induced

responses.

WearegratefultoProf.Dr. BerndJastorff for stimulating discus-sionsonthenature ofpartial antagonists, and we thank Dr. Peter Devreotesand Dr.Angelika Noegel for their kind gifts of cAR1 and CsA cDNAs. We further thank Martine van Ments-Cohen for measuring binding of 8-CPT-cAMPto cAK, and Johan Pinas for performingchemotaxis assays. This researchwassupported by the Foundation forBiological Research (BION), which is subsidized by the NetherlandsOrganization for Scientific Research (NWO).

1. Devreotes, P. (1989) Science 245, 1054-1058.

2. Mato, J. M.,Jastorff, B., Morr, M.&Konijn, T. M. (1978)Biochim.

Biophys.Acta544, 309-314.

3. VanHaastert, P. J. M.&Kien,E.(1983) J.Biol. Chem. 258,9636-9642. 4. Schaap,P.&VanDriel, R. (1985) Exp.CellRes.159, 388-398. 5. Oyama, M. &Blumberg, D. D. (1986) Proc.Nail. Acad. Sci.USA83,

4819-4823.

6. Haribabu,B.&Dottin, R. P. (1986) Mol. Cell. Biol. 6, 2402-2408. 7. Klein, P. S., Sun, T. J., Saxe, C. L., Kimmel, A. R., Johnson, R. L. &

Devreotes, P. N.(1988) Science 241, 1467-1472.

8. Dohlman,H.G., Caron,M.G. &Lefkowitz,R. J.(1987)Biochemistry

26, 2657-2664.

9. VanHaastert, P. J. M., Janssens, P. M. W. & Erneux, C. (1991)Eur.J. Biochem. 195, 289-303.

10. Saxe,C.L.,Johnson, R. L., Devreotes, P. N. &Kimmel, A. R. (1991) Genes Dev.5,1-8.

11. Kumagai,A.,Pupillo, M., Gundersen, R.,Miake-Lye,R., Devreotes, P. N.&Firtel,R. A.(1989) Cell57, 265-275.

12. Ross, F. M. &Newell,P.C.(1981)J. Gen.Microbiol.127,339-350.

13. Coukell,M.B., Lappano, S. &Cameron,A.M.(1983)Dev. Genet.3,

283-297.

14. Kesbeke,F.,Snaar-Jagalska,B. E.& VanHaastert, P. J. M.(1988)J. Cell Biol. 107,521-528.

15. Schaap,P., Van LookerenCampagne,M.M.,VanDriel, R., Spek, W.,

VanHaastert, P. J. M. &Pinas, J.(1986) Dev.Biol.118, 52-63. 16. Mann, S. K. O.,Pinko,C. &Firtel,R. A.(1988)Dev.Biol. 130, 294-303. 17. Hall, A. L., Warren, V. &Condeelis,J.(1989) Dev. Biol. 136, 517-525. 18. Frantz, C. E. (1980) Ph.D. thesis(Univ.ofChicago,Chicago).

19. Watts, D. J. &Ashworth,J. M.(1970)Biochem. J.119, 171-174. 20. Dingermann,T.,Reindl,N., Werner, H.,Hildebrandt, M.,Nellen, W.,

Harwood,A.,Williams,J.&Nerke,K.(1989) Gene 85, 353-362. 21. Pears, C. J. &Williams,J. G.(1987)EMBO J.6, 195-200.

22. DeWit,R. J.W., Arents, J. C. & VanDriel,R.(1982) FEBS Lett.145,

150-154.

23. Van Haastert, P. J.M., Dijkgraaf, P.A.M., Konijn, T. M., Abbad,

E.G., Petridis,G. &Jastorff,B.(1983)Eur.J.Biochem.131,659-666. 24. Snaar-Jagalska,B.E., DeWit,R.J. W. & VanHaastert, P. J.M.(1988)

FEBS Lett.232, 148-152.

25. Mann,S. K.0. &Firtel,R.A.(1987)Mol. Cell. Biol. 7, 458-469. 26. Sambrook, J.,Fritsch,E. F.&Maniatis,T.(1989)MolecularCloning:A

LaboratoryManual(ColdSpringHarborLab.,ColdSpring Harbor,NY).

27. Trung Nguyen,V., Morange, M. &Bensaude, 0. (1988)Anal.Biochem. 171,404-408.

28. VanHaastert,P.J. M.(1985) Biochim.Biophys.Acta845,254-260. 29. VanHaastert,P. J.M.(1989) Anal. Biochem. 177, 115-119.

30. VanHaastert,P. J. M.& DeWit, R.J. W.(1984)J.Biol.Chem.259,

13321-13328.

31. VanHaastert, P. J. M., DeWit,R. J.W.,Janssens, P. M. W.,Kesbeke,

F.& DeGoede,J.(1986)J.Biol. Chem. 261,6904-6911.

32. DeGunzburg,J.&Veron, M.(1982)EMBO J. 1, 1063-1068. 33. Malchow, D., Nagele, B., Schwartz, H.&Gerisch, G.(1972)Eur. J.

Biochem.28, 136-142.

34. Van Ments-Cohen, M., Genieser, H.-G., Jastorff, B., Van Haastert,

P.J. M.&Schaap,P.(1991)FEMS Lett. 82, 9-14.

35. VanMents-Cohen,M.& VanHaastert, P. J. M. (1989) J. Biol. Chem.

264,8717-8722.

36. Konijn,T. M.(1970)Experientia 26, 367-369.

37. Gerisch, G., Fromm, H., Huesgen, A. & Wick, U. (1975) Nature

(London)255, 547-549.

38. Darmon, M.,Brachet,P.& Pereira daSilva,L. H.(1975) Proc.NatI.

Acad. Sci. USA 72, 3163-3166.

39. Noegel, A.,Gerisch, G., Stadler,J.&Westphal,M.(1986) EMBO J. 5, 1473-1476.

40. Barklis, E.& Lodish,H. F.(1983) Cell 32, 1139-1148.

41. Pears, C. J.,Mahbubani,H. M.&Williams,J. G.(1985)Nucleic Acids

Res. 13,8853-8866.

42. VanHaastert, P. J. M. (1984) J.Gen.Microbiol. 130, 2559-2564. 43. Brenner,M.&Thoms,S. D.(1984) Dev. Biol. 101, 136-146. 44. Peters, D. J.M., Van Lookeren Campagne, M.M., Van Haastert,

P.J.M.,Spek,W.&Schaap,P.(1989)J.CellSci.93, 205-210. 45. Cassel, D. & Selinger, Z. (1976) Proc. Natl. Acad. Sci. USA 75,

4155-4159.

46. Ginsburg, G.&Kimmel,A. R.(1989) Proc. Natl. Acad. Sci. USA 86, 9332-9336.

47. Liu,G. &Newell,P.C. (1988) J.CellSci.90, 123-129.

48. Europe-Finner,G. N.&Newell,P.C.(1986)J. CellSci.82, 41-51. 49. Europe-Finner,G.N.&Newell,P.C. (1985) Biochem.Biophys. Res.

Commun.130,1115-1122.

50. Small,N.V.,Europe-Finner, G. N. &Newell,P.C. (1986) FEBSLett.

203,11-14.

51. Janssens,P. M. W., DeJong,C. C.C., Vink, A. A.&VanHaastert, P.J. M.(1989)J.Biol. Chem.264, 4329-4335.

52. Peters, D. J.M., Cammans, M., Smit, S., Spek, W., Van Lookeren

Campagne,M. M.&Schaap,P.(1991)Dev.Genet.12, 25-34. 53. Snaar-Jagalska,B.E., Kesbeke,F.& VanHaastert,P.J. M.(1988)Dev.

Genet.9,215-226.

54. Blumberg, D.D.,Comer,J. F.&Walton,E. M. (1989)Differentiation 41, 14-21.