/ . Einbryol. exp. Morph. Vol. 36, 2, pp. 431-442, 1976 43 \

Printed in Great Britain

Cell aggregation andsexual differentiation in pairs ofaggregation-deficient mutants of

Dictyostelium discoideum

By G. GERISCH1 AND A. HUESGEN2

From the Biozentrum der Universitdt Basel, andFriedrich-Miescher-Laboratorium der Max-Planck-Gesellschaft, Tubingen

SUMMARYA diffusible aggregation-stimulating factor (ASF) is released from a series of aggregation-

deficient mutants. Biochemical markers indicate that these ASF-donor mutants are blockedat a later step of cell differentiation than an ASF-requiring mutant. ASF is able to bridge theinitial block of differentiation in the latter mutant, such that development proceeds up toaggregation and even further. ASF is most probably neither identical with cyclic-AMP phos-phodiesterase nor with an inhibitor of this enzyme which both are released from donor strains.

In certain combinations of aggregation-deficient mutants, macrocysts, the sexual stages ofDictyostelium, are formed. Also, motile giant cells believed to be zygotes are observed inthese mutant combinations. The gamone known to be released from one mating type and toinduce macrocysts in the other, is probably not identical with ASF since this factor is pro-duced by mutants derived from either one of both mating types.

INTRODUCTION

One type of non-aggregating mutant in Dictyostelium discoideum is blockedin the initial step of cell differentiation from the growth-phase stage to aggrega-tion-competence. Mutants of this type are characterized by the absence of anyone of the membrane changes normally associated with the acquisition of aggre-gation-competence (Gerisch et al. 1975 a). They also do not form a macro-molecular inhibitor of cyclic-AMP phosphodiesterase which, in the wild-type, isreleased into the extracellular medium shortly after the end of exponentialgrowth (Gerisch et al. 1972). This type of mutant is represented by aggr 50-2(Malchow & Gerisch, 1974). Other aggregation-deficient mutants still form theinhibitor, indicating a later block of development (Riedel, Gerisch, Miiller& Beug, 1973). Like certain other mutants of this type, Wag-11 releasesa diffusible inducer of cell aggregation (ASF = aggregation-stimulating

1 Author's address: Biozentrum der Universitat Basel, 4056, Basel, Switzerland.2 Author's address: Friedrich-Miescher-Laboratorium der Max-Planck-Gesellschaft, 74Tubingen, Germany.

432 G. GERISCH AND A. HUESGEN

factor) which can be assayed by the specific requirement of certain aggr 50clones for this factor (Gerisch, Huesgen & Malchow, 1975b). In this paperwe report on the biochemical and morphogenetic phenotypes of ASF-donormutants.

Macrocysts are considered to be the sexual stages of D. discoidewn and re-lated species (Erdos, Nickerson & Raper, 1972; Clark, Francis, & Eisenberg,1973; Erdos, Raper & Vogen, 1975). In combination with Wag-11, certainclones of aggr 50-2 form macrocysts rather than normal aggregates. Wag-11 isa descendant of the wild-type strain NC-4, and aggr 50-2 is a descendant ofstrain v-12. NC-4 and v-12 belong to complementary mating types (Erdos,Raper & Vogen, 1973). The occurrence of macrocysts in the mutants indicatesthe expression of the mating type characteristics in spite of a block in cell aggre-gation.

METHODS

Cells were cultivated together with E. coli B/2 at 22-23 °C in 9 cm diameterpetri dishes on 20 ml Difco agar 2 %, containing 0-1 % bacteriological peptone(Oxoid), 0-1 % glucose and 0-017 M phosphate buffer pH 6-0. For the determina-tion of cyclic-AMP phosphodiesterase and inhibitor, both peptone and glucosewere increased to 0-4 %. Macrocyst formation was obtained on wet agar sur-faces either in darkness or at a yellow fluorescent lamp. Aggregation patternswere fixed and stained on the agar surface by placing the plates upside down andadding a few crystals of iodine to the cover.

For keeping mutant strains separate, a 0-15 (im pore size membrane filter(Sartorius, Gottingen) was inserted into the agar in such a way that the filterextended from the bottom of the agar layer to 5 mm beyond its surface. Controlswere run for all strains tested to ensure that within 8 days no cells would passthe filter.

Aggr 50 is a particularly unstable mutant. After periods of serial transfer itwas repeatedly re-cloned. First, aggr 50-2 (Riedel et al, 1973), was obtainedfrom aggr 50-4. Finally, aggr 50-2 was cloned by picking up single cells under adissection microscope. The primary cultures of these new clones were pre-served in liquid nitrogen. Aggr 50-2-5 was selected for its superb macrocystformation and aggr 50-2-8 for its ability to aggregate in response to Wag-11.Another clone, 50-2-7, showed no significant response. Part of the experi-ments were done by Huesgen (1973) with the parent clone, aggr 50-4, whichwas similar to aggr 50-2-8.

Extracellular cyclic-AMP phosphodiesterase and its inhibitor were deter-mined as described previously (Riedel et al. 1973) in the fluid extracted fromagar. To obtain the fluid, cells were stripped off the agar surface and pieces ofthe agar were put into bags of fine nylon gauze. The bags were fixed on top of atube and centrifuged for about 5 min at 400 g. The clear extract was collectedfrom the bottom of the tube. One phosphodiesterase unit was defined as the

Aggregation deficient mutants o/Dictyostelium 433amount of enzyme which hydrolysed 1 nmole cyclic AMP at 35 °C and pH 7-4 atsaturating substrate concentrations.

The factor for converting these units into the activities at the pH and tempera-ture of agar plates is 0-24 (Gerisch, 1976). One inhibitor unit inactivates understandard conditions two enzyme units by 50 % (Riedel et al. 1973).

Contact sites A were assayed by absorption of Fab directed against mem-branes of aggregating cells. Freshly washed cells harvested from suspensioncultures were used for absorption (Beug, Katz, Stein & Gerisch, 1973 b). Theaggregation-inhibiting activity of the absorbed Fab was re-titrated in the pre-sence of 10 mM EDTA using aggregation-competent V-12/M2 cells (Beug, Katz& Gerisch, 1973 a).

RESULTS

Induction of cell aggregation in aggr 50 lines

Pairs of aggregation-deficient mutants were inoculated about 3-5 cm aparton agar plates. Aggr 50-4 was the only mutant which, in combination withother mutants, gave rise to cell aggregates (Table 1). Also, in most combinations,rudiments of fruiting bodies were formed, in which no spores were detected,however. More recently, aggr 50-2-8 was selected as a clone that forms par-ticularly long aggregation streams in combination with certain Wag-11 clones(Fig. 1).

When aggr 50-4 was separated from the other mutants by a membrane filter,aggregation was observed only in the compartment that contained aggr 50-4(Table 3 and Gerisch et al. 1915b). The aggregation-stimulating effect extendedabout 1 cm distant from the filter. This indicates that aggregation in aggr 50-4is stimulated by a diffusible factor, ASF, which is released from a variety ofnon-aggregating mutants. No aggregation was obtained when aggr 50-4 wasseparated from donor strains by a dialysis membrane, suggesting that ASF is amacromolecule. The chemical nature requires further examination.

Biochemical phenotype of donor mutants

In all donor mutants tested, an extracellular inhibitor of cyclic-AMP phospho-diesterase was found (Table 1). In the wild type, the inhibitor is an early markerfor cell differentiation from the growth-phase stage to aggregation-competence.This result indicates that the donor mutants are blocked at a later stage ofdevelopment than aggr 50-4. In these studies, inhibitor formation has beendetermined in suspension cultures (Riedel et al. 1973). To ensure that under agarplate conditions development stops at the same stage as in suspension cultures,phosphodiesterase and the inhibitor were measured in the fluid extracted fromagar plates on which Wag-11 or aggr 50-2 clones had been grown. Table 2shows that Wag-11 forms inhibitor, whereas the two 50-2 clones showed notmore than background levels.

It should be emphasized that the phosphodiesterase inhibitor is most likely28-2

434 G. GERISCH AND A. HUESGEN

Aggregation deficient mutants of Dictyostelium 435not identical with ASF, although both are released from the cells at an earlystage of differentiation. No aggregation was observed after adding 480 units ofpartially purified inhibitor per plate along a line in front of the growth zone ofaggr 50-4. Several inhibitor-producing mutants did not act as ASF-donors(Table 1). Aggr 75 is a donor (Table 3) although in suspension cultures, it over-produces extracellular phosphodiesterase (Riedel et al. 1973). On agar plates,the enzyme activity was not so high. Nevertheless, an excess of extracellularphosphodiesterase between 20 and 68 units per ml was found during the wholeperiod between 3 h before and 9 h after the end of growth. (For definition ofunits see 'Methods'.) The excellent donor mutant Wag-11 also produces anexcess of extracellular phosphodiesterase in agar plate cultures (Table 2).

£ 60 - - 20

10 15 20Approximate time before and after

exhaustion of bacteria (h)

Fig. 2. Extracellular phosphodiesterase (•) and its inhibitor (O) in the donor mutantga 88. In agar plate cultures, phosphodiesterase and, after heating for lOmin at80 °C, total inhibitor were determined in the extracellular fluid. Data were collectedfrom two plates.

Extracellular phosphodiesterase is certainly not identical with ASF. This isdemonstrated by the action of ga 86 and 88 as donors (Table 3). In suspensioncultures, both mutants produce an excess of the inhibitor of extracellular phos-phodiesterase (Riedel & Gerisch, 1971). Similarly, ga 88 shows on agar platesmuch inhibitor and almost no activity of the enzyme (Fig. 2).

FIGURE 1.

Cell aggregation and sexual differentiation in aggr 50/ Wag-11 pairs. (A): Aggregatesat the boundary between aggr 50-2-8 (left) and Wag-11-26 colonies (right) on agar.(B), (C): A motile giant cell at the boundary of Wag-11-20 and aggr 50-2-5 colonies,at an interval of a few minutes. Normal-sized amoebae are also present. (D), (E):Rounded giant cell (D) and macrocyst cluster (E) at the zone of contact betweenaggr 50-2-5 and Wag-U-28.

28-3

Tab

le 1

/Ind

ucti

on o

f ag

greg

atio

n in

mut

ant

com

bina

tion

s

v-12 m

uta

nts

N

C-4

m

uta

nts

Mu

tan

t n

o.

...

aggr

20

-2

aggr

39

aggr

52-2

aggr

66-1

aggr

70-1

aggr

85

aggr

50-4

W

ag-2

-1

Wag-3

W

ag

-4

Wag-6

W

ag-8

W

ag-1

1

QP

D i

nh

ibit

or

(i+

) i+

i~

i~

(i

+)

(i+)

i"

(i+)

(i+)

i~

i+

(i+)

i+O

aggr

20

-2

__

__

__

__

__

__

_M

aggr

39

- -

- -

++

-

- -

- -

- 2

aggr

52

-2

__

__

__

__

__

_c

/3

aggr

66-1

_

__

__

__

__

_O

aggr

7

0-1

—

—

+

+—

—

—

—

—

_

ffi

aggr

85

- -

- -

- —

-

- >

aggr

50-4

-

++

+

+

- +

-

+ +

!_

.W

ag-2

-1

__

__

__

OW

ag-3

"

--

--

>W

ag-4

•

Wag-6

-

- -

XW

ag-8

-

CW

ag-1

1

- W O

Mu

tan

ts i

+ o

r i~

wer

e cl

assi

fied

on t

he

basi

s of

exp

lici

t d

eter

min

atio

ns

of c

ycli

c-A

MP

phosp

hodie

ster

ase

inh

ibit

or

in h

eate

d c

ult

ure

flu

ids

as d

escr

ibed

by R

iede

l et

al.

(19

73).

W

In t

he

case

of

aggr

50-4

, th

e ab

sence

of

inh

ibit

or

was

tes

ted i

n i

ts d

esce

ndan

t, a

ggr

50-2

. In

(i+

) la

bell

ed m

uta

nts

onl

y i

nact

ivat

ion o

f ph

osph

odie

ster

ase

at 5

or

9 h

aft

er t

he

^en

d o

f g

row

th w

as c

hec

ked

. E

xce

pt

in a

ggr

20-2

an

d a

ggr

70-1

, w

here

the

PD

act

ivit

y w

as r

educ

ed b

y 5

6 %

and 1

7 %

, res

pect

ivel

y, t

he

inhib

itio

n w

as >

90 %

.+

, A

ggre

gat

ion s

trea

ms

and p

seudopla

smodia

, i.

e. s

lug-

like

str

uct

ure

s fo

rmed

.+

+,

In a

dd

itio

n,

rud

imen

tary

fru

itin

g b

odie

s fo

rmed

.

Aggregation deficient mutants of Dictyostelium 437Using membrane filters for separation, aggregating strains including wild

types could be tested as ASF-donors (Table 3). Among three wild-type strains,ASF activity was observed only in the clone v-12/Ml. This strain is known forits slow development to aggregation-competence as tested in suspension-cul-tures (Gerisch, 1962). One possible reason for the low efficiency of wild-typestrains as donors could be the formation of ASF within a short transient period

Table 2. Cyclic AMP phosphodiesterase and its inhibitor in Wag-11and aggr 50-2 clones

PhosphodiesteraseStrain (u./ml) Inhibitor (u./ml)

Wag-11 233 87aggr 50-2-5 186 5aggr 50-2-8 11 0E. coli B/2 (control) 0 5

The inhibitor was determined after heating of the extracellular fluid 10 min at 80 °C. Thisprocedure inactivates the enzyme and liberates the inhibitor in an active state (Riedel et al.1973). The extracellular fluid was taken not earlier than 12 h after the exhaustion of foodbacteria.

Table 3. Development of aggr 50-4 induced by donor cells actingthrough a membrane filter

No.

V-12IM1V-12/M2Ax-2Wag-2-1Wag-3Wag-11

aggr 39

aggr 70-1aggr 75

ga 86 and

Donor strain

Phenotype

Wild typeWild typeAxenic wild typeNo aggregationNo aggregationNo aggregationexcept of travel-ling cell bands(Gerisch et al.1915b)

No aggregation

No aggregationFruiting bodies,no streamsformed duringaggregation

ga 88 Fruiting bodies,large aggregationterritories

RudimentaryAggregation Pseudo- fruiting

streams plasmodia bodies

i 1

_ _ _

_ _ _

+ + + + -+ + + + ++ + + + +

not ob- + + -served+ + —

+ + + + -

+ -i- —

+ and + + indicate arbitrary classifications into weak and strong responses.

438 G. GERISCH AND A. HUESGEN

between growth phase and the acquisition of aggregation competence. Onewould expect then efficient donor mutants to exist which are arrested at theearly stage of differentiation during which ASF is formed. This possibility wastested in the efficient donor, Wag-11, by the determination of contact sites A.These constituents of the cell membrane appear simultaneously with theability of aggregating cells to associate into streams (Beug, Katz & Gerisch,1973a). Fig. 3 shows that Wag-11 is defective in the expression of contact

0 2 4 6- 8 10Cell concentration used for absorption (x 107/ml)

Fig. 3. Contact sites A in two ASF donor mutants. Similar to aggregation-compe-tent wild-type Ax-2 cells (#), aggr 39 ( • ) absorbed Fab specific for contact-sites A(Beuget al. 1973). The same immunoassay of contact sites A showed that JVag-ll(A)equals wild-type Ax-2 cells of the growth-phase stage (O) where these sites arealmost absent. (The weak absorption found can be attributed to contact sites B,which are already present in growth phase cells and not fully blocked by 10 mMEDTA (Beug et al. 1973).) Mutant cells were harvested for Fab absorption at a timewhen their wild-types would have reached full aggregation-competence: aggr 39 at5 h, Wag-11 at 9 h after the end of growth.

sites A. This result indicates that certain donor mutants are blocked after in-hibitor formation and before the appearance of contact sites A. However,in other donors like aggr 39 contact sites A can be expressed (Fig. 3). De-velopmental arrest at a stage preceding contact-site A formation is thereforeconsistent with, but not indispensable for the action of a mutant as an efficientdonor.

Macrocyst formation.

Macrocyst formation in wild-type cultures is preceded by the development ofgiant cells growing-up by phagocytosis of normal cells (Filosa & Dengler, 1972).These cells are most probably zygotes (Maclnnes & Francis, 1974). Certainclones of aggr 50-2 formed macrocysts in combination with Wag-11 (Fig. 1E).In addition to mature macrocysts, actively motile giant cells were found (Fig.IB, C), as well as rounded giant cells containing multiple contractile vacuoles(Fig. ID). In the aggr 50-2-5/Wag-11 combinations shown in Fig. 1B-E, only

Aggregation deficient mutants of Dictyostelium 439sporadically aggregation streams were observed although these combinationsbelong to the best macrocyst forming ones. These mutants indicate, therefore,that all factors necessary for the mating reaction are expressed in the absence offull aggregation-competence. Giant cells can exist in these mutant combinationsin a free motile state obviously because of the lack of tightly adhering cellsaround them. In a wild-type strain it is difficult to isolate the giant cell from peri-pheral cells (Filosa, Kent & Gillette, 1975).

DISCUSSION

Synergism between aggregation-deficient mutants has been repeatedly ob-served (Sussman & Lee, 1955; Ennis & Sussman, 1958; Weber & Raper, 1971;Yamada, Yanagisawa, Ono & Yanagisawa, 1973) since its first investigation bySussman (1954). In most of these cases immediate contact between the mutantcells is required for synergism. The present case represents the other extreme:Aggregation is stimulated by a diffusible factor (ASF) which extends the in-fluence of the donor mutant over a longer distance. Since the responder strainretains the phenotype of a non-aggregating mutant when grown on washedE. coli B/r, and does so also when washed after the end of growth (Huesgen,1973), the synergism is not of the type reported by Weber and Raper (1971). Inthis case the responding mutants are able to develop normally in the absence of ahelper mutant when freed from bacterial metabolites.

ASF-donor strains exhibit a great variety of phenotypes (Table 3), so that theASF-donor activity cannot be associated with a specific morphogenetic charac-ter. The obvious reason is that the ASF-donor activity is expressed at an earlystage of cell differentiation, prior to the expression of any specific morphogeneticfunction. ASF-donors are found among mutants exhibiting opposite defects inthe regulation of cylic-AMP phosphodiesterase and its inhibitor (Fig. 2 andTable 2). This indicates that neither the enzyme nor the inhibitor activity isrequired for ASF action. The inhibitor, however, appears to be a fairly reliablemarker for the developmental stage which a mutant has either to reach or topass in order to produce ASF. This stage is reached before the acquisition offull aggregation-competence, as suggested by the ASF-donor activity of a mutantwhich does not express contact sites A (Fig. 3).

The absence of any differentiation marker in the ASF-requiring mutant haslead to the conclusion that cell differentiation from growth phase to aggregation-competence is blocked at an initial step in this mutant and, as far as aggrega-tion is concerned, ASF is able to bridge this deficiency (Gerisch et ah 1975 &).These results are summarized in Fig. 4. The effects of ASF and of cyclic-AMPpulses (Gerisch etal. 1975 a; Gerisch, Fromm, Huesgen & Wick, 1975 c; Darmon,Brachet & Pereira da Silva, 1975) show that in D. discoideum populations, celldifferentiation up to the aggregation stage is largely synchronized by chemicalcell communication. A third factor, folic acid, identified by Pan, Hall & Bonner,

440 G. GERISCH AND A. HUESGEN

(1972) as a chemotactic agent, controls cell differentiation in a similar way ascyclic AMP does (Wurster, 1976). It remains to be shown if folic acid pulses areproduced by developing D. discoideum cells.

The wild-type NC-4 releases a gamone that induces a sexual response in thewild-type strain v-12 (O'Day & Lewis, 1975; MacHac and Bonner, 1975). Theinduction of macrocysts by combination of two aggregation-deficient mutants, aNC-4 descendant and a v-12 derived one, indicates that Wag-ll produces thegamone and that aggr 50-2-5 is able to respond to it. Identity of the gamone

"SKr 391 ASF-donor type 1

Wnc-11* ' -\ ASF-donor type 2if ASF

aggr 50-2-8 „,-"""" ~ """" ^ . _V ^* > Responder

. PD inhibitor ^ AW->• —>•— contact sites A — > • aggregation

ASF

S t e p , ,

Differentiation

Fig. 4. Tentative scheme of developmental blocks in ASF-donors and in the responder.

with ASF is unlikely since ASF is produced in mutants of both wild-type strainsNC-4 and v-12 (Table 1). Consequently, ASF is not limited to one specificmating type. The highly motile giant cells shown in Fig. 1B and C pose thequestion of how the contractile system of Dictyostelium is constituted in order toallow polar cell movement in a 10 /im cell as well as in a cell of more than100/tmin length.

We thank Professor J. M. Ashworth, Colchester, for the Wag mutants, Miss E. Miillerand Miss S. Wicki for careful assistance. This work was supported by the ' SchweizerischerNationalfonds' and the 'Deutsche Forschungsgemeinschaft'.

REFERENCES

BEUG, H., KATZ, F. E. & GERISCH, G. (1973a), Dynamics of antigenic membrane sitesrelating to cell aggregation in Dictyostelium discoideum. J. Cell Biol. 56, 647-658.

BEUG, H., KATZ, F. E., STEIN, A. & GERISCH, G. (19736). Quantitation of membrane sites inaggregating Dictostelium cells by use of tritiated univalent antibody. Proc. natn. Acad. Sci.U.S.A. 70, 3150-3154.

CLARK, M. A., FRANCIS, D. & EISENBERG, R. (1973). Mating types in cellular slime molds.Biochem. biophys. Res. Comm. 52, 672-678.

DARMON, M., BRACHET, P. &. PEREIRA DA SILVA, L. H. (1975). Chemotactic signals inducecell differentiation in Dictyostelium discoideum. Proc. natn. Acad. Sci. U.S.A. 72, 3163—3166.

ENNIS, H. L. & SUSSMAN, M. (1958). Synergistic morphogenesis by mixtures of Dictyosteliumdiscoideum wild-type and aggregateless mutants. / . gen. Microbiol. 18, 433-449.

Aggregation deficient mutants of Dictyostelium 441ERDOS, G. W., NICKERSON, A. W. & RAPER, K. B. (1972). Fine structure of macrocysts in

Polysphondylium violaceum. Cytobiologie 6, 351-366.ERDOS, G. W., RAPER, K. B. & VOGEN, L. K. (1973). Mating types and macrocyst formation

in Dictyostelium discoideum. Proc. natn. Acad. Sci. U.S.A. 70, 1828-1830.ERDOS, G. W., RAPER, K. B. & VOGEN, L. K. (1975). Sexuality in the cellular slime mold

Dictyostelium giganteum. Proc. natn. Acad. Sci. U.S.A. 72, 970-973.FILOSA, M. F. & DENGLER, R. E. (1972). Ultrastructure of macrocyst formation in the

cellular slime mold, Dictyostelium mucoroides: Extensive phagocytosis of amoebae by aspecialized cell. Devi Biol. 29, 1-16.

FJLOSA, M. F., KENT, S. G. & GILLETTE, M. U. (1975). The developmental capacity of vari-ous stages of a macrocyst-forming strain of the cellular slime mold, Dictyostelium mucor-oides. Devi Biol. 46,49-55.

GERISCH, G. (1962). Zellfunktionen und Zellfunktionswechsel in der Entwicklung von Dictyo-stellium discoideum. IV. Der Zeitplan der Entwicklung. Wilhelm Roux1 Arch. EntwMech.Org. 153, 603-620.

GERISCH, G. (1976). Extracellular cyclic-AMP phosphodiesterase regulation in agar platecultures of Dictyostelium discoideum. Cell Diff. 5, 21-25.

GERISCH, G., MALCHOW, D., RIDEL, V., MULLER, E. & EVERY, M. (1972). Cyclic AMPphosphodiesterase and its inhibitor in slime mold development. Nature, Lond. 235,90-92.

GERISCH. G., MALCHOW, D., HUESGEN, A., NANJUNDIAH, V., Roos, W., WICK, U. & HULSER,D. (1975a). Cyclic-AMP reception and cell recognition in Dictyostelium discoideum. InDevelopmental Biology (ed. D. McMahon and C. F. Fox). ICN-UCLA Symposia onMolecular and Cellular Biology 2, 76-88. Menlo Park, Calif: W. A. Benjamin, Inc.

GERISCH, G., HUESGEN, A. & MALCHOW, D. (19756). Genetic control of cell differentiationand aggregation in Dictyostelium: The role of cyclic-AMP pulses. Proceedings of theTenth FEBS-meeting. Federation of European Biochemical Societies 38 (ed. G. Bernardi andF. Gros). Amsterdam and New York: North-Holland.

GERISCH, G., FROMM, H., HUESGEN, A. & WICK, U. (1975C). Control of cell-contactsites by cyclic AMP pulses in differentiating Dictyostelium cells. Nature, Lond. 255,547-549.

HUESGEN, A. (= VON STEIN) (1973). Diplomarbeit, Universitat Tubingen.MACHAC, M. A. & BONNER, J. T. (1975). Evidence for a sex hormone in Dictyostelium

discoideum. J. Bacteriol. 124, 1624-1625.MACINNES, M. A. & FRANCIS, D. (1974). Meiosis in Dictyostelium mucoroides. Nature,

Lond. 251, 321-323.MALCHOW, D. & GERISCH, G. (1974). Short-term binding and hydrolysis of cyclic 3': 5'-

adenosine monophosphate by aggregating Dictyostelium cells. Proc. natn. Acad. Sci. U.S.A.71, 2423-2427.

O'DAY, D. H. & LEWIS. K. E. (1975). Diffusible mating-type factors induce macrocystdevelopment in Dictyostelium discoideum. Nature, Lond. 754, 431-432.

PAN, P., HALL, E. M. & BONNER, J. T. (1972). Folic acid as second chemotactic substance inthe cellular slime molds. Nature New Biol. 237, 181-182.

RIEDEL, V. & GERISCH, G. (1971). Regulation of extracellular cyclic-AMP-phosphodiesteraseactivity during development of Dictyostelium discoideum. Biochem. biophys. Res. Comm., 42119-124.

RIEDEL, V., GERISCH, G., MULLER, E. & BEUG, H. (1973). Defective cyclic adenosine-3',5'-phosphate-phosphodiesterase regulation in morphogenetic mutants of Dictyosteliumdiscoideum. J. molec Biol. 74, 573-585.

SUSSMAN, M. (1954). Synergistic and antagonistic interactions between morphogeneticallydeficient variants of the slime mould Dictyostelium discoideum. J. gen. Microbiol. 10, 110—120.

SUSSMAN, M. & LEE, F. (1955). Interactions among variant and wild-type strains of cellularslime molds across thin agar membranes. Proc. natn. Acad. Sci. U.S.A. 41, 70-78.

WEBER, A. T. & RAPER, K. B. (1971). Induction of fruiting in two aggregateless mutants ofDictyostelium discoideum. Devi Biol. 26, 606-615.

442 G. GERISCH AND A. HUESGEN

WURSTER, B. (1976). Stimulation of cell development in Dictyostelium discoideum by folic acidpulses. Cold Spring Harbor Meeting on Dictyostelium, May 1976. Abstract.

YAMADA, T., YANAGISAWA, K. O., ONO, H. & YANAGISAWA, K. (1973). Genetic analysis ofdevelopmental stages on the cellular slime mold Dictyostelium purpureum. Proc. natn Acad.Sci. U.S.A. 70, 2003-2005.

(Received 15 June 1976)