Severe developmental defects in Dictyostelium null mutants foractin-binding proteins

Eleonora Pontea, Francisco Riverob, Marcus Fechheimerc, Angelika Noegelb,Salvatore Bozzaroa,*

aDipartimento di Scienze Cliniche e Biologiche, UniversitaÁ di Torino, Ospedale S. Luigi, 10043-Orbassano, ItalybInstitut fuÈr Biochemie I, UniversitaÈt zu KoÈln, Joseph-Stelzmann-strasse 52, 50931-KoÈln, Germany

cDepartment of Cellular Biology, University of Georgia, 724 Biological Sciences Building, Athens, GA 30602, USA

Received 27 August 1999; received in revised form 28 October 1999; accepted 29 October 1999

Abstract

The actin cytoskeleton is implicated in many cellular processes, such as cell adhesion, locomotion, contraction and cytokinesis, which are

central to any development. The extent of polymerization, cross-linking, and bundling of actin is regulated by several actin-binding proteins.

Knock-out mutations in these proteins have revealed in many cases only subtle, if any, defects in development, suggesting that the actin

system is redundant, with multiple proteins sharing overlapping functions. The apparent redundancy may, however, re¯ect limitations of

available laboratory assays in assessing the developmental role of a given protein. By using a novel assay, which reproduces conditions closer

to the natural ones, we have re-examined the effects of disruption of many actin-binding proteins, and show here that deletion of a-actinin,

interaptin, synexin, 34-kDa actin-bundling protein, and gelation factor affect to varying degrees the ef®ciency of Dictyostelium cells to

complete development and form viable spores. No phenotypic defects were found in hisactophilin or comitin null mutants. q 2000 Elsevier

Science Ireland Ltd. All rights reserved.

Keywords: Dictyostelium; Actin-binding proteins; Cell motility; Development

1. Introduction

In the lower eukaryote Dictyostelium, the concerted

action of chemotactic cell motility (Gerisch, 1987; Parent

and Devreotes, 1996) and intercellular adhesion (Bozzaro

and Ponte, 1995) transforms a monolayer of single cells into

multicellular three-dimensional aggregates, each of which

gives rise to a slug, a sausage-shaped unitary organism

capable of undergoing extended migration towards light

and temperature gradients, before culminating into a fruit-

ing body (Bonner, 1967; Loomis, 1975; Fisher, 1997).

Elaborate and polarized cell movements occur inside the

cell mass during slug and fruiting body formation, leading

to sorting out of differentiating cell populations, the prestalk

and the prespore cells; prestalk cells will eventually form a

slender stalk lifting in the air, and the prespore tissue gives

rise to a sorus of mature, encapsulated spores (Siegert and

Weijer, 1997; Williams, 1997). Because of the central role

of motility in the life cycle of Dictyostelium, the actin cytos-

keleton has been thoroughly studied. Several actin-binding

proteins have been isolated, mostly due to their interactions

in vitro with actin, or as cDNA based on sequence homo-

logies (Noegel and Luna, 1995). The in vivo role of actin-

binding proteins has been studied in mutants generated by

either gene disruption or gene overexpression. In some

cases, such as in mutants lacking coronin (Maniak et al.,

1995), myosin IB (Jung and Hammer, 1990), 120-kDa gela-

tion factor (Cox et al., 1996), pro®lin (Haugwitz et al., 1994)

or cortexillin (Faix et al., 1996), defects in phagocytosis

and/or in cytokinesis have been reported. In contrast, only

minor alterations in cell locomotion have been found in

mutants defective in these proteins, as well as severin

(Andre et al., 1989), cap32/34 (Hug et al., 1995), ponticulin

(Hitt et al., 1994), 34-kDa actin-bundling protein (Rivero et

al., 1996a), and a-actinin (Rivero et al., 1996b). Consistent

with these results, no or only minor developmental defects

have been reported for mutants in actin-binding proteins,

with the exception of pro®lin null mutant and of double

mutant a-actinin/gelation factor, which are blocked at the

tip stage (Witke et al., 1992; Haugwitz et al., 1994). Based

on these results, it has been proposed that actin-binding

proteins are part of a redundant network, i.e. several proteins

Mechanisms of Development 91 (2000) 153±161

0925-4773/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved.

PII: S0925-4773(99)00292-0

www.elsevier.com/locate/modo

* Corresponding author. Tel.: 139-011-6708106; fax: 139-011-

9038639.

E-mail address: sbozzaro@polito.it (S. Bozzaro)

might share similar activities or take over overlapping func-

tions. Genetic redundancy can be a driving evolutionary

force as it permits a process to continue under a wider

range of environmental conditions (Thomas, 1993). The

degree of redundancy is per de®nition relative to the envir-

onmental conditions, and thus to the laboratory assays used

to investigate developmental processes. We have suggested

that the standard experimental conditions to study Dictyos-

telium development do not re¯ect the stringent complexity

of the natural ones, and sometimes fail to reveal graded

phenotypic defects. In the natural environment, Dictyoste-

lium cells are exposed to rough, uneven surfaces made of

soil particles, decaying leaves and other debris with varying

physical and chemical properties. In the laboratory, the cells

can grow and aggregate in liquid media under submerged

conditions; postaggregative development and fruiting

require a solid substratum and contact with an air±water

interphase. Commonly used substrata to study development

include agar, wetted ®lter paper, sometimes glass coverslips

or polystyrene dishes. On all these substrata, cell aggrega-

tion results in a two-dimensional streaming pattern, while it

is three-dimensional on soil; diffusion patterns of secreted

chemoattractants are not as irregular as they may be on an

uneven ground, and cells do not have to `climb mountains or

cross valleys' to aggregate. Similarly, slug migration on a

smooth, even surface is very likely less dif®cult than on soil,

where slugs have to bridge chasms between soil particles

(Higuchi and Yamada, 1992). Thus, defects affecting cell

motility, cell±cell or cell±substratum adhesion will probably

interfere with cell aggregation and further development in a

more dramatic way on soil than under usual laboratory

conditions. We have devised a simple procedure to repro-

duce in the laboratory a soil substratum that allows a quan-

titative analysis of development under conditions closer to

the natural ones. By using this novel assay we have recently

shown that the adhesion molecule csA is essential for devel-

opment, while it is dispensable under more commonly used

developmental assays (Ponte et al., 1998). We have

extended this analysis to several mutants in actin-binding

proteins, and show here that deletion of many actin-binding

proteins impairs the ef®ciency and the morphology of devel-

opment.

2. Results

2.1. Timing and ef®ciency of development of null mutants for

actin-binding proteins on soil or agar

The deletion mutants that have been tested in this paper

are listed in Table 1. With the exception of mutants for

synexin, the others are all single or double mutants for

actin-binding proteins which stimulate either bundling

(ABP-34, comitin), cross-linking (a-actinin, ABP-120) or

membrane anchoring (hisactophilins, comitin, interaptin)

of actin. Comitin was identi®ed as a Golgi-bound protein

which interacts with, and bundles, actin (Weiner et al.,

1993; Jung et al., 1996). Hisactophilins are myristoylated

proteins present in membrane-bound and soluble form, and

supposedly anchor the actin cytoskeleton to the plasma

membrane at acidic pH (Scheel et al., 1989; Stoeckelhuber

et al., 1996). Synexins are the Dictyostelium isoforms of

annexin VII, and bind to membranes in the presence of

Ca21 ions (DoÈring et al., 1995). Interaptin is a new member

of the a-actinin family, which shares with a-actinin the

actin-binding domain at the NH2-terminus, while possessing

a carboxyl terminal domain responsible for binding to

membranes of the ER and nuclear envelope (Rivero et al.,

1998). The cross-linkers ABP-120, ABP-34 and a-actinin

are concentrated in the cell cortex, in pseudopodia, lamelli-

podia and focal adhesion points (Condeelis, 1995; Noegel

and Luna, 1995).

We ®rst compared the ability of wild-type or mutant cells

to undergo development and form fruiting bodies on non-

nutrient agar or `soil plates', i.e. Petri dishes containing a

layer of soil particles moistened with water, prepared as

described in the experimental procedures. In both cases,

cells were plated at the beginning of development at a

concentration of 1 £ 108 per ml. Two parameters were

analyzed: timing of development and ef®ciency of fruiting

body formation.

On agar, the time required for fruiting bodies to form was

slightly longer for mutants 342, aA2, abpD2 and syn2,

while it was comparable to the wild-type for the remaining

mutants tested (Fig. 1). Under the conditions used, we did

not appreciate changes in shape and size of fruiting bodies

formed on agar in any of the analyzed mutants, but the

ef®ciency in fruiting body formation, compared to the

wild-type, was reduced by 20±35% in some cases (Fig. 2).

On soil plate, the deletion mutants fell into two distinct

groups: comitin2, H1/H22, and the double mutant comi-

tin2/H1/H22 formed fruiting bodies of normal shape and

size with timing and ef®ciency that were comparable to

wild-type cells; the remaining strains, with the exception

of mutant 1202, started forming fruiting bodies with a slight

delay in the case of 342, and a more pronounced one for

mutants aA2, 342/aA2, abpD2 and syn2 (Fig. 1). The

latter mutants, as well as mutant 1202, were strongly

impaired in their ef®ciency at fruiting, which varied from

less than 1% of the wild-type for 342/aA2 mutant to a

maximum of 45 ^ 12% for syn2 mutant (Fig. 2). Interest-

ingly, the double mutation in 34-kDa actin-bundling protein

and a-actinin resulted in a much stronger impairment of

fruiting body formation than in the single mutants, whereas

no additive effect was found in the double mutation 342/

1202 (Fig. 2).

2.2. Altered phenotypes of null mutants for actin-binding

proteins on soil

Most of the mutants that were less ef®cient at fruiting also

showed an abnormal phenotype at the end of development

E. Ponte et al. / Mechanisms of Development 91 (2000) 153±161154

on soil, which was not evident on agar. The fruiting bodies

formed by mutant syn2 were characterized by having smal-

ler sori on top of longer stalks (Fig. 3). Mutant 1202, double

mutant 342/1202 and, to a lesser extent, mutant aA2

formed small fruiting bodies, consisting of normal size

sori but very short, sometimes thick, stalks (Fig. 3). Mutant

abpD2 formed fruiting bodies heterogeneous in size;

remarkably, this mutant gave rise to slugs with multiple

enlargements all over the body (Fig. 3B,C), and releasing

a large amount of slime sheath during migration (Fig. 3A).

The double mutant 342/aA2 was the most impaired one,

being almost totally blocked at slug stage, forming extre-

mely small slugs rarely able to transform into minute fruit-

E. Ponte et al. / Mechanisms of Development 91 (2000) 153±161 155

Fig. 1. Timing of development of Dictyostelium wild-type cells (AX2) or

null mutants for actin-binding proteins on agar or soil plates. At the begin-

ning of starvation, a total of 2.5 £ 107 cells of the indicated strain in 0.2 ml

of 0.017 M SoÈrensen Na/K phosphate buffer were plated on agar or soil

plates to cover an area of about 2 cm2 and incubated at 238C. Development

was monitored at regular intervals with a stereomicroscope. The time

elapsed from the beginning of starvation for appearance in the colonies

of the ®rst compact aggregates (dotted lines) or fruiting bodies (dashed

lines) is shown in the ordinate. The average times (^SD) of ®ve experi-

ments are shown (for fruiting body formation only). For further details, and

for preparation of soil plates, see Section 4.

Fig. 2. Ef®ciency in fruiting body formation of null mutants for actin-

binding proteins on agar or soil plates. The number of fruiting bodies per

colony formed by each strain was scored and expressed as a percentage of

AX2 fruiting bodies formed on each plate. Mean values (^SD) of three

experiments are shown. Experimental conditions as in Fig. 1.

Table 1

strains useda

Strain Genotype Description

AX2 Wild-type strain AX2-214, derivative of NC-4

342 abpB2/hygR 34 kDa actin-bundling protein-de®cient cell line generated from AX2 by gene

replacement (Rivero et al., 1996a)

aA2 abpA2/G418R a-Actinin-de®cient cell line generated from AX2 by gene replacement (Rivero

et al., 1996b)

1202 abpC2/G418R Gelation factor (ABP-120)-de®cient cell line generated from AX2 by gene

disruption (Rivero et al., 1996b)

342/aA2 abpA2/abpB2/hygR/G418R 34 kDa protein and a-actinin-de®cient cell line generated from 342 by gene

disruption (Rivero et al., 1999)

342/1202 abpB2/abpC2/hygR/G418R 34 kDa protein and gelation factor-de®cient cell line generated from 342 by

gene disruption (Rivero et al., 1999)

abpD2 abpD2/bsrR Interaptin-de®cient cell line generated by gene disruption (Rivero et al., 1998)

Syn2 nxnA2/G418R Synexin-de®cient cell line generated by gene disruption (DoÈring et al., 1995)

H12/H22 hatA2/G418R Hisactophilin I- and II-de®cient cell line obtained by gene disruption

(Stoeckelhuber et al., 1996)

Com2 comA2/bsrR Comitin-de®cient cell line obtained by gene disruption (Krempelhuber et al.,

unpublished data)

Com2/H12/H22 comA2/hatA2/G418R/bsrR Comitin- and hisactophilins-de®cient cell line generated from H12/H22 mutant

by gene disruption (Krempelhuber et al., unpublished data)

a The nomenclature of Loomis and Kuspa (1997) has been followed for the loci corresponding to the actin-binding proteins tested. The resistance cassettes

used were hygromycin (hygR), neomycin (G418R) and blasticidin (bsrR).

ing bodies (Fig. 3). The double mutation 342/aA2, but not

342/1202, displays a clearly additive phenotypic effect. In

fact, the double mutant 342/aA2 has a much more

pronounced morphological defect than the single mutant

E. Ponte et al. / Mechanisms of Development 91 (2000) 153±161156

Fig. 3. Phenotypes of wild-type or null mutants in actin-binding proteins at the end of development on soil plates. Fruiting bodies and/or slugs of the indicated

strains are shown. For abpD2 mutant, examples of slugs (B,C) and of slime sheath meshwork (A) are shown. For comparison, slugs formed by the wild-type

AX2 are also shown. Wild-type fruiting bodies resemble those of com2, HIH22 and com2/HIH22. Bars 0.25 mm.

aA2, while mutant 342 forms normal fruiting bodies,

though reduced in number. In contrast, the fruiting bodies

formed by 342/1202 are morphologically similar to those of

the single mutant 1202.

2.3. Spore yields in null mutants for actin-binding proteins

developing on soil or agar

Estimating the fruiting body number on soil has the draw-

back that small slugs and fruiting bodies that do not manage

to reach the soil surface escape scoring. The mutants may,

thus, appear to be inef®cient at fruiting, though their spore

yield could be similar to the wild-type. We therefore deter-

mined the spore yield in the whole colonies for the seven

mutants that were at the greatest disadvantage when plated

on soil. Soil particles, or the piece of agar, corresponding to

the entire colony area were collected in equivalent volumes

of buffer solution, potentially entrapped amoebae were

lysed with SDS, and an equal volume of the solution

containing spores was plated on bacterial lawns. As

shown in Fig. 4, the percentage of colonies formed by spores

originating from agar-plated mutant cells was comparable to

the wild-type in mutants 342, 1202 and 342/1202, but

reduced by about 50% for syn2, aA2 and 342/aA2, and

by 75% for abpD2 cells. For colonies developed on soil, the

spore yield amounted in all cases to less than 20% of the

wild-type, except for syn2, where it was of about 40% (Fig.

4). Again, the double mutation 342/aA2, not however 342/

1202, displayed an additive effect. Thus, the determined

spore yields on soil were dramatically lower than on agar,

consistent with the soil conditions being more stringent, and

correlated with the reduced ef®ciency to form fruiting

bodies.

2.4. Spore viability in null mutants for actin-binding

proteins

A reduced spore yield in the colony may result from

either a lower proportion of cells entering the fruiting bodies

and maturing into SDS-resistant spores or the spores being

less viable or both. To test the possibility that a defect in

actin-binding proteins might affect spore viability, we

plated a de®ned number of SDS-treated, morphologically

intact spores on bacterial lawns and determined the viability

index by calculating the proportion of colonies formed per

plated spores. As summarized in Table 2, the spore viability

index of mutants 342, 1202 and 342/1202 was comparable

to the wild-type, while it was reduced by about 25% in syn2,

and by 50±70% in mutants aA2, 342/aA2 and abpD2.

Remarkably, with the exception of syn2, for the remaining

three mutants these values are in the same range as their

spore yields on agar. We conclude that the lower spore yield

of mutants aA2, 342/aA2 and abpD2 on agar re¯ects the

reduced spore viability only, whereas on soil it depends on

both lower spore viability and a lower number of cells enter-

ing the fruiting bodies.

3. Discussion

3.1. Phenotypes on soil and agar of mutants in actin-binding

proteins

The more stringent conditions of the developmental assay

on soil plates have revealed developmental effects of null

mutations in actin-binding proteins that are not evident, or

only barely detectable, on agar plates. As developmental

parameters we have investigated timing, ef®ciency of fruit-

ing body formation, altered phenotype and spore yield. We

have also examined the possibility that a reduced spore yield

might re¯ect a reduced spore viability consequential to dele-

tion of a given actin-binding protein. Depending on the

severity of the effects, the mutants can be grouped in four

classes: (1) deletion of a-actinin or interaptin strongly

impairs the viability of SDS-resistant spores as well as the

ef®ciency to form fruiting bodies or spores on soil and, to a

much lower extent, on agar; (2) deletion of synexin reduces

the ef®ciency of fruiting body formation on soil, but not on

agar, affects partially the spore yield both on agar and soil,

and reduces slightly spore viability; (3) deletion of the 34-

E. Ponte et al. / Mechanisms of Development 91 (2000) 153±161 157

Fig. 4. Spore yields of selected null mutants for actin-binding proteins on

agar or soil plates. At the end of development, the agar piece or the soil

particles of each whole colony were transferred to 50-ml test tubes contain-

ing 10-ml SoÈrensen phosphate buffer. After repeated vortexing, the larger

soil particles were allowed to sediment and 5 ml of the suspension was

transferred to 15-ml test tubes and treated with 0.5% SDS for 5 min. A 1-ml

aliquot of the suspension containing SDS-resistant spores was then trans-

ferred to a new 15-ml tube, and centrifuged at 1000 £ g for 5 min. The pellet

was resuspended in 10-ml SoÈrensen phosphate buffer, and samples contain-

ing spores were diluted 1000- or 10 000-fold; from each dilution aliquots of

1, 10 or 25 ml were plated on bacterial lawns. The number of colonies

formed per each dilution was scored and expressed as a percentage relative

to colonies formed by wild-type spores originating from agar- or soil-plated

cells. The number of germinating AX2 spores was 25 ^ 7 and 5 ^ 2 per ml

on agar or soil, respectively.

kDa actin-bundling protein or the 120-kDa gelation factor

also affects the ef®ciency to form fruiting bodies and spores

on soil, with only slight effects on agar, and no effect on

spore viability; (4) single or double deletions of comitin and

hisactophilins have no effect whatsoever on development or

fruiting body shape.

With regard to the phenotype of fruiting bodies in the

de®cient mutants, there is a clear difference between syn2

and abpD2 mutants, from one side, and mutants 1202 and

aA2, on the other side. Synexin and interaptin-minus

mutants give rise to fruiting bodies with, respectively, elon-

gated stalk and small sorus or heterogeneous size; the fruit-

ing bodies of mutants 1202 and aA2 are characterized by

having a short stalk and normal-size sori. The latter pheno-

type is quite evident on soil, and much less pronounced, if

not absent, on agar. Interestingly, the double mutation 342/

aA2 results in clearly cumulative effects with regard to the

phenotype as well as ef®ciency in fruiting body formation

and spore yield; developmental timing and spore viability

instead resembled those of aA2 mutant. In contrast, the

double mutation 342/1202 does not display any additive

effects, the developmental parameters of the double mutant

being identical to that of 1202 mutant.

3.2. The altered phenotypes in mutants for actin-cross-

linking proteins are related to defects in cell motility

affecting aggregation and culmination

A delay in developmental timing and low ef®ciency at

fruiting on soil in mutants defective in actin-cross-linking

proteins, such as 34-kDa bundling protein, a-actinin or gela-

tion factor, can be explained by alterations in cell motility.

Altered pseudopod formation, reduced cell motility and

chemotaxis have been reported for gelation factor null

mutant (Cox et al., 1996). A slight but signi®cant reduction

in chemotactic speed on glass has also been recently

reported for double mutants 342/1202 and 342/aA2, but

not however for the single mutants 342 and aA2 (Rivero

et al., 1999). The latter mutant showed a slightly impaired

adhesion to polystyrene dishes (data not shown). We have

shown that the stringent conditions which cells are exposed

to on soil accentuate defects in cell motility and/or adhesion

to substrate, which are barely detectable on agar or plastic

surfaces (Ponte et al., 1998). By altering cell motility,

disruption of actin-cross-linking proteins, such as a-actinin,

the 34-kDa bundling protein or the 120-kDa gelation factor,

will lead to a lower number of cells undergoing aggregation,

and thus to a reduced number of fruiting bodies. In this

respect, all three cross-linkers seem to contribute equally

well to the aggregation process.

A defect in cell motility can also explain the most

common phenotype of fruiting bodies formed on soil by

null mutants for a-actinin and gelation factor, namely

small fruiting bodies with a short, sometimes thick, stalk.

A highly dynamic cytoskeleton is required for migration of

prestalk cells inside the cell mass of the culminating slug,

and to provide them the necessary strength for lifting the

spore mass into the air (Shelden and Knecht, 1995; Rivero et

al., 1996b). The minute slugs formed by 342/aA2 mutant

and their extreme dif®culty to culminate can also be

explained by a defect in cell motility impairing aggregation

®rst and culmination later on. The cumulative effects of the

342/aA2 double mutation, in contrast with the double

mutation 342/1202, and the reported cumulative effects

on development of the double mutation aA/1202 (Rivero

et al., 1996b) suggest that a-actinin is the most prominent

regulator of actin-linked motility during development.

3.3. Synexin and interaptin may regulate differentiation and

patterning

The fruiting bodies of synexin-minus mutants consist of

elongated stalks and relatively small sori. Clearly, culmina-

tion is not a problem in the mutant, but there could be a

patterning problem, with relatively more cells differentiat-

ing along the stalk pathway. Synexin has been involved in

the regulation of intracellular Ca21 homeostasis (DoÈring et

al., 1995), and it is tempting to speculate that the observed

phenotype could be linked to this activity. Indeed, several

lines of evidence suggest that raising intracellular Ca21

levels shifts the stalk/spore ratio toward the stalk pathway

(Cubitt et al., 1995).

The phenotype of interaptin null mutant on soil differs not

only in the heterogeneity in fruiting body size and shape, but

also in the remarkable enlargements distributed all over the

migrating slug body, and the high amount of released slime

sheath. Interaptin is developmentally regulated and

expressed mainly in the tip and in anterior-like cells

(ALC) of the slug (Rivero et al., 1998). It is possible that

interaptin, by regulating membrane traf®cking and secretion

E. Ponte et al. / Mechanisms of Development 91 (2000) 153±161158

Table 2

Spore viability in null mutants for actin-binding proteinsa

Strains Plated

spores (n)

Colonies

formed (n)b

Viability index

(colony/spore)c

AX2 200 168 ^ 20 0.84 ^ 0.01

342 200 160 ^ 17 0.80 ^ 0.09

1202 200 169 ^ 20 0.84 ^ 0.01

342/

1202200 154 ^ 5 0.77 ^ 0.03

aA2 200 90 ^ 7 0.45 ^ 0.03

342/

aA2200 81 ^ 6 0.40 ^ 0.03

abpD2 200 62 ^ 6 0.31 ^ 0.03

syn2 200 128 ^ 19 0.64 ^ 0.10

a At the end of development on agar, spores were randomly collected

with a loop and transferred in SoÈrensen phosphate buffer containing 0.5%

SDS. After 5 min SDS-resistant spores were diluted 1000-fold, counted to

determine the concentration, and an aliquot containing 200 spores per each

strain was plated in triplicate on bacterial lawn.b The number of colonies formed per plate was scored, and reported in

column 3 (^SD).c Viability index, proportion of colonies formed per plated spores (^SD).

in the tip (Rivero et al., 1998), could be involved in proper

cell differentiation and sorting. If so, the enlargements of the

slugs could re¯ect a defect in tip dominance leading to slug

fragmentation, which could explain the heterogeneity in

fruiting body size.

3.4. Spore viability: possible roles of actin-binding proteins

Deletion of a-actinin or interaptin, and to a lower extent

synexin, results in lower spore viability. Our experiments do

not allow us to discriminate whether the spores fail to differ-

entiate properly, despite the presence of a spore coat, or to

germinate. If the former is true, many seemingly intact

spores could be killed by SDS treatment. Evidence for

involvement of the actin cytoskeleton in spore germination

is supported by the ®nding that actin is strongly phosphory-

lated in dormant spores, and its dephosphorylation is the

®rst temporal event upon activation of germination (Kishi

et al., 1998). A transient increase in intracellular Ca21

levels, followed by activation of calmodulin, is also required

prior to spore swelling (Lydan and Cotter, 1995). Both a-

actinin and synexin are regulated by Ca21; synexin, in turn,

regulates Ca21 homeostasis. Thus, Ca21-regulated actin

cross-linking and/or anchoring to the membrane could be

required for optimal swelling and germination. Mutations in

actin-cross-linking proteins could affect germination by

altering the visco-elasticity of cells. In this context, it is

suggestive that aA2 cells are less resistant against deforma-

tion (Eichinger et al., 1996), and that mutants aA2/1202

and aA2/342 are more sensitive to osmotic stress (Rivero

et al., 1996b, 1999).

We do not have an explanation for the lower spore viabi-

lity in interaptin null mutants. As mentioned above, the

protein is expressed at high levels in the ALC cells, and

only at low levels in prespore cells, which might suggest a

role for the protein during slug migration and culmination,

but not in spore germination. However, a low level expres-

sion of interaptin has been found in vegetative cells; thus, it

would be interesting to examine whether changes in the

expression of interaptin are correlated with spore germina-

tion.

The introduction of soil plates to study development has

evidenced developmental defects in gene disruption mutants

for proteins involved in cell adhesion (Ponte et al., 1998)

and motility (this paper), allowing to de®ne better their

degree of relative redundancy. The ®nding that mutants

for comitin and hisactophilins develop normally on soil is

further evidence for the validity of the experimental system.

Comitin is supposed to play an intracellular role in vesicle

traf®c, which may not be linked with development (Jung et

al., 1996). Hisactophilin interacts with actin only at acidic

pH (Stoeckelhuber et al., 1996). It is possible that a potential

role of hisactophilin in development could be revealed by

exposing hisactophilin null and overexpressing mutants to

acidic soil particles.

4. Materials and methods

4.1. Cell cultures and growth conditions

AX2 wild-type and mutant cells were cultivated in axenic

medium under shaking at 150 rev./min and 238C in a

KuÈhner (Birsfelden, Switzerland) climatic cabinet as

described (Watts and Ashwort, 1970; Peracino et al.,

1998). They were harvested at a density of no more that

4 £ 106 cells/ml. The single and double mutants used in

this work were all generated by gene disruption as described

in the references shown in Table 1.

4.2. Preparation of soil plates

Soil plates were prepared as described by Ponte et al.

(1998). Brie¯y, commercially available garden soil (type

universal, neutral pH) was sieved to obtain particles of

homogenous size (#0.4 cm in diameter), and was auto-

claved. The autoclaved soil can be stored for several

weeks in a closed container at room temperature, without

apparent contamination. Soil plates were prepared freshly

by distributing aliquots of 20 g soil homogeneously on 90-

mm in diameter Petri dishes, and moistening with 0, 5 or 10

ml of sterile water to obtain three types of soil plates with

different degrees of moisture. Dictyostelium development is

dependent on the degree of humidity, thus preliminary

experiments with AX2 wild-type cells were performed on

all three types of plates when a new batch of soil was

prepared to ®nd out the soil plate in which the developmen-

tal timing was closer to that of agar plates. For most of the

experiments described in this paper this corresponded to soil

plates moistened with 5 ml of water.

4.3. Development on soil or agar plates

Cells were washed free of axenic medium, and concen-

trated to 1 £ 108/ml in 0.017 M SoÈrensen Na/K phosphate

buffer (pH 6.0), and aliquots of 0.25 ml were pipetted onto

soil or non-nutrient agar plates to cover an area of approxi-

mately 2 cm2. Four samples can be accommodated easily in

one plate with enough space left in between to avoid any

interference between colonies. The plates were covered with

a lid and were incubated at 238C. Development and fruiting

body formation were monitored with a Wild M3Z stereo-

microscope (Heerbrugg, Switzerland) equipped with over-

head optical ®bers (Intralux 6000, Volpi Heerbrugg,

Switzerland).

4.4. Determination of spore yield and spore viability

To determine the spore yield on soil or agar, starving cells

were plated as described above. After 48 h, the agar piece or

the soil particles of each whole colony were transferred in

50-ml test tubes containing 10-ml SoÈrensen phosphate

buffer. After repeated vortexing, the larger soil or agar parti-

cles were let to sediment, and 5 ml of the suspension was

E. Ponte et al. / Mechanisms of Development 91 (2000) 153±161 159

immediately transferred to 15-ml test tubes and treated with

0.5% SDS for 5 min (Bozzaro et al., 1987). A 1-ml aliquot

of the suspension containing spores was then transferred to a

new 15-ml test tube and was centrifuged at 1000 £ g for 5

min. The pellet was resuspended in 10-ml SoÈrensen phos-

phate buffer, and samples containing spores were diluted

either 1000 or 10 000-fold; from each dilution aliquots of

1, 10 or 25 ml were plated on nutrient agar with Escherichia

coli B/2 as food source. The number of plaques formed per

each dilution was scored and expressed as a percentage

relative to plaques formed by wild-type spores originating

from agar- or soil-plated cells to estimate the spore yield of

each mutant colony.

To determine spore viability, cells were let develop on

agar plates as described above. Spore were randomly

collected with a loop and transferred in SoÈrensen phosphate

buffer containing 0.5% SDS for 5 min. SDS-resistant spores

were diluted 1000-fold, counted to determine the concentra-

tion, and an aliquot containing 200 spores was plated in

triplicate on bacterial lawn. The number of plaques formed

per plate after 60 h from plating was scored, and the ratio of

colonies formed per plated spores was calculated to deter-

mine the viability index.

Acknowledgements

This work was supported by funds of MURST and CNR

to S.B.

References

AndreÂ, E., Brink, M., Gerisch, G., Isenberg, G., Noegel, A.A., Schleicher,

M., Segall, J.E., Wallraff, E., 1989. A Dictyostelium mutant de®cient in

severin, and F-actin fragmenting protein shows normal motility and

chemotaxis. J. Cell Biol. 108, 985±995.

Bonner, J.T., 1967. The Cellular Slime Molds, Princeton University Press,

Princeton, NJ, pp. 1±204.

Bozzaro, S., Ponte, E., 1995. Cell adhesion in the life cycle of Dictyoste-

lium. Experientia 51, 1175±1188.

Bozzaro, S., Hagmann, J., Noegel, A., Westphal, M., Calautti, E., Bogliolo,

E., 1987. Cell differentiation in the absence of intracellular and extra-

cellular cyclic AMP pulses in Dictyostelium discoideum. Dev. Biol.

123, 540±548.

Condeelis, J., 1995. Elongation factor 1 alpha, translation and the cytoske-

leton. Trends Biochem. Sci. 20, 169±170.

Cox, D., Wessels, D., Soll, D.R., Hartwig, J., Condeelis, J., 1996. Re-

expression of ABP-120 rescues cytoskeletal, motility, and phagocytosis

defects of ABP-120(2) Dictyostelium mutants. Mol. Biol. Cell 7, 803±

823.

Cubitt, A.B., Firtel, R.A., Fischer, G., Jaffe, L.F., Miller, A.L., 1995.

Patterns of free calcium in multicellular stages of Dictyostelium expres-

sing jelly®sh apoaequorin. Development 121, 2291±2301.

DoÈring, V., Veretout, F., Albrecht, R., MuÈhlbauer, B., Schlatterer, C.,

Schleicher, M., Noegel, A.A., 1995. The in vivo role of annexin VII

(synexin): characterization of an annexin VII-de®cient Dictyostelium

mutant indicates an involvement in Ca21-regulated processes. J. Cell

Sci. 108, 2065±2076.

Eichinger, L., KoÈppel, B., Noegel, A.A., Schleicher, H., Schliwa, M.,

Weijer, C.J., Witke, W., Janmey, P.A., 1996. Mechanical pentuzbation

elicits a phenotypic difference between Dictyostelium wild-type cells

and cytoskeletal mutants. Biophys. J. 70, 1054±1060.

Faix, J., Steinmetz, M., Boves, H., Kammerer, R.A., Lottspeich, F., Mintert,

U., Murphy, J., Stock, A., Aebi, U., Gerisch, G., 1996. Cortexillins,

major determinants of cell shape and size, are actin-bundling proteins

with a parallel coiled-coil tail. Cell 86, 631±642.

Fisher, P.R., 1997. Directional movement of migrating slugs. In: Maeda,

Y., Inouye, K., Takeuchi, I. (Eds.). Dictyostelium: A Model System for

Cell and Developmental Biology, UAP, Tokyo, pp. 437±454.

Gerisch, G., 1987. Cyclic AMP and other signals controlling cell develop-

ment and differentiation in Dictyostelium. Annu. Rev. Biochem. 56,

853±879.

Haugwitz, M., Noegel, A.A., Karakesisoglou, J., Schleicher, M., 1994.

Dictyostelium amoebae that lack G-actin-sequestering pro®lins show

defects in F-actin content, cytokinesis, and development. Cell 79,

303±314.

Higuchi, G., Yamada, T., 1992. Behaviour and differentiation of cellular

slime mold: a clari®ed mode of life on soil (Cine document, Tokyo,

Japan), 16-mm color, sound ®lm.

Hitt, A.L., Hartwig, J.H., Luna, E.J., 1994. Ponticulin is the major high

af®nity link between the plasma membrane and the cortical actin

network in Dictyostelium. J. Cell Biol. 126, 1433±1444.

Hug, C., Jay, P.Y., Reddy, I., NcNally, J.G., Bridgman, P.C., Elson, E.L.,

Cooper, J.A., 1995. Capping protein levels in¯uence actin assembly and

cell motility in Dictyostelium. Cell 81, 591±600.

Jung, E., Hammer, J.A., 1990. Generation and characterization of Dictyos-

telium cells de®cient in a myosin I heavy chain isoform. J. Cell Biol.

110, 1955±1964.

Jung, E., Fucini, P., Stewart, M., Noegel, A.A., Schleicher, M., 1996. Link-

ing micro®laments to intracellular membranes: the actin-binding and

vesicle-associated protein comitin exhibits a mannose-speci®c lectin

activity. EMBO J. 15, 1238±1246.

Kishi, Y., Clements, C., Mahadeo, D., Cotter, D.A., Sameshima, M., 1998.

High levels of actin tyrosine phosphorylation: correlation with the

dormant state of Dictyostelium spores. J. Cell Sci. 111, 2923±2932.

Loomis, W.F., 1975. Dictyostelium discoideum. A Developmental System,

Academic Press, New York, pp. 1±214.

Loomis, W.F., Kuspa, A., 1997. The genome of Dictyostelium discoideum.

In: Maeda, Y., Inouye, K., Takeuchi, I. (Eds.). Dictyostelium: A Model

System for Cell and Developmental Biology, UAP, Tokyo, pp. 15±30.

Lydan, M.A., Cotter, D.A., 1995. The role of Ca21 during spore germina-

tion in Dictyostelium. Autoactivation is mediated by the mobilization of

Ca21 while amoebal emergence requires entry of external Ca21. J. Cell

Sci. 108, 1921±1930.

Maniak, M., Rauchenberger, R., Albrecht, R., Murphy, J., Gerisch, G.,

1995. Coronin involved in phagocytosis: dynamics of particle-induced

relocalization visualized by a green ¯uorescent protein tag. Cell 83,

915±924.

Noegel, A.A., Luna, E., 1995. The Dictyostelium cytoskeleton. Experientia

51, 1135±1143.

Parent, C.A., Devreotes, P.N., 1996. Molecular genetics of signal transduc-

tion in Dictyostelium. Annu. Rev. Biochem. 65, 411±440.

Peracino, B., Borleis, J., Jin, T., Westphal, M., Schwartz, J.-M., Wu, L.,

Bracco, E., Gerisch, G., Devreotes, P., Bozzaro, S., 1998. G protein b

subunit-null mutants are impaired in phagocytosis and chemotaxis due

to inappropriate regulation of the actin cytoskeleton. J. Cell Biol. 141,

1529±1537.

Ponte, E., Bracco, E., Bozzaro, S., 1998. Detection of subtle phenotypes:

the case of the cell adhesion molecule csA in Dictyostelium. Proc. Natl.

Acad. Sci. USA 95, 9360±9365.

Rivero, F., Furukawa, R., Noegel, A.A., Fechheimer, M., 1996a. Dictyos-

telium discoideum cells lacking the 34 000-dalton actin-binding protein

can grow, locomote and develop, but exhibit defects in regulation of cell

structure and movement: a case of partial redundancy. J. Cell Biol. 135,

965±980.

Rivero, F., KoÈppel, B., Peracino, B., Bozzaro, S., Siegert, F., Weijer, C.J.,

Schleicher, M., Albrecht, R., Noegel, A.A., 1996b. The role of the

E. Ponte et al. / Mechanisms of Development 91 (2000) 153±161160

cortical cytoskeleton: F-actin crosslinking proteins protect against

osmotic stress, ensure cell size, cell shape and motility, and contribute

to phagocytosis and development. J. Cell Sci. 109, 2679±2691.

Rivero, F., Kuspa, A., Brokamp, R., Matzner, M., Noegel, A.A., 1998.

Interaptin, an actin-binding protein of the a-actinin superfamily in

Dictyostelium discoideum, is developmentally and cAMP-regulated

and associates with intracellular membrane compartments. J. Cell

Biol. 142, 735±750.

Rivero, F., Furukawa, R., Fechheimer, M., Noegel, A.A., 1999. Three actin

cross-linking proteins, 34 kDa actin-bundling protein, a-actinin, and

ABP-120, have both unique and redundant roles in growth and devel-

opment of Dictyostelium. J. Cell Sci. 112, 2737±2751.

Scheel, J., Ziegelbauer, K., Kupke, T., Humbel, B.M., Noegel, A.A.,

Gerisch, G., Schleicher, M., 1989. Hisactophilin, a histidine-rich

actin-binding protein from Dictyostelium discoideum. J. Biol. Chem.

164, 2832±2839.

Shelden, E., Knecht, D., 1995. Mutants lacking myosin II cannot resist

forces generated during multicellular morphogenesis. J. Cell Sci. 108,

1105±1115.

Siegert, F., Weijer, C., 1997. Control of cell movement during multicellular

morphogenesis. In: Maeda, Y., Inouye, K., Takeuchi, I. (Eds.). Dictyos-

telium: a Model System for Cell and Developmental Biology, UAP,

Tokyo, pp. 425±436.

Stoeckelhuber, M., Noegel, A.A., Eckerkorn, C., KoÈhler, J., Rieger, D.,

Schleicher, M., 1996. Structure/function studies on the pH-dependent

actin-binding protein hisactophilin in Dictyostelium mutants. J. Cell

Sci. 109, 1825±1835.

Thomas, J.H., 1993. Thinking about genetic redundancy. Trends Genet. 9,

395±399.

Watts, D.J., Ashwort, J.M., 1970. Growth of myxamoebae of the cellular

slime mould Dictyostelium discoideum in axenic culture. Biochem. J.

(Tokyo) 119, 171±174.

Weiner, O.H., Murphy, J., Grif®ths, G., Schleicher, M., Noegel, A.A., 1993.

The actin-binding protein comitin (p24) is a component of the Golgi

apparatus. J. Cell Biol. 123, 23±34.

Williams, J., 1997. Prestalk and stalk cell heterogeneity in Dictyostelium

discoideum. In: Maeda, Y., Inouye, K., Takeuchi, I. (Eds.). Dictyoste-

lium: A Model System for Cell and Developmental Biology, UAP,

Tokyo, pp. 293±304.

Witke, W., Schleicher, M., Noegel, A.A., 1992. Redundancy in the micro-

®lament system ± abnormal development of Dictyostelium cells lacking

two F-actin cross-linking proteins. Cell 68, 53±62.

E. Ponte et al. / Mechanisms of Development 91 (2000) 153±161 161