Effects of cytochalasin B on cell movements and chemoattractant-elicited actin changes of Dictyostelium

Experimental Cell Research 160 (1985) 275-286

Effects of Cytochalasin B on Cell Movements and Chemoattractant-elicited Actin Changes of Dictyostelium STUART J. McROBBIE’, ’ and PETER C. NEWELL’, * ‘Department of Biochemistry, Universi@ of Oxford, Oxford 0x1 3Qll, and ‘Mi?C Laboratory of Molecular Biology, Cambridge, CB2 ZQH, UK

The actin-binding drug cytochalasin B (CB) was employed to study the stability and role of cytoskeletal actin following chemotactic stimulation of Dictyostelium discoideum. Intact amoebae were found to be impermeable to this drug, as shown by lack of inhibition of chemotactic movement in its presence and failure of t3H]CB to bind to intact amoebae. However, there were approx. 150000 high afIinity CB-binding sites per cell detectable after cell breakage and preparation of Triton-insoluble cytoskeletons. The effect of CB on cytoskeletons was to destabilize the second (2545 set) and third (60 set) chemotacticallyinduced peaks of cytoskeletal actin accumulation and to reduce the actin levels to the low prestimulus amount. In contrast, the drug had no such action on the rapid (3-5 set) actin peak. This peak appeared to be stable in the presence of CB added before or simultaneously with lysis of the cell. It was also observed that the instability of the second and third peaks to CB gradually decreased after cell lysis (as did the number of CB binding sites) such that if CB was added 5 min after lysis of the chemotactically stimulated amoebae it had no destabilizing effect. Evidence was obtained from experiments employing centrifugation of cytoskeletons at 100000 g and from the use of the DNase I inhibition assay for Gactin, that the first (3-5 set) actin peak of accumulation involved polymerization rather than just cross-linking of short filamentous actin fragments. The significance of these actin accumulation peaks is discussed and their timing correlated with events involved in chemotaxis. @ 1985 Academic Press, Inc.

Many eukaryotic cells can be seen to move across a surface by extending pseudopodal projections of their cell periphery. In recent years much effort has been expended in unravelling the mechanisms of such motility. With the discovery of the involvement of actomyosin in the sliding filament mechanism of force generation in striated muscle [l] and the subsequent detection of actin and myosin in eukaryotic cells [2] most cellular research is now based on the premise that actin and myosin, together with their regulatory proteins, provide a universal motor for a wide range of motile functions. The polymerization of actin into filaments capable of bearing tension is a prerequisite for its role in cell structure and motility. In muscle the polymerization occurs only once and results in nearly 100% of the cells’ actin being in the filamentous form. In contrast, non-muscle actin filaments are considerably more labile than muscle actin filaments and large pools of unpolymerised actin are * To whom offprint requests should be addressed. 19858340

Copyright @ 1985 by Academic Press, Inc. All rights of reproduction in any form reserved 0014-4827/85 $03.00

276 McRobbie and Newell usually found in non-muscle cells under non-stimulated conditions [3]. Consequently temporal and spatial regulation of actin polymerization may be central to the structural and motile activities of non-muscle cells. Such a possibility has been suggested for several cell types including thrombin-stimulated platelets [4-71, neutrophils stimulated with chemotactic peptide [g-10] and during construction of the acrosomal process of Thyone sperm [l 11. Actin-filament capping drugs, the cytochalasins, have been widely used to study the role of actin polymerization in cell motility [8, 9, 12-151. These drugs (of which cytochalasin B (CB) is the most widely used) have been reported to inhibit actin polymerization by preventing monomer addition at the ‘barbed’ end of actin filaments, i.e., the preferred end for monomer addition [16, 171, and to cause depolymerisation of actin filaments already formed [12, 181. We have recently reported changes in actin organization in the cellular slime mould Dictyostelium discoideum following chemotactic stimulation [ 19, 201. These changes may be associated with cell motility in this organism. The existence of two well characterized, developmental stage-specific chemoattractants, folic acid [21] and CAMP [22] allows us to study such processes as stimulusmediated cytoskeletal assembly with great precision in this primitive eukaryote. Dictyostelium actin has many of the properties of skeletal muscle actin [23-251 and can exist in at least two states of assembly, G (monomer) and F (filamentous polymer), which are in equilibrium and subject to an ATP-driven steady-state exchange [26]. In this paper we show that CB has markedly different destabilizing effects on different peaks of accumulated actin in isolated cytoskeletons from amoebae of Dictyostelium after chemotactic stimulation. METHODS Growth and Development of Cells D. discoideum, strain NC-4, was maintained on SM nutrient agar [27] in association with Klebsiella aerogenes. Cells were prepared for experiments by growth as mass plates on SM agar under conditions allowing uniform clearing of the bacterial lawn by the amoebae. Development was initiated by harvesting the clearing plates with 17 mM phosphate buffer (pH 6.15) and washing the amoebae free of bacteria by centrifugation at 190 g for 2 min. After three such washes the cells were resuspended in the same buffer at a density of 2x10’ ml-’ and shaken at 170 rpm-’ at 22°C on an orbital shaker until the desired time of development was reached.

Cell Motility Assays Motility competence of cells was determined by their ability to move outwards from their original position in a high population density cell spot on agar containing chemoattractant [28, 291. These assays were done at 22°C in 24well Linbro tissue culture plates with 0.5 ml of buffered agar (2% purified agar in 17 mM phosphate buffer) containing either 10e5 M folate, 10M5M CAMP or no chemoattractant, and CB as detailed in the text. Cells were harvested from the growth margin of a single clone on SM agar, washed free of bacteria and resuspended at a density of 109 ml-‘. In each tissue culture well a single 1 ul drop of this suspension was deposited on the agar surface. The size of each spot of cells was determined immediately under the microscope and the degree of movement outward from the spots was scored 7 h later. Exp Cell Res 160(1985)

Cytochalasin B, actin and cell motility in Dictyostelium

277

Optical Density Changes in Cell Suspensions Cells at the desired point in development were resuspended at 10sml-’ in 17 mM phosphate buffer and placed in a 3 ml spectrophotometer cuvette. Cells were oxygenated and kept in suspension by passing 95 % Or : 5 % CO2 at 40-50 cm3 mitt- ’ through two fine syringe needles anchored down one side of the cuvette. Changes in optical density were continuously recorded at 405 nm in a Unicam SP1800 spectrophotometer. This is basically the method described by Gerisch & Hess 1301.

Isolation of Cytoskeletal Proteins Cytoskeletal proteins were isolated as those insoluble in ‘hiton X-100 as described previously [19, 201. Cells at a density of 10s ml-’ in 17 mM phosphate buffer (pH 6.15) were distributed in 150 pl ahquots to 1.5 ml microcentrifuge tubes shaken at 1400 r-pm-‘. Chemoattractant (10 pl) was then added to each tube and allowed to react for a preset time before the reaction was terminated by addition of 160 pl of a 2% ‘It&on solution 119, 201. It is important to note that in CB pretreatment experiments the drug was present throughout this process. Tubes were then placed on ice for 10 min. allowed to warm to room temperature for a further 10 min, then centrifuged at 8000 g (unless otherwise specified) for 4 min. Supematants were discarded and each pellet washed in 300 pl of 2 % ‘hiton solution diluted 1: 1 with 17 mM phosphate buffer. The tubes were recentrifuged, supematants again discarded and the tubes inverted to dry. The ‘Riton-insoluble cytoskeletal pellets were then prepared for gel electrophoresis [19, 201. Cells employed in DNase I inhibition assays were lysed in ‘Iliton and used immediately after the 10 min period on ice without further centrifugation or washing.

Quantitation of Changes in Cytoskeletal Actin Content Changes in cytoskeletal total actin were assessed by scarming densitometry of stained actin-protein bands in cytoskeleton samples separated on one-dimensional SDS-polyacrylamide slab gels as described previously [19, 201.

Binding of [‘H]Cytochalasin

B to Whole Cells and Cytoskeletons

Cells at the required time of development were suspended at 5x10’ ml-’ in 17 mM phosphate buffer (pH 6.15). Cytoskeletons were prepared as described above at a density of 5x 10’ ml-‘. Portions of these suspensions were incubated, as described in the text, at room temperature in a 250 pl reaction volume containing 10d6 M [‘HICB (Amenham International PLC) with or without a lOOfold excess of unlabelled CB and containing 10’ cells or cytoskeletons. A portion of each reaction volume (200 pII)was placed in a 500 pl microcentrifuge tube and bound ligand was separated from free ligand by centrifugation at 8000 g for 4 min. Supematants unere.discarded and tubes carefully dried out. The tip of each tube was then cut off, placed in a scintillation vial and soaked overnight in cocktail T scintillant (BDH) before counting. Specific binding was calculated by subtracting non-specific radiolabel bound (in presence of excess unlabelled hgand) from total bound (no excess unlabelled ligand).

DNase Inhibition Assay Changes in G-actin concentration were determined by the ability of fresh ‘Il-iton-lysed amoebal preparations to inhibit the activity of DNase I. In mixtures of monomeric and fdamentous actin, only the monomeric form is measured as DNase inhibitor [31]. The modification of the basic assay described by Harris et al. [32] was employed. Total actin was determined by assaying samples in a depolymerizing concentration of guanidine HCI [3 11.

RESULTS Efsects of Cytochalasin B on Cell Movement Cell movement of Dictyostelium amoebae was examined in two ways. First, we observed the outward expansion of high density cell spots on buffered agar with

278 McRobbie and Newell

AE

4

0.1

t

60 set

I

I

Fig. 1. Optical density changes at 405 nm seen in response to pulses of chemoattractant applied to oxygenated cell supsensions at 22°C. Cells were pretreated with either 1% DMSO (control) or 10m5 M CB in 1% DMSO for at least 3 min before stimulation. Arrow indicates addition of stimulus. (a) 0 h (preaggregative) cells in the presence of 1% DMSO, stimulated with lo-’ M folate; (b) 0 h (preaggregative) cells in the presence of 10d5 M CB in 1% DMSO, stimulated with 10-j M folate; (c) 8 h (aggregation-competent) cells in the presence of 1% DMSO, stimulated with lo-’ M CAMP; (d) 8 h (aggregation-competent) cells in the presence of 10d5 M CB in 1% DMSO, stimulated with lo-’ M CAMP. These traces are representative examples of at least four separate experiments.

or without added chemoattractant (lo-‘) and in the presence of either 10e5 M CB in 1% DMSO, or 1% DMSO alone. Such assays are thought to function by enzymatic degradation of chemoattractant by the amoebae in their localized spots with subsequent response to the gradient of attractant thereby created [28, 291. Using this test we found that CB had no effect on the movement of amoebae towards either CAMP or folate. Secondly, we examined the effects of lo-’ CB on optical density changes induced by chemoattractants in oxygenated amoebal suspensions. These changes are thought to be a consequence of shape changes associated with chemotactic movement [30, 33, 341. The results showed that CB had no effect on these changes (fig. 1). These results suggest that either intact amoebae are insensitive to CB (even after prolonged contact in the agar assays) or that changes in the polymeric state of actin are not necessary for cell movement. These possibilities are investigated below. Effect of Cytochalasin B on Prestimulus Actin Content of the Cytoskeleton Fig. 2 shows the effect of exposing whole cells to CB for O-300 set before preparing ‘Ifiton-insoluble cytoskeletons and measurement of resting actin levels. Note that the drug was present throughout both the preincubation and cytoskeleErp Cell Res 160 (1985)

Cytochalasin B, actin and cell motility in Dictyostelium

279

g-yy-, x

0

60

120

160

240

300

120

b 5x165Mtolale

Fig. 2. Effect of CB on pre-stimulus actin levels. Cells were pretreated with either 0, 1% DMSO

(control); or A, IO-’ M CB in 1% DMSO for the indicated times before being lysed with Biton buffer. Changes in cytoskeletal actin content are expressed as percentage difference from an untreated control. (a) 8 h (aggregation-competent) cells. Results are the mean of four separate experiments; (b) 0 h (preaggregative) cells. Results are the mean of two separate experiments. Fig. 3. Chemoattractant-mediated cytoskeletal actin accumulation in cells pretreated with either 0, 1% DMSO (control); or A, 1O-5 M CB in 1% DMSO for 3 min before stimulation. Changes in cytoskeletal actin content are expressed as percentage difference from an untreated prestimulus control. (a) 8 h (aggregation-competent) cells stimulated with 5X lo-* M CAMP. This is a representative example of eight separate experiments; (b) 0 h (preaggregative) cells stimulated with 5~ lop5 M folate. This is a representative example of seven separate experiments.

ton isolation procedure. Exposure for 300-600 set gave similar results (data not shown). In aggregation-competent (8 h). D. discoideum cells (fig. 2 a) there was a decrease in the amount of actin recovered in the Triton-insoluble cytoskeleton following treatment with 10m5M CB. It is worth noting (fig. 2a) that treatment with 1% DMSO alone actually resulted in a small increase in the actin content of the cytoskeleton, confirming reports [35, 361 that DMSO can cause a reorganization of filamentous actin in D. discoideum. In contrast to these effects with aggregation-competent cells, CB or DMSO treatment appeared to have little or no effect on the actin content of cytoskeletons from preaggregative (0 h) cells (fig. 2b). This may reflect some difference in the accessibility of the cytoskeleton to CB during development. CB at 10m6M gave very similar results (zero time values in fig. 4). Although, from these results, CB clearly had an effect on the assembly state of actin in the cytoskeleton, the observation that CB had an almost constant effect on aggregation-competent cells, regardless of pretreatment time (or at least over O-180 set) suggested that it may have been having.its effect on the cytoskeleton only during or after cell lysis and not on the whole cells. ExpCell Res MO (I 985)

200 McRobbie and Newell

5a 00 ‘i

c

5~10-~M

CAMP

20 40 60 80 Time after stimulatb3n(sec)

100

Fig. 4. Chemoattractant-mediated

cytoskeletal actin accumulation in cells pretreated with either 0, 1% DMSO (control), or A, 10e6 M CB in 1% DMSO for 3 min before stimulation. Changes in cytoskeletal actin content are expressed as percentage difference from an untreated prestimulus control. (a) 8 h (aggregation-competent) cells stimulated with 5x 10vg M CAMP. (b) 0 h (preaggregatin) cells stimulated with 5X lo-’ M folate. Each curve is a representative example of two separate experiments. Fig. 5. Chemoattractant-mediated cytoskeletal actin accumulation in cells treated with either 0, 0.5 46 DMSO (control); or A, 5 x 10v6 M CB in 0.5 % DMSO (final concentration) 5 min after lysis of cells in Witon buffer. Changes in cytoskeletai actin content are expressed as percentage difference from an untreated prestimulus control. (a) 8 h (aggregation-competent) cells stimulated with 5x 10-s M CAMP; (b) 0 h (preaggregative) cells stimulated with 5x low5 M folate. Each curve is a representative example from three separate experiments.

CB Efiects on Chemoattractant-Mediated

Actin Changes

Initially the effects of CB on actin accumulation in response to chemotactic stimulation were determined using a preincubation time of 3 min in the presence of CB. It was found that aggregation-competent cells treated in this way with lo-’ M CB underwent the expected decrease in resting cytoskeletal actin levels (fig. 3a), and this concentration (and lower concentrations such as 10m6M, fig. 4 a) completely abolished the second and third peaks of actin accumulation which followed CAMP stimulation and reduced the actin levels during these responses to the very low prestimulus amount. In contrast, no effect was observed on the first (5 set) peak of actin incorporation and indeed this response began from the lower starting levels and reached the same maximum as control cells pretreated with DMSO. When preaggregative amoebae were stimulated with folate after pretreatment with 10m5M or lO-‘j M CB they showed a similar absence of a second and third actin accumulation response, while the 5-set peak was again observed to be normal (fig. 3 b, 4 b respectively). In all cases, treatment with 1% DMSO in the absence of CB appeared to have little or no effect on responses. Erp Cell Rrs 160 (1985)

Cytochalasin B, actin and cell motility in Dictyostelium

281

Table 1. Effect of various treatments with 5~10~~ M CB final post-lysis concentration) on prestimulus actin content of cytoskeletons from aggregationcompetent amoebae Reatment 3 min CB pretreatment (and present during lysis) CB treatment simultaneous with cell lysis CB added 1 min after lysis CB added 3 min after lysis CB added 5 min after lysis

% difference in actin content from control

-58.2 -63.3 -25.3 -10.1 <+0.1

If CB was having these effects on the cytoskeletons during their isolation rather than during the preincubation of the whole amoebae, then one would expect that the simultaneous addition of CB and Triton lysis buffer would give similar results to those just described. In the experiments described above, CB present at 10e5 M before cell lysis was actually at a concentration of 5 x 10m6M at the moment of lysis due to the doubling of the reaction volume during this process. Addition of 5x 10m6M CB (final concentration) simultaneously with Triton lysis buffer was indeed found to give identical results to those shown in fig. 3 (data not shown). As a further control, cells pretreated with CB or DMSO for 3 min were washed free of these compounds before being stimulated. Cytoskeletons isolated from such cells showed no effect of CB on their actin accumulation responses (data not shown). These data further support the suggestion that CB affects the cytoskeleton after cell lysis and has no effect on intact amoebae. CB Effects on Cytoskeletal Actin Accumulation when Added at Various Times after Cell Lysis The effect on prestimulus actin content of adding CB at 0, 1, 3 and 5 min after cell lysis was examined. The results (table 1) reveal that CB has a decreasing effect with time when added after cell lysis and if added at 5 min the effect was negligible. Further investigation of this property involved the examination of chemoattractant-elicited actin changes in cytoskeletons to which CB was added 5 min after cell lysis. The results of these experiments (fig. 5) reveal that 5 x 10m6M (final concentration) CB added after this interval had no effect on the actin accumulated in response to chemotactic signals in contrast to the effects seen when CB was present at the instant of cell lysis. These data suggest that cytoskeletal actin becomes gradually inaccessible or insensitive to CB after cell lysis. Exp Cell Res 160 (1985)

282 McRobbie and Newell Table 2. Binding of 10e6 M [3H]CB to aggregation-competent cytoskeletons

whole cells and

Each value is the mean + SE from three separate experiments Preparation and treatment Whole cells Cytoskeletons, CB present during cell lysis Cytoskeletons, CB added 5 min after cell lysis

Molecules bound per cell or cytoskeleton 0 149 OSOf14 100 96 950+7 300

Binding of Cytochalasin B to Cytoskeletons Employing a simple centrifugation binding assay it was found that there were no detectable [3H]CB-binding sites on aggregation-competent whole cells (table 2). However, using 10m6M [3H]CB it was found that there were a considerable number of CB-binding sites in cytoskeletons prepared from such cells and about one-third of these became inaccessible to label 5 min after cell lysis (table 2). (This decrease may, in fact, be even greater than revealed by these figures, since table 1 shows that cytoskeletons with CB present during lysis have less actin than those with CB added 5 min after lysis.) Preliminary experiments indicate that the decline in site number after cell lysis is a gradual process to some extent resembling the declining sensitivity to CB shown in table 1. These results also tend to confirm that intact amoebae are impermeable to CB and do not possess high affinity binding sites for the drug on their cell surfaces. Polymerisation Contributes to the First Peak of Cytoskeletal Actin Accumulation The observation that only the second and third peaks of actin accumulation were sensitive to CB (figs 3, 4) cast some doubt on the involvement of actin polymerization in the first (3-5 set) peak. An alternative mechanism which could account for the 3-5 set response is an increased cross-linking to the ‘Bitoninsoluble cytoskeleton (exclusively composed of filamentous actin [37]) by short fragments of F-actin which are not normally sedimented at 8000 g which we routinely employ. We tested this possibility by comparing the actin changes in cytoskeletons prepared by 8000 g centrifugation to those pelleted at 100000 g using a Beckman Airfuge for 30 min, a centrifugal force sufficient to remove short F-actin fragments from the ‘kiton-soluble fraction 138, 391. The actin accumulation responses that we observed in cytoskeletons 5 set after a 5 X lo-* M CAMP pulse were very similar whichever of the two g forces was employed. In 8000 g-treated cytoskeletons the increase in actin content was 37.8f 1.6 % (mean f SEM of ten measurements), whilst in 100000 g preparations Exp Cell Res 160(1985)

Cytochalasin B, actin and cell motility in Dictyostelium

283

the increase was 31.4k2.7 % (mean + SEM of nine measurements). This suggested that actin polymerisation was the major contributor to the S-set actin increase, since no increase would be expected in 100000 g cytoskeletons if increased crosslinking of short filaments were the mechanism. This conclusion was further supported by experiments employing the DNase I inhibition assay [31]. In mixtures of monomeric and tilamentous actin, only the monomeric form is measured by its DNase inhibitory activity, and this activity was found to fall by as much as 30% during the first 3-5 set following a chemotactic stimulus whilst no changes in the total actin were observed during this time. This was equivalent to a rise in the F/G actin ratio from approx. 1.28 before stimulation to a peak at 3-5 set of 2.05 (mean of two experiments) coincident with the cytoskeletal actin accumulation seen previously. This result strongly supports the hypothesis that the actin change observed in the first peak after chemotactic stimulation represents a loss of G-actin in favour of cytoskeletal F-actin. DISCUSSION Following our previous reports describing the series of rapid changes in cytoskeletal actin content elicited by chemoattractants [ 19, 201we attempted to relate the actin changes to cellular movement by studying the effects on both phenomena of the actin-filament capping drug CB. We found that CB had no effect on cell movement towards CAMP or folate, or on chemoattractant-induced optical density changes (fig. l), which are thought to be a consequence of shape changes associated with chemotactic movements [30, 33, 341. However, the pronounced effects of CB on basal actin levels (fig. 2 a) and on chemoattractant-elicited actin changes when present during cell lysis (fig. 3) indicated that the drug affected the assembly state of actin that was associated with the cytoskeleton. These results suggested that either actin polymerisation was not involved in Dictyostelium cell movement or that CB was ineffective against intact cells and had its effects only during cytoskeleton isolation. The latter possibility was supported by the finding that 10m6M [3H]CB did not bind to whole cells but did bind to a considerable number of CB-binding sites on cytoskeletons (table 2). Such a result may seem unusual, since CB has been reported to have almost as great an affinity for membrane glucose transport proteins as it has for the cytoskeleton in other systems [40]. However, Dictyostelium has no glucose transport system [411. A low-affinity binding site for CB (& 10m5M) has been reported for whole D. discoideum amoebae [421 but in our studies CB concentration (10T6 M) below the affinity of this site have been used. The Kd of D. discoideum actin for CB has been calculated to be approx. 6x lo-* M [43]. It is interesting that Lin et al. [42] using intact D. discoideum amoebae were unable to detect such high-aflinity sites and this suggests, as do the data presented here, that D. discoideum cells are impermeable to CB. Such a conclusion was further supported by our finding that the drug had no effect on the Exp Cell Res 160 (1985)

284 McRobbie and Newell cytoskeleton if the cells were washed before cell lysis. It is worth noting, however, that CB has been reported to affect development of D. discoideum at very high (25&500 PM) concentrations [44]. The most intriguing finding with CB concerned the drug’s ability to abolish the second and third peaks of actin accumulation at 10V6- 10e5 M but not the first (5 set) peak (figs 3, 4). The effects of higher CB concentrations on the first actin accumulation peak were not investigated because higher concentrations of DMSO would have had to be employed which might have interfered with actin filament assembly [35, 361 and because low5 M CB was greatly in excess of the concentration required for optimal binding to D. discoideum actin (6x IO-* M [431). These different sensitivities of the first actin peak and the other peaks to CB treatment need cautious interpretation, since CB appears to have no effect on whole cells and therefore has its effect only after the actin increases have occurred, i.e., during the period of cytoskeletal isolation. It follows that the different sensitivities to the presence of CB are due to differential stability of the polymerized actin and it seems likely that such differences reflect differences in the polymeric structure of the actin. We propose that the second and third actin peaks may arise by simple ‘barbed-end’ polymerization and consequently be unstable in the presence of CB, but that some other actin assembly mechanism is involved in the first actin peak, rendering this actin in some unknown way resistant to depolymerisation. The finding that the first actin peak is equally detectable in cytoskeletons prepared at 100000 and 8000 g argues strongly against the possibility that the assembly mechanism of the first peak just involves cross-linking of short filamentous fragments without polymerization, as at the higher centrifugal force all such fragments would we pelleted even without crosslinking. From the experiments using the DNase I inhibition assay it was also clear that a massive shift from G to F actin occurs during this first peak. A further interesting finding was the apparent stabilization of the second and third actin peaks after cell lysis to some CB-insensitive condition (fig. 5). This appeared to be a gradual process (as measured by the effects on prestimulus actin content, table 1) and was associated with a decrease in the number of [3H]CBbinding sites on cytoskeletons (table 2). Whether this reflects a genuine decrease in the ability to bind CB or a change in the accessibility of the cytoskeletons remains to be resolved. The significance of the chemotactically induced actin changes that have been observed is still a matter for speculation but they seem likely to be related to amoeba1 movement. It has been noted that only one of the iso-actins (the most acidic Ai form) takes part in the rapid cytoskeletal accumulation [45]. The first (3-5 set) actin peak corresponds in its timing with the observed rapid formation of a pseudopodium by an amoeba that has been stimulated with a pulse of chemoattractant from a microelectrode [46]. This response may be a chemotactic orientating mechanism rather than a mechanism of amoeba1 translocation, as likp Cell Res 160 (1985)

Cytochalasin B, actin and cell motility in Dictyostelium

285

within 15-25 set the amoeba rounds up (or ‘cringes’) [33, 341, and only then elongates and moves in the direction of the initial pseudopodium formation. The cringe response, which has been correlated with the first optical density response [34] may be seen in lawns of synchronously aggregating amoebae under high resolution darktield optics as a narrow very dark band between the dark stationary band and the bright movement band [47]. The contraction and rounding involved in the cringe does not correspond in its timing with any of the peaks of actin accumulation on the cytoskeleton but from numerous experiments it correlates well with the trough in the actin response that follows rapidly after the first (3-5 set) peak, with the bottom of the trough commonly dipping below the prestimulated starting value. Such a correlation is explicable in terms of the solation-contraction hypothesis of amoeboid movement, as a decrease in actin gel structure (such as may be occurring during the trough in the actin curve) favours contraction and rounding up of the cell [48]. We thank Hank Tillinghast, Nick Europe-Finner, Ian Crandall and Hugh Huxley for their critical reading of the manuscript and Frank Caddick for the drawings. We are grateful to Alan Weeds for his helpful suggestions and for the gift of [‘Hlcytochalasin B. The Science and Engineering Research Council and the Medical Research Council provided financial support.

REFERENCES 1. 2. 3. 4.

Huxley, H E, Science 164 (1969) 1356. Pollard, T D & Weihing, R R, CRC crit rev biochem 2 (1974) 1. Kom, E D, Proc natl acad sci US 75 (1978) 588. Carlsson, L, Markey, F, Blikstad, I, Persson, T & Lindberg, U, Proc natl acad sci US 76 (1979) 6376. 5. Phillips, D R, Jennings, L K & Edwards, H H, J cell bio186 (1980) 77. 6. Jennings, L K, Fox, J E B, Edwards, H H & Phillips, D R, J biol them 256 (1981) 6927. 7. Markey, F, Persson, T & Lindberg, U, Cell 23 (1981) 145. 8. Rao, K M K & Varani, J, J immunol 129 (1982) 1605. 9. White, J R, Naccache, P H & Sha’afi, R I, J biol them 258 (1983) 14041. 10. Howard, T H & Meyer, W H, J cell biol 98 (1984) 1265. 11. Tilney, L G, Bonder, E M, Coluccio, L M & Mooseker, M S, J cell biol 97 (1983) 112. 12. Casella, J F, Flanagan, M D & Lin, S, Nature 293 (1981) 302. 13. - Actin: structure and function in muscle and non-muscle cells (ed C G dos Remedios & J A Barden) p. 227. Academic Press, Australia (1983). 14. Fox, J E B & Phillips, D R, Semin hematol20 (1983) 243. 15. Detmers, P A, Goodenough, U W Bt Condeelis, J, J cell bio197 (1983) 522. 16. Ma&ran-Fletcher, S & Pollard, T D, Cell 20 (1980) 329. 17. Mabuchi, I, J biochem 94 (1983) 1349. 18. Brenner, S L & Kom, E D, J biol them 254 (1979) 9982. 19. McRobbie, S J & Newell, P C, Biochem biophys res commun 115 (1983) 351. 20. - J cell sci 68 (1984) 139. 21. Pan, P, Hall, E M & Bonner, J T, Nature new biol237 (1972) 181. 22. Konijn, TM, Van de Meene, J G C, Bonner, J T & Barkley, D S, Proc natl acad sci US 58 (1%7) 1152. 23. Woolley, DE, Arch biochem biophys 150 (1972) 519. 24. Spudich, J A, J biol them 249 (1974) 6013. 25. Uyemura, D G, Brown, S S & Spudich, J A, J biol them 253 (1978) 9088. 26. Simpson, P A & Spudich, J A, Proc natl acad sci US 77 (1980) 4610. Exp Cell Res 160 (1985)

286 McRobbie and Newell 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46.

Sussman, M, Methods cell physiol 2 (1966) 397. Bomer, J T, Kelso, A P & Gillmor, R G, Biol bull 130 (1966) 28. Bonner, J T, Hall, E M, Sachsenmaier, W & Walker, B K, J bacterial 102 (1970) 682. Gerisch, G & Hess, B, Proc natl acad sci US 71 (1974) 2118. Blikstad, I, Markey, F, Carlsson, L, Persson, T & Lindberg, U, Cell 15 (1978) 935. Harris, H E, Bamburg, J R, Bernstein, B W & Weeds, A G, Anal biochem 119 (1982) 102. Futrelle, R P, ‘Baut, J & McKee, W G, J cell bio192 (1982) 807. Futrelle, R P, Pescitelli, M J & McNamara, G T, J cell bio195 (1982) 7a. Fukui, Y, J cell biol76 (1978) 146. Yumura, S & Fukui, Y, J cell biol % (1983) 857. Bray, D & Thomas, C, J mol biol 105 (1976) 527. Stossel, T P & Hartwig, J H, J cell bio168 (1976) 602. Rubin, R W, Warren, R H, Lukeman, D S & Clements, E, J cell bio178 (1978) 28. Rampal, A L, Pinkofsky, H B & Jung, C Y, Biochemistry 19 (1980) 679. Faust, R G & Filosa, M F, J cell camp physiol 54 (1959) 297. Lin, S, Santi, D V & Spudich, J A, J biol them 249 (1974) 2268. Brown, S S & Spudich, J A, J cell biol88 (1981) 487. Cappuccinelli, P & Ashworth, J M, Exp cell res 103 (1976) 387. McRobbie, S J & Newell, P C, FEBS lett 181 (1985) 100. Gerisch, G, Malchow, D, Huesgens, A, Nanjundiah, V, Roos, W, Wick, U & Htilser, D, ICN-UCLA Symp molec cell biol 2 (1975) 76. 47. Ross, F M & Newell, P C, J gen microbial 127 (1981) 339. 48. Hellewell, S B & Taylor, D L, J cell biol 83 (1979) 633. Received May 2, 1985

Exp Cell Res 160 (1985)