J. Cell Sci. 19, 169-182 (:97s) 169

Printed in Great Britain

CIRCUS MOVEMENT IN DISSOCIATED

EMBRYONIC CELLS OF A TELEOST,

ORYZIAS LATIPES

NOBORU FUJINAMI AND TETSUO KAGEYAMA

Laboratory of Developmental Biology,Zoological Institute, Faculty of Science,University of Kyoto, Sakyo-ku, Kyoto 606, Japan

SUMMARY

The dissociated early embryonic cells of the fresh water fish, Oryzias latipes, protrudehyaline lobopodia, which tend to rotate around the cell circumference in a propagatingwave. Cells from late blastula or gastrula continuously show this 'circus movement', whilemost cells up to early blastula are rounded. The linear velocity of the lobopodium wasestimated by means of time-lapse cinemicrography. The velocity increases slightly as celldiameter increases.

The effects of pH, temperature and osmotic pressure of the immersion media on themovement were also quantitatively investigated. Cells become rounded and do not formlobopodial blebs when immersed in media below pH 5. The velocity is reduced by decreasingtemperature, but the movement continues even at 5 °C. Cells placed in hypertonic saltsolutions become crenated and do not continuously demonstrate the circus movement.

INTRODUCTION

The process of circus movement involves blebbing of a hyaline pseudopodium,spreading of the bleb, and thus propagating of the lobopodium around the cellcircumference. This movement has been observed in various cells such as amoebae(Rhumbler, 1898), dissociated gastrodermal cells of planaria (Betchaku, 1967), eggsof a surf-clam (Rebhun, 1963), amoebocytes in the circulating blood of a crab(Loeb, 1928), embryonic teleostean cells in vivo (Trinkaus, 1973; Kageyama, 1975)and in vitro (Sirakami, 1963), dissociated embryonic amphibian cells (Roux, 1894;Holtfreter, 1946; Karasaki, 1957; Sirakami, 1959, 1963; Johnson, 1970; Kauffman,1974), and chick fibroblasts in culture (Dornfeld & Owczarzak, 1958). Thus, thecircus movement may be regarded as common behaviour of many diversified animalcells, based on the fundamental properties of their cell surfaces and motile activities.In addition to the cytological interest as a form of amoeboid movement, this behaviourof metazoan embryonic cells also attracts attention due to its embryological aspects

This paper is dedicated to the memory of our Professor Ken-Iti Sirakami (1913-1974) whohad been interested in the circus movement for many years. He measured the velocity ofcircus movement in various species of amphibians and attempted to associate those valueswith the progressive velocities of the surface cleavage furrow and the overall embryonicdevelopmental sequence. His efforts are beneficial to the understanding of Dettlaff's description(1964) and it will offer us a prospect of seeing embryological processes in terms of activitiesof individual cells sub specie moti.

170 N. Fujinami and T. Kageyama

such as its relationship to the morphogenetic movements of the embryo or to thecell differentiation. The process of circus movement remains poorly understood,due in part to the lack of quantitative description such as those carried out on thebehaviour of fibroblasts in culture (Abercrombie, Heaysman & Pegrum, io,7oa-c).In his cyto-embryological studies of amphibians, Sirakami (1963, 1972) proposeda method for quantitating the circus movement and stated some features of thismovement. These included: the antagonistic relationship that exists between thecircus movement and cytokinesis, the height of the lobopodial wave and the velocityof the movement which remain constant irrespective of cell size, and changespropagating in cell surface which may regulate the pace of the circus movement orthe cleavage and developmental speed of the embryo.

The present study describes the circus movement of dissociated embryonic cellsof a fresh water teleost, Oryzias latipes. These cells were observed by time-lapsecinemicrography in order to demonstrate their intrinsic and extrinsic factors forthe movement. Because of the relative transparency of the teleostean embryos, wecan also make use of in vivo studies on the behaviour of their cells (Trinkaus, 1973;Kageyama, 1975).

MATERIALS AND METHODS

The developing eggs of the orange-red variety of the cyprinodont fish, Oryzias latipes(Japanese Medaka) were used. Details, with respect to the nature of this fish, have beenpreviously reported (Yamamoto, 1967). The fertilized eggs attached in mass to the abdomenof females were collected each morning. The eggs were transferred to dechlorinated tapwater and allowed to develop at room temperature (approximately 25 °C) until the initiationof the experiments. Basically, the embryos were staged according to the chart of Matsui(1949), with some of the modifications of Kageyama (1975). The eggs were washed and thentransferred into Cal+-free Yamamoto solution (0-75 % NaCl, 002 % KC1, and 0002 % NaHCO,at pH 73). In this solution eggs were dissected and blastoderms were mechanically dissociatedinto their constituent cells with watchmaker's forceps. Unless otherwise stated, cells werecollected by centrifugation at 200 g for 10 min, resuspended in rinse solution, collected againand resuspended once more in Ca*+-free Yamamoto solution. Although these proceduresappear harsh, a suspension composed mostly of active individual cells was readily obtainable.Immersion medium (0-5 ml) was added to each depression slide in which 005 ml of cellsuspension had been previously placed. In order to investigate the effects of pH on themovement, cells were immersed in Cat +- and NaHCOj-free Yamamoto solution of variouspH values containing Britton & Robinson's universal buffer mixtures employing 5 min acidmixtures (phosphoric acid, boric acid and acetic acid) adjusted to the required pH by NaOH.In some experiments the solution was buffered by Michaelis' buffer (CH,COOH-CH3COONafor p H 4 , 5; KHiPO1-Nai,HPO4 for pH 6, 7, 8; NH,-NH4C1 for pH 9, 10, 11). In thetemperature experiments the cell suspensions were placed in a glass cylinder which wasfixed to a Petri dish by means of paraffin wax. After cells were allowed to settle for 30 minthe temperature was changed by altering the water temperature within the Petri dish. Thetemperature was frequently checked and maintained at constant values during the study.Dissociated cells were immersed in Ca1+-free solutions of various strengths following themethods described by Yamamoto (1941) in order to determine the osmotic effects on thecircus movement. For stock solution, 1 M NaCl 100 parts+1 M KC1 2 parts was prepared.This solution was diluted with a 0-002 % NaHCO3 solution at pH 7-3.

After 30—90-min incubations in the various types of test solutions maintained at 25 CC, thecells showing circus movement were photographed on 16-mm film by a Nikon CFMA at aninterval between frames of 2 or 4 s. The films were analysed frame by frame with the use of

Circus movement of embryonic cells 171

an analytical projector, NAC Dynamic Frame, by tracing the paths of the lobopodial move-ment of each cell. As proposed by Sirakami (1972) the velocity of the circus movement canbe represented as the linear velocity of displacement of the lobopodial wave front when thewave unilaterally propagates around the cell circumference in a plane relatively parallel tothe substratum. Thus, the method for the estimation of the velocity in the present study(Fig. 1) is essentially the same as that reported in embryonic amphibian cells by Kageyama,Satoh & Sirakami (1975). The difference between these methods comes from the fact thatthe spherical shape of the motionless cell core is more distinct in amphibian cells than inteleostean cells. The velocity of the circus movement of each cell was expressed as a meanvalue obtained from analysing four or more frames of the film.

Fig. 1. The linear velocity of the movement is given as Ijt if the front of the ad-vancing wave moves from Po to P, through the distance, /, in a considerably shortinterval, t.

RESULTS

Cell behaviour at different developmental stages

Cells from each embryo at different stages (morula to mid-gastrula) were dissociatedin a deep depression slide and observed without washing or collecting by centri-fugation. Immediately after cell dissociation, hyaline blebs or lobopodia bulge outrapidly and are inclined to rotate around the cell circumference. Formation of blebs,if any, seems delayed in dividing cells. These changes in cell surface immediatelyafter cell dissociation increase in amplitude and in the number of cells involvedaccording as development progresses. In a few minutes, most individual cellsisolated from morula retract their blebs and become spherical. Synchronous celldivisions are frequently observed among these cells and a lobopodium sometimesspreads near the cleavage furrow on the surface of dividing cells. Blebs of cellsfrom early blastula are also retracted. Some cells protrude blebs again, but theseblebs remain rather stationary or intermittently propagate around the cell circum-ference. Cells from late blastula and gastrula show the surface activity more thancells from early blastula. After cells changing from a knotty or even a vermiform intoa spherical shape (Fig. 10), blebs reappear on cells from late blastula or gastrula andcontinuously show various types of movements as follows: (1) blebbing, (2) jumping,

172 N. Fujinami and T. Kageyama

or hopping of blebs around the cell circumference, (3) the circus movement,which may be regarded as a successive jumping of blebs. A lobopodial wave ofthe circus movement usually propagates with a frequent reversal of its directionof propagation. In some cells, however, a lobopodial bleb was seen to pass uni-laterally (2-48 times) around the entire cell circumference (Fig. 11). The timerequired for a lobopodium to propagate around the cell circumference is fairlyconstant for each revolution (Table 1).

Table 1. The unilateral circus movement of the cells from early gastrula

Cell

Diameter,approx.,

fim Revolutions Direction

Averagetime/revolution*,

ABC

IS20

20

481417

ClockwiseCounter clockwiseClockwise

63 ±37O ±566 ±2

• The time required for a lobopodium to propagate around the cell circumference for onerevolution (expressed as the mean ± standard deviation).

Morula Blastula Gastrula stage

Fig. 2. The percentage of cells exhibiting a propagating bleb at different stages. Eggsof each brood were allowed to develop and dissociated at intervals of 2 h at roomtemperature (23-25 °C). The percentage of cells exhibiting a propagating bleb wasdetermined at 1 h after cell dissociation and plotted for each brood.

As shown in Fig. 2, the percentage of cells exhibiting a propagating bleb increasesduring late blastula and gastrula stage. Under longer observations, more blebscome to propagate around in dissociated cells from late blastula and gastrula, whileblebs usually remain stationary in cells from early blastula. Therefore, the differencebetween early blastula and later stages is underestimated in Fig. 2. Dissociatedcells from early blastula were observed until embryos of the same brood developed

Circus movement of embryonic cells 173

15 -

T, 1 0 -

E

05 -

000

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0 0

0

0

0

CO

0

0

Time, h

Fig. 3. The velocity of the circus movement of several cells from late blastula at25 °C against the indicated hours after cell dissociation.

2 0 r-

1-5

•S 1-0o

0-5

I I I //I I I

15 20 2S

Cell diameter, //m

70

Fig. 4. The velocity of the circus movement of cells at 25 °C is plotted against celldiameter. Some cells ( • ) from early blastula in an alkaline solution show thecircus movement. O» cells from late blastula; %, cells from early gastrula. Theregression line calculated is y = O'O4# + O'io (y, the velocity of the circus move-ment, pirns'"1; x, cell diameter, /4m). A correlation coefficient of 057 (the numberof cells measured = 63) is obtained.

174 N. Fujinami and T. Kageyama

to the late blastula stage. During these in vitro observations, there was no increasein the number of cells showing the circus movement. Generally speaking, cellsremain healthy and active for about 3 h or more, and only a few cells flatten on theglass substratum.

The velocity of the circus movement decreases slightly over the course of 3 h(Fig. 3) with no significant difference in the velocity whether or not cells were washedfree of yolk. However, the velocity slightly increases as the cell diameter increases(Fig. 4). Generally, the circus movement was determined on cells of diameteri5-2O/<m in the experiments described below. It is interesting that some cells frommorula or early blastula show the continuous circus movement in a hypotonic oran alkaline solution. Their velocity of the movement is slightly faster than those ofcells from late blastula or gastrula in a neutral and isotonic solution, though theirmovements do not last as long and seem to be somewhat pathological.

Effects of pH on the circus movement

The motile activities and cell shapes were observed in the range between pH 3-0and i2-o. Cells from late blastula or early gastrula were washed and immersed inthe solution of various pH values. Initially, most cells are lobopodial. Cells in solutionof pH between 6-0 and 9-0 show the active circus movement. There is no significantdifference in the velocity of the movement in this range of pH (Fig. 5). Almost allcells in solution below pH 5-0 become spherical by retracting pseudopodia and beginto lose viability at pH 3-0 or 4-0. With increasing pH, some of these cells bulge outpseudopodia and show circus movement again. Immediately after immersion insolution of about pH 12, large hyaline lobes form and spread around, the cellcircumference. All of these cells become spherical, swell up and then show cytolysis,but at pH io-o some cells remain lobopodial. These results and the schematicrepresentation of pH-dependency of cell shape are shown in Fig. 6. Filopodia, whichcan move but eventually disappear, are also induced by immersing cells in alkalinesolutions. A few cells sometimes show both the fiJopodial movement and the circusmovement.

Effects of temperature on the circus movement

Experiments were done with cells from late blastula or gastrula after washing andtransfer to the Ca^-free Yamamoto solution. Although the pH of the immersionmedia is also changed slightly by temperature, the velocity of the movement is notso affected by this minute change of pH as already shown in Fig. 5. Some cellsfrom the same cell suspension were left at 25 °C and their velocities of movementwere measured as controls. The results are shown in Table 2 and Fig. 7. The velocityof circus movement increases with temperature. After the temperature was shiftedback to 25 °C, recovery of the initial velocity of movement was observed. The changesin the velocity of circus movement for the same cell were also determined by raising(Fig. 8 A) and then lowering (Fig. 8 B) temperature. The velocity of the movementin each cell fluctuates somewhat irregularly. Generally speaking, however, it changesas expected from Fig. 7. Biological zero of the circus movement (i.e. temperature

Circus movement of embryonic cells

1Sr

175

10E

£ 0-5

epH

10

Fig. 5. The velocity of the circus movement of celh from early gastrula at 25 °Cplotted against pH value of medium from 6-10 adjusted with Britton & Robinson'suniversal buffer (O) or Michaelis' buffer ( • ) .

3 4 5 6 7 8 9 10 11 12

cytolyse

Fig. 6. Schematic representation of pH-dependency of cell shape and the percentageof cells exhibiting a propagating bleb. Each plot represents the mean of duplicatecell counts after cells were immersed for 1 h in solutions of various pH valuesadjusted with Britton & Robinson's universal buffer. O, cells from late blastula;9 , cells from early gastrula.

at which the movements are arrested by cold) may be below 5 °C, because cellsshowed the movement at this temperature when analysed at accelerated speed bythe projector.

Effects of osmotic pressure on the circus movement

The osmotic pressure of the interior of Oryzias eggs was described by Yamamoto(1941) to be equivalent to 0-134 M or 1̂ 1/7-5 NaCl solution. Cell suspensions fromlate blastula or early gastrula were washed and transferred to the salt solutions ofvarious concentrations as described in Materials and Methods (M/3, M/5, M/7, M/Q,,M/I I , M/13, M/15 and M/I7 salt solution). Cells placed in M/3 salt solution do not

i 7 6 N. Fujinami and T. Kageyama

Table 2. Effects of temperature on the velocity of circus movement, fun s-x

5 °C 0-1010-03 N* = I I10 °C 0-3710-12 iV = 915 °C o-6i 10-19 iV = 920 °C 0-5310-12 AT = 825 °Cf o-8o ± 0-27 iV = 6230 CC 1-3110-24 N = 1635 °C 1-3810-31 N = 7

(076 ±0-26 N = 20)(078 ±0-31 N = 9)(0-9010-25 N = 7)(0-6910-23 iV = 10)

(0-71 ±0-27 N = 7)(0-97 ±0-29 iV = 10)

The average velocity is expressed as the mean with standard deviation. The velocity of thecontrol cells at 25 °C (see text) is in parentheses.

• N, the number of cells measured,f The average velocity of the control cells of each experiment.

20 1-

1-5 -

>; 10 -

0J

OS -

0

0

0

0

°§8

W

8

0

0

00

000 »

0

Io*

i

00

80

0

n

I

•

0

00

0

0o>t0 *

1 1

1

s

ft

80

80

0

1

•

•

•

8

8P90

0 0

0

1

10 15 20 25

Temperature,°C

30 35

Fig. 7. The velocity of the circus movement of cells from early gastrula plottedagainst temperature. 0 , cells at the indicated temperature; O» cells at 25 °C ascontrol from the same suspension.

show the typical circus movement but shrink, become crenated and sometimes showblebbing or jumping. In highly hypotonic solution such as M/15 and M/17, largehyaline blebs bulge out. Eventually the cells swell up and die. These effects of osmoticpressure on circus movement are reversed by exchanging the immersion fluid. Thevelocity of the movement does not seem to vary with osmotic pressure (Fig. 9).However, in hypotonic solutions, propagating blebs are larger than in hypertonicsolutions.

20

15

ic 10

0-5

Circus movement of embryonic cells

20 r

1-5

177

10

>

0-5

15 20 25 30Temperature, °C

35 15 20 25 30Temperature, °C

35

Fig. 8. The changes in velocity of the circus movement for the same individual cellsobtained by raising (Fig. 8 A) up to 35 °C and then lowering (Fig. 8 B) temperature.

101-

t>

00

0

0

0

00

0

00

0

80

0

1

0

0

0

1

0

00

1

1/5 1/7 1/9 1/11

Concentration, M

1/13

Fig. 9. The velocity of the circus movement of cells from early gastrula at 25 °Cplotted against the concentration of salt solution (see Materials and methods) in thesurrounding medium.

DISCUSSION

When dissociated in an appropriate medium, embryonic cells show unique andremarkable movements. On embryonic teleostean cells in the present study, theirlobopodia tend to rotate around the cell circumference. Other types of cell behaviourare observed rather in rare cases but immediately after cell dissociation, such as

12 C E L 19

178 N. Fujinami and T. Kageyama

peristaltic movement or 'tug' movement described by Sirakami (1963) and Kageyamaet al. (1975).

Because it is a lobopodial activity of cell surfaces, the circus movement is notobserved unless the immersion media are suitable for the formation of pseudopodiaFor instance, the circus movement is reversibly prevented in a solution below pH 5or in a hypertonic salt solution. Harris (1973) also stated that blebbing of fibroblastswas inhibited in a strongly hypertonic medium. These effects of pH and osmoticpressure on circus movement are essentially similar to those described by Holtfreter(1948). A highly hypotonic solution and a strong alkaline solution cause cells toform large hyaline lobopodia, even to cytolyse. In these solutions, some early blastulacells show continuous circus movement for a while, but this is not commonly observedin these cells while maintained in isotonic and neutral solutions. The velocity ofcircus movement does not change much if only a lobopodium is formed (Figs. 5, 9).However, Taylor (1962) reported that elevation of pH accelerated cell movementssuch as pinocytosis and the motion of cellular granules. The environmental conditionssuch as pH or osmotic pressure of the immersion fluid are suspected to affect thedeformability of cell surface (Braatz-Schade, Haberey & Stockem, 1973; Weiss &Clement, 1969). We propose that the deformability of cell surface may be relatedto the mobility of cells in a complicated manner, which is discussed later.

The velocity of the circus movement is reduced by lowering temperature. Theinconstancy of Q10 with temperature or phase separation may be suggested in thepresent study (Table 2), but its relation to the temperature-dependency of thefluidity of membrane (Noonan & Burger, 1973) remains uncertain.

The importance of thiol-disulphide exchange in the circus movement is suggested(Fujinami, 1975). This was previously reported in amoeboid movement in Amoebaproteus, flagellar movement in Pandorina, and gliding movement in Oscillatoria(Abe, 1963), in zeiotic bubbling in murine tumour cells (Belkin & Hardy, 1961) orin cell division of sea-urchin eggs (Sakai & Dan, 1959). The circus movement isreversibly inhibited by cytochalasin B (Fujinami, 1975). These considerations leadus to relate circus movement to normal cytokinesis, which is another kind of cellmovement. It is well known that the most active and irregularly shaped cells alsoassume a spherical shape at the time of cell division. Sirakami (1963) suggested thatcytokinesis is antagonistic to all other forms of movement exercised in the life ofa cell. He proposed that when a cell begins to divide, cytoplasmic motility is lostwith surface movements being effectively suppressed. If cytokinesis is suppressed,the cell may perform other types of activity. Recently, Sanger (1974) suggested thatcellular actin could be recycled to perform different functions (cytoskeletal, amoeboidmovement and furrowing) during the cell cycle. If so, the difference between earlyblastula and later stages in Fig. 2 may be explained by the switching of this basicmotile mechanism which is normally 'on' to cell division in cells of early blastula.

Cell surface changes of embryonic cells of Fundulus during the process of develop-ment have been described: increased deformability (Tickle & Trinkaus, 1973) andincreased adhesivity (Trinkaus, 1963). We have observed that bleb formation,immediately after cell dissociation, appears to be associated with the embryonic

Circus movement of embryonic cells 179

developmental stages. Cell surface changes tend to increase in amplitude and in thenumber of cells involved as development progresses, which is probably due to theincreased deformability of cell surfaces during development. Because the environ-mental conditions in the present study are unsuitable for cells to adhere to othercells or to the glass substratum, we have not observed increased adhesivity, butdissociated cells of Oryzias latipes also show an increase in degree of reaggregationwith respect to their stage of development in an appropriate medium (Yokoya,1966). Johnson (1970) discussed the antagonistic relationship between 'limicolamovements' and cell-substratum adhesions and Epperlein (1974) regarded thecircus movement as a characteristic process of attachment of cells to the glasssubstratum. Therefore, the increased mobility of cells during their early develop-mental stages (Fig. 2) may depend on increased surface activities, which will appearas increased adhesivity if cells are observed in media suitable for adhesion. Theincreased surface activities might be based principally on increased deformability.In this respect, it is of interest that trypsin increases cell deformability (Weiss,1966) and is able to promote the spreading of macrophages (Rabinovitch & DeStefano,1Q-73)-

Holtfreter (1947, 1948) described in amphibian cells that the surface activity wasrestricted or reduced with progressive differentiation. Preliminary experiments inour laboratory show that no blebs form on liver cells from adult fish, when they aredissociated by trypsin while embryonic cells treated in the same manner show circusmovement. Therefore, the cessation of lobopodial formation can be used as a con-venient criterion of the onset of differentiative processes (Wilde, 1961).

The activities of pseudopodia or amoeboid movement during morphogenesis ofembryos have been frequently reported in other species (Gustafson, 1964; Hamano,1964; Wourms, 1972; Nakatsuji, 1974). Circus movement of deep cells in vivo wasreported in the Fundulus blastoderm (Trinkaus, 1973) and in the Oryzias blastoderm(Kageyama, 1975), where their velocity was approximately the same as that observedin our study. The behaviour of the deep cells in vivo corresponds to our findingsof isolated cells, but the role of the circus movement in morphogenesis remains tobe explained.

REFERENCES

ABE, S. (1963). The effect of p-chloromercuribenzoate on amoeboid movement, flagellarmovement and gliding movement. Biol. Bull. mar. biol. Lab., Woods Hole 124, 107-114.

ABERCROMBIE, M., HEAYSMAN, J. E. M. & PEGRUM, S. M. (1970a). The locomotion of fibro-blasts in culture. I. Movements of the leading edge. Expl Cell Res. 59, 393-398.

ABERCROMBIE, M., HEAYSMAN, J. E. M. & PEGRUM, S. M. (19706). The locomotion of fibro-bla8t8 in culture. II. 'Ruffling'. Expl Cell Res. 60, 437-444.

ABERCROMBIE, M., HEAYSMAN, J. E. M. & PEGRUM, S. M. (1970c). The locomotion of fibro-blasts in culture. III . Movements of particles on the dorsal surface of the leading lamella.Expl Cell Res. 62, 389-398.

BELKIN, M. & HARDY, W. G. (1961). Relation between water permeability and integrity ofsulfhydryl groups in malignant and normal cells. J. biophys. biochem. Cytol. 9, 733-745.

BETCHAKU, T. (1967). Isolation of planarian neoblasts and their behaviour in vitro with someaspects of the mechanism of the formation of regeneration blastema. J. exp. Zool. 164,4O7-434-

180 N. Fujinatni and T. Kageyama

BRAATZ-SCHADE, K., HABEREY, M. & STOCKEM, W. (1973). Correlation between membranepotential, cell shape, and motile activity in Amoeba proteus. Expl Cell Res. 80, 456—458.

DETTLAFF, T. A. (1964). Cell divisions, duration of interkinetic states and differentiation inearly stages of embryonic development. In Advances in Morphogenesis, vol. 3 (ed. M. Aber-crombie & J. Brachet), pp. 323-362. New York and London: Academic Press.

DORNFELD, E. J. & OWCZARZAK, A. (1958). Surface responses in cultured fibroblasts elicitedby ethylenediamine-tetraacetic acid. J. biophys. hiochem. Cytol. 4, 243-259.

EPPERLEIN, H. H. (1974). The ectomesenchymal-endodermal interaction-system (EEIS) ofTriturus alpestris in tissue culture. I. Observations on attachment, migration and differentiationof neural crest cells. Differentiation 2, 151-168.

FUJINAMI, N. (1975). Studies on the mechanism of circus movement in dissociated embryoniccells of a teleost, Oryzias latipes (in manuscript).

GUSTAFSON, T. (1964). The role and activities of pseudopodia during morphogenesis of thesea urchin larva. In Primitive Motile Systems in Cell Biology (ed. R. D. Allen & N. Kamiya),PP- 333~3S°- New York and London: Academic Press.

HAMANO, S. (1964). A time-lapse cinematographic study on gastrulation in the zebra fish.Ada Embryol. Morpli. exp. 7, 42-48.

HARRIS, A. K. (1973). Cell surface movements related to cell locomotion. In Locomotion ofTissue Cells, Ciba Fdn Symp. 14, pp. 3—20. Amsterdam, London, New York: Associationof Scientific Publishers.

HOLTFRETER, J. (1946). Structure, mobility and locomotion in isolated embryonic amphibiancells. J. Morph. 79, 27-62.

HOLTFRETER, J. (1947). Observations on the migration, aggregation and phagocytosis ofembryonic cells. J . Morph. 80, 25-55.

HOLTFRETER, J. (1948). Significance of the cell membrane in embryonic processes. Ann. N.Y.Acad. Sci. 49, 709-760.

JOHNSON, K. E. (1970). The role of changes in cell contact behaviour in amphibian gastrulation.J. exp. Zool. 175, 391-428.

KAGEYAMA, T . (1975). Motility and locomotion of embryonic cells of the medaka, Oryziaslatipes, during early development (in manuscript).

KAGEYAMA, T., SATOH, N. & SIRAKAMI, K I. (1975). Shape and motility of dissociatedembryonic cells in Xenopus laevis (in manuscript).

KARASAKI, S. (1957). On the mechanism of the dorsalization in the ectoderm of Triturusgastrulae caused by precytolytic treatments. I. Cytological and morphogenetic effects ofvarious injurious agents. Embryologia 3, 317-334.

KAUFFMAN, S. (1974). Phase-dependent lobopodial contact paralysis. Expl Cell Res. 86,217-221.

LOEB, L. (1928). Amoebocyte tissue and amoeboid movement. Protoplasma 4, 596-625.MATSUI, K. (1949). Illustration of the normal course of development in the fish, Oryzias

latipes (in Japanese). Zikkenkeitaigaku (Experimental Morphology) 5, 33-42.NAKATSUJI, N. (1974). Studies on the gastrulation of amphibian embryos. Pseudopodia in

the gastrula of Bufo bufo japonicus and their significance to the gastrulation. J. Embryol.exp. Morph. 32, 795-804.

NOONAN, K. D. & BURGER, M. M. (1973). The relationship of concanavalin A binding tolectin-initiated cell agglutination. J. Cell Biol. 59, 134-142.

RABINOVITCH, M. & DESTEFANO, M. J. (1973). Macrophage spreading in vitro. I. Inducersof spreading. Expl Cell Res. 77, 323~334-

REBHUN, L. I. (1963). Induced amoeboid movement in eggs of the surf-clam Spisula solidissima.Expl Cell Res. 29, 593-602.

RHUMBLER, L. (1898). Physikalishe Analyse von Lebensercheinungen der Zelle. I. Bewegung,Nahrungsaufnahme, Defakation, Vacuolen-Pulsation und Gehausebau bei lobosenRhizopoden. Arch. EntwMech. Org. 7, 103-350.

Roux, W. (1894). Ober den ' Cytotropismus' der Furchungszellen des Grasfrosches {Ranafusca). Arch. EntivMech. Org. 1, 43-68.

SAKAI, H. & DAN, K. (1959). Studies on sulfhydryl groups during cell division of sea urchinegg. Expl Cell Res. 16, 24-41.

SANGER, J. W. (1974). Changing patterns of actin localization during cell division. J. CellBiol. 63, 297 a.

Circus movement of embryonic cells 181

SIRAKAMI, K . I . (1959). Cyto-embryological studies of amphibians. II . On 'tug-cells' andtheir bearings upon the gastrulation and other morphogenetic movements. Mem. Fac.Lib. Arts & Ed. Yamanashi Univ. 10, 125-129.

SIRAKAMI, K. I. (1963). Cyto-embryological studies of amphibians. IV. Behaviours of isolatedectodermal cells from blastula stage embryos of Bufo vulgaris. Mem. Fac. Lib. Arts & Ed.Yamanashi Univ. 10, 132-140.

SIRAKAMI, K. I. (1972). Circus movement of dissociated embryonic cells of amphibians (inJapanese). Kagaku to Seibutsu {Chemistry and Biology) 10, 547-549.

TAYLOR, A. C. (1962). Responses of cells to pH changes in the medium. J. Cell Biol. 15,201-209.

TICKLE, C. A. & TRINKAUS, J. P. (1973). Change in surface extensibility of Fundulus deepcells during early development. J. Cell Sci. 13, 721-726.

TRINKAUS, J. P. (1963). The cellular basis of Fundulus epiboly. Adhesivity of blastula andgastrula cells in culture. Devi Biol. 7, 513—532.

TRINKAUS, J. P. (1973). Surface activity and locomotion of Fundulus deep cells during blastulaand gastrula stages. Devi Biol. 30, 68-103.

WEISS, L. (1966). Studies on cell deformability. II. Effects of some proteolytic enzymes.J. Cell Biol. 30, 39-43.

WEISS, L. & CLEMENT, K. (1969). Studies on cell deformability. Some rheological considerations.Expl Cell Res. 58, 379-387-

WILDE, C. E., JR. (1961). The differentiation of vertebrate pigment cells. In Advances inMorphogenesis, vol. 1 (ed. M. Abercrombie & J. Brachet), pp. 267-300. New York andLondon: Academic Press.

WOURMS, J. P. (1972). The developmental biology of annual fishes. II. Naturally occurringdispersion and reaggregation of blastomeres during the development of annual fish eggs.J. exp. Zool. 182, 169-200.

YAMAMOTO, T. (1941). The osmotic properties of the egg of fresh-water fish, Oryzias latipes.J. Fac. Sci., Tokyo Imp. Univ. Section IV, 5, 461-472.

YAMAMOTO, T. (1967). Medaka. In Methods in Developmental Biology (ed. F. H. Wilt &N. K. Wessells), pp. i o i - m . New York: Thomas Crowell.

YOKOYA, S. (1966). Cell dissociation and reaggregation in early stage embryo of a teleost,Oryzias latipes. Scient. Rep. Tohoku Univ. Ser. IV (Biol.), 32, 229-236.

(Received 10 March 1975)

182 N. Fujinami and T. Kageyama

Fig. 10. The surface change of cells from late blastula (stage 12) immediately aftercell dissociation and 1 and 4 min later, respectively, observed in the same field. Theknotty-shaped or even vermiform cells become spherical owing to retracting blebs.

o .0 o. o oo o o 0 0

Fig. 11. A cell from early gastrula (stage 13) shows the stable circus movement inCa1+-free Yamamoto solution at 25 °C (sequence of frames from a 16-mm cin6 filmexposed at intervals of 4 s). This unilaterally propagating lobopodium continuouslypassed around the entire cell circumference clockwise for 48 revolutions. The shapeof the motionless cell core is deformed by blebbing and spreading of a lobopodialbleb if the lobopodium is large enough as compared with the cell. The scale markrepresents 10 /tm.