/ . Embryol. exp. Morph. Vol. 36, 1, pp. 209-223, 1976
Printed in Great Britain
209
Incorporation of tritiated
cytochalasin B into ecto dermal explants of
the amphibian gastrula
By M. M. PERRY, 1 G. G. SELMAN 2 AND J. JACOB 2
From the Institute of Animal Genetics, University of Edinburgh
SUMMARY
Ectodermal explants from gastrulae of Triturus alpestris were exposed to tritiated cytochalasin B (5 and 10/tg/ml) for -£-2^ h. The flat explants failed to heal into compact spheres
and the component cells gradually rounded up. Radioactivity in the fixed and sectioned
material was analysed by light and electron microscopic autoradiography. Silver grains were
found predominantly over the yolk platelets, but were also scattered over other cellular
components. Areas containing pigment granules or the feltwork material, present in the
apices of dissociating cells of the superficial layer, exhibited higher activity than areas containing mitochondria or cytoplasmic matrix. Lipid droplets and nuclei showed comparatively
little activity. The results are discussed in relation to previous findings on the uptake of
tritiated cytochalasin D by cell fractions.
INTRODUCTION
The cytochalasins, a group of related fungal metabolites, inhibit diverse cell
activities ranging from motility to transport mechanisms. The literature on the
subject has recently been summarized by Miranda, Godman, Deitch & Tanenbaum (1974a). Cultured cells show a rapid response to the drug by undergoing
changes in shape within minutes of the application and removal of cytochalasin
B (CB) (Everhart & Rubin, 1974) or the related but more potent drug cytochalasin D (CD) (Miranda et al. 1914a). The time course of the morphological
response of embryonic tissues is however considerably longer, taking up to 2 h
in the case of salivary gland epithelium (Spooner & Wessells, 1972) and chick
neural plate (Karfunkel, 1972). This difference in the rate of response raises the
question of the accessibility to the drug of cells that are closely compacted in a
mass or sheet.
Components of the cell periphery, namely the plasma membrane and cortical
systems of microfilaments, possibly representing the contractile apparatus of the
cell, have been implicated as sites of action of the drugs (Everhart & Rubin,
1974; Miranda, Godman & Tanenbaum, 1914b). Evidence in support of these
1
2
14
ARC Poultry Research Centre, West Mains Road, Edinburgh EH9 3JS, U.K.
Institute of Animal Genetics, West Mains Road, Edinburgh EH9 3JN, U.K.
EMB 36
210
M. M. PERRY, G. G. SELMAN AND J. JACOB
proposals has been presented by Tannenbaum et al. (1975) who demonstrated
that, in fractions of cultured mammalian cells exposed to tritiated CD, the
highest binding activity was exhibited by the plasma membrane fraction. These
authors have also provided indirect proof, from an autoradiographic analysis of
whole cells, that CD is incorporated into the cell in vivo and is not merely bound
to the cell surface.
An accompanying report deals with the permeability of the amphibian egg at
first cleavage to short exposures of tritiated CB (Selman, Jacob & Perry, 1976).
The present work is concerned with the localization of CB in the ectoderm of the
amphibian gastrula after longer periods of treatment. Uptake of the tritiated
compound was analysed by means of light and electron microscopic autoradiography. It has been shown that, in ectodermal explants, treatment with
CB inhibits the healing process and eventually leads to cellular dissociation
together with a disorganization of the microfilamentous bands normally present
at the external cell junctions (Perry, 1975).
MATERIALS AND METHODS
1. Treatment with cytochalasin B
Tritiated cytochalasin B [3H]CB, specific activity 1-0 Ci/mM, was prepared
and purified by a modification of the method of Lin, Santi & Spudich (1974) as
described in an accompanying paper (Selman et al. 1976). A stock solution
containing [3H]CB in dimethyl sulphoxide (1 mg/ml) was diluted with Holtfreter's solution (0-06 M-NaCl, 0-006 M-KCI, 0-001 M-MgCl2, 0-001 M-CaCl2, in
0-001 M-phosphate buffer, pH 7-2) to give a final concentration of 5 or 10 /*g/ml
[3H]CB. Similar solutions of unlabelled CB (Aldrich Chemical Co.) were also
prepared.
Pieces of bilayered ectoderm about 1 mm2 were removed from the animal
pole region of early gastrulae (crescent-shaped blastopore lip) of Triturus
alpestris and cultured in CB as follows:
Expt. I. [3H]CB (5 /*g/ml) for \ h, rinsed twice in glutaraldehyde fixative for
3 min.
Expt. II. [3H]CB (5 /tg/ml) for 2\ h, rinsed twice in CB (5 /*g/ml) for 3 min.
Expt. III. [3H]CB (5 /ttg/ml) for 2\ h, rinsed twice in glutaraldehyde fixative for
3 min.
Expt. IV. [3H]CB (10/tg/ml) for 2 h, rinsed twice in CB (10/ig/ml) for
3 min.
The explants were then fixed in a solution of 2-5 % glutaraldehyde in 0-05M-Na
cacodylate buffer, pH 7-2, for 1 h at room temperature, rinsed rapidly in the
buffer and post-fixed in 1 % OsO4, in the same buffer containing 3 % sucrose,
for 2 h at 4 °C. After dehydration in a graded series of alcohols the material was
transferred through epoxy propane and embedded in an Epon-Araldite mixture.
Tritiated cytochalasin B in the amphibian ectoderm
211
The negligible amount of radioactivity in the processing fluids measured in a
parallel experiment (Selman et al. 1976) indicates that after fixation there is no
loss of the tritiated compound from early embryonic amphibian material during
the dehydration procedure, except in epoxy propane which gave a count slightly
above background level.
Some explants were treated with unlabelled CB (5 and 10 /*g/ml) and observed
with a dissecting microscope in order to compare the effects of the two samples
of CB. Others were cultured in Holtfreter's solution, or in Holtfreter's solution
containing 1 % dimethyl sulphoxide as controls.
2. Autoradiography
For light microscopy, \-/im sections of blocks from Expts. I-IV were placed
on slides and coated with Uford L4 emulsion by dipping. After 10 weeks'
exposuret he autoradiographs were developed in Kodak D19b developer and
examined by phase contrast microscopy. Some preparations were stained with
3 % basic fuchsin in 70 % alcohol for 2-3 min, prior to making permanent
mounts.
For electron microscopy material was prepared from Expts. II-IV. Thin sections with pale gold interference colours were mounted on gilded nickel grids,
previously coated with Formvar and carbon, and stained with alcoholic uranyl
acetate and lead citrate. The sections were then coated with another layer of carbon to prevent the possibility of chemographic effects and to avoid loss of stain
during photographic processing. Monolayers of Ilford L4 emulsion (deep purple
interference colour) were pregelled and applied to the preparations by the loop
method of Caro and van Tubergen (see details in Jacob, 1971), exposed for 16
weeks and developed in Kodak D19b developer.
Preliminary studies (Selman et al. 1976) employing tritiated CB of low specific
activity, which showed no silver grains in light microscope autoradiographic
preparation of early embryonic amphibian material after 9 months' exposure,
were taken as controls for the present experiments.
3. Analysis of the autoradiographs
Estimates of the grain density in the light microscope preparations were
obtained by counting the number of silver grains in graticule squares of 5-3 /tm2
for yolk platelets and 21-2 jum2 for nuclei and yolk-free cytoplasm. Three such
areas were counted in each of three cells sampled from 18 sections in each
experimental group. The grain density per unit area of 100/*m2 was calculated
from the total figures and is given in Table 1.
In electron microscope preparations, areas containing mostly yolk-free cytoplasm were examined from sections on 12 grids in experimental groups II-IV.
A total area of about 22000 ^m 2 was analysed from the three groups. The
analysis was based on the 'probability circle' method (Williams, 1969; Kent &
Williams, 1974) which enables a quantitative estimate to be made of the labelling
14-2
212
M. M. PERRY, G. G. SELMAN AND J. JACOB
patterns in preparations where silver grains are distributed over various cellular
organelles.
Tabulation of the data involved three steps. Firstly, the various components
in the micrograph were divided into a number of items that were generally
recognizable as separate functional entities in the cell. Some of the items were
larger than the size of the circle used in the analysis, some were smaller, and in
some instances the circles overlapped profiles of two or more items (see Table
2). Secondly, an estimate of the relative area occupied by each item (relative
effective area) was obtained by superimposing a regular array of circles on each
micrograph and allocating each circle to an item or items. The radius of the
circle used was equal to the value of autoradiographic resolution in the preparations analysed. In a preparation with a given set of parameters, the resolution
obtainable depends on whether radiation emanates from a point or a linear
source. In our preparations the half-distance (HD) value, a direct measurement
of resolution for a linear source of radioactivity, can be estimated (Salpeter,
Bachmann & Salpeter, 1969) to be 150 nm. The corresponding value (HR) for
a point source can be calculated (Salpeter et al. 1969; Salpeter & McHenry,
1973) to be 255 nm, from the formula HR = 1-7 x HD. The main analysis in the
present study was done with circles of radius 255 nm, i.e. circles with radii of
3-75 mm or 5 mm respectively for examining prints enlarged to final magnifications of 14500 (Expt. IV) or 20000 (Expts. II and III). The electron microscope
was calibrated using a diffraction-grating replica. The circles of appropriate
size were printed on to a transparent screen at a density of one circle per 30 mm2
for enlargements of 14500 and one circle per 50 mm2 for enlargements of
20000.
Data from one set of experiments (Expt. Ill) were also analysed using a circle
size equal to the resolution obtainable for linear sources (150 nm) rather than
point sources. The circle was consequently smaller, with a radius of 3 mm, for
examining micrographs at a final magnification of 20000.
Thirdly, data on the distribution of silver grains were collected by superimposing a circle on the appropriate size on each grain, such that the centre of
the circle coincided with the mid-point of the longest axis of the grain. The grain
was then allocated to an item or items contained within the circle.
An estimate of the relative radioactivity of each item or groups of items could
then be obtained by referring grain counts to relative effective area. This was
expressed as relative specific activity and denoted as a ratio (see Table 2). There
was a high correlation between these values in all the three experimental groups
and the statistical means did not differ significantly; the data were therefore
pooled (Table 2). Furthermore, items were grouped together where the number of
grains or circles was small, and/or the items were of similar relative specific
activity and biological significance. It was found that the values for the relative
specific activity of various items comprising the yolk-free cytoplasm were
similar; the data for these items were therefore subjected to further statistical
Tritiated cytochalasin B in the amphibian ectoderm
213
analysis. The chi-squared test was applied, to determine the extent to which the
frequency distribution of silver grains on these items diverged from random
(Table 3). The expected distribution, if random, was calculated by multiplying the number of circles for each item by the ratio of total grains/total circles.
RESULTS
1. General morphology of the ectodermal explants
At the time of explantation the tissue pieces were slightly curved, with the
pigmented, superficial ectodermal cell layer forming their convex surfaces and
the unpigmented inner ectodermal cell layer forming their concave surfaces.
In the control solution the explants rapidly curled up, with the pigmented surface
outermost, and in 2 h healed into compact, smooth-surfaced spheres. In CB
(5 /tg/ml) the explants flattened in 10 min, then, commencing at the periphery,
the cells began to round up and lose their intimate contact with each other. This
process continued for 2 h until all the cells were spherical and adhered solely
at points around their peripheries. On transfer to Holtfreter's solution the explants formed into smooth-surfaced spheres after a period of 6 h. A similar,
but more rapid response was elicited with the higher concentration of CB
(10 ^g/ml), and recovery was slower. In comparison with the unlabelled compound, the preparation of [3H]CB was less effective in bringing about cellular
disaggregation, although the initial effect of inhibition of explant healing was
similar for both samples of the drug. In [3H]CB (5 ^g/ml) only the cells of the
inner layer and the peripheral cells of the superficial layer had rounded up after
2h, whereas the cells throughout the superficial layer were affected after a
similar period in [3H]CB (10/tg/ml).
2. Light microscope autoradiographs
At the lowest dose (5 ^g/ml) for \ h, the superficial cells at the periphery of
the explant were spherical, or pear-shaped with tapering apical regions, whereas
the centrally placed superficial cells remained cuboidal. In contrast to the situation in the intact gastrula ectoderm, intercellular spaces of considerable dimensions were visible in both layers except along the flat, pigmented surface of the
superficial layer. The depth of the explants was 100-150 /im; the component
cells were 40-60 fim in width. Silver grains were present over all the cells in the
sections regardless of the shape or position of the cells in the explant. Explants
treated with the same dose for a longer period presented a similar appearance,
although the silver grain density was three times higher (cf. Expts. I and III,
Table 1). At the higher drug concentration a dispersal of the pigment granules
from the external surface, together with a rounding up of the cells, led to a loss
of identity of the superficial layer. The grain density was not however proportionally greater (cf. Expts. II and IV, Table 1). This result may mean that the CB
binding sites are saturated at the lower dose, but that a faster rate of saturation
214
M. M. PERRY, G. G. SELMAN AND J. JACOB
Table 1. Grain density in light microscope sections o/Triturus gastrula
ectoderm cells labelled with [3H]cytochalasin B. Counts/100 ju,m2
Expt. no.
I
II
III
IV
Treatment
Nucleus
Yolk-free
cytoplasm
Yolk
5 /*g/ml for ^ h
5 /ig/ml for 2± h
rinsed 'cold' CB
5 /tg/ml for 2\ h
10/tg/mlfor 2h
rinsed 'cold' CB
5
12
16
47
51
133
17
12
51
52
161
150
N
Fig. 1. Light microscope (phase contrast) autoradiograph of a section through an
ectodermal cell, after 2 h in [3H]CB (10 /*g/ml). Many grains are present over the yolk
platelets, few occur over the nucleus (N). x 1100.
Fig 2. A cell, treated as in Fig. 1, exhibiting clusters of grains (arrow) in an area of
yolk-free cytoplasm near the nucleus (TV). Section stained with basic fuchsin. x 1100.
in the initial period of treatment is required to bring about total disaggregation
of the superficial cell layer.
Preparations rinsed in unlabelled CB for a short period before fixation, in
order to avoid the artifactual penetration of residual [3H]CB in the fixative,
gave a slightly lower count than unrinsed preparations (cf. Expts. II and III,
Table 1). The relative grain density in peripheral (cytoplasmic) and central
(nuclear) areas of the cells did not differ appreciably between the two groups,
suggesting that there was little exchange of CB across the plasma membrane
Tritiated cytochalasin B in the amphibian ectoderm
215
€
Fig. 3. An electron microscope autoradiograph to illustrate the ultrastructural
features characteristic of ectodermal cells. Glycogen granules, which usually occupy
the spaces observed in the cytoplasm, have been extracted during processing.
[3H]CB (5 /ig/ml) for 2\ h. Yolk platelet (Y), pigment granule (P), lipid droplet (L),
myelin body (M). x 20000.
216
M. M. PERRY, G. G. SELMAN AND J. JACOB
Fig. 4. To show the typical appearance of groundplasm in a non-extracted cell. Areas
containing smooth-surfaced vesicles (V), masses of dense glycogen granules (G) and
small patches of less dense ribosomes are placed in the 'cytoplasm' category listed
in Table 2. x 20000.
Fig. 5. Feltworks of fibro-granular material appearing in the apices of the superficial ectodermal cells at the periphery of the explant after 2\ h in [3H]CB (5 /tg/ml).
Here the cells are still connected by their apical junctions, which follow a winding
course between adjacent cells, x 20000.
Tritiated cytochalasin B in the amphibian ectoderm
217
within a 3-min period. Evidence of extraction of radioactivity during the embedding process was indicated by the presence of a few grains on the Araldite
in the proximity of the cells; otherwise the background grain count was
negligible.
The untreated cells possessed multilobed nuclei and their cytoplasm consisted
entirely of yolk platelets interspersed with lipid droplets and some pigment
granules. In the superficial cells the pigment granules were concentrated in a
narrow strip of cortical cytoplasm at the external surface. Cells most severely
affected by CB contained fairly large areas of yolk-free cytoplasm, often located
in the vicinity of the nucleus or, in the columnar superficial cells, in the apical
regions. Groups of pigment granules were frequently seen in these yolk-free
areas. The yolk platelets were clearly well labelled (Fig. 1) and the rest of the
cytoplasm seemed to be fairly well labelled (Fig. 2). A comparison of the grain
density in the three major cellular compartments showed that the activity of the
yolk, yolk-free cytoplasm and nuclei differed by a factor of three respectively
(Table 1). Few grains were present on the lipid droplets, or on the chromosomes.
3. Electron microscope autoradiographs
The ultrastructural features of the treated cells, depicted in Fig. 3, generally
corresponded with those of normal cells, They consisted of mitochondria,
composite pigment granules, lipid droplets and yolk platelets embedded in a
heterogeneous groundplasm. The components of the groundplasm, grouped into
the 'cytoplasm' category for the purposes of this analysis, were smooth-surfaced
vesicles, traces of rough endoplasmic reticulum, small clusters of ribosomes
and masses of glycogen granules (Fig. 4). The glycogen granules, distinguished
from the ribosomes by their greater size and electron opacity, were absent in
some cells (for example see Fig. 3) presumably because they had been extracted
during the processing procedures. The few myelin bodies probably represent
sections through the edges of yolk platelets that are undergoing utilization, as
demonstrated by the presence of concentric layers of membranes at the platelet
margin in several instances.
The main difference between normal and treated cells concerned the structure of the apical cell junctions in the superficial layer. In the intact ectoderm
these have been shown to consist of occluding junctions, bounded by 8 nm
microfilaments, which are aligned in a ring in the cortical cytoplasm (Perry,
1975). A thin meshwork of fibro-granular material, which is also found in the
cortical layer of the dividing zygote (Selman & Perry, 1970), can be discerned
beneath the external apical surfaces in some cells. In regions of the CB-treated
explants where the apical junctions were in the process of rupturing, felt-like
masses of compact fibro-granular material were observed in the cell apices
(Fig. 5). Clumps of similar material were also present in small surface protuberances of the more completely dissociated cells. Bundles of parallel microfilaments known to occur in association with these cytoplasmic masses (Perry,
218
M. M. PERRY, G. G. SELMAN AND J. JACOB
Table 2. Grain and circle analysis of EM autoradiographs obtained 2-2\ h
after labelling with [zH]cytochalasin B, using circles of 255 nm radius*"
Item
no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Item
Yolk
Yolk/cytoplasm
Myelin bodies/cytoplasm
Yolk/cytoplasm/pigment
Yol k/cy toplasm/mitochondria
Yolk/cytoplasm/lipid
Pigment
Pigment/cytoplasm
Pigment/cytoplasm/extracellularf
Pigment/cytoplasm/mitochondria
Pigment/cytoplasm/lipid
Feltwork
Feltwork/cytoplasm
Feltwork/extracellularf
Mitochondria
Mitochondria/cytoplasm
Mitochondria/cytoplasm/nucleus
M itochondria/cytoplasm/lipid
Cytoplasm
Cytoplasm/Golgi
Golgi
Cytoplasm/extracellularf
Cytoplasm/lipid
Cytoplasm/chromosomes
Cytoplasm/chromosomes
Nucleus
Chromosomes
Lipid
Extracellular
Total
No. silver
grains
376 376
424\
Relative
specific
activity
No. random % grains
% circles
circles
315 315
567\
21
8
14) 493
23 648
21
13;
17
9
\
470
19) 537
32
V
91
91
771
67/
1
319
2
11
1109]
6M121
6J
159 159
741
24 [ 101
3J
3
J | 38
16
32
5-5
16
32
3441
33;
46\
1655
70)1857
64
22;
320 320
239 \
294/ 5 3 3
8'
1544
13
54
7256]
23[7286
l)
1034 1034
528]
195 [ 742
19J
391
3 61427
/4
224 224
997 997
16002
3-5
1-3
1-3
1.1
X J
A.Q
U7
0-7
0-7
0-6
0-4
0-3
01
* Combined data from Expts. II-IV.
f These categories include intercellular regions.
1975) were seldom observed in the present preparations. The feltwork of flbrogranular cytoplasm referred to above was placed in a separate category, because
such CB-evoked masses are of common occurrence and could be regarded as
potentially specific binding sites.
An indication of the radioactivity of the cellular components is given in Table
2, in which items consisting of one or more components are listed in order of
Tritiated cytochalasin B in the amphibian ectoderm
219
Table 3. Observed and random silver grain distribution on grouped items in
the yolk-free cytoplasm 2-2\ h after labelling with [3H]cytochalasin B
Item
no.
7-11
12
13-14
15-18
19-21
22-25
Item
Pigment/
associated areas
Feltwork
Feltwork/
associated areas
Mitochondria/
associated areas
Cytoplasm/Golgi
Cytoplasm/
associated areas
Total
Expected no. (Observed —
Actual no.
silver grains expected)2/
silver grains No. circles if random
expected
537
1857
345
106-8
91
144
320
533
59
99
17-3
20-4
333
1619
301
3-4
1121
260
7286
1776
1352
330
39-5
14-8
2486
13391
2486
201-9
2
X = 201-9. Probability < 0001 for 4 degrees of freedom.
decreasing relative specific activity. The somewhat higher relative activity of the
yolk, as compared with the data from the light microscope autoradiographs, is
probably due to the greater accuracy afforded by the electron microscope in
counting grains over areas of high grain density. Otherwise the figures from
the two kinds of analysis were broadly compatible. The yolk platelets had high
values, the lipid droplets and nuclei had low values, and the remaining items,
comprising the majority, had intermediate values, and were of broadly similar
magnitude. No difference was noted between the' cytoplasm' at the cell periphery,
to a depth of 0-25 /on (Item 22) and the 'cytoplasm' at a deeper level (Items
19-21). Neither was there any difference in the activity between the 'cytoplasm'
of glycogen-extracted cells and non-extracted cells in samples of the micrographs
covering areas of 800 /tm2 and 900 /on2 respectively. On applying the chi-square
test to the data for those items in the yolk-free cytoplasm with closely similar
values (Items 7-25), the distribution of grains amongst them was found to be
highly non-random, with a probability of less than 1:1000 of the distribution
being uniform (Table 3). There was an excess of grains over areas enclosing
pigment, feltwork and mitochondria, and a deficiency over areas enclosing the
'cytoplasm'.
These results were corroborated by the data from the separate analysis of
Expt. Ill, using a smaller circle size as described earlier. The relative specific
activities of the items did not differ significantly from those obtained using a
50 % probability circle size (for a point source), and showed a high correlation
with the data in Table 2. This finding indicates that, for analyses of this type,
the actual value of autoradiographic resolution may not be a very critical
factor.
220
M. M. PERRY, G. G. SELMAN AND J. JACOB
In summary, the results show that, when administered at high doses for
prolonged periods, CB is bound to many components of amphibian ectodermal
cells, in particular to the yolk platelets and, to a lesser extent, to areas associated
with pigment granules and CB-evoked feltworks of fibro-granular material.
DISCUSSION
The main points to arise from the autoradiographic study are firstly that CB
is incorporated in vivo into most of the cellular constituents, and secondly that
CB permeates throughout the embryonic tissue fragments within a period of
thirty minutes, indicating that a differential penetration of the drug cannot
account for the slow rate of disintegration of the central regions of the explants.
The limitation of the autoradiographic analysis is such that for grouped
items, such as pigment and its associated items, the source of radiation may
either lie inside or in the immediate vicinity of the main structure listed. Within
this limitation of the method, the values obtained for the relative activities of
the various cellular components are in general agreement with those of Tannenbaum et al. (1975) for the binding of CD to cellular fractions in vitro. They
recorded the highest activity in the plasma membrane fraction, moderate
activity in the microsomal (endomembrane) fraction and little binding in the
mitochondrial and nuclear fractions. In these amphibian ectodermal cells the
' cytoplasmic' component, which is broadly comparable with the microsomal
fraction becauuse it contains considerable quantities of smooth-membraned
vesicles, and the mitochondrial component, exhibited twice the activity of the
nucleus. The cell periphery, in regions containing the feltwork of fibro-granular
material, was roughly twice as active as the 'cytoplasmic' component, but elsewhere in the periphery the value was similar to that of the deeper cytoplasm,
which is contrary to the finding of Tannenbaum et al. (1975) on the plasma
membrane fraction. This discrepancy may be due to the different distribution
of cortical microfilamentous material in the two kinds of cell. In cultured
mammalian cell lines, microfilaments in the form of meshworks, or in parallel
arrays, are prominent features of the cortical cytoplasm as well as the plasma
membrane fraction, whereas in the ectodermal cells such material is generally
restricted to the apical junctions of the external surface and, to a lesser extent,
to the apical ectoplasm. In CD-treated cells a hypercontraction of the microfilaments is considered to give rise to the compact, felt-like masses in the cell
periphery (Miranda et al. 1914 b) Traces of attached feltwork material might
also account for the activity exhibited by pigment granules, which are initially
sited in the apical ectoplasm.
The low activity of the lipid droplets is unexpected in view of the lipophilic
properties of the cytochalasins. Certainly some preliminary evidence indicates
that lipids are not involved in CD-binding of the plasma membrane (Tannenbaum et al. 1975). The possibility remains, however, that the lipid may be
extracted during processing procedures, although the double-fixation method
Tritiated cytochalasin B in the amphibian ectoderm
221
employed here has been shown to be satisfactory for the retention of a high proportion of many kinds of lipid (Williams, 1969). Since the lipid content of the
amphibian egg is 25 % of its dry weight (see Balinsky, 1970), and since there was
little activity in the processing fluids from CB-treated eggs after fixation (Selman
et al. 1976), a massive extraction of labelled lipid from the lipid droplets can be
discounted.
The preferential localization of CB in the yolk platelets, the major protein
reserve of the developing embryo, is seemingly unrelated to the biological effects
of the drug. In this context the finding that serum proteins have a high capacity
for binding CB in the presence of high drug concentrations (Lin et al. 1974) is
of some relevance. However, whilst the yolk proteins are incorporated into the
ovary from the bloodstream during oogenesis, they have been shown to be
distinct from other serum proteins (Wallace & Jared, 1969). The alternative
explanation, that yolk proteins have some cytochalasin-binding sites in common
with those of the plasma membrane and endomembranes, deserves attention.
The crystalline core of the platelet consists of a lipoprotein component (lipovitellin) and a phosphoprotein component (phosvitin) in the ratio of 1:2
(Wallace, 1963). Concentric arrays of membranes, considered to be derived
from yolk lipoproteins, frequently occur around the yolk platelets at the time
they are being utilized during embryogenesis (Karasaki, 1963; Jurand & Selman,
1964). An investigation of the binding characteristics of yolk platelets over
a range of CB concentrations would help to elucidate the question of whether
their binding sites are of the low affinity (non-specific) type present in serum
proteins.
Thus, from a consideration of the quantitative aspects of CB uptake, the
results per se yield little information on the specific binding sites concerned
with the biological activity of cytochalasin. Both low and high affinity binding
sites have been distinguished in intact mammalian cells (Lin et al. 1974) and in
the plasma membrane fraction (Tannenbaum et al. 1975). These authors suggest
that the low, as well as the high affinity sites, may be required for the cellular
responses elicited by the cytochalasins. The dose used in the present work was
comparable with that required for low affinity binding and in consequence a
masking of high affinity sites, possibly present in small amounts, is to be
expected. It is nevertheless of some interest that only those regions of the cortex
containing the compact masses of felt-like material had a higher grain density
than the deeper cytoplasm. In this regard the observations of Puszkin, Puszkin,
Lo & Tanenbaum (1973) on the binding of CD to the contractile proteins of
muscle and blood platelets are pertinent. Their demonstration that CD binds to
free myosin, but not to the actomyosin complex, indicates that cytochalasin
binding in vivo may depend on the physiological state of the contractile system.
222
M. M. PERRY, G. G. SELMAN AND J. JACOB
In amphibian ectoderm cells there is some evidence for the presence of actinlike microfilaments, which may be identical with the microfilaments at the apical
junctions (Perry, 1975). Although myosin has not been detected in these cells,
the recent observation that myosin occurs in fibres coincident with those
containing actin-like microfilaments in other non-muscle cells (Weber & Groeschel-Stewart, 1974) tends to suggest that this protein may also be associated
with the rings of apical microfilaments. Thus the binding activity of the feltwork
may conceivably be attributed to the presence of dissociated myosin. This
speculation may be extended further to explain the slow response in that myosin
sites may only become available in a gradual manner for interaction with
CB.
The results offer no clue regarding the initial effect of CB, i.e. the inhibition
of explant curling. The process of curling involves active cell movements based
on the formation of lobopodia (Perry, 1975). Presumably the appropriate
CB-binding sites are so sparsely distributed that they cannot be detected by the
methods employed.
We are grateful to Dr R. C. Roberts for assistance with the statistical analysis. This
investigation was supported by the Agricultural Research Council (M.M.P.) and the Medical
Research Council (G.G.S. and J. J.)- The authors wish to acknowledge the debt of gratitude
they owe to the late Professor C. H. Waddington for the encouragement he gave to our work
on cleavage, morphogenesis and embryonic development.
REFERENCES
BALINSKY, B. I. (1970). An Introduction to Embryology.
EVERHART, L. P. & RUBIN, R. W. (1974). Cyclic changes
Philadelphia: W. B. Saunders.
in the cell surface. II. The effect of
cytochalasin B on the surface morphology of synchronised Chinese hamster ovary cells.
/. Cell Biol. 60, 442-447.
JACOB, J. (1971). The practice and application of electron microscope autoradiography. Int.
Rev. Cyt. 30, 91-m.
JURAND, A. & SELMAN, G. G. (1964). Yolk utilization in the notochord of newt as studied by
electron microscopy. / . Embryol. exp. Morph. 12, 43-50.
KARASAKI, S. (1963). Studies on amphibian yolk. V. Electron microscope observations on the
utilisation of yolk platelets during embryogenesis. /. Ultrastruc. Res. 9, 225-247.
KARFUNKEL, P. (1972). The activity of microtubules and microfilaments in neurulation in the
chick. /. exp. Zool. 181, 289-302.
KENT, C. & WILLIAMS, M. A. (1974). The nature of hypothalamo-neurohypophyseal neurosecretion in the rat. A study by light and electron microscope autoradiography. /. Cell Biol.
60, 554-570.
LIN, S., SANTI, D. V. & SPUDICH, J. A. (1974). Biochemical studies on the mode of action of
cytochalasin B. Preparations of [3H] cytochalasin B and studies on its binding to cells.
/. biol. Chem. 249, 2268-2274.
MIRANDA, A. F., GODMAN, G. C , DEITCH, A. D. & TANENBAUM, S. W. (1974a). Action of
cytochalasin D on cells of established lines. I. Early events. /. Cell Biol. 61, 481-500.
MIRANDA, A. F., GODMAN, G. C. & TANENBAUM, S. W. (19746). Action of cytochalasin D on
cells of established lines. II. Cortex and microfilaments. / . Cell Biol. 62, 406-423.
PERRY, M. M. (1975). Microfilaments in the external surface layer of the early amphibian
embryo. /. Embryol. exp. Morph. 33, 127-146.
PUSZKIN, E., PUSZKIN, S., Lo, L. W. & TANENBAUM, S. W. (1973). Binding of cytochalasin D
to platelet and muscle myosin. /. biol. Chem. 248, 7751-7761.
Tritiated cytochalasin B in the amphibian ectoderm
223
M. M., BACHMANN, L. & SALPETER, E. E. (1969). Resolution in electron microscopic autoradiography. /. Cell Bio I. 41, 1-20.
SALPETER, M. M. & MCHENRY, F. A. (1973). Electron microscope autoradiography: analysis
of autoradiograms. In Advanced Techniques in Biological Electron Microscopy (ed. J. K.
Koehler), pp. 113-152. Berlin: Springer-Verlag.
SELMAN, G. G. & PERRY, M. M. (1970). Ultrastructural changes in the surface layers of the
newt's egg in relation to the mechanism of its cleavage. /. Cell Sci. 6, 207-227.
SELMAN, G. G., JACOB, J. & PERRY, M. M. (1976). The permeability to cytochalasin B of
the new unpigmented surface in the first cleavage furrow of the newt's egg. /. Embryol. exp.
Morph. 36 (In the Press).
SPOONER, B. S. & WESSELLS, N. K. (1972). An analysis of salivary gland morphogenesis: role
of cytoplasmic microtubules and microfilaments. Devi Bio I. 27, 35-54.
TANNENBAUM, J., TANENBAUM, S. W., LO, L. W., GODMAN, G. C. & MIRANDA, A. F. (1975).
Binding and sub-cellular localisation of tritiated cytochalasin D. Expl Cell Res. 91, 47-56.
WALLACE, R. A. (1963). Studies on amphibian yolk. IV. An analysis of the main body component of the yolk platelets. Biochim. biophys. Ada 74, 505-518.
WALLACE, R. A. & JARED, D. W. (1969). Studies on amphibian yolk. VIII. The estrogen
induced hepatic synthesis of a serum lipophosphoprotein and its selective uptake by the
ovary and transformation into yolk platelet proteins in Xenopus laevis. Devi Biol. 19,
498-526.
WEBER, K. & GROESCHEL-STEWART, U. (1974). Antibody to myosin: the specific visualisation
of myosin-containing filaments in nonmuscle cells. Proc. natn. Acad. Sci. U.S.A. 71,
4561-4564.
WILLIAMS, M. A. (1969). The assessment of electron microscopic autoradiographs. In
Advances in Optical and Electron Microscopy, vol. 3 (ed. R. Barer & V. E. Cosslett),
pp. 219-272. London: Academic Press.
SALPETER,
{Received 22 March 1976)