Cell asymmetries in development
151
Lateral mobility of plasma membrane lipids in Xenopus eggs: regional
differences related to animal/vegetal polarity
/. G. Bluemink*1, W. J. A. G. Dictus, E. J. J. van Zoelen \ P. A. T. Tetteroo, L. G. J. Tertoolen 1 andS.
W. deLaatl. lHubrecht Laboratory, International Embryological Institute, Uppsalalaan 8, 3584 CT Utrecht,
The Netherlands. 2Department of Biology, V. U., Workgroup Histology and Electron Microscopy, P. 0. Box
7161,1007 MC Amsterdam, The Netherlands.3Central Laboraotry of the Blood-transfusion Service,
Plesmanlaan 125,1066 CX Amsterdam, The Netherlands
Regional differences in the lateral mobility properties of plasma membrane lipids were studied in
unfertilized and fertilized eggs and early embryos of Xenopus laevis by fluorescent photobleaching recovery
(FPR) measurements. The aminofluorescein-labelled fatty acids HEDAF and TEDAF appear to distribute
themselves in the plasma membrane under all conditions used. These molecules show partial recovery of
fluorescence upon photobleaching, indicating the existence of lipidic microdomains in the membrane. In the
unfertilized egg the mobile fraction 8of plasma
membrane lipid (~ 50 %) has a 5-fold smaller lateral
diffusion coefficient (D = 1-5 x 10~ crrr.s"1) in the animal than in the vegetal polarity within the egg
plasma membrane. Upon fertilization this polarity is strongly (> lOOx) enhanced, leading to the formation
of two distinct macrodomains within the plasma membrane. On the animal side
of2 the
egg the lipids are
10
-1
completely immobilized on the time scale of FPR 8measurements
(D
«
10~
cm
.s
),
whereas on the
vegetal side D is only slightly reduced (D = 2-8xlO~ cm2 s"1). The transition from the non-fluid animal to
the fluid vegetal domain is sharp and lies on the equator. This animal/vegetal difference in lateral mobility
properties of plasma membrane lipids is maintained through the period of cellulation, as shown by FPR
measurements on the morula (st. 6i) and blastula (st. 8). At stage 6£ the transition from the animal to the
vegetal domain across the equator is two cells wide, lipid mobility changing stepwise from one cell to the
next. Preliminary FPR measurements on the mid-gastrula (st.lOi) show that the dorsal and ventral
blastoporal lips are within the fluid domain. Current research is aimed at investigating the morphogenetic
role of the animal/vegetal difference in plasma membrane mobility characteristics, using embryos which
have developed under conditions when the blastopore should appear in the non-fluid domain of the animal
half (eggs turned 180 ° with respect to gravity)
Electrical phenomena and their possible significance in vitellogenic
follicles of Drosophila melanogaster
J. Bohrmann \ U.-R. Heinrich \ A. Dorn 2, K. Sander 1 andH. Gutzeit*1. lInstitutfurBiologieI
(Zoologie), Albertstrasse 21a, D-7800 Freiburg, West Germany. 2Botanisches Institut I, Kaiserstrasse 2,
D-7500 Karlsruhe, West Germany
We have measured extracellular currents in vitellogenic follicles of Drosophila with a vibrating probe. In
stage 10 follicles current enters at the anterior half of the follicle (nurse cells) and leaves at the posterior half
(oocyte). Stage 11 follicles show a variable pattern of current flow. Often a strong current (either inward or
outward) was measured in the region of the centripetally migrating follicle cells. Some of these cells are
specialized with respect to their ion content: when cations (in particular Ca++) are precipitated with
pyroantimonate and the formed precipitates viewed in the electron microscope, some cells with a
particularly high number of precipitates per unit area can be identified. These cells increase in number
during stage 10B and extend from the area of the ring canals to the outer face of the follicle. These
precipitate-rich cells may be part of an intra-follicular current loop whose existence in Cecropia follicles was
inferred by extracellular potential measurements (Jaffe and Woodruff, 1979).
The physiological role of the electrical phenomena is still unclear. Electrophoretic migration of charged
proteins (Woodruff and Telfer, 1980) in the follicle may occur but positive evidence with endogenous
proteins is still lacking and this mechanisms probably does not account for the site-specific deposition of
molecules within the follicle. The rapid influx of nurse cell cytoplasm into the oocyte during stage 10B to 12
appears not to be related to the electrical phenomena, since some follicles which for unknown reasons did
not generate any extracellular currents continued cytoplasmic streaming in vitro. Our observation that ion
asymmetries in Drosophila follicles can first be detected with the pyroantimonate method when yolk uptake
begins (stage 7) and then increase up to the stage of nurse cell breakdown (stage 10B) suggests that electrical
phenomena may be related to vitellogenesis.
JAFFE, L. F. & WOODRUFF, R. I. (1979). Large electrical currents traverse developing Cecropia follicles.
Proc. Natl. Acad. Sci USA 76, 1328-1332.
WOODRUFF, R. I. & TELFER, W. H. (1980). Electrophoresis of proteins in intercellular bridges. Nature 286,
84-86.
152
Cell asymmetries in development
Gytoskeletal control of endosome polarisation in mouse preimplantation
embryos
Peter M. Cannon and Tom P. Fleming, Department of Anatomy, University of Cambridge, Downing Street,
Cambridge CB2 3 DY
The 8-cell stage of mouse development is characterized by a process of cell polarisation which includes, in
the majority of cells, the clustering of endosomes from a uniform distribution into an apical cytoplasmic
region, subtending the microvillous pole. The endosome pole is present in outer blastomeres following
division to the 16-cell stage.
The role of the cytoskeleton during the formation and subsequent stabilization of endosome poles has
been investigated in 8- and 16-cell embryos pre-incubated in horseradish peroxidase to label the endocytotic
apparatus. The formation of endosome poles is significantly reduced if newly-formed 8-cell embryos (or
natural 2/8 couplets) are incubated in cytochalasin D (CCD: 0-5 /i.g/ml) during the initial period of the
fourth cell cycle, demonstrating that intact microfilaments are required for endosome relocation. In
contrast, pre-existing endosome poles within compacted 8-cell embryos are insensitive to CCD treatment
but are significantly disrupted by colcemid (50 /ng/ml) incubation, indicating that microtubules, but not
microfilaments, are responsible for stabilizing the asymmetric pattern of endosome cytolocation. Colcemidinduced loss of endosome polarity is blocked by CCD suggesting that relocation of endosomes is
microfilament-dependent.
The occurrence of endosome poles within outer blastomeres at the 16-cell stage is significantly greater
than within compacted 8-cell blastomeres and they consist of more tightly-packed clusters of endocytotic
vesicles. In addition, endosome poles at the 16-cell stage are insensitive to both colcemid and CCD,
demonstrating a phase of stabilization in the generation of endocytotic polarity.
Concanavalin A as a probe for the polar organization of the plasma
membrane in a molluscan egg cell
M. R. Dohmen, J. E. Speksnijder and K. J. Teerds, Zoological Laboratory, University of Utrecht, Padualaan
8, Utrecht, The Netherlands
The polarity of egg cells is an important factor in early development. The determinants of the spatial
organisation of cells are largely unknown. Most likely, the plasma membrane plays a central role in this
respect. The results of experiments with conA support the view that a polar structure exists at the level of
the plasma membrane. When fertilized uncleaved eggs of the gastropod Nassarius reticulatus are treated
with conA the plasma membrane/cytoskeleton complex starts being affected at the vegetal pole of the egg
and this reaction progressively spreads towards the animal pole. Irrespective of the moment of incubation
with conA, the reaction of the egg begins when the first polar lobe is being formed, a process immediately
preceding first cleavage. As soon as the polar lobe constriction appears, the microvilli start disappearing at
the vegetal pole. Concomitantly, the vitelline membrane which is attached to the egg surface disappears at
the same rate and to the same extent as the microvilli. This process stops short of the constriction of the
polar lobe. The position of the polar lobe constriction may also be strongly influenced by conA treatment.
At low doses the constriction is situated in its normal position, but an abnormally long and narrow stalk is
formed. At high doses the position of the constriction may shift extremely towards the animal or vegetal
pole of the egg. A cleavage furrow starts being formed at the animal pole, but its progression is inhibited and
it soon regresses. Nuclear division is not impaired. After some time most of the eggs lyse, a process which
occurs invariably at the vegetal pole.
It is concluded that conA exerts a propagated transmembrane effect on the cortical cytoskeleton via
receptors present in the plasma membrane at the vegetal pole. The phenomenon that the effect of conA is
triggered by the onset of cytokinesis may be due to cell cycle dependent modulation of these receptors or
their relay systems.
Cell asymmetries in development
Maturation and polarisation of the endocytotic apparatus in outer
blastomeres of the preimplantation mouse embryo
153
Tom P. Fleming, Department of Anatomy, University of Cambridge, Downing Street, Cambridge CB2 3DY
During cleavage, outer blastomeres undergo a progressive maturation and polarisation of the endocytotic
apparatus. These incremental phases in the generation of a polarised epithelium (trophectoderm) have been
studied using various ultrastructural techniques.
Oocytes to early 8-cell stages possess apolar clusters of endosomes predominantly confined to the cortical
cytoplasm. In compacted 8-cell embryos, endosomes become preferentially localised in the apical cytoplasm
beneath the microvillous pole. Incubation in the endocytotic tracer, horseradish peroxidase (HRP; 5 mg/ml)
results in pinocytic vesicle (PV) formation and rapid fusion with apical endosomes from 5-20 m. PV
formation (expressed as PV number/unit length plasma membrane) is significantly greater at the apical
(outer) PM than at the basolateral PM in both compacted and decompacted 8-cell embryos; no such
distinction in pinocytic activity is evident in apolar, early 8-cell embryos.
A polar cytolocation of endosomes is also evident in outer blastomeres at the early 16-cell stage (9 h
post-compaction). These cells display an increased preferential endocytotic activity at the apical PM face.
During the late 16- and early 32-cell stages, outer blastomeres generate and accumulate electron dense
secondary lysosomes (SLs) in the basal cytoplasm. SLs show trimataphosphatase activity, and become
labelled with HRP by 30 m incubation. They are derived from autolytic digestion of cytoplasmic granules
and transformation of endosomes. Following zonular tight junction formation at the 32-cell stage, outer cells
(trophectoderm) possess transcellular and recycling endocytotic pathways (visualized by HRP and/or
cationized ferritin labelling), which collectively appear responsible for maintaining PM asymmetry and cell
polarity.
The maturation of a polarised endocytotic apparatus in outer
blastomeres during preimplantation development in the mouse
Tom P. Fleming* and Peter M. Cannon, Department of Anatomy, University of Cambridge, Downing Street,
Cambridge
Morphological studies have shown that:
(a) the endocytotic apparatus within the trophectoderm epithelium of the blastocyst is polarised with
respect to organelle distribution and constituent processing pathways.
(b) the development of endocytotic asymmetry occurs incrementally in outer blastomeres during
preceeding stages of cleavage.
The important phases in the generation of endocytotic polarity in outer cells can be summarised as
follows:
1) Oocyte - early 8-cell stage. Endocytotic organelles consist primarily of endosomes (or multivesicular
bodies) which are in apolar distribution and occur mainly in the cortical cytoplasm. No difference in
pinocytic activity is evident between apical (outer) and basolateral membrane faces.
2) Compacted 8-cell stage. Endosomes become preferentially localised in the apical cytoplasm underlying
the microvillous pole. Relocation is sensitive to cytochalasin D; pre-formed endosome poles are disrupted
by colcemid. Pinocytic activity is significantly greater at the apical rather than the basolateral membrane
face.
3) Early 16-cell stage. Endosomes remain polarised and become stable to either colcemid or cytochalasin
D treatments. Preferential pinocytic activity at the apical membrane face is increased.
4) Late 16- to early 32-cell stages. Secondary lysosomes, showing trimetaphosphatase activity, accumulate
in the basal cytoplasm.
5) Trophectoderm. Apical and basolateral membrane faces are separated by zonular tight junctions.
Transcellular endocytotic pathways via endosomes operate in both directions; these, in conjunction with
short-circuit endosome recycling pathways, are likely to preserve the asymmetry of cell membrane domains.
154
Cell asymmetries in development
A three-step scheme of early grey crescent formation in the Axolotl
oocyte
J. Gautier 1>2 andJ. C. Beetshcen *. lLaboratoire de Biologie ge"n&rale, University Paul Sabatier,
31062 Toulouse, France. 2D6partement de Biologie moUculaire, University Libre de Bruxelles,
1640 Rhode-Saint-Genese, Belgium
It has been shown that inhibition of protein synthesis can elicit, within a few hours, the precocious
appearance of the grey crescent (GC) in Axolotl oocytes, artificially maturing in vitro (1,2). However, such
a symmetry reaction does not occur when the oocyte has been enucleated before progesterone-induced
maturation. The ability to form a GC is re-established in enucleated oocytes by the injection of nucleoplasm
from8 a normal oocyte, either before or after injection of the protein synthesis inhibitor (diphtheria toxin,
10~ M) (3). In the latter case, the GC appears even though the protein synthesis level is as low as one-tenth
that of the control enucleated oocyte, and it is formed very quickly (after 30 min in 50 % of the injected
oocytes). Finally, it has been established that the interactions oetween nuclear factor(s) and cytoplasm are
possible only when cytoplasmic maturation has been proceeding for 10 h at least.
Our results thus suggest a three-step scheme for establishing preliminary conditions essential to Axolotl
oocyte symmetrization (GC formation):
1) Cytoplasmic maturation.
lS Nucleoplasmic factor(s) mix(es) with cytoplasm after germinal vesicle breakdown.
3) Inhibition of protein synthesis.
Steps 2 and 3 can be experimentally inverted, but step 1 must always occur first. A similar sequence of
events should apply to normal GC formation in the fertilized egg, since a sharp decrease of the protein
synthesis level (about 50 %) was observed in uncleaved eggs before appearance of the grey crescent.
The active nuclear fraction of the normal oocyte is under investigation. It rapidly migrates out of isolated
germinal vesicles but it is present in the soluble fraction from whole oocytes extracts, in the supernate
(150,000xg, 3 h). It is not found in the extract of enucleated oocytes. The active fraction is heat-resistant,
RNAase-resistant and trypsin-sensitive.
1) GRINFELD, S. & BEETSCHEN, J. C. (1982). Roux's Arch. Devi. Biol., 191, 215-221.
2) GAUTIER, J. & BEETSCHEN, J. C. (1983). Ibid, 192, 196-199.
3) GAUTIER, J. & BEETSCHEN, J. C. (1983). C.R. Acad. Sc. Paris, s6r. III. 296, 815-818.
Displacement of cellular components during the maturation of the Xenopus
oocyte
P. Hausen, Ya-Hui Wang, and C. Dreyer, Max-Planck-Institutfur Entwicklungsbiologie,
Spemannstrasse 351V, 7400 Tubingen, West Germany
Oocyte maturation was induced in vitro by progesterone. The movements and translocations of the
intracellular components in the course of the maturation process were followed first by standard histological
techniques at the light microscope level. By use of monoclonal antibodies directed against various oocyte
components, their fate during egg maturation was followed. Special reference has been made to proteins of
the germinal vesicle and their allocation to the different regions of the egg.
Cell asymmetries in development
155
The mitochondrial cloud of Xenopus oocytes is the source of germinal
granule material; its breakdown localises this material in the vegetal
pole
Janet Heasman and C. C. Wylie, Department of Anatomy, St. George's Hospital Medical School, Cranmer
Terrace, London, SW17ORE
The mitochondrial cloud is a prominent mass in the cytoplasm of previtellogenic oocytes of Xenopus
laevis. We report here that it contains electron-dense granulo-fibrillar material (GFM) as well as
mitochondria. Using a combination of light microscopical, fluorescence, time-lapse filming and electron
microscopical techniques, the ontogeny of these components has been studied. We find that the cloud is
stationary in previtellogenic stages, and fragments into islands of mitochondria and GFM at the time of
onset of vitellogenesis. These islands become localized in the peripheral cytoplasm at one pole of the small
oocyte. By studying successive stages, we find that GFM remains localized at one pole; in larger oocytes,
where the animal/vegetal axis becomes obvious due to pigment accumulation, this is found to be the vegetal
pole. Furthermore GFM bears a striking resemblance in position, appearance and association with
mitochondria to the 'germinal granules' found in the vegetal pole of unfertilized eggs. Germinal granules
have been shown by others to become incorporated into germ-line cells. We conclude that GFM is the
precursor of this material and that it accumulates in the mitochondrial cloud of previtellogenic oocytes. The
dispersing cloud may provide a mechanism for its localization in the vegetal pole of the oocyte and egg, and
is the first indication of polarity in the Xenopus oocyte.
Segregation of informational macromolecules in Ascidian embryos
W. Jeffrey
No abstract for publication
156
Cell asymmetries in development
Mechanisms for generating and stabilizing cell asymmetries
M. H. Johnson*, Department of Anatomy, Downing Street, Cambridge CB23DY
The generation of cell diversity by the differentiative division of an asymmetrically organised cell is well
established in embryology. This Symposium is not concerned with describing more examples of this
phenomenon but rather in probing the way in which cell asymmetries are set up, elaborated and stabilized.
The Symposium will concentrate on mechanisms. Many, if not all, cell asymmetries are set up, and the
orientation of the reorganisation determined, by external signals. Early, spatially-defined responses to these
signals include changes in electrical permeability, local membrane properties and cytoskeletal organisation.
Subsequently, major redistributions of cell organelles and of specific informational molecules such as
mRNAs and proteins occur. Careful analysis of the temporal sequence in which these changes occur, and
the use of specific reagents that affect individual components of the sequence, can help elucidate the nature
of the primary response to the signal for asymmetry, and how this response is transmitted throughout the
cell. Similarly, once asymmetry is established, manipulations on the cell, followed by observation of the
immediate and medium term regulative capacity of the cell, can give clues as to whether a specific 'memory'
of the asymmetry is present in the cell and if so, where it resides. This latter feature is of particular
importance to the cell during division if non-equivalent progency are to be produced. It is the mechanisms
underlying these various events that concerns this Symposium.
An EM study of germ plasm in normal and inverted eggs of Xenopus
laevis
A. Jurand * and K. E. Dixon 2. 1Edinburgh University. 2The Flinders University of South Australia,
Bedford Park, 5042, Australia
Germ plasm in unfertilised eggs of anuran amphibians is distributed throughout the cortex in small,
isolated islets containing mitochondria and germinal granules (Czolowska, 1972). During the first two
cleavage divisions, the individual islets move towards the vegetal pole, coalescing as they do so to form large
aggregates identified in the light microscope by their yolk-free, granular structure. In 8-cell embryos in
normal orientation, the region identified as the germ plasm in the light microscope is found at the
ultrastructural level to be composed of a number of organelles - numerous mitochondria, many germinal
granules, vesicles, ribosomes, glycogen and some microfilaments, ER and Golgi bodies. The mitochondria
adhere together to form chains which presumably provide structural cohesion (ie. a cytoskeletal function)
for the whole germ plasm region which can therefore be considered largely autonomous within the
cytoplasm of the cell containing it. The germinal granules are aggregated together, and intimate structural
associations are observed between the aggregates and the mitochondria. When fertilised eggs were rotated
through 90 ° or inverted during the first cleavage cycle and then examined at the 8-cell stage, the
mitochondria in the germ plasm were more dispersed and adhesions between them were rare. The
aggregates of germinal granules were less compact and large spaces due to expansion and thinning of their
texture were common, resulting in their exhibiting a sponge-like appearance. The failure of the
mitochondria to adhere together and the abnormal structures formed by the germinal granules may be
causally related to the decreased number of germ cells which enter the genital ridges in rotated and inverted
embryos (Wakahara et al., 1984; Cleine and Dixon, unpublished).
R. (1972). Thefinestructure of the 'germinal cytoplasm' in the egg of Xenopus laevis. Wilhelm
Roux' Archiv. 169, 335-344.
WAKAHARA, M., NEFF, A. W. & MALACINSKI, G. M. (1984). Topology of the germ plasm and development
of primordial germ cells in inverted amphibian eggs. Differentiation, in press.
CZOLOWSKA,
Cell asymmetries in development
Microtubules, centrioles, and cell asymmetry
157
Marc Kirschner*, Eric Karsenti and Tim Mitchison, Department of Biochemistry and Biophysics, University
of California San Francisco, San Francisco, California 94143, U.S.A.
Microtubules play an important role in establishing and maintaining the asymmetry of somatic cells and
eggs. The organization of microtubules depends in part on the disposition and activity of microtubule
organizing centers which in most animal cells consists of a pair of centrioles surrounded by osmiophillic
material. These organizing centers function in both interphase and mitosis. We have recently studied the
mechanics by which the organizing centers spatially organize microtubules in cells in three types of
experiments: a) the function of organizing centers in anchoring microtubules in interphase somatic cells, b)
the mechanisms by which microtubule arrays are converted from interphase to mitosis using microinjection
into frog eggs, and c) the underlying molecular mechanisms by which organizing centers function to nucleate
microtubules, by studying the assembly of pure tubulin on isolated centers in vitro. The results of the latter
studies suggest that microtubules may be much more dynamic than previously thought. They also suggest
that nucleation itself may be a simple process but the unusual physical and chemical properties of
microtubule polymerization can provide mechanisms by which the cell can impose a pattern of microtubule
organization on a nucleated array. The conversion of microtubule arrays from mitosis to interphase suggests
that such unusual mechanisms operated in vivo.
The role of microtubules during compaction of the mouse 8-cell embryo
B. Maro, M. H. Johnson and S. J. Pickering, Department of Anatomy, Downing Street,
Cambridge CB23DY
The role of microtubules during compaction of the mouse 8-cell stage embryo was investigated using the
drugs Taxol (which leads to a non controlled polymerization of tubulin) and Nocodazole (which causes
depolymerization of microtubules). Taxol inhibits compaction in most non-compacted embryos and reverses
it in already compacted embryos. These effects were observed on cell flattening (as judged by phase contrast
microscopy), on cell polarization (as judged by scanning electron microscopy and the surface binding of
fluorescent concanavalin A) and on polarization of intracellular organelles (as judged by immunofluorescent
staining of actin and clathrin). In contrast, Nocodazole does not inhibit cell flattening but rather accelerates
its completion. Nocodazole influences the detailed organization of the surface poles and appears to reduce
the incidence of surface polarization but does not reverse (to a significant extent) polarity once established.
Nocodazole is also able to inhibit and reverse the redistribution of intracellular organelles which takes place
during compaction. We conclude that microtubules exercise a constraining role during compaction,
regulating the changes in cell shape and cell organisation, and being involved in the timing of compaction
rather than constituting part of the mechanism of cell flattening and cell polarization.
158
Cell asymmetries in development
Assembly and topogenesis of the spectrin-based membrane-cytoskeleton
during erythroid development
Randall T. Moon* and Elias Lazarides, Division of Biology, California Institute of Technology,
Pasadena, CA 91125, U.S.A.
Components of multisubunit protein complexes are frequently synthesized at one or more cytoplasmic
sites, and assembled at another site. As a model for the assembly of localized multisubunit complexes we
have studied the biogenesis of the membrane skeleton of erythroid cells. In erythrocytes a/J-spectrin is
linked to the plasma membrane through ankyrin, which binds to both the /3-subunit of spectrin and to a
subset of the transmembrane anion transporters. Development of in vivo and in vitro reconstitution assays
which distinguish assembled from unassembled subunits has revealed that during chicken erythroid cell
development a- and /S-spectrin and ankyrin are synthesized simultaneously but in excess of the amount of
each polypeptide assembled. Assembly of all three polypeptides onto the membrane occurs rapidly after
synthesis and at the stoichiometry detected at steady state. Assembly proceeds by a series of limiting steps:
the availability of membrane binding sites limits the amount of ankyrin assembled, which in turn limits the
amount of ar-spectrin assembled. Pulse-chase experiments reveal that assembled subunits are stable while
unassembled subunits turn over with a half-life of approximately 45 min. These studies support a scheme for
the assembly of spectrin and ankyrin where high affinity receptors localized on the plasma membrane
(presumably the anion trnasporter) mediate post-translationally the stable assembly of the three cytoskeletal proteins in their correct stoichiometry, thus also defining their spatial localization at the plasma
membrane. Failure of subunits to assemble results in their degradation, suggesting that synthesis and
assembly of these polypeptides although concurrent, are not tightly coupled events. These principles of the
assembly of the spectrin-based membrane cytoskeleton also apply to the assembly of many other spatially
segregated multisubunit complexes in both plants and animals. These principles may, therefore, reflect
widespread mechanisms involved in the establishment of cell asymmetries.
Strikingly different distributions of nuclear constituents during fibroblast
cell division revealed by monoclonal antibodies
E. A. Nigg, C. F. Lehner, V. KurerandH. M. Eppenberger, Institute for Cell Biology, ETH-Hoenggerberg,
CH-8093 Zurich, Switzerland
Monoclonal antibodies were raised against nuclear extracts prepared from chicken embryonic tissues.
According to immunofluorescent localizations in fibroblast interphase nuclei, four different classes of
antigens were distinguished: i) envelope-associated lamin proteins (67-70K), ii) nucleolar constituents
(90,40K), iii) nucleoplasmic proteins (mw range: 25-130K) and iv) antigens apparently confined to discrete
nuclear substructures (e.g. 40K). Using immunofluorescence microscopy, antibodies against selected
antigens were used to study the process of nuclear disassembly and re-formation during cell division.
Following breakdown of the nuclear envelope in late prophase most nuclear antigens become diffusely
distributed throughout the entire cell. Lamin proteins are detectable in partial association with condensed
chromosomes up to early metaphase, but no longer in anaphase. Nucleolar proteins are the first antigens to
re-associate with chromosomes in late anaphase. Concomitant with the onset of envelope re-formation in
telophase, nucleoplasmic antigens begin to accumulate in the forming daughter nuclei and completely
disappear from the cytoplasm within an impressively short time. Undoubtedly most interesting is the
behaviour of those antigens which are located within a characteristic substructure of interphase nuclei.
These antigens begin to display a 'patchy' distribution in anaphase, suggesting the formation of a relatively
large supramolecular complex, e.g. a vesicular structure. This 'patchy' distribution persists in the cytoplasm
of daughter cells for a considerable time into early interphase. The nature of the antigens showing this
particular delayed type of re-entry and the reason for their temporary extranuclear sequestration are
currently under study. (Supported by the Swiss NSF and an ETH doctoral grant to C.F.L.).
Cell asymmetries in development
159
Aggregation and movement of germ plasm, the cytoplasmic determinant
for germ cells, in early embryos of Xenopus laevis
R. E. Ressom and K. E. Dixon, School of Biological Sciences, The Flinders Unversity of South Australia,
Bedford Park, 5042, Australia
Early embryos of X. laevis contain a microscopically visualisable region of cytoplasm which is segregated
by cleavage divisions to a small number of cells (4 in a 5000-cell blastula) which are the progenitors of the
gametes. Examination of sections of early embryos showed that the germ plasm can first be distinguished
about 2 h after fertilisation as small islets in the subcortical region of the vegetal hemisphere. As
development proceeds, the islets aggregate together forming an average 3-4 patches per embryo at 4 h
postfertilisation (stage 6). The mechanisms controlling aggregation and movement of germ plasm were
investigated in fertilised eggs and eggs artificially activated with calcium ionophore A23187. Three processes
were distinguished: (i) aggregation of the islets into large masses, a process dependent on microtuoules but
independent of cleavage; (ii) internalisation of these aggregates as part of a general cytoplasmic flow from
the vegetal pole region towards the interior; (iii) capture of the aggregates by cleavage asters. The result of
the first of these processes is that the germ plasm is partitioned between the first four blastomeres.
Movement internally permits capture of the germ plasm by the cleavage asters and in turn this immobilises
the germ plasm and ensures that it is segregated unequally at each subsequent cleavage division.
Animal vegetal polarity and the sea urchin cortex
Christian Sardet*, ER250 CNRS, Station Marine, F06230 Villefranche-sur-mer, France
Sea urchin eggs have animal and vegetal poles. Successive cleavages are oriented with respect to these
poles and development proceeds according to this animal/vegetal polarity.
In the egg of the sea urchin Paracentrotus liyidus, polarity is apparent in the cortex before fertilization as a
subequatonal band of pigment granules. This structure is preserved as part of the isolated cortex. I will
present evidence that similar vesicular bodies (acidic vesicles) are present in all sea urchin eggs and exhibit
animal/vegetal polarity as well as apical/basal polarity in blastomeres.
I will describe other components that remain as parts of the isolated cortex and in particular an intricate
network of rough endoplasmic reticulum, that is tightly anchored to the plasma membrane. I will discuss the
functions of these components and their relationship to the animal/vegetal polarity.
160
Cell asymmetries in development
Cytoplasmic control of cell-cycle timing in early embryos of
Caenorhabditis elegans
Einhard Schierenberg*1 and William B. Wood, Department of Molecular, Cellular and Developmental
Biology, Boulder, Colorado, 80309, USA. 1Present address•; Max-Planck-Institute for experimental
Medicine, H. Rein-Str. 3, D-3400 Goettingen, West Germany
During early cleavage in the C. elegans embryo, a series of asymmetric divisions of the germ line cell
generates five somatic founder cells. The cell lineages derived from each of these founder cells, including the
fate of each cell are known (1). They are invariant. Lineage-specific differences in cell cycle timing appear to
be important in establishing the correct spatial patterns of cells as indicated by mutants with altered cell
cycle rhythms (2) and laser-induced retardations of cell cycle timing within specific cell lines (3). To examine
the role of nuclei and cytoplasm in controlling cell cycle timing in C. elegans embryos, we have developed a
technique for extruding either substantial amounts of cytoplasm without nuclei or nuclei with small amounts
of cytoplasm from individual blastomeres. This allows us to observe the cycling behavior of blastomeres with
altered nuclear/cytoplasmic ratios and of enucleated cytoplasts. By laser-induced fusion (4) of a cytoplast
with an adjacent blastomere, we have been able to observe effects of exposing a nucleated cell from one
lineage to cytoplasm of another, as well as effects of changing the nuclear/cytoplasmic ratio within the same
lineage. We find that (a) nuclei in a common cytoplasm divide synchronously; (b) enucleated blastomeres
retain a cycling period characteristic of their lineage; (c) cycling period is not substantially affected by
changes in the ratio of nuclear to cytoplasmic volumes or the DNA content per cell; (d) the cycling period of
a cell from one lineage can be substantially altered by introduction of cytoplasm from a cell of another
lineage with a different period; and (e) short-term effects of foreign cytoplasm on the timing of the
subsequent mitosis differ with time in the cell cycle of the donor cell. Based on our results we suggest a
model for the action of cytoplasmic factors in controlling cell-cycle timing.
SULSTON et al. (1983). Devi. Biol. 100, 64-119.
SCHIERENBERG et al. (1980). Devi. Biol. 76, 141-159.
unpublished results.
SCHIERENBERG (1984). Devi. Biol. 101, 240-245.
Cell culture of renal medullary, (thick ascending limb of Henle's loop)
cells
D. M. Scott, K. Zierold, E. Kinne-Saffran and R. Kinne, Max-Planck-Institutfuer systemphysiologie,
4600 Dortmund 1, West Germany
The nephron of the kidney is made up of a variety of epithelial cells that vary in their morphology,
function and location. To date the majority of in vitro cell culture studies of kidney cell function have been
carried out with established cell lines of unknown derivation. Consequently, in order to overcome the
obvious limitations of such systems we have attempted to establish in culture cells of defined origin. Such
homogeneous cell populations were derived from the thick ascending limb of Henle's loop (TALH) of
rabbit kidney and were obtained by initial collagenase/hyaluronidase digestion of medullary fragments,
followed by sequential trypsin digestion and cell separation by density gradient centrifugation (Eveloff, J.,
Haase, W., and Kinne, R., (1980) /. Cell. Biol. 87, 672-681). These cells were then resuspended in
Dulbecco's Modified Essential Medium supplemented with 10 % foetal calf serum, non-essential amino
acids, pyruvate and antibiotics, and were maintained at 37 °C in a 5 % CO2/95 % air atmosphere. Medium
was initially changed at 24 h and subsequently every 48 h. Primary cultures exhibited typical epithelial
morphology (as demonstrated by light and electron microscopy) and expressed biochemical and functional
characteristics of TALH cells. Cultures routinely showed high levels of Na-K-ATPase activity as 3would be
expected of such cells and their respiration was markedly inhibited (ca. 60 %) by Furosemide (10~ M). This
diuretic inhibits salt transport in TALH cells and has been shown to be a specific inhibitor of these cells
(Eveloff et al., 1980). Only low levels of brush border enzymes such as alkaline phosphatase were observed.
Na-K-ATPase activity was shown to increase in activity during the period in primary culture, although this
activity was markedly reduced (by up to 85 %) by subculture of these cells, even when performed in the
absence of proteases. These studies demonstrate the feasibility of the growth in culture of defined kidney
epithelial cell populations and their subsequent detailed study should provide us with an important system
for the study of the expression and regulation of TALH cell specific functions.
Cell asymmetries in development
161
Enveloped viruses as tools to study cell surface polarity in epithelial cells
K. Simons* and K. Matlin, European Molecular Biology Laboratory, Postfach 10.2209, D-6900 Heidelberg,
West Germany
Our present studies are directed towards elucidating the mechanisms of establishment of cell surface
polarity in epithelial cells using enveloped viruses as tools. We are studying the Madin Darby Canine kidney
(MDCK) cell line which grows in culture as an epithelium. In MDCK cells infected with influenza, the viral
hemagglutinin behaves as an apical plasma membrane glycoprotein. To find out where the hemagglutinin
was sorted, the domain of the plasma membrane, apical or basolateral, where newly synthesized
hemagglutinin first appears was studied by a novel approach. Cells were cultured on Millipore filters to
make cell surface domains independently accessible and infected with influenza virus. Hemagglutinin was
pulse-labeled, chased and detected on the plasma membrane with a sensitive trypsin assay (Matlin, K. and
K. Simons (1983) Cell 34, 233-243). Under all conditions tested, newly made hemagglutinin appeared in
large excess on the apical domain, but there was always a fraction on the basolateral domain. Trypsin
continously present on the basolateral surface during the chase failed to cleave an amount of hemagglutin
equivalent to that transported apically. In addition specific antibodies against the hemagglutinin placed
basolaterally had no effect on transport to the apical domain. These observations suggested that apically
destined hemagglutinin did not transiently appear on the basolateral surface and that sorting, therefore,
occurs intracellularly prior to arrival of the glycoprotein on the cell surface.
Lateral mobility of plasma membrane lipids during early molluscan
development
/. E. Speksnijder, M. R. Dohmen, E. J. J. van Zoelen, L. G. J. Tertoolen, J. G. Bluemink and S. W. de
Laat, Zoological Laboratory, University of Utrecht, Padualaan 8, 3508 TB Utrecht and the Hubrecht
Laboratory, Uppsalalaan 8,3584 CT Utrecht, The Netherlands
The lateral diffusion of the lipid analog C14-dil (3',3'-dihexadecylindocarbocyanine iodide) was measured
in the plasma membrane of early embryos of the mollusc Nassarius reticulatus using the FPR-(Fluorescence
Photobleaching Recovery) method. At almost all stages measured (from fertilized egg up to 8-cell stage) the
diffusion coefficient (D) of the mobile fraction (MF) of Ci4-dil is significantly higher in the plasma
membrane of the vegetal pole area as compared to the plasma membrane of the animal half of the embryo.
These results demonstrate the presence of an animal/vegetal polarity in the plasma membrane of the embryo
of Nassarius, possibly related with the polar localization of morphogenetic factors.
The lateral diffusion of the lipid probe in the plasma membrane of the vegetal pole area shows a cell
cycle-dependent
modulation; the highest mean values for D are reached during S-phase (ranging from 7-0 to
7-8x10"^ cm2/sec in the three cycles 9measured),
while at the end of G2-phase mean values for D have
rates in the animal membrane of the embryo are
decreased to a range of (5 0-5-9) xlO~ cm2/sec.9 Diffusion
rather constant (D ranging
from 4-4 to 5-0xl0~ cm2/sec), except for a peak during the S-phase of the first
9
cycle (D = 6-0 xlO" cm^sec).
At third cleavage, four small micromeres are formed
at the animal pole. Both D and MF decrease
9
2
considerably in these cells (D = 2-9 ± 0-2xl0~
cm
/sec, MF = 0-51 ± 0.02) as compared to the
9
2
corresponding macromeres (D = 4-9 ± 0-3X10" cm /sec, MF = 0-78 ± 0-02). The significance of this
sudden change in lateral diffusion of membrane lipids in the micromeres is unknown.
162
Cell asymmetries in development
Control of morphological and physiological cell asymmetries in early
chick development
Claudio D. Stern*, Dept. Anatomy & Embryology, University College London, Gower Street,
London WC1E6BT
Using a combination of ultrastructural, histochemical and physiological techniques it was shown that the
epiblast of the chick embryo at primitive streak stages has defined dorso-ventral polarity. This polarity
manifests itself in the position of nuclei, intercellular junctions, microfilament bundles and the basal lamina
associated with the cells. Physiologically, cells are capable of uni-directional sodium and fluid transport, and
the sodium pumps (NaK-ATPase) are localized towards the basal surfaces of the tissue. As a result of
uni-directional sodium transport, the epiblast maintains a potential across itself amounting to about 20 mV
(ventral side positive).
Experimental application of a potential of similar magnitude but reverse polarity to that measured (about
25 mV, dorsal side positive) leads to rapid and stable reversal of all the morphological and physiological
markers studied. This is not due to simple electrophoresis of these markers across the tissue.
Based on these results, a simple model will be outlined to explain the formation and maintenance of the
embryonic axis, the main factors are postulated to be sodium, calcium and fluid homeostasis by the epiblast
and interactions between the forming mesoderm layer and the overlying epiblast.
Compartmentation of the nucleotides pool in eggs of the polar lobe forming
mollusc Nassarius reticulatus
C. A.M. van Dongen, H. Goedemans, J. Wes, Department of Experimental Embryology, Zoological
Laboratory, State University of Utrecht, Padualaan 8,3508 TB Utrecht, The Netherlands
Early determinative events in differentiation processes during development are controlled by
extranuclear maternally derived morphogenetic determinants. These as a rule are localized in particular
regions of the egg and are parcelled out to particular cell lines during cleavage. In molluscs, in particular the
vegetal region of the egg is an important morphogenetic compartment. In polar lobe forming species, e.g.
Nassarius, this compartment is temporarily set apart during fist cleavage in the form of a cytoplasmic
protuberance, the so-called polar lobe, which can be removed from the egg either by surgical or chemical
methods. In lobeless embryos, differentiation in a number of specific cell lines follows an aberrant, although
as such highly typical, course. The effects of polar lobe removal are strictly reproducible and at present are
well defined. The restricted pattern of developmental deficiences in lobeless embryos is analogous to the
phenotypic expressions of a maternal effect genetic mutant. Polar lobe forming species hence are an
excellent model for studying maternal control of early developmental processes. In an attempt to
characterize the molecular contents of the lobe compartment, we have analyzed the low molecular weight
constituents by means of capillary isotachophoresis (an ultrasensitive micromethod). We have found, that
the lobe is highly enriched in its nucleotides contents, and that its nucleotides spectrum is charcterized by
several prominent compounds in particular (e.g. GTP).
Cell asymmetries in development
Pole cell formation in Drosophila melanogaster
163
R. M. Warn*, L. Smith and A. Warn, School of Biological Sciences, University of East Anglia,
Norwich, NR4 7TJ
The mode of pole cell formation was studied in whole mounts stained with rhodaminyl-lysine-phallotoxin
(RLP), a specific stain for F-actin, and the nuclear fluorescent stain 4-6 diamidine-2-phenylindole
dihydrochloride (DAPI). Polar surface caps were found to emerge first after the eighth cleavage, one cycle
ahead of the somatic caps which form elsewhere, and to divide twice before pole cells formed. The caps,
syncytial surface protrusions, had an F-actin rich cortex, distinct from that surrounding them and an interior
somewhat less brightly stained. The first division cycle of the polar caps was similar to that previously
described for the somatic caps (Warn et al. 1984). Caps were found to swell out and then flatten whilst
continuing to expand. They then split and new caps bubbled out from both ends. Prior to and during
splitting the region of cleavage did not show a distinctive contractile ring structure but became depleted in
F-actin as compared with the forming daughter caps. The second division cycle of the polar caps followed
the pattern of the first until the last stages of mitosis, when a distinctive hoop of RLP staining formed in the
centre of each cap. This moved down with the plasmalemmas. 'Haloes' of RLP staining were then found at
the bases of each forming cell. These became smaller and disappeared. Possible mechanisms of cleavage are
discussed as is the role of cytoplasmic determinants in controlling the organization of regionally localized
F-actin microfilament structures.
WARN, R. M., MAGRATH, M. & WEBB, S. (1984). Distribution of F-actin during cleavage of the Drosophila
syncytial blastoderm. /. Cell Biol. 98, 156-162.