/. Embryol. exp. Morph. Vol. 41, pp. 23-32, 1977
Printed in Great Britain © Company of Biologists Limited 1977
23
On the differentiation of
prospective ectoderm to a ciliated cell pattern in
embryos of Ambystoma mexicanum
By ULF LANDSTROM 1
From the Department ofZoophysioIogy, University of Umed
SUMMARY
The differentiation of the ectoderm in Ambystoma mexicanum (Harrison stage 26-27) was
examined under in vivo and in vitro conditions by scanning electron microscopy under
different experimental conditions. About one out of three flank epidermal cells was found
to be ciliated in the undisturbed or control embryos. The shape of ciliated cells in the explants
from the animal region was only slightly affected. In no case was it possible to find two
adjacent ciliated cells, implying that these cells prevent the appearance of cilia in the cells
in direct contact. Transformation to ciliated cells is suppressed by hypertonicity but favoured
in a hypotonic medium. The differentiation of epidermis is also dependent upon the synthesis
of RNA and some kind of sulphated glucosaminoglycan, corroborated by the inhibitory
effect of actinomycin and selenate.
The differences between the test series and the controls are discussed with regard to factors
controlling embryonic epidermal differentiation.
INTRODUCTION
By means of vital staining and other techniques it has been shown that the
part of surface ectoderm located between the union of the neural folds in the
median dorsal line and the median ventral line of the embryo, may be regarded
as prospective skin epidermis (Holtfreter, 1936). In a balanced physiological
salt solution the ectoderm isolated from the animal pole at the blastula stage
differentiates into typical epidermal tissue (e.g. Holtfreter, 1931, 1947). This
differentiation capacity of prospective epidermal and prospective medullary
areas is an expression of the state of determination at the time of isolation.
At neurula stage, ciliated cells frequently appear in the outer layer but not in
the underlying cell layers, which are exposed to the internal medium. Inner and
outer layers are originally derived from the same type of cells through tangential cleavage at the early blastula stage. The different cell traits indicate some
kind of fundamental polarity with respect to the factors involved in cell transformation. As an important factor we may mention the polarity in osmotic
1
Author's address: Department of Zoophysiology, University of Umea, S-901 87 Umea,
Sweden.
24
U. LANDSTROM
activity. According to Tuft (1957) the blastocoel and archenteron contents are
hypertonic compared to the outer medium. It has also been shown by Willmer
(1961) that exposure to a dilute medium promotes the transformation from
amoebocytes to epitheliocytes. Thus in protozoa, e.g. Naegleria gruberi, where
this transformation process is reversible, the flagellate-epitheliocyte form is
preponderant in the hypotonic medium.
On the basis of these observations we have tested the effect of different
tonicity on the differentiation of epidermal cilia. Another approach has been
to investigate the differentiation of epitheliocytes in relation to the synthesis of
RNA and sulphated glucosaminoglycans.
MATERIALS AND METHODS
The experiments were carried out on axolotl embryos (Ambystoma mexicanum), obtained from naturally mating adults raised in the laboratory. The
fertilized eggs were kept in 7-5 % frog Ringer at room temperature (23 °C) until
appropriate stages were reached. In vivo analysis of epidermal differentiation
was made on embryos of Harrison stage 26-27. The jelly and the vitelline membrane were removed with forceps after which the embryos were prepared for
steroscan studies as described below.
Isolation of explants. In vitro analyses were made on explants isolated from
embryos at stage 8 (late blastula). The embryos were decapsulated by means of
a pair of watch-maker's forceps and rinsed in sterilized 7-5 % Ringer. Pieces
from the ectoderm layer at the animal pole (about 30 cells) were dissected out
by means of a glass needle and a hair loop. The explants were rinsed and transferred to a Petri dish (diameter 4 cm) filled with 5 ml standard growth medium
containing benzylpenicillin (25 i.u./ml) and streptomycin (25 /Jg/ml). The composition of the balanced saline standard solution (SS) used in the present experiments was described by Barth & Barth (1959). The hypertonic and hypotonic
media were obtained by changing the concentration of salts 1-5 x and 0-5 x
respectively.
About 1 h after dissection the ectodermal explants assumed the shape of
spherical vesicles by fusion of their free ends. The isolated fragments were
reared for 3 days under aseptic conditions at 23 ± 1 °C.
Scanning electron microscopy. Embryos of stage 26-27 and explants cultured
for 3 days were fixed in 3 % glutaraldehyde and 3 % paraformaldehyde in 0-1 M
Tris-buffer, pH 7-3, for 24 h at 23 °C. After being rinsed in several changes of
buffer and distilled water they were dehydrated in a graded ethanol series (30 %,
50 % 70 %, 80 % 90 %, 95 % and absolute ethanol). The ethanol was then substituted with amylacetate in a graded series 25 %, 50 %, 75 % to 100 % amylacetate. The explants were dried in an Anderson critical point drying apparatus,
coated with a layer of gold and examined by scanning electron microscopy at
a beam voltage of 15 kV over the magnification range of x 30-2000.
Ciliated cell pattern in Ambystoma embryos
25
Autoradiography. In an experimental series aimed to analyse the RNA synthesis of control explants and explants grown in actinomycin, hypertonicity and
selenate, tritiated uridine was added to the culture medium. The final concentration of [5-3H]uridine was 0-04 nM and the radioactivity 1 /tCi/ml. The incorporation of radioisotope was allowed to proceed for 24 h at three different
intervals, during the first, second or third day of culturing. At the end of each
incorporation period the explants were rinsed in growth medium without tritiated uridine. After 3 days in culture the aggregates were placed in a Ca-free
medium for dissociation (Barth & Barth, 1959). The disaggregated cell cultures
were fixed in Carnoy for 1 h and postfixed in glutaraldehyde (2% in 0-1 M
phosphate buffer, pH 7-8) for 10 h. After fixation the cell cultures were plated
out on slides by means of a Spemann micropipette. The cells were dried on the
slides, covered with a stripping film Kodak AR-10, and finally exposed for
3 weeks (Appleton, 1972). The number of grains per nucleus was counted in
selected samples of cells (Landstrom, Lovtrup-Rein & Lovtrup, 1976). Background counts were determined on similarly incubated cells, but in absence of
[5-3H]uridine.
RESULTS
In vivo and in vitro differentiation of prospective ectoderm
to ciliated epidermal ceils
The ectodermal ciliary pattern of the normal embryo (stage 26-27) is shown
in Fig. 1. The ciliated cells are apparently distributed in different densities but
present in almost every surface region. It is also to be noticed that, when
analysed in closer detail, cells with differentiated cilia are always surrounded
by non-ciliated 'neighbours'. Figure 2, a higher magnification of the flank
region below somites 4-8, represents the highest population density (one in
three). Of 162 cells counted in this region, 54 were found to be ciliated. A
schematic model of the distribution of hexagonal cells in this area is shown in
Fig. 3.
After 3 days in culture the epidermis from the animal region also contains
many ciliated cells, distributed all over the explant (Fig. 4). The shape of the
ciliated cells, as well as the non-ciliated ones, does not differ much from those
which appear in the normal embryo of the same stage. The regular alignment
and the cell boundaries of the epidermal cells are less pronounced under in
vitro conditions and the individual cells have a slightly convex contour compared to the flat surface of the intact embryo. The distribution of ciliated cells
varies between different explants, probably because they were isolated from
different areas in the animal pole. As previously mentioned the distribution also
varies in different areas of the embryo. The ciliated cells in the explants are
always surrounded by non-ciliated ones (Fig. 5).
26
U. LANDSTROM
Fig. 1. Whole larva, stage 26-27, with ciliary tufts distributed over the body, x 30.
Fig. 2. Epidermal area from the flank of the larva (the area shown by the rectangle
in Fig. .1) showing the 1 in 3 distribution of ciliated cells on a flat surface with distinct cell boundaries, x 550.
Ciliated cell pattern in Ambystoma embryos
27
Fig. 3. Diagram showing the theoretical 1 in 3 distribution of ciliated
cells in a hexagonal arrangement (ciliated cells black).
Suppression of the differentiation of ciliated epidermal cells
From macroscopic observations with the stereomicroscope it was noticed
that rotation was almost completely suppressed in explants grown in increased
ionic concentration (1-5 x standard solution). This observation was confirmed
by the steroscan study. In contrast to the control explants grown in normal
medium, there are no ciliated cells in the explants cultured in hypertonic medium
(Fig. 6). These cells are also curved convexly like hemispheres and the individual cells are separated from their 'neighbours' by large grooves (Fig. 7). The
formation of an atypical non-spherical aggregate and the reduced packing of
cells may indicate that the intercellular adhesivity is reduced.
A similar result was obtained if the explants were cultured in actinomycin or
selenate. In the presence of actinomycin the cells were completely scattered,
while with selenate the effect was similar to that of the hypertonic cultures.
Selenate has been reported to inhibit the sulphatation of glucosaminoglycans
(Wilson & Bandurski, 1958). Hypotonicity (0-5 x standard solution), on the
contrary, did not suppress the differentiation of ciliated cells (Table 1).
The difference between the incorporation of uridine in the nuclei of animal
epidermal control cells and inhibited cells is shown in Table 2. The difference
between controls and the actinomycin-inhibited cells becomes quite significant
after the first day (P < 0-05). Selenate and hypertonicity, on the contrary, have
hardly any effect on RNA synthesis (P > 0-1).
DISCUSSION
The present investigation using the scanning electron microscope confirms
that the animal blastula cells are determined to form ciliated epidermal structures. Under appropriate conditions only small differences can be observed in
the surface ultrastructure of epidermal cells differentiated under in vivo and in
vitro conditions. The appearance of cilia does not seem to be affected by isolation. From the fact that isolated cells may gradually assume new diversification patterns, it is possible to conclude that the responsible induction occurs
28
U. LANDSTROM
Fig. 4. Three-day culture of an animal explant, grown in a balanced medium,
showing ciliated cells in a spherical aggregate, x 220.
Fig. 5. Cell surface from an animal explant grown in a balanced medium showing
the distribution of ciliated cells, x 880.
Ciliated cell pattern in Ambystoma embryos
29
Fig. 6. Three-day culture of an animal explant grown in a hypertonic medium,
showing an unspherical aggregate, x 220.
Fig. 7. Cell surface from an animal explant grown in a hypertonic medium, showing
unciliated convexly curved cells and large intercellular grooves, x 880.
EMB 41
30
U. LANDSTROM
Table 1. Normal and suppressed differentiation of animal explants showing
the effect of tonicity, actinomycin and selenate at different concentrations
(SS = standard solution. Only moving explants were counted as being ciliated.)
Number of
explants
Ciliated explants
Culture medium
0-5 xSS
1-5 xSS
SS +10 /*M actinomycin
SS +1 /*M actinomycin
SS + 01 JAM actinomycin
SS + 10 mM selenate
SS +1 mM selenate
SS + 01 mM selenate
225
310
310
50
50
50
50
50
50
71
75
7
—
SS
(%)
12
38
—
30
56
Table 2. Incorporation of [5-3H]uridine in the nuclei of cells of animal control
explants and explants cultured in actinomycin, selenate and hypertonic medium
(SS = standard solution.)
Period for
incorporation (days)
Grains per nucleus
(mean ± S.E.)
SS
SS
0-1
1-2
SS
2-3
1-5 xSS
1-5 xSS
1-5 xSS
Actinomycin 10 [im
Actinomycin 10 /*m
Actinomycin 10 /(m
Selenate 10 mM
Selenate 10 mM
Selenate 10 mM
Background
0-1
1-2
2-3
0-1
1-2
2-3
0-1
1-2
2-3
0-1
:4±3
77±4
100 ±5
56 ±3
66±3
92±4
20±l
9+1
9±1
50±4
69±3
89±4
5±1
Culture medium
spontaneously in the individual cells. We may call it self-transformation or selfdifferentiation through endogenous induction.
So far, the present results are in agreement with what has previously been
shown by Billett & Courtenay (1973) and Grunz, Multier-Lajous, Herbst &
Arkenberg (1975). Recent scanning electron microscope analyses by Lofberg
(1974) on the larvas of Ambystoma mexicanum have shown that the ciliated
cells are distributed in different densities over almost the whole embryo. We
also found that the regular alignment of ciliated cells varies in different areas of
the larva. At stage 26-27, the highest density of ciliated cells is found in the
Ciliated cell pattern in Ambystoma embryos
31
flank region below the somites 4 and 8, where one in three of the epidermal cells
is ciliated. Tn no area was it possible to find two adjacent ciliated cells. Apparently this cannot be incidental but implies that some kind of contact inhibition is
involved. Ciliated cells may appear spontaneously in the outer epidermal layer
but not in direct contact with each other. They are always separated by one or
several non-ciliated cells. If so, the one in three distribution of ciliated hexagonal
cells is also the theoretically highest possible.
The fact that the cells exposed to the external medium often become epitheliocytes, whereas many cells in the interior are amoebocytes, may indicate
that the exposure to an external hypotonic medium favours the transformation
to epidermal cells (Lovtrup, 1974). In most embryos the formation of epitheliocytes is almost simultaneous in the superficial cells, suggesting the involvement
of a normal environmental polarity. Our findings support the contention that
contact with a hypotonic external medium may further the transformation to
ciliated epidermal cells. The hypertonic internal medium of the blastocoel is
important for the properties of amoebocytes.
A comparable involvement of the environmental medium in cell transformation has recently been shown by Monroy, Baccetti & Denis-Donini (1976).
In this scanning electron microscopic investigation it was shown that the outer
cells in the blastopore region have their exposed surface covered with microvilli.
The shape of the cells, however, changes once they have gone through the
blastoporal lip and become part of the archenteron roof or the archenteron
floor. The cells round up, lose their mutual adhesiveness and they also become
free of microvilli. According to Monroy et al. (1976), translocation from an
outer position in contact with the external medium (perivitelline fluid), to the
new environment of the blastocoelic fluid, alters the molecular organization and
therefore also the properties of the cell membrane. It may also show that the
microvillus polarization, determined by the cytoskeleton of the microfilaments
in the hyaloplasm, is labile and dependent upon external influences.
We could not measure any significant effect of tonicity on the amount of
RNA synthesized. That RNA synthesis is involved in the diversification process
is corroborated by the fact that actinomycin inhibits the appearance of ciliated
cells. This is in agreement with a recent investigation by Yoshizaki (1976),
where it is also shown that the DNA-dependent RNA synthesis, involved in
cilogenesis, occurs during a limited period at stage 13 (in Rana japonica).
Rather little is known about the mechanism responsible for the appearance
of epidermal ciliated cells. It has been suggested that one of the changes, leading
from amoebocyte to epitheliocyte, is concerned with cell adhesivity. The
chemical basis for this adhesivity is not definitely established, but there is some
reason to presume that the adhesive substances are sulphated glucosaminoglycans. When these are present in the cytoplasma and the surface coat, different
cell properties may be obtained. Residing in the surface coat they may thus be
responsible for the characteristic trait of epitheliocytes, namely their formation
3-2
32
U. LANDSTROM
of cell aggregates. When present in the cytoplasma the adhesive glucosaminoglycans may act as a cement between the tensible linear proteins (microfilaments and microtubules), and might thereby be involved in the formation of
microvilli and cilia (Porter, 1966; L0vtrup, 1974). The inhibitory effect of
selenate supports these contentions and suggests that some sulphated glucosaminoglycan substance or substances are involved in the transformation of
epidermal tissue. Since selenate interferes with the sulphate-transferring system
it has no or little effect on RNA synthesis.
REFERENCES
T. C. (1972). Stripping film autoradiography. In Autoradiography for Biologists
(ed. P. B. Gahan), pp. 33-49. London: Academic Press.
BARTH, L. G. & BARTH, L. J. (1959). Differentiation of cells of the Rana pipiens gastrula in
unconditioned medium. /. Embryol. exp. Morph. 7, 210-222.
BIIXETT, F. S. & COURTENAY, T. H. (1973). A steroscan study of the origin of ciliated cells in
the embryonic epidermis of Ambystoma mexicanum. J. Embryol. exp. Morph. 29, 549-558.
GRUNZ, H., MULTIER-LAJOUS, A.-M., HERBST, R. & ARKENBERG, B. (1975). The differentiation of isolated amphibian ectoderm with or without treatment with inductor. Wilhelm
Roux Arch. EntwMech. Org. 178, 277-284.
HOLTFRETER, J. (1931). t)ber die Aufzucht isolirter Teile des Amphibienkeimes. II. Ziichtung
von Keimen und Keimteilen in Salzlosung. Arch. EntwMech. Org. 124, 404-466.
HOLTFRETER, J. (1936). Regionale Induktionen in xenoplastisch zusammengesetzten Explantaten. Wilhelm Roux Arch. EntwMech. Org. 134, 466-550.
HOLTFRETER, J. (1947). Changes of structure and kinetics of differentiating embryonic cells.
/. Morph. 80, 57-92.
LANDSTROM, U., LOVTRUP-REIN, H. & L0VTRUP, S. (1976). On the determination of the
dorso-ventral polarity in the amphibian embryo. Suppression by lactate of the formation
of Ruffini's flask-cells. /. Embryol. exp. Morph. 36, 343-354.
LDVTRUP, S. (1974). Epigenetics. London: John Wiley.
LOFBERG, J. (1974). Preparation of amphibian embryos for scanning electron microscopy of
the functional pattern of epidermal cilia. ZOON 2, 3-11.
MONROY, A., BACCETTI, B. & DENIS-DONINI, S. (1976). Morphological changes of the surface
of the egg of Xenopus laevis in the course of development. III. Scanning electron microscopy
of gastrulation. Devi Biol. 49, 250-259.
TUFT, P. H. (1957). Changes in the osmotic activity of the blastocoel and archenteron contents during the early development of Xenopus laevis. Proc. R. phys. Soc, Edin. 26, 42-48.
PORTER, K. R. (1966). Cytoplasmic microtubules and their functions. In Principles of Biomolecular Organization (ed. G. E. W. Wolstenholme and M. O'Conner), pp. 308-356.
London: Churchill.
WILLMER, E. N. (1961). Amoeba flagellate transformation. Expl Cell Res., Supp. 8, 32-46.
WILSON, L. G. & BANDURSKI, R. S. (1958). Enzymatic reactions involving sulfate, sulfite,
selenate and molobdate. /. biol. Chem. 233, 975-981.
YOSHIZAKI, N. (1976). Effect of actinomycin D on the differentiation of hatching gland cell
and cilia cell in the frog embryo. Develop., Growth and Different. 18 (2), 133-143.
APPLETON,
(Received 3 January 1977)