/. Embryol. exp. Morph. Vol. 61, pp. 249-258, 1981
Printed in Great Britain © Company of Biologists Limited 1981
249
Ionic induction of the frog cement-gland cell from
presumptive ectodermal tissues
By NORIO YOSHIZAKI 1
From the Department of Biology, Gifu University, Japan
SUMMARY
Cells of the superficial layer which had been explanted from the presumptive ectoderm
of Rana japonica early gastrulae at stage 10 differentiated into cement-gland cells (CGCs)
when cultured in Barth's solution containing 90-130 mM-NaCl., and into common epidermal
cells and cilia cells when cultured in a solution containing 20-40 mM-NaCl. They failed to
differentiate, however, when cultured in a solution in which NaCl is 15 IHM or lower.
The optimum condition for inducing the differentiation of CGC was stimulating them
with a solution containing 130mM-NaCl for 6-10 h at 18 °C, followed by culturing in a
solution containing 15-40mM-NaCl for 7 days. The greatest ability to react to the CGCinducing stimuli resided in the superficial layer of the presumptive ectoderm of the embryo
at stages 10-11. Under the optimum condition, the total volume of CGCs induced amounted
to about 85% of the explanted tissue. High percentage comparable to this was obtained
with stimulation by KC1, RbCl, sucrose or mannitol.
INTRODUCTION
That cell differentiation is induced and regulated by ions is an idea proposed
by Barth & Barth. After extensively investigating the differentiation of cell
types from amphibian presumptive ectoderm using treatments with various
reagents, they have arrived at the hypothesis that the inducing compounds
have as a common factor an alteration of membrane properties, resulting in
the internal release of inorganic ions to the level necessary for induction
(Barth & Barth, 1969, 1972, 1974). At present, we have no direct evidence
for the ionic induction of cellular differentiation during normal development,
but if one specific type of cell can be induced by a simple, well-known ion, this
technique can be made useful in investigating and inferring the normal process
of differentiation after the induction.
Given this situation, Picard (1975) and Yoshizaki (1979) aimed to produce
an experimental model in which large amounts of cell masses homogeneous
with respect to cell population can be obtained for use in the biochemical
study of embryonic differentiation. Under the optimum condition of each
model, 80-90% of the total explanted tissue differentiated into cement-gland
1
Author's address: Department of Biology, Faculty of General Education, Gifu University,
Gifu 502, Japan.
250
N. YOSHIZAKI
cells (CGCs) in Xenopus (Picard, 1975) and 70% of it differentiated into
hatching-gland cells in Rana (Yoshizaki, 1979) after stimulation by 10 mM-NH4Cl
and 70 mM-LiCl, respectively. The CGC of the amphibian sucker is one of
the useful experimental subjects because its differentiation can be monitored
under the dissecting microscope by the appearance of adhesive substances on
its surface. The present study provides a better method for inducing CGCs
from the presumptive ectoderm by NaCl or related alkali metals or sugars.
MATERIAL AND METHODS
Matured grass frogs, Rana japonica, were purchased from a dealer in Tokyo
during the hibernation period and maintained at 4 °C until use. Fertilized eggs
and embryos were obtained as previously described (Yoshizaki, 1976). The
stages of development were determined according to Tahara (1959). The jelly
coat was digested by immersion in a papain-cysteine mixture (Dawid, 1965)
and the vitelline coat was removed manually in sterile Barth's standard salt
solution (Barth & Barth, 1959; 88mM-NaCl, 1 mM-KCl, 0-8mM-MgSO4,
0-3 mM-Ca(NO3)2, 0-4 mM-CaCl2, 2-3 mM-NaHCO3, 0-2 mM-Na2HPO4,0-3 mMKH2PO4) in which neither serum nor antibiotics were included. All solutions
and instruments used thereafter were sterile.
The technical details were shown in the previous paper (Yoshizaki, 1979).
The explanted portion of the presumptive ectoderm (ca. 0-3 x 0-3 mm) was an
area about an animal pole of a blastula and an area exactly opposite to the
blastopore of a gastrula. Except when otherwise specified, the presumptive
ectoderm from a stage-10 embryo (early gastrula) was used. Explants were
transferred to Ca-free De Boer's solution (HOmM-NaCl, 2-2 mM-KCl, pH 8-0
with NaHCO3) and after 15 min of immersion, the pigmented superficial layer
was separated from the basal layer by pipetting out the solution surrounding
the explants. The explants composed of superficial-layer tissue were then
transferred into test solutions containing various concentrations of sodium
chloride or related metal ions or sugars which were substituted for sodium
chloride in Barth's standard solution. All the stimulating and culturing solutions
were based on Barth's solution. After incubation in the test solutions for
appropriate periods, the explants were washed and cultured in a modified
standard solution containing 20 mM-NaCl for 7 days at 18 °C.
At the end of the culturing, the explants were fixed in Bouin's solution, and
serial paraffin sections, 8 /*m thick, were stained with PAS (periodic acid-Schiff)
after salivary treatment and Mayer's hematoxylin. The cement-gland cell could
be distinguished from the other cell types by the accumulated secretory granules
stained with PAS. The volumes of individual tissues were determined by
multiplying a thickness of the histological sections by an area of them measured
by the point counting planimetry of Hennig & Meyer-Arendt (1963). The
results were given as means ± standard errors.
Ionic induction of frog cement-gland cell
251
Table 1. Differentiation of cement gland cell (CGC) in presumptive ectodermal
explants cultured in Earth's standard solution containing various concentrations
ofNaCl
Concentration
of
NaCl (mM)
No. of
explants
Percent volume
of CGC
Total volume
of explants
(lOVm3)
15
20
40
90
130
11
12
16
16
18
0
3±2
25±6
68±4
64 + 5
26±2
21 ±2
25±1
30±l
26±1
Results are expressed as the mean ± S.E.
Explants were prepared for electron microscopy as described previously
(Yoshizaki & Katagiri, 1975). They were fixed in 5% glutaraldehyde in 0-1 Mcacodylate buffer (pH 7-3) at 4 °C for 3 h, washed in the buffer and postfixed
in 1 % osmium tetroxide in cacodylate buffer for 1 h. After dehydration through
graded series of ethanol, they were embedded in Epon 812, and ultrathin
sections were stained in uranyl acetate and lead citrate and viewed with a
Hitachi HS-8 electron microscope.
RESULTS
I. Differentiating cell types depending on the
concentration of NaCl
When the explants from stage-10 embryos were cultured in modified solutions
for 7 days, they differentiated into three types of cells, viz., cement-gland cell
(CGC), cilia cell (CC) and common epidermal cell (CEC), depending on the
concentration of NaCl in the solutions (Table 1). In the solutions containing
NaCl higher than 90 mM, they differentiated mostly into CGCs. Concentrations
of NaCl higher than 130 mM, however, seemed harmful, since the explants
in the solution containing 150 mM-NaCl dissociated during the culture period.
The percentage of CGC differentiation decreased in the explants in the solution
containing 20-40 mM-NaCl and the culture resulted mostly in the differentiation
of CCs and CECs. Some explants were composed of only CECs when cultured
in the solution containing 20 mM NaCl. When the explants were cultured in
the standard solution containing 15 mM or lower concentration of NaCl, no
differentiation was observed and the cells were filled with yolk platelets. Ultrastructurally, the differentiated cells had few or no yolk platelets after 7 days
culture but did possess each characteristic structure in the form of secretory
granules and secreted matter in CGC (Fig. 1), mucous vesicles in CEC and
cilia in CC (Fig. 2).
252
N. YOSHIZAKI
Ionic induction of frog cement-gland cell
253
IT. Determination of optimum condition to induce CGC by NaCl
(a) NaCl concentration and duration of treatment
Since longer periods of exposure to higher concentrations of NaCl is unfavourable to the viability of the cells, the explants were first treated with the
standard solution containing 130 mM-NaCl for 8 h and then cultured in the
solutions containing decreasing concentrations of NaCl. As shown in Fig. 3,
the percentage of CGC differentiation increased as the concentration of NaCl
in the culture medium decreased, and about 85 % of the CGC differentiation
occurred when the explants were cultured in the solutions containing 15-40 raMNaCl.
Tn the next experiment, the explants were treated with the standard solutions
containing various concentrations of NaCl, ranging from 20 to 150 HIM for
8 h, and cultured in the solution containing 20 mM-NaCl. The highest incidence
of CGC differentiation was obtained when the explants were treated with
130mM-NaCl(Fig. 4).
The effect of the duration of stimulation by 130mM-NaCl is shown in
Fig. 5. The significantly high percentage of CGC differentiation was obtained
when the explants were stimulated for 6-10 h which correspond to an approximate duration of gastrulation.
(b) Temporal and regional differences in the ability to react to NaCl stimulation
Explants of the superficial layer of presumptive ectoderm taken from embryos
ranging from blastulae (stage 8) to late gastrulae (stage 12) were stimulated
by 130 mM-NaCl for 8 h and cultured in the standard solution containing
20 mM-NaCl for 7 days. As seen in Fig. 6, the percentage of CGC differentiation
was low for stages 8 and 9, reached a maximum for stages 10 and 11, and
decreased thereafter.
An attempt was made to define the regional difference with respect to the
ability to react to NaCl stimulation. Several different parts from the superficial
layer of presumptive ectoderm of stage-10 embryos were treated in the manner
mentioned above, but no significant difference appeared (data not shown).
FIGURES 1 AND 2
Fig. 1. Electron micrograph showing cement-gland cells in an explant treated
with the standard solution containing 130 mM-NaCl for 8 h and cultured for 7 days
in the standard solution containing 20 mM-NaCl. These cells are characterized by
the accumulation of secretory granules (SG) and the presence of secreted matter
(arrow) near the cell surface. L, lipid droplet; P, pigment granule; Y, yolk platelet.
x2100.
Fig. 2. Electron micrograph showing common epidermal cells (CEC) and cilia
cell (CC) in an explant cultured for 7 days in the standard solution containing
20 mM-NaCl. The CEC is characterized by the presence of a layer of mucous
vesicles (MV) beneath the apical plasma membrane, x 2100.
Q
EMB 6l
N. YOSHIZAKI
100 \-
50
15 20 40
90 110 130
Concentration of NaCl (ITIM)
20
40
'
90 110 130 150
Concentration of NaCl (mM)
Fig. 3. Frequency of induction of the cement-gland cell (CGC) as a function of
concentration of NaCl in the culture media. Explants (whose numbers are given)
from early gastrulae (stage 10) were treated with standard solution containing
130mM-NaCl for 8h and cultured in the standard solution containing various
concentrations of NaCl for 7 days. The average percent volume of CGC and the
standard error are presented.
Fig. 4. Frequency of induction of the cement-gland cell (CGC) as a function of
concentration of NaCl in the test solutions. Explants from stage-10 embryos were
treated with the standard solution containing various concentrations of NaCl for
8 h and cultured in standard solution containing 20 mM-NaCl.
The next experiment was carried out in order to elucidate the difference in this
ability between superficial and basal layers of presumptive ectoderm. For
convenience in handling, the explants of each layer were dissected from stage-11
embryos, stimulated by 130mM-NaCl for 8 h and cultured in the standard
solution containing 40 mM-NaCl. The results given in Table 2 show that the
basal layer is inferior to the superficial one in its ability to react to the stimulation. The major part of the former explants was occupied by the nerve
cells. Thus the optimum reaction to the CGC-inducing stimuli occurs in the
superficial layer of the presumptive ectoderm of stages-10 and -11 embryos.
III. Induction by various alkali metal ions
or sugars replacing Na+
The explants from stage-10 embryos were stimulated by the chlorides of
alkali metal ions comparable with Na + (K+, Li + , Rb+ and Cs+) or sugars for
8 h. As a buffer system, 5 mM-Tris-HCl (pH 7-8) was substituted for sodium
bicarbonate and phosphates in the test solutions. As shown in Table 3, a high
percentage of CGC differentiation comparable to that by NaCl was obtained
Ionic induction of frog cement-gland cell
255
100
"30
30
E 50
0
1
10
Time of treatment (h)
Fig. 5. Frequency of induction of the cement-gland cell (CGC) by NaCl as a function
of time (h). Explants (whose numbers are given) from early gastrulae (stage 10)
were treated with the standard solution containing 130 mM NaCl for various hours,
and cultured in standard solution containing 20 mM-NaCi for 7 days. The average
percent volume of CGC and standard error are presented.
100 i—
50
8
9
10
11
Stage of development
12
Fig. 6. Frequency of induction of the cement-gland cell (CGC) by NaCl as a
function of developmental stages. Explants from embryos from blastulae (stage 8)
to gastrulae (stage 12) were treated with the standard solution containing 130 mMNaCl for 8 h, and cultured in standard solution containing 20 mM-NaCl.
9-2
256
N. YOSHIZAKI
Table 2. Difference between two layers of presumptive ectoderm in the
ability to react to stimuli inducing cement-gland cells {CGC)
Region
Superficial
Basal
No. of
explants
Total
Percent volume volume of explants
of CGC
(108/*m3)
30
88 ±1
26
23 ±3
Results are expressed as the mean ± S.E.
27 ±1
27 ±2
Table 3. Induction of cement-gland cell (CGC) by 8 h treatment
with various kinds of chlorides or sugars
Concentration
Inductor
Li
Na
K
Rb
Cs
Mannitol
Sucrose
(mM)
No. of
explants
Percent volume
of CGC
35
0
130
130
31
87±3
130
26
86±3
130
32
69±4
47 + 5
130
30
220
31
64±5
220
25
77±3
Results are expressed as the mean ± S.E.
Total volume of
explants
(10 Vm8)
22+1
36±1
29±2
35±1
29±1
24±1
26±2
with stimulation by 130 mM-KCl and RbCl, and 220 mM-mannitol and sucrose.
The percentage was low with stimulation by CsCl. No differentiation of CGC
was observed with LiCl, but the pigment cells differentiated. The explants
dissociated during the culture period when they were stimulated by 130 mMNH4C1.
DISCUSSION
Sequential induction of the cement-gland cell (CGC), cilia cell (CC) and
common epidermal cell (CEC) from the superficial layer of presumptive
ectoderm could be made only by changing the concentration of Na + in the
culture medium. A similar phenomenon was reported by Barth (1965, 1966)
and Barth & Barth (1963) in regard to the induction of the nerve cell, pigment
cell, cilia cell and epithelial cell from the basal layer of presumptive ectoderm
by sucrose, K + , Li + , Mg ++ , Ca ++ or Na + . Attempts to obtain cell masses
homogeneous with respect to cell population have already been made with
CGC in Xenopus (Picard, 1975) and with the hatching-gland cell in Rana
(Yoshizaki, 1979), in which the presumptive ectoderm was stimulated by NH 4 +
and Li + , respectively. Under the optimum condition defined here of 6-10 h
Ionic induction of frog cement-gland cell
257
treatment with the standard solution containing 130mM-NaCl, about 85%
of the total explanted tissue was induced to differentiate into CGC. The exact
mechanism of the action of these inducing compounds on the receptive cells
remains an open question, but that they act to keep the intracellular concentration of the ions at the level necessary for induction of each cell type
(Barth & Barth, 1974) might be one of the possible explanations for the
results. Barth & Barth (1972) have suggested the essential role of intracellular
Na + for normal development based on the finding of a specific increase of
22
Na uptake after the beginning of gastrulation.
The effectiveness of sucrose or mannitol in the induction of CGC might
be explained by the tonicity that it withdraws the water from the cell to elevate
the concentration of intracellular ions to an appropriate level. The tonicity
alone, however, seems insufficient to explain the effects by alkali metal ions
because of the greatness of the difference of inducing ability among used ions
and the occurrence of occasional higher induction by ions than that by sugars.
There is as yet no direct evidence of ionic induction of differentiation in the
normal development of amphibian embryos. Kostellow & Morrill (1968) have
reported that in Rana the intracellular sodium passes from the cells into the
blastocoel fluid during the blastocoel formation, and the blastocoel fluid comes
to contain a high concentration of sodium, about 71 mM which was calculated
by Barth & Barth (1974). The concentration is about 100 mM in Xenopus
(Slack, Warner & Warren, 1973). Thus the concentration of sodium in the
blastocoel fluid of the late blastulae is less than the optimum concentration
for CGC induction (130 mM), but within the range in which some induction
can occur. The sodium concentration of the fluids both in the blastocoel and
the intercellular spaces to which the presumptive cells are exposed will change
with the advance of gastrulation.
Except for Li + , the alkali metal ions used were more or less effective in
inducing CGC differentiation in Rana, whereas they were ineffective in Xenopus
(Picard, 1975). This difference may be due to different potencies of presumptive
ectoderm in two genera. A difference in potency was observed between different
layers of Rana ectoderm, too, the basal layer having a lower potency to react
to CGC-inducing stimuli than the superficial layer.
The yolk platelets of the amphibian embryonic cells are important organelles
as a reservoir of both nutrients and inorganic ions (Morrill, Kostellow &
Murphy, 1971). There is an evidence that yolk-platelet breakdown is correlated
with embryonic induction and cellular differentiation. As Karasaki (1963)
observed, the disappearance of the superficial layer of platelets is one of the
first changes in induced cells. The same situation was true in the present study
in that the differentiated cells had few or no platelets after 7 days culture,
whereas the undifferentiated cells in the standard solution containing 15 mM
or lower NaCl were filled with platelets still possessing the superficial layer.
Remarkable in the latter cells was an abnormal accumulation of ribosome-like
258
N. YOSHIZAKI
particles in a regular arrangement around the nucleus (Yoshizaki, unpublished).
Once stimulated, however, presumptive ectodermal cells can embark on a programme leading to differentiation even in the solution containing 15 mM-NaCl,
as shown in Fig. 3. On the other hand, such cells from embryos later than the
late gastrula stage could continue their development in this solution without
artificial stimulation (Yoshizaki, unpublished). One cause of the arrest of
development and the resultant absence of differentiation must be the lack of
available precursors for the synthesis of protein due to an inadequate intracellular environment for yolk lysis. Analysis of the arrested cells is now under
way to reveal the basic metabolism in amphibian embryonic cells.
REFERENCES
BARTH, L. G. (1965). The nature of the action of ions as inductors. Biol. Bull. 129, 471-481.
BARTH, L. G. (1966). The role of sodium chloride in sequential induction of the presumptive
epidermis of Ranapipiens gastrulae. Biol. Bull. 131, 415-426.
BARTH, L. G. & BARTH, L. J. (1959). Differentiation of cells of the Rana pipiens gastrula
in unconditioned medium. / . Embryol. exp. Morph. 7, 210-222.
BARTH, L. G. & BARTH, L. J. (1963). The relation between intensity of inductor and type
of cellular differentiations of Rana pipiens presumptive epidermis. Biol. Bull. V2A, 125-140.
BARTH, L. G. & BARTH, L. J. (1969). The sodium dependence of embryonic induction.
Devi Biol. 20, 236-262.
BARTH, L. G. & BARTH, L. J. (1972). 22Sodium and *6calcium uptake during embryonic
induction in Rana pipiens. Devi Biol. 28, 18-34.
BARTH, L. G. & BARTH, L. J. (1974). Ionic regulation of embryonic induction and cell
differentiation in Rana pipiens. Devi Biol. 39, 1-22.
DAWID, I. B. (1965). Deoxyribonucleic acid in amphibian eggs. /. molec. Biol. 12, 581-599.
HENNIG, A. & MEYER-ARENDT, J. R. (1963). Microscopic volume determination and probability. Lab. Invest. 12, 460-464.
KARASAKI, S. (1963). Studies on amphibian yolk. 5. Electron microscopic observations on
the utilization of yolk platelets during embryogenesis. / . Ultrast. Res. 9, 225-247.
KOSTELLOW, A. B. & MORRILL, G. A. (1968). Intracellular sodium ion concentration changes
in the early amphibian embryo and the influence on nuclear metabolism. Expl Cell Res.
50, 639-694.
MORRILL, G. A., KOSTELLOW, A. B. & MURPHY, L. B. (1971). Sequential forms of ATPase
activity correlated with changes in cation binding and membrane potential from meiosis
to first cleavage in Rana pipiens. Expl Cell Res. 66, 289-298.
PICARD, J. J. (1975). Xenopus laevis cement gland as an experimental model for embryonic
differentiation. I. In vitro stimulation of differentiation by ammonium chloride. J. Embryol.
exp. Morph. 33, 957-967.
SLACK, C , WARNER, A. E. & WARREN, R. L. (1973). The distribution of sodium and potassium in amphibian embryos during early development. / . Physiol. 232, 297-312.
TAHARA, Y. (1959). Tables of the normal developmental stages of the frog, Rana japonica.
I. Early development (stages 1-25). Jap. J. exp. Morph. 13, 49-60.
YOSHIZAKI, N. (1976). Effect of actinomycin D on the differentiation of hatching gland
cell and cilia cell in the frog embryo. Develop. Growth Different. 18, 133-143.
YOSHIZAKI, N. (1979). Induction of the frog hatching gland cell from explanted presumptive
ectodermal tissue by LiCl. Develop. Growth Different. 21, 11-18.
YOSHIZAKI, N. & KATAGIRI, CH. (1975). Cellular basis for the production and secretion of
the hatching enzyme by frog embryos. /. exp. Zool. 192, 203-212.
{Received 6 May 1980, revised 29 August 1980)