/. Embryol. exp. Morph. Vol. 61, pp. 1-13, 1981
Printed in Great Britain © Company of Biologists Limited 1981
\
A definite number of aphidicolin-sensitive
cell-cyclic events are required for acetylcholinesterase
development in the presumptive muscle
cells of the ascidian embryos
ByNORIYUKI SATOH1 AND SUSUMU IKEGAMP
From the Department of Zoology, Kyoto University,
and Asamushi Marine Biological Station
SUMMARY
In order to determine whether or not a crucial number of DNA replications are prerequisite
for cellular differentiation, we have studied development of a tissue-specific enzyme, muscle
acetylcholinesterase (AChE) in the presumptive muscle cells of cleavage-arrested ascidian
embryos. Embryos were cleavage-arrested with cytochalasin B (an inhibitor of cytokinesis)
and aphidicolin (an inhibitor of DNA synthesis). The 64-ceIl-stage embryos which had been
permanently cleavage-arrested with cytochalasin B developed AChE in all the eight presumptive muscle cells, but the same stage embryos which had been prevented from undergoing further divisions by simultaneous treatment with aphidicolin and cytochalasin did not
produce AChE at all. Cytochalasin-arrested 76-cell-stage embryos were able to differentiate
AChE in the ten presumptive muscle cells, while aphidicolin-cytochalasin-arrested 76-cell
stages in as many as four cells. The early gastrulae which had been arrested with cytochalasin
B produced AChE in all the sixteen presumptive muscle cells, while the same stage embryos
arrested with aphidicolin and cylochalasin in as many as twelve cells. Cytochalasin-arrested
late gastrulae developed AChE in twenty blastomeres, while aphidicolin-cytochalasinarrested late gastrulae in eighteen cells.
The presumptive muscle cells at these four stages consist of those of three different (seventh,
eighth, and ninth) generations, and the relative positions of the blastomeres in the cleavagearrested embryos remained fixed. Judging from the relative positions of the blaslomeres, the
AChE-producing cells in aphidicolin-cytochalasin-arrested embryos were always at eighth or
ninth generation, while those with no AChE activity were certainly at seventh generation.
Based on these findings it was supposed that aphidicolin-sensitive cell-cyclic events, presumably DNA replication, are closely associated with AChE development and that the eighth
cleavage cycle may be 'quanta!' for the histospecific enzyme development.
INTRODUCTION
At the time a cell line is segregated it is made up of a small number of cells,
far smaller than the number of cells eventually to constitute the final tissue or
organ. Hence the lineage cells are bound to divide a number of times before the
1
Author's address: Department of Zoology, Faculty of Science, Kyoto University,
Kyoto 606, Japan.
a
Author's address: Department of Applied Biochemistry, Horoshima University, Fukuyama 720, Japan.
2
N. SATOH AND S. IKEGAMI
phenotypic character of the differentiated state is overtly expressed. This raises
the important question as to whether or not the lineage cells have to undergo
a definite number of DNA replication cycles or a definite number of cell
divisions in order to express their differentiated condition. Since the answer to
this question may have important implications for the interpretation of the
regulatory mechanisms of gene expression in differentiation, the roles and
significance of DNA replication in cellular differentiation are worthy of intense
scrutiny (Holtzer, Weintraub, Mayne & Mochan, 1972; Rossi, Augusti-Tocco
& Monroy, 1975).
Acetylcholinesterase (AChE) is a tissue-specific enzyme of the muscle cells of
the tail of the developing ascidian embryos (Durante, 1956; Whittaker, 1973;
Ohmori & Sasaki, 1977; Meedel & Whittaker, 1979). AChE activity is first
detected histochemically and biochemically in the presumptive muscle cells of
the peurula, and the enzyme activity increases dramatically with development
time (Whittaker, 1973; Meedel & Whittaker, 1979). An inhibitor of protein
synthesis (puromycin) prevents the occurrence of AChE activity in embryos
treated continuously with it from the neural-plate stage onwards, while those
treated continuously with it from the neurula stage onwards develop slight
traces of enzyme activity (Whittaker, 1973, 1977; Meedel & Whittaker, 1979;
Satoh, 1979a). Since this puromycin-sensitivity period coincides with the time
that the enzyme is first detected, embryos probably begin to synthesize the
enzyme at this time (cf. Fig. 1). Actinomycin D inhibits the development of
AChE in embryos reared in the drug from the early gastrula stage onwards,
while those from the late gastrula stage onwards eventually develop a low level
of the enzyme activity (Whittaker, 1973, 1977; Meedel & Whittaker, 1979;
Satoh, 1979 a). This result suggests that new RNA synthesis, presumably including mRNA synthesis, which begins the early and late gastrula stages is
needed for AChE development (Fig. 1). Since embryos which are permanently
cleavage-arrested with cytochalasin B or with colchicine can develop the enzyme,
neither cytokinesis of the blastomeres nor the so-called nuclear division is
required for AChE development (Whittaker, 1973; Satoh, 1979 a; Satoh &
Ikegami, 1980).
As to the significance of DNA synthesis, aphidicolin (an inhibitor of DNA
synthesis) prevents the development of AChE in embryos treated continuously
with it from the 64-cell stage onwards, but those from the 76-cell stage onwards
eventually differentiate the enzyme, suggesting that DNA replications might be
prerequisite for the histospecific enzyme development (Satoh & Ikegami, 1980;
Fig. 1). The aim of the present study is to determine whether or not a crucial
number of DNA replication cycles are needed for the enzyme development.
Fortunately, muscle AChE development of ascidian embryos offers a very advantageous experimental system for studying this question. The presumptive
muscle cells of the gastrula consist of those of three different generations. If, in
the gastrulae which are prevented from undergoing further DNA replication by
•+•
(2) (2) (4)
(6)
(10)
" 1:
(8)
| 8-cell | 32-cell 64-cell 76-cell
4-cell 16-cell
•+•
(16)
early
gastrula
(16)
(20)
middle late
gastrula gastrala
•+•
10
neural
plate
II
neural;:
12
13
I
tail-bud
Acetylcholinesterase
synthesis
30
Hatching
H
Fig. 1. Diagram summarizing acetylcholinesterase development in the presumptive muscle cells of the ascidian embryo.
(Upper) Relationship between the time of first synthesis of acetylcholinesterase, its actinomycin D-sensitivity period,
aphidicolin-sensitivity period and the embryonic stages of Halocynthia roretzi. (Middle) The number of presumptive
muscle cells examined in cleavage-arrested embryos. (Lower) The timing and frequency of cell divisions in the muscle
lineage blastomeres (left half embryo). The division of B8-15 to B9-29 and B9-30 is arbitrary. Vertical dotted lines indicate
the presumptive muscle cells at the four stages subjected to the present inquiry. Constructed according to data of
Conklin (1905), Ortolani (1955), Whittaker (1973), Satoh (1979a, b), Satoh & Ikegami (1980) and the present study. The
lineage of muscle cells is discussed in the text.
-CQ
(2)
Number of
presumptive
muscle cells
Lineage of
muscle cells
(half embryo)
2-cell
Embryonic
stage
Developmental
time(h)
Actinomycin Dsensitivity period
•Aphidicolin-sensitivity period
'
5"
a
8"
•S,
i
Co
4
N. SATOH AND S. IKEGAMI
continuous treatment with aphidicolin, the presemptive muscle cells of a certain
generation do not develop AChE activity, but the cells of the next generation
produce the enzyme, then the critical number of DNA replications which are
inevitable for the enzyme synthesis must be determined. In other words, if
AChE development is closely associated with a definite number of DNA
replication cycles, both presumptive muscle cells with AChE activity and those
without AChE activity will appear in one gastrula which is treated with aphidicolin.
MATERIALS AND METHODS
Embryos. Halocynthia roretzi (Drasche) was obtained during two breeding
seasons (November through April) in the vicinity of the Marine Biological
Station of Asamushi, Aomori, Japan. Naturally spawned eggs, about 280 fim
in diameter, were fertilized artificially and reared in filtered sea water at
15±0-2°C using a constant temperature water bath. Only batches in which
cleavage occurred in more than 95 % of the eggs were used. Development of
eggs from different animals fertilized at the same time is essentially synchronous.
The timing of the developmental stages at 15 °C is shown in Fig. 1.
Enzyme histochemistry. Acetylcholinesterase activity was detected histochemically in whole embryos by the direct-colouring thiocholine method of
Karnovsky & Roots (1964). During Halocynthia embryogenesis histochemical
staining of AChE activity was first detected at 12 h of development, and staining
intensity of the cells increased dramatically with development time (Satoh,
1979 a). Since enzyme activity in cleavage-arrested embryos was examined
usually at 26-28 h of development, the distinction between occurrence and nonoccurrence of AChE reaction was clear enough to exclude the possibility of
misjudgement (Fig. 2). In addition, because the presumptive muscle cells of the
gastrulae are large, about 50 /tm in diameter, the number of AChE-producing
blastomeres could be counted exactly using a dissecting microscope.
Cleavage inhibition. Cytochalasin B (Aldrich Chem. Co.) and aphidicolin
were used as cleavage inhibitors. Cytochalasin B (2 /^g/ml) completely prevents
cytokinesis of the blastomeres, but the nucleus in the cell of cytochalasintreated embryos divides in good synchrony with that of normal embryos.
Aphidicolin at 2 /tg/ml blocks ascidian development, presumably due to inhibition of DNA synthesis by interfering with the activity of DNA polymerase-a
(Ikegami et al. 1978). But aphidicolin allows one more cleavage in eggs kept in
aphidicolin and stops divisions thereafter, no matter how soon a cleavage it is
applied (Satoh & Ikegami, 1980). In cytochalasin-arrested embryos the relative
positions of the blastomeres remain fixed. The presumptive muscle cells of
aphidicolin-arrested gastrulae, however, do not remain stationary as they do in
cytochalasin-arrested gastrulae, but tend to migrate the inside of an embryo.
Therefore, it was difficult to count precisely the number of AChE-producing
blastomeres in aphidicolin-arrested embryos (Satoh & Ikegami, 1980). In this
Significance of DNA replication for cell differentiation
5
study, to get rid of this difficulty, aphidicolin was always used together with
cytochalasin B.
Twelve separate experimental series of total 829 embryos were examined.
RESULTS
A ChE development in embryos cleavage-arrested with cytochalasin B and frequency
of cell divisions in the muscle lineage blastomeres
Whittaker (1973) has successfully shown that ascidian embryos which are
permanently cleavage-arrested with cytochalasin B differentiate AChE activity
only in their muscle lineage blastomeres. The maximum numbers of AChEproducing blastomeres at each cleavage-arrested stage are one at 1-cell stage,
two at 2-celI stage, two at 4-cell stage, two at 8-cell stage, four at 16-cell stage,
six at 32-cell stage, and eight at 64-cell stage, respectively. Applying this method
to the later stages, the numbers of presumptive muscle cells at the 64-cell stage,
76-cell stage, early gastrula stage, and late gastrula stage were determined
respectively. As shown in Table 1, all of cytochalasin-arrested 64-cell and later
stages developed the enzyme in some blastomeres. Cytochalasin-arrested 64-cellstage embryos produced AChE usually in eight blastomeres (Figs 2a, 3a), and
in arrested 76-cell stages as many as ten blastomeres could be found differentiating the enzyme (Figs 2 c, 3 b). Cytochalasin-arrested early gastrulae usually
developed AChE in sixteen blastomeres (Figs 2e, 3c), and arrested late gastrulae
formed the enzyme in up to twenty blastomeres (Figs 2g, 3d).
The cell lineage for larval muscle-cell development in ascidian embryos has
been studied (Conklin, 1905; Ortolani, 1955; Mancuso, 1969; Whittaker, 1973).
Based on Conklin's description (1905), Ortolani's correction (1955) and the
present result, the timing and frequency of cell divisions in the muscle lineage
blastomeres are summarized in Fig. 1. As is noticeable in Fig. 1, eight presumptive muscle cells of the 64-cell-stage embryo are all at the seventh generation, counting the unsegmented egg as the first generation according to Conklin
(1905). The presumptive muscle cells of the 76-cell-stage embryo, early gastrula
and late gastrula, however, consist of those of different generations; i.e. the
76-cell-stage embryo has ten presumptive muscle cells of two different (seventh
and eighth) generations, while the early and late gastrulae those of three different
(seventh, eighth, and ninth) generations.
An early gastrula which contains 16 presumptive muscle cells of three different
generations is shown in a scanning electron micrograph (Fig. 4; Satoh, 19796).
AChE development in embryos cleavage-arrested by simultaneous treatment with
;* '
aphidicolin and cytochalasin B
The 64-cell-stage embryos which had been permanently cleavage-arrested
with aphidicolin and cytochalasin did not develop AChE activity at all (Table 1,
Figs 2b, 3d).
N. SATOH AND S. IKEGAMI
100 urn
Significance of DNA replication for cell differentiation
7
Table 1. Acetylcholinesterase development in embryos cleavage-arrested with
cytochalasin B or by simultaneous treatment with aphidicolin and cytochalasin B
Cytochalasin B (2 /*g/ml)
+ aphidicolin
Arrested
embryonic stage
Cytochalasin B
(2/*g/ml)
2 [igfm\
10/*g/ml
64-cell
76-cell
Early gastrula
Late gastrula
75/75(100%)
81/81 (100%)
78/78(100%)
76/76(100%)
0/84(0%)
27/61(44%)
59/59(100%)
32/32(100%)
0/86(0%)
26/74(35%)
76/76(100%)
47/47(100%)
The 76-cell-stage embryos treated with these drugs, however, developed
a distinct enzyme reaction in about 40 % of the embryos, although the number
of reacting embryos varied from one batch of eggs to another (Table 1, Fig. Id).
There was a range in the number of enzyme-producing blastomeres, but about
a half of the reacting embryos produced four enzyme-containing blastomeres
(Figs Id, 3b). The four AChE-producing cells usually consisted of two clusters
of blastomeres, each with two cells, apart from each other (Fig. Id).
All of the cleavage-arrested embryos in the early and late gastrula stages
developed AChE activity (Table 1), but the number of AChE-producing
blastomeres in aphidicolin-cytochalasin-arrested gastrulae was quite different
from that of cytochalasin-arrested gastrulae (Fig. 3 c, d). In most of aphidicolincytochalasin-arrested early gastrulae twelve blastomeres could be found
differentiating AChE (Figs If, 3 c). The twelve blastomeres, almost without
exception, consisted of two rows of blastomeres, each containing six cells,
between which several cells with no AChE activity were present (Fig. If).
Aphidicolin-cytochalasin-arrested late gastrulae usually produced AChE in as
many as eighteen blastomeres (Figs 2h, 3d).
FIGURE 2
Acetylcholinesterase (AChE) development in the presumptive muscle cells of
cleavage-arrested embryos, (a, c, e, and g) AChE activity in cytochalasin-arrested
embryos. Eight blastomeres of the arrested 64-cell stage (a), ten cells of the arrested
76-cell stage (c), sixteen blastomeres of the arrested early gastrula (e), and twenty cells
of the arrestedlate gastrula (g), respectively, develop AChE activity, (b, d,f, and h)
AChE development in embryos cleavage-arrested by simultaneous treatment with
aphidicolin and cytochalasin B. The arrested 64-cell stage does not develop AChE
activity (b). However, four cells of the arrested 76-cell stage (d), twelve blastomeres
of the arrested early gastrula (/), and eighteen cells of the arrested late gastrula (h),
respectively, produce the enzyme. The four AChE-containing cells of the arrested
76-cell stage consist of two groups of cells apart from each other (d), and the twelve
reacting blastomeres of the arrested early gastrula two clusters of cells between which
some cells of no-AChE activity are always present (/). Embryos which are treated
with aphidicolin and cytochalasin do not become so flattened as those arrested with
cytochalasin.
N. SATOH AND S. IKEGAMI
100 T
80
60
40
20
0
Iu
(a) 64-cell
(b) 76-cell
100
cytochalasin B
80
.10Mg/ml
aphidicolin
60
cytochalasin B
40
I
JLa
6
n
7 8
(c) Early gastrula
100
20
3
100
r
4
8
9
10
(d) Late gastrula
80
80.
1
60
40
60
40
20
20
5
0
10
0
12
14 15 16
16
18
Number of blastomeres containing acetylcholinesterase
1
20
Fig. 3. Frequency of embryos relating to the number of blastomeres
containing acetylcholinesterase.
ISU-16
>-
B9-14
6815
RR-16'
Fig. 4. A scanning electron micrograph of an early gastrula of Halocynthia roretzi
showing that sixteen presumptive muscle cells consist of those of seventh (B7-5 and
B7-6), eighth (B8-15 and B8-16), and ninth (B9-13, B9-14, B9-15, and B9-16)
generations. (From Satoh, 1979 b.)
Significance of DNA replication for cell differentiation
9
Interpretation of the results
Cytochalasin B completely prevents cytokinesis of the blastomeres, but not
nuclear division or DNA replication. Aphidicolin blocks ascidian development,
but allows one more cleavage in eggs kept in aphidicolin and stops divisions
thereafter; i.e. if embryos are treated with aphidicolin soon after '«—l'th
division, 'n'th division occurs but not 'w + l'th division. This may presumably
be due to an overlap in time of '« — l'th mitosis with Vth S-phase. However,
aphidicolin must block 'n + l'th DNA synthesis, because 'w + l'th division is
blocked with it.
The 64-cell stage. As shown in Figs 1 and 5 a, eight presumptive muscle cells
of the 64-cell-stage embryos are all at the seventh generation (i.e. the cells have
divided six times up to this stage). The 64-cell-stage embryos which had been
permanently cleavage-arrested with cytochalasin B developed AChE activity in
the eight blastomeres, while the same stage embryos which had been continuously arrested with aphidicolin and cytochalasin B did not develop AChE
at all. It is probable that seventh DNA replication might be completed in the
aphidicolin-cytochalasin-arrested blastomeres (Fig. 5a). Therefore, the result
seems to indicate that seven times DNA replications are not enough for AChE
development.
The Id-cell stage. Among ten presumptive muscle cells of this stage, B8-7,
B8-7, B8-8, and B8-8 are at the eighth generation, while the other cells are at the
seventh generation (Figs 1, 5b). The 76-cell-stage embryos which had been
arrested with cytochalasin B developed AChE activity usually in ten blastomeres. On the other hand, the same stage embryos which had been arrested by
simultaneous treatment with aphidicolin and cytochalasin produced enzyme in
as many as four blastomeres. Judging from the relative positions of the AChEproducing blastomeres in aphidicolin-cytochalasin-arrested embryos, the reacting cells must be B8-7, B8^7, B8-8, and B8^ (Fig. 5b). These four blastomeres
might have finished eighth DNA replication cycle, while the six other cells
might have completed but seventh DNA replication (Fig. 56). Therefore, it is
likely that eight times DNA replications may lead to develop AChE in the
presumptive muscle cells.
The early gastrula. The early gastrula has sixteen presumptive muscle cells of
three different generations; B7-5, B7-5, B7-6, and B7-6 are on the seventh
generation, B8-15, B8-15, B8-16, and B8-16 on the eighth generation, and
B913,B9-13, B9-14, B9-14, B9-15, B9-15TB916, andB9-16on the ninth generation, respectively (Figs 1, 4, 5 c). The early gastrulae which had been arrested
with cytochalasin B were able to differentiate AChE activity in all these sixteen
blastomeres. If the early gastrulae are prevented from further development by
concomitant treatment with aphidicolin and cytochalasin B, the arrested
embryos form 12 AChE-producing blastomeres. Judging from their relative
positions of the AChE-containing blastomeres in aphidicolin-cytochalasin-
AChE development
I
cytochalasin B
aphidicolin
(b) 76-cell stage
cytochalasin B
+
aphidicolin
cytochalasin B
~>~
\
B7-5
(c) Early gastrula
\
aphidicolin
+
cytochalasin B
Fig. 5. Illustration demonstrating the relationship between the number of DNA replication cycles and acetylcholinesterase (AChE)
development in the presumptive muscle cells at three embryonic stages of the ascidian embryos. The numerals within blastomeres
indicate the supposed number of DNA replications. Blastomeres containing AChE activity are dotted. See the text for details.
I
(a) 64-cell stage
o
C/3
X
H
O
Significance of DNA replication for cell differentiation
11
arrested embryos, it is sure that the presumptive muscle cells of eighth and ninth
generations developed AChE activity, while the cells of seventh generation did
not differentiate the enzyme (Figs 2/, 5 c). The cells of eighth and ninth generations might have completed eighth DNA replication in aphidicolin-cytochalasinarrested embryos, while those of seventh generation seventh DNA replication.
The result strongly suggests that eight times DNA replications are required for
the enzyme development.
The late gastrula. The late gastrulae which had been arrested with cytochalasin
B developed AChE activity usually in as many as twenty blastomeres, while the
same stage embryos which had been arrested with aphidicolin and cytochalasin
showed as many as eighteen AChE-containing blastomeres. Conklin (1905)
described that B7-6 and B7-6 do not divide until at least the late gastrula stage.
If B7-6 and B7-6 in aphidicolin-cytochalasin-arrested embryos do not develop
AChE, thus one can explain the difference in the number of AChE-producing
blastomeres between cytochalasin-arrested embryos and aphidicolin-cytochalasin-arrested embryos.
These results are consistent in implying that eight times DNA replications
may be needed for AChE development in the presumptive muscle cells of the
ascidian embryos.
DISCUSSION
Details of the lineage for larval muscle-cell development in ascidian embryos
have not fully been established yet. Originally Conklin (1905) described that
the blastomeres derived from B6-3 were mesenchyme cells. However, Ortolani
(1955) followed the muscle cell lineage by marking the blastomeres with granules
of coloured chalk and she concluded that the descendants of B6-3 (i.e. both
B7-5 and B7-6) were muscle cells. On the other hand, based on the ultrastructural
characteristics of the presumptive muscle cells such as existence of numerous
mitochondria, Mancuso (1969) asserted that B7-5 was a muscle cell, while B7-6
was a mesenchyme cell. The results of the present study, support Ortolani's
proposal (1955), and the interpretation of the results has been done according
to it.
A significant concept in cellular differentiation is that a 'quantal cell cycle',
proposed by Holtzer and his co-workers (Holtzer et al. 1972, 1975). The basic
tenet of this model is that a unique cell-cyclic event, called a quantal cell cycle,
occurs when a cell becomes programmed from an undifferentiated to a differentiated state. During this quantal cell cycle, extensive changes in the genetic
programme can occur which are needed for differentiation. The decision to differentiate is presumably made during a critical S phase of the cell cycle. The
finding reported here may be a demonstration of a quantal cell cycle for cellular
differentiation during early embryonic development. In addition, the present
results imply that the quantal DNA replication may occur after a definite
number of DNA replications. If so, the time at which the changes in the genetic
12
N. SATOH AND S. IKEGAMI
programme responsible for this differentiation begin, may be counted by the
cycle of DNA replication and determined by a crucial number of DNA replication cycles. Relating to this clock mechanism, Holliday & Pugh (1975) presented
a model that DNA itself may count the number of its replications by the enzymatic modification of specific bases in repeated DNA sequences. However,
details of properties of the clock mechanism are subjects of further studies.
It has been proposed that the eggs of some animal groups have cytoplasmic
determinants localized in particular regions of the egg and that these agents are
segregated by a determinate cleavage pattern into certain cell lineages where
they appear to play a role in programming the differentiation pathways of the
cells (Wilson, 1925; Davidson, 1976). Since the cell lineage studies of ascidian
embryonic development by Conklin (1905) and his direct observations on
segregation during cleavage of visible yellow crescent region of myoplasm into
muscle lineage blastomeres, the muscle cell development of ascidian embryos
may be one of the best-known examples of cytoplasmic determinants for
cellular differentiation. Whittaker's discovery (1973) that cells which showed
AChE activity in cleavage-arrested embryos are always, at each cleavage stage,
the presumptive muscle cells, strongly suggests the presence of cytoplasmic
information for the tissue-specific enzyme development. Therefore, the changes
in the state of the genome during the supposed quantal cell cycle might involve
the interaction of the cytoplasmic determinants to the genome (Davidson &
Britten, 1971; Whittaker, 1973, 1979). Very recently, Whittaker (1980) has successfully induced AChE development in extra cells by changing the distribution
of myoplasm during the third cleavage of the ascidian egg, suggesting that nuclear
lineages are not responsible for muscle AChE development. However, this
result (Whittaker, 1980) does not essentially mean that DNA replication is not
needed for AChE development. At present, we may speculate that muscle AChE
development is as follows: (a) The cytoplasmic determinants for larval musclecell development are localized in particular region of the egg and segregated by
a determinate cleavage pattern into the muscle lineage blastomeres. (b) The egg
counts the time of initiation of differentiation by the cycle of DNA replication,
(c) During the eighth DNA replication cycle, the interaction between the cytoplasmic determinants and the genome occurs which brings about activation of
the genome responsible for muscle cell development, (d) Following the gene
activation, mRNA synthesis related to the protein synthesis occurs from the
early gastrula stage onwards, and AChE synthesis begins from the neurula stage
onwards.
We wish to thank Dr T. Numakunai and all other members of Asamushi Marine Biological
Station of Tohoku University for affording us opportunities to utilize their facilities. Thanks
are also due to Profs V. Mancuso and G. Ortolani of Palermo University for discussing the
lineage for larval muscle-cell development.
This study was supported in part by a Grant-in-Aid from the Ministry of Education,
Science and Culture, Japan.
Significance of DNA replication for cell differentiation
13
REFERENCES
E. G. (1905). The organization and cell lineage of the ascidian egg. J. Acad. nat.
Sci.Philad. 13, 1-119.
DAVIDSON, E. H. (1976). Gene Activity in Early Development, 2nd ed. New York: Academic
Press.
DAVIDSON, E. H. & BRITTEN, R. J. (1971). Note on the control of gene expression during
development. /. theor. Biol. 32, 123-130.
DURANTE, M. (1956). Cholinesterase in the development of Ciona intestinalis (Ascidia).
Experientia 12, 307-308.
HOLLIDAY, R. & PUGH, J. E. (1975). DNA modification mechanisms and gene activity during
development. Science, N. Y. 187, 226-232.
CONKLIN,
HOLTZER, H., RUBINSTEIN, N., FELLINI, S., YEOH, G., CHI, J., BIRNBAUM, J. & OKAYAMA, M.
(1975). Lineages, quantal cell cycles, and the generation of cell diversity. Quart. Rev.
Biophys. 8, 523-557.
HOLTZER, H., WEINTRAUB, H., MAYNE, R. & MOCHAN, B. (1972). The cell cycle, cell lineages,
and cell differentiation. Current Topics in Devi. Biol. 7, 229-256.
TKEGAMI, S., TAGUCHI, T., OHASHI, M., OGURO, M., NAGANO, H. & MANO, Y. (1978).
Aphidicolin prevents mitotic cell division by interfering with activity of DNA polymerase-a.
Nature, Lond. 275, 458-460.
KARNOVSKY, M. J. & ROOTS, L. (1964). A 'direct-coloring' thiocholine method for cholinesterases. J. Histochem. Cytochem. 12, 219-221.
MANCUSO, V. (1969). L'uovo di Ciona intestinalis (Ascidia) osservato al microscopio elettronico. I. 11 cell-lineage. Acta Embryol. Exp. (1969), 231-255.
MEEDEL, T. H. & WHITTAKER, J. R. (1979). Development of acetylcholinesterase during
embryogenesis of the ascidian Ciona intestinalis. J. exp. Zool. 210, 1-10.
OHMORI, H. & SASAKI, S. (1977). Development of neuromuscular transmission in a larval
tunicate. J. Physiol. 269, 221-254.
ORTOLANI, G. (1955). The presumptive territory of the mesoderm in the ascidian germ.
Experientia, 11, 445-446.
Rossi, M., AUGUSTI-TOCCO, G. & MONROY, A. (1975). Differential gene activity and segregation of cell lines: an attempt at a molecular interpretation of the primary events of embryonic development. Quart. Rev. Biophys. 8, 43-119.
SATOH, N. (1979 a). On the 'clock' mechanism determining the time of tissue-specific enzyme
development during ascidian embryogenesis. 1. Acetylcholinesterase development in
cleavage-arrested embryos. J. Embryol. exp. Morph. 54, 131-139.
SATOH, N. (19796). Visualization with scanning electron microscopy of cleavage pattern of
the ascidian eggs. Bull. Marine Biol. St Asamushi, Tohoku Univ. 16, 169-178.
SATOH, N. & TKEGAMI, S. (1980). On the 'clock' mechanism determining the time of tissuespecific enzyme development during ascidian embryogenesis. II. Evidence for association
of the clock with the cycle of DNA replication. J. Embryol. exp. Morph. (submitted).
WHITTAKER, J. R. (1973). Segregation during ascidian embryogenesis of egg cytoplasmic
information for tissue-specific enzyme development. Proc. natn. Acad. Sci., U.S.A. 70,
2096-2100.
WHITTAKER, J. R. (1977). Segregation during cleavage of a factor determining endodermal
alkaline phosphatase development in ascidian embryos. /. exp. Zool. 202, 139-154.
WHITTAKER, J. R. (1979). Cytoplasmic determinants of tissue differentiation in the ascidian
egg. Tn Determinants of Spatial Organization (ed. S. Subtelny & I. R. Konigsberg), pp.
29-51. New York: Academic Press.
WHITTAKER, J. R. (1980). Acetylcholinesterase development in extra cells caused by changing
the distribution of myoplasm in ascidian embryos. /. Embryol. exp. Morph. 55, 343-354.
WILSON, E. B. (1925). The Cell in Development and Heredity, 3rd ed. New York: Macmillan.
{Received 6 May 1980, revised 12 August 1980)