/ . Embryol. exp. Morph. Vol. 34, 3, pp. 687-694, 1975
Printed in Great Britain
687
Mitotic activity of germ cells during normal
development of Xenopus laevis tadpoles
ByKEN-ICHI IJIRI 1 AND NOBUO EGAMI1
From the Zoological Institute, University of Tokyo
SUMMARY
Data on the spatio-temporal pattern of germ cell proliferation in Xenopus laevis tadpoles
were obtained, tracing the germ cells from the cloacal position forward. This spatial pattern
in germ cell distribution and its change during normal development clearly coincided with
histological observations of germ gland development. By application of regression lines to
the analysis of this complex pattern, an interesting conclusion about the mitotic activity of
germ cells was suggested. While the mitotic activity of germ cells before sexual differentiation
shows a regional difference along the germ-cell-containing ridge (GCCR), the doubling
time of sexually differentiated gonia seems to show a uniform value over the whole GCCR.
INTRODUCTION
It is now generally accepted that in Anura primordial germ cells originate in
the endoderm, and segregation of germ cells takes place at an early embryonic
stage (Blackler, 1958; review by Franchi, Mandl & Zuckerman, 1962; Ijiri &
Egami, 1976). Most authors agree that, after segregation, mitotic arrest of the
primordial germ cells occurs until some time after their arrival at the gonadal
area (Bounoure, 1925; Cheng, 1932; Blackler, 1970). One interesting problem
concerns the mitotic activity of germ cells after the resumption of mitosis. The
temporal change in mitotic activity of the germ cells seems to reflect the interactions between the germ cells and some somatic elements. The present paper
tries to focus attention on the proliferation kinetics of germ cells, in relation
to the influences from somatic elements. In this article, the numerical relation
between the number of germ cells and developmental stage was obtained, considering both spatial and sex differences in Xenopus laevis.
MATERIALS AND METHODS
Rearing tadpoles and fixation
To minimize fluctuations in rate of tadpole development, the following procedure was undertaken.Tadpoles of Xenopus laevis were reared from one batch
1
Authors' address: Zoological Institute, Faculty of Science, University of Tokyo, Hongo,
Tokyo 113, Japan.
688
K. IJIRI AND N. EGAMI
of eggs obtained by injecting human chorionic gonadotrophic hormone into
a pair of mature toads (Gurdon, 1966), and were staged after Nieuwkoop &
Faber (1956) (N. & F.). Animals were raised at 20 ± 1 °C at an initial concentration of 20 tadpoles/1, and from stage 45 were fed daily on the diet for goldfish.
At stage 49~, they were placed at a concentration of three tadpoles/1. Animals
were examined daily, and only the most advanced tadpoles were retained for
study.
Up to stage 49~, all animals to be fixed were examined for size, external
characteristics and intestinal revolutions. At the early stages, the intestinal
revolutions served as one of the useful criteria in deciding the developmental
stages. After stage 50, besides the size, external characteristics, especially the
shape of hindlimb-bud, were adopted. From stage 53~, the shape of hindlimb
together with the forelimb became decisive in determining the tadpole developmental stages.
All tadpoles of stages 45-55 were fixed in Bouin's solution. These were
serially sectioned transversely in paraffin at 7 jam thickness, and stained with
Delafield's hematoxylin and eosin.
Germ-cell counting
Direct germ-cell counting was performed by tracing all the germ cells through
serial sections from the cloacal position rostrally until the genital ridges disappeared. Great care was exercised to obtain accurate counts. At stages 45 and
46, because of the mass of yolk platelets, the accuracy of counting was within
an error of one or two cells. Two germ-cell counts were performed for each
tadpole and the average was taken as the number of germ cells of that tadpole.
From stage 47~ to 53, the two germ-cell counts usually coincided in their
number. When they did not, counting was performed until there remained no
doubt about the number obtained. At stage 55, the accuracy of counting was
within an error of about five cells. The counting was performed twice for each
tadpole at this stage, and the average number was adopted as the number of
germ cells.
Sexual difference was judged at stages 53 and 55 from the histological
observations of the germ cell arrangement in the gonads, the shape of the germ
cell nucleus and the existence of cavities in the gonads. The anterior portion of
the gonads showed these characteristics most clearly.
Spatio-temporal pattern
To get a spatio-temporal pattern of germ cell proliferation in a tadpole, the
following procedure was undertaken. Each section was numbered from the
first section which revealed the first posterior germ cell rostrally. Then the germ
cells were traced from that first section rostrally. Whenever a new germ cell
was encountered, its existence was recorded with the corresponding number of
the section which contained the maximum portion of the nucleus, together with
Mitotic activity of germ cells o / X e n o p u s
689
its appearance in the left or right gonad. In the early stages, however, frequently
one could not definitely decide whether a germ cell lay either in the left or right
gonad, since its situation was somewhat ambiguous. In such a case the germ
cell was recorded as situated in the middle portion.
Table 1. Germ cells during the normal development of tadpole*
Average
Days after Developmental
fertilization stage (N. & F.) diameter of
germ eel 1st
(/tm)
7-5
8-5
9-5
12
16
18-25
21-25
26-25
28
32
39
45
46
47"
47+
4950
51
5253"
Number of
embryos
examined
Number of
Number of
germ cells
embryos
(mean ±S.E.) with mitotic
germ cells
10
231 ± 1-7
11
220 ±1-9
10
22-9 ±1-6
11
22-4 ±2-0
30-3 ±1-9
19
10
35-0 ± 3 0
40-7 ±4.2
13
13
55-3 ±6-3
66-9 ±4-5
14
14
1051 ±9-2
114-5 ± 1 4 0
53
8
15(12-23)
\ (?)
6
92-7 ±9-5
1 (<?)
284-1 ±42-4
f(9&<?) 16
55
9
349-7 ±65-3
15 (12-23
] (?)
7
199-9 ±28-8
I (<?)
* All tadpoles were obtained from one batch.
t Range: of diameter is shown inside the parentheses.
26 (20-33)
25 (20-33)
20 (14-27)
20 (12-25)
18 (12-23)
18 (12-23)
20 (12-25)
20 (15-23)
15 (12-18)
.
0
0
0
0
0
0
8
6
4
8
6
2
12
8
4
Regression lines and doubling time
Since resumption of mitosis and the onset of sexual differentiation (Nieuwkoop
& Faber, 1956) are the two important events, the normal development of germ
cells up to stage 55 was investigated by dividing this period into three parts with
these two events as critical points. These three parts were termed states 0,1 and
II, respectively. State 0 may correspond to the mitotically dormant state, state
II to the gonial state, and state I may be regarded as an intermediate period
between the resumption of mitosis and the onset of sexual differentiation.
Analysis of experimental data was performed by applying the following regression lines to these states. Expressing the common logarithm of germ cell
number N as y, namely log10 N = y, the regression line y = ax + b was applied
to states I and II. The regression line for state I was calculated from the data of
stages 49", 50, 51, 52"; state II from stages 53~, 53, 55. Here the abscissa x was
expressed in days after fertilization. alt bi, an and 6 n being the regression
coefficients, the doubling time Td of states I and II can be expressed as:
and
Td(l) = log
respectively.
10
= log 10 2jaLI,
690
K. IJIRI AND N. EGAMI
RESULTS
Number of germ cells during the development of the tadpole
The result of direct counting of germ cells during normal development of
the tadpole is summarized in Table 1. Morphological sex difference was
discernible at stages 53 and 55. Table 1 shows that female germ cell numbers are
60 H
50-
-3000
40-
-2500
e 30-
- 2000
u
u
20-
- 1500 O
10-
- 1000
0-
- 500
10
47-
20
30
Days after fertilization (20 °C)
47+
49" 50
51
52" 53"
53
Developmental stage (N.and F.)
40
55
Fig. 1. Development of body length and the GCCR. The average value of body
length was obtained from 25 tadpo'esfixedfor study at each stage. The measurement
was performed before fixation.
larger than those of male tadpoles, therefore indicating the existence of sexual
difference in germ cell number at these stages.
Spatio-temporal pattern of germ cell number
One of the prominent features of gonadal development is the existence of the
spatial difference, especially the longitudinal difference in its development.
From histological observations, it seems that the development proceeds from
the front backwards.
Mitotic activity of germ cells o/Xenopus
691
Data on the spatio-temporal pattern of "germ cell proliferation were
obtained, tracing the germ cells from the cloacal position rostrally. The tadpoles
examined here were chosen randomly from the batch of Table 1. The length of
gonad in which all the germ cells were contained, i.e. the length from the first
posterior germ cell to the last anterior germ cell, was denoted here as
'germ-cell-containing ridge (GCCR)'.
The results of the germ cell tracing are presented in Figs. 1 and 2. Fig. 1
represents the development of the GCCR and the body length. In this batch of
tadpoles, during the stages examined, body length and the length of the GCCR
both seem to have developed in proportion to the days elapsed. No lateral
spatial difference was observed in germ cell number: at every stage, left and
right gonads contained almost the same number of germ cells.
Fig. 2 shows the average number of germ cells contained in each portion of
the GCCR at each stage of development. The GCCR was divided into 20
segments, described from posterior to anterior as 0-5 %, 5-10%, 10-15 %, and
so on. Every germ cell traced through the serial sections was assigned to its
proper segment of the GCCR. To divide the length of the GCCR of each
tadpole into 20 segments almost impartially, the following procedure was
adopted. Suppose there exist 143 sections from the first posterior cell to the
last anterior cell, then these 143 sections must be divided into 20 segments. The
best way is seven sections to each 17 segments, and eight sections to each of
three segments. This was accomplished by choosing these three segments by
use of tables of random numbers.
As seen in the spatio-temporal pattern of germ cell number shown in Fig. 2,
there exist no striking spatial differences in germ cell distribution at stages 47~
and 47+. However, as development proceeds, one can obviously notice the
prominent spatial difference. The anterior part of the GCCR shows quite a
large number of germ cells, while the posterior part keeps few germ cells. Such
a spatial difference pattern in germ cell distribution, and its change of pattern
during normal development, clearly coincide with histological observation of
germ gland development itself (Fig. 3 A-C).
Analysis of mitotic activity from the spatio-temporal pattern
In the hope of obtaining a relatively simple picture of the mitotic activity of
germ cells through this complex spatio-temporal pattern, the application of
regression lines to each of 20 segments of the GCCR in Fig. 2 was performed.
As for the doubling time of state I, i.e. 77/(1) in each of 20 segments, although
they take somewhat excessively large values at 0-5 %, 20-25 %, and 95-100%,
the rest of the values seem to exhibit a regional difference along the GCCR:
gradually decreasing in their value from 17 days in the posterior to 10 days in
the anterior part.
In contrast to Td(J), the doubling time of state II, i.e. Td(ll) was found to
have a rather constant value over the whole GCCR. For instance, germ cells
44
E M B 34
692
K. IJIRI AND N. EGAMI
20 -i
100
St. 47(10)
10 0
80
60
2 0 -i
20
0
0
20 -
St. 49"
(10)
10 -
120
100
0
80
1 4-0 -.
I 3-0St. 50
(10)
10-
I0
ng
40
ber c
60
*o 2-0 H
I
St. 53
(8)
40
St. 47 +
(10)
10-
0
St. 53
20
S
3
z
40 30 -
St. 51
(10)
20 -
1A A
10 0
h
St. 55
200-
50 -i
40 St. 52"
(10)
30 20-
10 0 -
10 0
0
(Posterior)
50
100%
(Anterior)
0
(Posterior)
Part of GCCR
50
100%
(Anterior)
Part of GCCR
Fig. 2(A). Legend on facing page,
of the female, or oogonia, exhibit a constant Td(IT) of about 4 days, while the
values for spermatogonia take a constant value around 6 days.
DISCUSSION
A spatio-temporal pattern was drawn for the germ cell tracing over the
whole length of the germ-cell-containing ridge (GCCR). So far as the germ
cell count was concerned, there was no difference in the left and the right
gonads. As shown in Fig. 2, SL spatio-temporal pattern of germ cell number
coincided well with histological observations of gonadal development.
Mitotic activity of germ cells o/Xenopus
693
150 -A
100 ..,
St. 53
50 -
40-0 A
I
Z 300 St. 55
200-
100 -
o
I
50
(Posterior)
100%
(Anterior)
Part of GCCR
Fig. 2CB).
Fig. 2. Spatio-temporal pattern of germ cell number. (A) Pattern for the mingled
set of both sexes. Number of embryos examined is shown inside the parentheses.
(B) Sex difference pattern at stages 53 and 55.
Through application of regression lines to the analysis of the complex spatiotemporal pattern obtained above, an interesting conclusion about the doubling
time Td or the cell-cycle time of germ cells was reached. The doubling time of
state I shows a spatial difference in its value, while that of state II shows a
uniform value over almost the whole GCCR. This result may suggest that the
distinction between state I and state II comprises not only a difference of
mitotic rate of germ cells, but also a difference in the nature and degree of the
way they react to the influences from the gonadal somatic elements.
However, this tentative analysis of the doubling time is solely based on germ
cell counts, and neglects several possibilities such as cell migration within the
germ gland itself, unequal rate of germ cell divisions, and so on. Careful and
44-2
694
K. I J I R I AND N. EGAMI
Fig. 3. Spatial difference in the development of gonad along the GCCR. Sections
of ovary from a tadpole at stage 55. (A), (B) and (C) correspond to the posterior,
middle and anterior part, respectively.
elucidative investigations remain to be performed to substantiate the above
statement.
The authors wish to thank Mr K. Kobayashi in Sendagi, Tokyo, for his valuable advice
and encouragement.
REFERENCES
BLACKLER, A. W. (1958). Contribution to the study of germ cells in the Anura. /. Embryol.
exp. Morph. 6, 491-503.
BLACKLER, A. W. (1970). The integrity of the reproductive cell line in the Amphibia. Current
Topics in Developmental Biology 5, 71-87. New York: Academic Press.
BOUNOURE, L. (1925). L'origine des gonocytes et revolution de la premiere ebauche genitale
chez les Batraciens. Annls Sci. nat. (Zool.) 10e ser, 8, 201-278.
CHENG, T. H. (1932). The germ-cell history of Rana cantabrigensis Baird. I. Germ-cell
origin and gonad formation. Z. Zellforsch. mikrosk. Anat. 16, 497-541.
FRANCHI, L. L., MANDL, A. W. & ZUCKERMAN, S. (1962). The development of the ovary
and the process of oogenesis. The Ovary, vol. i, pp. 1-88. London: Academic Press.
GURDON, J. B. (1966). African clawed frogs. Methods in Developmental Biology, pp. 75-84.
New York: Crowell.
IJIRI, K. & EGAMI, N. (1976). A mathematical model for germ cell determination process
and effect of ultraviolet light on the process. 7. theor. Biol. (in the Press).
NIEUWKOOP, P. D. &FABER, J. (1956). Normal table o/Xenopus laevis (Daud.). Amsterdam:
North-Holland Publishing Co.
(Received 2 June 1975)