- Email: [email protected]
Pattern regulation in defect embryos of Xenopus laevis
DEVELOPMENTAL
BIOLOGY
101,
410-415 (1984)
Pattern Regulation in Defect Embryos of Xenopus laevis H. KAGEURA Deparhent
of Bib, Received
Fac&y April
AND K. YAMANA
of Science,
12, 198s; accepted
Kmhu
University $8, Fukuoka,
in revised form August
812 Japan
18, 1983
Defect embryos of 24 series were prepared by removing increasing numbers of blastomeres from an 8-cell embryo of Xenopus Latvia They were cultured and their development was examined macroscopically when controls reached a tailbud stage or later. Results show that most of defect embryos of 12 series develop normally, and some of them become normal frogs. Each of these defect embryos contain at least two animal blastimeres, one dorsal, and one ventral blastomere of the vegetal hemisphere. This suggests that a set of these four blastomeres of the three types is essential for complete pattern regulation.
removed manually. The embryos were then transferred into 50% Leibovitz (L-15) medium supplemented with 10% fetal calf serum (L-15 FCS) in a petri dish which had been coated with 2% agar. A blastomere or blastomeres were removed by repeatedly flipping the cleavage furrow with a glass needle. Defect embryos of series 13,14,18, and 23 consisted of two animal and two vegetal blastomeres; one of the former lying on one of the latter. In these cases, the other animal blastomere was put
INTRODUCTION
In our previous isolation experiments on Xenqmhs embryos (Kageura and Yamana, 1983), it was shown that most of the lateral halves at the 2-cell stage develop into normal tailbud embryos, but most or almost all the dorsal and ventral halves of the 4-cell embryos and all the animal and vegetal halves of the g-cell embryos fail in undergoing normal or nearly normal development, giving rise to characteristically abnormal embryos. This fact supports the idea that for complete pattern regulation some of the blastomeres of an early embryo are essential but the others are not. The purpose of the present study is to find out which blastomeres are essential for complete pattern regulation. A blastomere or blastomeres were removed from an &cell embryo, and the resulting defect embryos of 24 series were cultured and their development examined. An additional six series of defect experiments have already been described (Kageura and Yamana, 1983). Results obtained in the present study and in a previous one showed that the blastomeres of an &cell embryo are of three types which distinctly differ in their developmental capacity, and that at least a total of four blastomeres of these three types is necessary for complete pattern regulation.
a
b
Ventral
Left MATERIALS
Fertilized eggs from laboratory-raised X laevis were dejellied, and then allowed to develop to the 8-cell stage. Embryos with regular and symmetrical patterns of cleavage and pigmentation were selected. Figure la is a photograph of such a living embryo, and Fig. lb is a diagram indicating the positions of blastomeres to be removed in the defect embryos in Tables 1-5. Embryos were sterilized, and the vitelline membranes 0012-1606/84 $3.00 Copyright All righta
Right
AND METHODS
Q 1984 by Academic Press, Inc. of reproduction in any form reserved.
FIG. 1. An 8-cell embryo. (a) An embryo, viewed from the animal pole. The upper blastomeres are ventral and the lower ones are dorsal. (b) A diagram schematizing the composition of blastomeres of a defect embryo. The inner and outer circles represent the animal and vegetal blastomeres, respectively. Removed blastomeres are represented by filled ones in the tables. 410
KAGEURA
AND YAMANA
Defect Experiments TABLE
DEFECT
Series number
6
EMBRYOS
411
in Xenspus
1
THAT OFTEN DEVELOPED INTO NORMAL TAILBUD EMBRYOS
Normal embryo
Abnormal
Incompletely invaginated embryo
embryo
32
Head deformity (4)=
12
31
Small head (8)
9
37
Head deformity (5)
1
34
Large head and small tail (7) Head deformity (3)
5
20
Head deformity (9)
21
Small secondary axial structures (15) Head deformity (1) Short body (1)
Early dead
16 9
24
23
29
Small right half of the head (13)
5
36
Large head and small tail (7)
4
10
37
No head (5) Large head (1) Body curved dorsally (2)
2
11
20
Small right half of the head (5) Neural plate not close (1)
21
3
12
12
Head deformity (6) Exogastrula (1)
13
18
13
17
Body curved dorsally (2) Small head (6) No head (1)
14
10
14
18
Double embryo (1) Exogastrula (2)
15
14
Note. See Fig. lb. a Figures in parentheses represent the numbers of embryos.
onto the other vegetal blastomere. The defect embryo was then rotated 90” to bring its wounded side down, and allowed to stand for a while. All the procedures employed were described previously (Kageura and Yamana, 1983). Embryos deprived of the vitelline membrane served as controls. Defect embryos and controls were put into a tissue-culture plate (Falcon-3034), the surface of which was coated with 2% agar. They were cultured in L-15 FCS, which was gradually changed to 10% Steinberg solution. When controls reached stage 26 (Nieuwkoop and Faber, 1967), embryos derived from defect embryos were examined macroscopically. Normal embryos were allowed to develop further, and examined after they became frogs.
Each series of experiments consisted of 50 defect embryos. Of 50 controls 2 were arrested at early stages of development and 48 normal tailbud embryos arose. RESULTS
AND
DISCUSSION
The results of 24 series of defect embryos are shown in Tables l-5, according to the types of embryos derived from them. Defect embryos of six series which have been described in a previous paper (Kageura and Yamana, 1983), are reproduced here. Defect embryos of four series were prepared by removing one of the four blastomeres of the right half (series l-4) (Table 1). Only a few of them were arrested at early stages, whereas 62-74s became tailbud embryos which appeared normally proportioned. No serious ab-
412
DEVELOPMENTAL
BIOLOGY
VOLUME
TABLE DEFECX
EMBRYOS
THAT
OFTEN
DEVELOPED
2
INTO TAILBUD
Normal embryo
Series
number
101,1984
WITH A LARGE
EMBRYOS
Abnormal
5
Large Head
tail
(42)”
2
Large head and small tail No head (1) Secondary axial structures
(38)
16
Large head and small No head (14)
tail
(23)
Large
represent
the numbers
head and small deformity (1)
and small
tail
(1’7)
Large head Exogastrula
and small (9)
tail
(14)
THAT
OFTEN GAVE
Series number
RISE
TO HEADLESS
Normal embryo
20
3
(1) One dorsal and one ventral imal hemisphere (series 7).
TAILBUD
Abnormal
AND
VESICLES
Vesicle (3)” No head (28) Small head (4) Nearly normal embryo
with
a vesicle
22
0
Vesicle (23) Vesicle with a dorsoventral Small axial structures (5)
1
Vesicle (7) Vesicle with axial structures No head (4) Small head (17); Exogastrula
1
Vesicle with small No head (19) Small head (6)
represent
the numbers
1
3
3
8
3
19
14
17
8
of embryos.
blastomere
axial
WITH
OR WITHOUT
AXIAL
of the an-
STRWTURES
Incompletely invaginated embryo
embryo
Vesicle with axial structures (4) No head (7) Small head (30) Small head with a vesicle (5)
Note. See Fig. lb. ’ Figures in parentheses
1
3 EMBRYOS
0
24
Early dead
developmental capacity to a great extent. The remaining four series (7-10) of defect embryos were prepared and their development was examined in the present study. Results show that removal. of the following sets of blastomeres did not prevent defect embryos from undergoing normal or nearly normal development (Table 1).
21
23
TAIL
of embryos.
TABLE EMBRYOS
A SMALL
(3)
head
normalities were observed, and this suggests that pattern regulation was complete in these defect embryos. There are a total of eight series of defect embryos lacking two blastomeres that have been examined. Four series of them have already been described (Kageura and Yamana, 1983), and are reproduced in series 5, 6, 15, and 20 (Tables 1 and 2). The results of series 5 and 6 show that removal of two dorsal or two ventral blastomeres of the animal hemisphere does not reduce the
DEFECT
AND
Incompletely invaginated embryo
embryo
15
Note. See Fig. lb. “Figures in parentheses
HEAD
axis
Early dead
1
5
0
4
0
3
0
10
0
8
(6)
(19)
(10) (1)
structures
(16)
KAGEURA AND YAMANA
Defect Experiments
in Xenqtms
413
TABLE 4 DEFECT EMBRYOS THAT OFTEN GAVE RISE TO EXOGASTRULAE AND ABNORMAL TAILBUD EMBRYOS Normal embryo
Series number
Abnormal
embryo
Incompletely
invaginated
Early dead
embryo
25
0
0
Dorsal
side not covered with epidermis
(43)=
7
26
0
0
Dorsal
side not covered with epidermis
(36)
14
2-tb
0
Exogastrula
28
0
Vesicle (2)
29
0
Exogastrula
9
0
(124)’ Lateral or posterior epidermis (45)
side not covered with
Lateral or dorsal side not covered with epidermis (37)
(2)
3 11
Note. See Fig. lb. ’ Figures in parentheses represent the numbers of embryos. b This series consists of 133 defect embryos.
(2) One animal and one vegetal dorsal blastomere (series ES),or one animal and one vegetal ventral blastomere (series 9). (3) One dorsal and one ventral blastomere of the vegeta1 hemisphere (series 10). In these series, except series 7, rather few defect embryos died at early stages or invaginated incompletely. The frequency and the anatomical proportions of these embryos were comparable to those of defect embryos lacking only one blastomere. In series ‘7,on the contrary, although few died at early stages, about half of the defect embryos underwent incomplete invagination, probably due to the lack of two animal blastomeres. Forty-eight to seventy-four percent in series 7-10 developed into normal or nearly normal tailbud embryos. They showed no significant retardation of development. The results of series 5-7 indicate that two animal biastomeres are su&cient for complete pattern regulutim and that the two may be both dorsal or ventral, or dorsal and ventral. This means that the four animal blastomeres are roughly equivalent in their developmental capacity. On the other hand, no normal development occurred when two dorsal or two ventral blastomeres of the veg-
eta1 hemisphere were removed (series 15 and 20) (Tables 2 and 3). A comparison between the results of series 15 and 20 and those of series 10 clearly shows that at least one dorsal and one ventral blastomere of the vegetal hemisphere are necessa~ fw complete pattern regulatim The absence of these two blastomeres, together with others, explains the failure in development of defect embryos of series 16-19 (Table 2), 21-24 (Table 3), and 30 (Table 5). Abnormalities observed in these series differed, depending on the blastomeres retained (or removed). Most defect embryos of series 16-19, lacking the two ventral blastomeres, developed tailbud embryos that had a large head and a small tail. These embryos resembled those derived from the dorsal halves of 4-cell embryos (Kageura and Yamana, 1983). On the other hand, most of the defect embryos of series 21-24; that is, those containing no dorsal blastomeres of the vegetal hemisphere gave rise to tailbud embryos with a small head or no head, and also to vesicles with or without axial structure. Few or no normal embryos arose in these series. However, the frequency of incomplete invagination was very low. The embryos in these series were similar to those derived from ventral halves of 4-cell embryos (Kageura and Yamana, 1983). Defect embryos of series 30, animal
TABLE 5 DEFECT EMBRYOS THAT OFTEN GAVE RISE TO VESICLES WITH OR WITHOUT A LONG PROJECTION Series number
3ob
Normal embryo
63
0
Abnormal
embryo
A vesicle (62)” A vesicle with a long projection
Note. See Fig. lb. ‘Figures in parentheses represent the number of embryos. b This series consists of 123 defect embryos.
(60)
Incompletely invaginated embryo
Early dead
0
1
414
DEVELOPMENTAL BIOLOGY
halves, became vesicles with or without a long projection, as described previously (Kageura and Yamana, 1983). The vesicles consisted of epidermis and a cement gland, and the projection contained muscle and melanophores (Table 5). Defect embryos of series 25 and 26 had lost three animal blastomeres, though retaining all the four vegetal blastomeres. No normal embryos arose in these series (Table 4). Most of the defect embryos invaginated incompletely and then developed into embryos whose surface was only partially covered with epidermis, probably due to the lack of animal blastomeres which would have given rise to epidermis. In these embryos endoderm and mesoderm were exposed at a dorsal or lateral side of the body (Figs. 2a,b). Essentially the same results were obtained with defect embryos of series 28 and 29, which had lost three animal blastomeres, together with two vegetal ones (Table 4). The absence of all the animal blastomeres did not permit normal development of defect embryos (series 27): 93% became exogastrulae and ‘7% died at early stages (Table 4). As described above, most of the defect embryos lacking one or two animal blastomeres developed normal tailbud
VOLUME 101.1!?&1
embryos, whereas none of the defect embryos that had lost more than two animal blastomeres underwent normal development. This shows that at least two animal blast-a
are necessary for come
pattern regulutim
The present results are consistent with those of Votquenne (1933) and Vintemberger (1936) (both cited in Gerhart, 1980) who removed or killed two animal dorsal blastomeres at the 8-cell stage in Rana and, nonetheless, obtained normal development. There are six other possible series of defect embryos that can be prepared by removing three blastomeres from an &cell embryo; however, they have not been examined. Defect embryos of series 11-14 (Table 1) had lost four blastomeres, but still retained the other four that have been shown to be necessary for complete pattern regulation. About 50-60s of these defect embryos were arrested at early stages of development or invaginated incompletely. This might be due to some damage made by removal of as many as four blastomeres which amounted to half an embryo. Furthermore, the operation was not simple in series 12-14. Irrespective of the high frequencies of early arrest and incomplete invagination,
a
FIG. 2. Embryos derived from defect embryos. (a, b) Abnormal “tailbud embryos” (30 hr after fertilization) derived from defect embryos of series 25 and 26, respectively (lateral and dorsal view). They were normal in size; however, endodermal and mesodermal cells were exposed at a dorsal side. (c, d) Normal tadpoles derived from defect embryos of series 11 and 12, respectively (stage 29/30 and 33/W, lateral view). The speed of development of these larvae did not differ from that of controls, and are completely normal in their appearance. (The upper tadpole in (c) is a control.)
KAGEURA AND YAMANA
Defect Experiments
FIG. 3. A normal frog derived from a defect embryo of series 13. This shows no retardation of development and is normal in all respects as far as external morphology and behavior are concerned.
most of the embryos that developed beyond gastrulation became normal. They showed no significant retardation of development, although some were slender. Figures 2c, d, and Fig. 3 show some tadpoles and a frog obtained in series 11-13. These larvae and this frog are normal in all respects as far as external morphology and behavior are concerned. Several defect embryos gave nearly normal tailbud embryos whose right half of the head and the right eye were smaller. However, such abnormalities disappeared soon. No other remarkable abnormalities were found. Defect embryos of four series that had been made after removal of five blastomeres were examined (series 19, 24, 28, and 29). They necessarily lacked one of the four essential blastomeres and, in fact, all but three developed abnormally, as shown above.
in Xenopus
415
An &cell embryo is bilaterally symmetric, and consists of the minimum number of blastomeres which represent the animal-vegetal and dorsal-ventral axes. The present study shows that a set of four blastomeres of such an embryo are essential for complete pattern regulation. The blastomeres are of three types. This implies that blastomeres of each type have “differentiated” or have been destined to play a definite role to such an extent that a blastomere of a type can no longer compensate for the loss of a blastomere of a different type. The role of each blastomere can be inferred from the results of our previous isolation experiments (Kageura and Yamana, 1983), and the present defect experiments, as well as from those of earlier workers. Furthermore, an arrangement of four blastomeres of the three types can exactly define the axes of an embryo. This work was supported by Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture, Japan. REFERENCES GERHART,J. C. (1980). Mechanisms regulating pattern formation in the amphibian egg and early embryo. In “Biological Regulation and Development” (R. F. Goldberger, ed.), Vol. 2, pp. 133-316. Plenum, New York/London. KAGEURA,H., and YAMANA, K. (1933). Pattern regulation in isolated halves and blastomeres of early Xenqpus laevis. J. EmbryoL Exp. Mmyhol
74.221-234.
NIEUWKOOP,P. D., and FABER, J. (1967). “Normal Table of Xenopus Zoevis (Daudin).” North-Holland, Amsterdam. VINTEMBERGER,P. (1936). Sur le dbveloppement compare des micromeres de l’oeuf de Rum fusca divise en huit: a) apres isolement, b) ap& transplantation sur un socle de cellules vitellines. C. R Sot. Bid (Paris) 122,927-930. VOTQUENNE,M. (1933). La disposition g&&ale des Qbauches presomptives dans l’oeuf de Grenouille divise en huit blastomeres et les consequences de la destruction d’un micromere dorsal. C. R. Sot Bid (Paris) 113,1531-1533.
BIOLOGY
101,
410-415 (1984)
Pattern Regulation in Defect Embryos of Xenopus laevis H. KAGEURA Deparhent
of Bib, Received
Fac&y April
AND K. YAMANA
of Science,
12, 198s; accepted
Kmhu
University $8, Fukuoka,
in revised form August
812 Japan
18, 1983
Defect embryos of 24 series were prepared by removing increasing numbers of blastomeres from an 8-cell embryo of Xenopus Latvia They were cultured and their development was examined macroscopically when controls reached a tailbud stage or later. Results show that most of defect embryos of 12 series develop normally, and some of them become normal frogs. Each of these defect embryos contain at least two animal blastimeres, one dorsal, and one ventral blastomere of the vegetal hemisphere. This suggests that a set of these four blastomeres of the three types is essential for complete pattern regulation.
removed manually. The embryos were then transferred into 50% Leibovitz (L-15) medium supplemented with 10% fetal calf serum (L-15 FCS) in a petri dish which had been coated with 2% agar. A blastomere or blastomeres were removed by repeatedly flipping the cleavage furrow with a glass needle. Defect embryos of series 13,14,18, and 23 consisted of two animal and two vegetal blastomeres; one of the former lying on one of the latter. In these cases, the other animal blastomere was put
INTRODUCTION
In our previous isolation experiments on Xenqmhs embryos (Kageura and Yamana, 1983), it was shown that most of the lateral halves at the 2-cell stage develop into normal tailbud embryos, but most or almost all the dorsal and ventral halves of the 4-cell embryos and all the animal and vegetal halves of the g-cell embryos fail in undergoing normal or nearly normal development, giving rise to characteristically abnormal embryos. This fact supports the idea that for complete pattern regulation some of the blastomeres of an early embryo are essential but the others are not. The purpose of the present study is to find out which blastomeres are essential for complete pattern regulation. A blastomere or blastomeres were removed from an &cell embryo, and the resulting defect embryos of 24 series were cultured and their development examined. An additional six series of defect experiments have already been described (Kageura and Yamana, 1983). Results obtained in the present study and in a previous one showed that the blastomeres of an &cell embryo are of three types which distinctly differ in their developmental capacity, and that at least a total of four blastomeres of these three types is necessary for complete pattern regulation.
a
b
Ventral
Left MATERIALS
Fertilized eggs from laboratory-raised X laevis were dejellied, and then allowed to develop to the 8-cell stage. Embryos with regular and symmetrical patterns of cleavage and pigmentation were selected. Figure la is a photograph of such a living embryo, and Fig. lb is a diagram indicating the positions of blastomeres to be removed in the defect embryos in Tables 1-5. Embryos were sterilized, and the vitelline membranes 0012-1606/84 $3.00 Copyright All righta
Right
AND METHODS
Q 1984 by Academic Press, Inc. of reproduction in any form reserved.
FIG. 1. An 8-cell embryo. (a) An embryo, viewed from the animal pole. The upper blastomeres are ventral and the lower ones are dorsal. (b) A diagram schematizing the composition of blastomeres of a defect embryo. The inner and outer circles represent the animal and vegetal blastomeres, respectively. Removed blastomeres are represented by filled ones in the tables. 410
KAGEURA
AND YAMANA
Defect Experiments TABLE
DEFECT
Series number
6
EMBRYOS
411
in Xenspus
1
THAT OFTEN DEVELOPED INTO NORMAL TAILBUD EMBRYOS
Normal embryo
Abnormal
Incompletely invaginated embryo
embryo
32
Head deformity (4)=
12
31
Small head (8)
9
37
Head deformity (5)
1
34
Large head and small tail (7) Head deformity (3)
5
20
Head deformity (9)
21
Small secondary axial structures (15) Head deformity (1) Short body (1)
Early dead
16 9
24
23
29
Small right half of the head (13)
5
36
Large head and small tail (7)
4
10
37
No head (5) Large head (1) Body curved dorsally (2)
2
11
20
Small right half of the head (5) Neural plate not close (1)
21
3
12
12
Head deformity (6) Exogastrula (1)
13
18
13
17
Body curved dorsally (2) Small head (6) No head (1)
14
10
14
18
Double embryo (1) Exogastrula (2)
15
14
Note. See Fig. lb. a Figures in parentheses represent the numbers of embryos.
onto the other vegetal blastomere. The defect embryo was then rotated 90” to bring its wounded side down, and allowed to stand for a while. All the procedures employed were described previously (Kageura and Yamana, 1983). Embryos deprived of the vitelline membrane served as controls. Defect embryos and controls were put into a tissue-culture plate (Falcon-3034), the surface of which was coated with 2% agar. They were cultured in L-15 FCS, which was gradually changed to 10% Steinberg solution. When controls reached stage 26 (Nieuwkoop and Faber, 1967), embryos derived from defect embryos were examined macroscopically. Normal embryos were allowed to develop further, and examined after they became frogs.
Each series of experiments consisted of 50 defect embryos. Of 50 controls 2 were arrested at early stages of development and 48 normal tailbud embryos arose. RESULTS
AND
DISCUSSION
The results of 24 series of defect embryos are shown in Tables l-5, according to the types of embryos derived from them. Defect embryos of six series which have been described in a previous paper (Kageura and Yamana, 1983), are reproduced here. Defect embryos of four series were prepared by removing one of the four blastomeres of the right half (series l-4) (Table 1). Only a few of them were arrested at early stages, whereas 62-74s became tailbud embryos which appeared normally proportioned. No serious ab-
412
DEVELOPMENTAL
BIOLOGY
VOLUME
TABLE DEFECX
EMBRYOS
THAT
OFTEN
DEVELOPED
2
INTO TAILBUD
Normal embryo
Series
number
101,1984
WITH A LARGE
EMBRYOS
Abnormal
5
Large Head
tail
(42)”
2
Large head and small tail No head (1) Secondary axial structures
(38)
16
Large head and small No head (14)
tail
(23)
Large
represent
the numbers
head and small deformity (1)
and small
tail
(1’7)
Large head Exogastrula
and small (9)
tail
(14)
THAT
OFTEN GAVE
Series number
RISE
TO HEADLESS
Normal embryo
20
3
(1) One dorsal and one ventral imal hemisphere (series 7).
TAILBUD
Abnormal
AND
VESICLES
Vesicle (3)” No head (28) Small head (4) Nearly normal embryo
with
a vesicle
22
0
Vesicle (23) Vesicle with a dorsoventral Small axial structures (5)
1
Vesicle (7) Vesicle with axial structures No head (4) Small head (17); Exogastrula
1
Vesicle with small No head (19) Small head (6)
represent
the numbers
1
3
3
8
3
19
14
17
8
of embryos.
blastomere
axial
WITH
OR WITHOUT
AXIAL
of the an-
STRWTURES
Incompletely invaginated embryo
embryo
Vesicle with axial structures (4) No head (7) Small head (30) Small head with a vesicle (5)
Note. See Fig. lb. ’ Figures in parentheses
1
3 EMBRYOS
0
24
Early dead
developmental capacity to a great extent. The remaining four series (7-10) of defect embryos were prepared and their development was examined in the present study. Results show that removal. of the following sets of blastomeres did not prevent defect embryos from undergoing normal or nearly normal development (Table 1).
21
23
TAIL
of embryos.
TABLE EMBRYOS
A SMALL
(3)
head
normalities were observed, and this suggests that pattern regulation was complete in these defect embryos. There are a total of eight series of defect embryos lacking two blastomeres that have been examined. Four series of them have already been described (Kageura and Yamana, 1983), and are reproduced in series 5, 6, 15, and 20 (Tables 1 and 2). The results of series 5 and 6 show that removal of two dorsal or two ventral blastomeres of the animal hemisphere does not reduce the
DEFECT
AND
Incompletely invaginated embryo
embryo
15
Note. See Fig. lb. “Figures in parentheses
HEAD
axis
Early dead
1
5
0
4
0
3
0
10
0
8
(6)
(19)
(10) (1)
structures
(16)
KAGEURA AND YAMANA
Defect Experiments
in Xenqtms
413
TABLE 4 DEFECT EMBRYOS THAT OFTEN GAVE RISE TO EXOGASTRULAE AND ABNORMAL TAILBUD EMBRYOS Normal embryo
Series number
Abnormal
embryo
Incompletely
invaginated
Early dead
embryo
25
0
0
Dorsal
side not covered with epidermis
(43)=
7
26
0
0
Dorsal
side not covered with epidermis
(36)
14
2-tb
0
Exogastrula
28
0
Vesicle (2)
29
0
Exogastrula
9
0
(124)’ Lateral or posterior epidermis (45)
side not covered with
Lateral or dorsal side not covered with epidermis (37)
(2)
3 11
Note. See Fig. lb. ’ Figures in parentheses represent the numbers of embryos. b This series consists of 133 defect embryos.
(2) One animal and one vegetal dorsal blastomere (series ES),or one animal and one vegetal ventral blastomere (series 9). (3) One dorsal and one ventral blastomere of the vegeta1 hemisphere (series 10). In these series, except series 7, rather few defect embryos died at early stages or invaginated incompletely. The frequency and the anatomical proportions of these embryos were comparable to those of defect embryos lacking only one blastomere. In series ‘7,on the contrary, although few died at early stages, about half of the defect embryos underwent incomplete invagination, probably due to the lack of two animal blastomeres. Forty-eight to seventy-four percent in series 7-10 developed into normal or nearly normal tailbud embryos. They showed no significant retardation of development. The results of series 5-7 indicate that two animal biastomeres are su&cient for complete pattern regulutim and that the two may be both dorsal or ventral, or dorsal and ventral. This means that the four animal blastomeres are roughly equivalent in their developmental capacity. On the other hand, no normal development occurred when two dorsal or two ventral blastomeres of the veg-
eta1 hemisphere were removed (series 15 and 20) (Tables 2 and 3). A comparison between the results of series 15 and 20 and those of series 10 clearly shows that at least one dorsal and one ventral blastomere of the vegetal hemisphere are necessa~ fw complete pattern regulatim The absence of these two blastomeres, together with others, explains the failure in development of defect embryos of series 16-19 (Table 2), 21-24 (Table 3), and 30 (Table 5). Abnormalities observed in these series differed, depending on the blastomeres retained (or removed). Most defect embryos of series 16-19, lacking the two ventral blastomeres, developed tailbud embryos that had a large head and a small tail. These embryos resembled those derived from the dorsal halves of 4-cell embryos (Kageura and Yamana, 1983). On the other hand, most of the defect embryos of series 21-24; that is, those containing no dorsal blastomeres of the vegetal hemisphere gave rise to tailbud embryos with a small head or no head, and also to vesicles with or without axial structure. Few or no normal embryos arose in these series. However, the frequency of incomplete invagination was very low. The embryos in these series were similar to those derived from ventral halves of 4-cell embryos (Kageura and Yamana, 1983). Defect embryos of series 30, animal
TABLE 5 DEFECT EMBRYOS THAT OFTEN GAVE RISE TO VESICLES WITH OR WITHOUT A LONG PROJECTION Series number
3ob
Normal embryo
63
0
Abnormal
embryo
A vesicle (62)” A vesicle with a long projection
Note. See Fig. lb. ‘Figures in parentheses represent the number of embryos. b This series consists of 123 defect embryos.
(60)
Incompletely invaginated embryo
Early dead
0
1
414
DEVELOPMENTAL BIOLOGY
halves, became vesicles with or without a long projection, as described previously (Kageura and Yamana, 1983). The vesicles consisted of epidermis and a cement gland, and the projection contained muscle and melanophores (Table 5). Defect embryos of series 25 and 26 had lost three animal blastomeres, though retaining all the four vegetal blastomeres. No normal embryos arose in these series (Table 4). Most of the defect embryos invaginated incompletely and then developed into embryos whose surface was only partially covered with epidermis, probably due to the lack of animal blastomeres which would have given rise to epidermis. In these embryos endoderm and mesoderm were exposed at a dorsal or lateral side of the body (Figs. 2a,b). Essentially the same results were obtained with defect embryos of series 28 and 29, which had lost three animal blastomeres, together with two vegetal ones (Table 4). The absence of all the animal blastomeres did not permit normal development of defect embryos (series 27): 93% became exogastrulae and ‘7% died at early stages (Table 4). As described above, most of the defect embryos lacking one or two animal blastomeres developed normal tailbud
VOLUME 101.1!?&1
embryos, whereas none of the defect embryos that had lost more than two animal blastomeres underwent normal development. This shows that at least two animal blast-a
are necessary for come
pattern regulutim
The present results are consistent with those of Votquenne (1933) and Vintemberger (1936) (both cited in Gerhart, 1980) who removed or killed two animal dorsal blastomeres at the 8-cell stage in Rana and, nonetheless, obtained normal development. There are six other possible series of defect embryos that can be prepared by removing three blastomeres from an &cell embryo; however, they have not been examined. Defect embryos of series 11-14 (Table 1) had lost four blastomeres, but still retained the other four that have been shown to be necessary for complete pattern regulation. About 50-60s of these defect embryos were arrested at early stages of development or invaginated incompletely. This might be due to some damage made by removal of as many as four blastomeres which amounted to half an embryo. Furthermore, the operation was not simple in series 12-14. Irrespective of the high frequencies of early arrest and incomplete invagination,
a
FIG. 2. Embryos derived from defect embryos. (a, b) Abnormal “tailbud embryos” (30 hr after fertilization) derived from defect embryos of series 25 and 26, respectively (lateral and dorsal view). They were normal in size; however, endodermal and mesodermal cells were exposed at a dorsal side. (c, d) Normal tadpoles derived from defect embryos of series 11 and 12, respectively (stage 29/30 and 33/W, lateral view). The speed of development of these larvae did not differ from that of controls, and are completely normal in their appearance. (The upper tadpole in (c) is a control.)
KAGEURA AND YAMANA
Defect Experiments
FIG. 3. A normal frog derived from a defect embryo of series 13. This shows no retardation of development and is normal in all respects as far as external morphology and behavior are concerned.
most of the embryos that developed beyond gastrulation became normal. They showed no significant retardation of development, although some were slender. Figures 2c, d, and Fig. 3 show some tadpoles and a frog obtained in series 11-13. These larvae and this frog are normal in all respects as far as external morphology and behavior are concerned. Several defect embryos gave nearly normal tailbud embryos whose right half of the head and the right eye were smaller. However, such abnormalities disappeared soon. No other remarkable abnormalities were found. Defect embryos of four series that had been made after removal of five blastomeres were examined (series 19, 24, 28, and 29). They necessarily lacked one of the four essential blastomeres and, in fact, all but three developed abnormally, as shown above.
in Xenopus
415
An &cell embryo is bilaterally symmetric, and consists of the minimum number of blastomeres which represent the animal-vegetal and dorsal-ventral axes. The present study shows that a set of four blastomeres of such an embryo are essential for complete pattern regulation. The blastomeres are of three types. This implies that blastomeres of each type have “differentiated” or have been destined to play a definite role to such an extent that a blastomere of a type can no longer compensate for the loss of a blastomere of a different type. The role of each blastomere can be inferred from the results of our previous isolation experiments (Kageura and Yamana, 1983), and the present defect experiments, as well as from those of earlier workers. Furthermore, an arrangement of four blastomeres of the three types can exactly define the axes of an embryo. This work was supported by Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture, Japan. REFERENCES GERHART,J. C. (1980). Mechanisms regulating pattern formation in the amphibian egg and early embryo. In “Biological Regulation and Development” (R. F. Goldberger, ed.), Vol. 2, pp. 133-316. Plenum, New York/London. KAGEURA,H., and YAMANA, K. (1933). Pattern regulation in isolated halves and blastomeres of early Xenqpus laevis. J. EmbryoL Exp. Mmyhol
74.221-234.
NIEUWKOOP,P. D., and FABER, J. (1967). “Normal Table of Xenopus Zoevis (Daudin).” North-Holland, Amsterdam. VINTEMBERGER,P. (1936). Sur le dbveloppement compare des micromeres de l’oeuf de Rum fusca divise en huit: a) apres isolement, b) ap& transplantation sur un socle de cellules vitellines. C. R Sot. Bid (Paris) 122,927-930. VOTQUENNE,M. (1933). La disposition g&&ale des Qbauches presomptives dans l’oeuf de Grenouille divise en huit blastomeres et les consequences de la destruction d’un micromere dorsal. C. R. Sot Bid (Paris) 113,1531-1533.