/. Embryol. exp. Morph. Vol. 33, 1, pp. 177-185, 1975
Printed in Great Britain
177
The development of trophoblast
in vitro from blastocysts containing varying
amounts of inner cell mass
By J. D. ANSELL 1 AND M. H. L. SNOW2
From Institute of Animal Genetics, Edinburgh
SUMMARY
When intact mouse blastocysts are cultured in vitro in medium supplemented with foetal
calf serum, trophoblast cells proliferate and undergo giant cell transformation such as occurs
in vivo. If the amount of inner cell mass in the blastocyst is decreased by culture with [3H]thymidine then giant cell transformation occurs normally but proliferation is reduced. In
the absence of inner cell mass no proliferation occurs, and giant cell transformation is more
rapid than in undamaged blastocysts.
INTRODUCTION
During the early development of mouse embryos, two distinct cell populations have arisen by the 16-cell stage-those enclosed on the inside of the
embryos and those on the outside. By the late morula-early blastocyst stage
those on the inside have become the cells of the inner cell mass, whilst those on
the outside tend to differentiate into trophoblast cells (Hillman, Sherman &
Graham, 1972), the majority of which become giant. The nuclear enlargement
associated with the development of trophoblast giant cells has been shown to
involve the formation of polytene chromosomes (Snow & Ansell, 1974). At the
16-cell stage differences have been observed in the cell cycle characteristics of
the two cell populations (Barlow, Owen & Graham, 1972) and Snow (1973 a, b)
has noted that the inside population is selectively damaged at this stage by
treatment of embryos in vitro with tritiated thymidine. The population of outside
cells continues to cleave normally, however, giving rise to blastocysts enclosing
a significantly reduced number of inner cell mass cells. At a critical concentration
of tritium 'blastocysts' composed entirely of trophoblast are produced. We
have studied the further development of such trophoblast vesicles both in vitro
and by transfer to ectopic sites.
1
Author's address: ARC Unit of Reproductive Physiology and Biochemistry, 307
Huntingdon Road, Cambridge CB3 OJQ, U.K.
2
Author's address: MRC Mammalian Development Unit, University College London,
Wolfson House, 4 Stephenson Way, London, NW1 2HE, U.K.
12
EMB 33
178
J. D. ANSELL AND M. H. L. SNOW
MATERIALS AND METHODS
Blastocysts or trophoblast vesicles were taken from their previous culture
medium, washed in phosphate buffered saline, and placed in 0-1 ml drops of a
modified Brinster's medium (Bowman & MacLaren, 1970), supplemented with
5 % foetal calf serum (Gwatkin, 1966) under paraffin oil. All cultures were
gassed with 10 % CO2 in air at atmospheric pressure and incubated at 37 °C.
In this culture system the embryos remain as blastocysts for approximately 24 h
before hatching - the blastocyst bursting through its zona pellucida; the hatched
blastocyst then attaches itself to the surface of its culture vessel and trophoblast
giant cells grow out as a monolayer of cells, generally leaving a nodule of inner
cell mass in its centre. These events are recorded in Fig. 4.
Mouse eggs from the randomly bred ' Q ' strain of mice were grown from the
2-cell to blastocyst stage in three concentrations of tritiated thymidine ([3H]Tdr);
0-01, 0025 and 0-05 ^Ci/ml (Snow, 1973a). Methyl-[3H]thymidine (sp. act.
17 Ci/mM, the Radiochemical Centre) was used throughout. Each group of
embryos then transferred into the outgrowth medium was subdivided into two
classes - those removed from tritiated thymidine before cavitation (i.e. morulae)
and those removed, as blastocysts, to test the effects of any residual [3H]Tdr
remaining enclosed in the blastocoel cavity after transfer to the outgrowth
medium.
The embryos were monitored over a 5-day culture period in the outgrowth
medium and scored as having remained in their zonas, hatched, or stuck to
their culture vessel and outgrown.
Groups of control blastocysts and trophoblast vesicles grown at 0-05 /*Ci/ml
of [3H]Tdr were also outgrown on glass coverslips. After 5 days of culture the
trophoblast outgrowths, still firmly adhering to the coverslips, were washed in
phosphate buffered saline and immersed in Carnoy's fixative. The preparations
were stained with haematoxylin and eosin, mounted in Canada Balsam and an
estimate made of the number of cells in each outgrowth.
Blastocysts and vesicles from all treatments were also transferred singly or in
groups to two ectopic sites in the mouse; beneath the kidney capsule or into the
experimentally cryptorchidized testis.
Ten days after surgery, the kidney and testis were dissected out, fixed in
Sanfelice's fluid, embedded in paraffin wax and sectioned. All sections were
stained with haematoxylin and eosin and scored for evidence of trophoblast
proliferation.
RESULTS
No differences were observed in any group between the outgrowths from
those eggs kept continuously in [3H]Tdr until the blastocyst stage and those
transferred to fresh medium before cavitation. These two groups were, therefore,
pooled for the ensuing analyses. Table 1 shows the mean angular response
Development of trophoblast in vitro
179
Table 1. Mean angular response to hatching (H) and outgrowth (O) of control and
tritiated thymidine treated blastocysts over 4 days culture in outgrowth medium
Day 1
Group
Control
Mean
S.E.
Mean
001
S.E.
0025
Mean
S.E.
005
Mean
S.E.
Day 4
Day 3
Day 2
(H)
(H)
(O)
(H)
(O)
(H)
(O)
36-8
±7-6
31-3
±6-5
15 00
±4-20
5-6
+ 3-7
59-8
±5-6
51-5
±8-2
29-9
±7-9
190
±8-2
17-4
±7-1
14-4
±7-1
18-5
±6-3
8-4
±5-7
65-7
±5-7
51-3
±4-4
29-4
+ 2-2
33-1
±6-5
46-6
±5-3
48 1
±5-9
27-6
±2-1
31-9
±6-6
63-8
+ 4-5
55-5
±4-8
301
±3-1
35-9
±7-0
61-3
±4-2
55-5
±4-8
301
±3-1
35-9
±7-0
80
70
60
005
I 0025
Days in culture
Fig. 1. The angular response to hatching over 4 days in culture of control and
[3H]thymidine treated blastocysts. The vertical bars represent standard errors.
D, Control (11 replicates; 110 blastocysts), 0-001 /tCi/ml [3H]Tdr (9 replicates;
84 blastocysts). x , 0025/iCi/ml [3H]Tdr (11 replicates; 130 blastocysts). • ,
005 ^Ci/ml [3H]Tdr (9 replicates; 65 blastocysts).
(Biggers & Brinster, 1965) to hatching and outgrowth over the five days in
culture. Figs. 1 and 2 are the graphical representations of these results. In both
of these figures the control and 0-01 /tCi/ml blastocysts behave similarly to each
other as do the 0-025 and 0-05 /tCi/ml groups. The numbers of blastocysts
hatching on each day in the control and 0-01 groups are significantly larger than
those in the 0-05 /tCi/ml group (P < 0-01 for each day). This difference is not
180
J. D. ANSELL AND M. H. L. SNOW
Days in culture
Fig. 2. The angular response to outgrowth, over 4 days in culture, of the blastocysts
hatched from Fig. 1.
reflected in the outgrowth data until day 4, a significant difference in the numbers
outgrowing between control and O05/*Ci/ml treated outgrowths not arising
until then. Since hatching appears to be the crucial stage in the outgrowth
process, all of the blastocysts that hatch tending to outgrow (see Table 1), this
suggested a difference in the rate of outgrowth between groups. This is expressed
as the percentage of those blastocysts hatched that have outgrown per day in
culture and is shown in histogram form in Fig. 3. All but a very few of those
blastocysts hatched in the 0-05/^Ci/ml group have completed outgrowth by
day 3 in culture, whilst outgrowth in the control and 0-01 /tCi/ml groups is only
about two-thirds complete at this stage, taking a further 24 h to complete
outgrowth.
Typical preparations from outgrowths of the control and 0-05 /^Ci/ml groups
which were fixed and stained are shown in Fig. 5. Whilst control outgrowths
have an appreciable quantity of inner cell mass material in the centre of the
growth, the vesicle outgrowth is composed entirely of giant trophoblast cells.
Metaphase plates were often seen in control outgrowths, but never in those
from vesicles. A comparison of maximum nuclear sizes in both types of outgrowth showed no difference in the extent to which their nuclei had become
giant. Cell counts from these preparations are shown in Table 2. It was possible
to distinguish between those cells which were obviously giant and those which
were diploid or in the early stages of enlargement. Blastocyst cell number data
were obtained from embryos grown in vitro from the 2-cell to blastocyst stage
with and without [3H]Tdr treatment (see Snow, 1973a). All of the cells in the
outgrowths from vesicles are giant and there is no significant proliferation of
Development of trophoblast in vitro
181
50
40
30
20
10
Day 2
Day 3
Day 4
Fig. 3. The rate of outgrowth of blastocysts expressed as the number of hatched
blastocysts beginning outgrowth on each day in culture.
D
Fig. 4. The events during in vitro outgrowth of mouse blastocysts. (A) Normal 3|-day
blastocyst enclosed within the zona pellucida (ZP). The inner cell mass (ICM) is
visible. (B) The blastocyst has ruptured the ZP and is in the process of hatching.
(C) The hatched blastocyst attaches itself to the substrate and begins outgrowth (O).
(D) Trophoblast giant cells with polyploid nuclei (N) have outgrown as a monolayer.
ICM can still be identified.
cells through the hatching and outgrowth process. The cell number significantly
increases in control outgrowths over that found in the blastocyst from a mean
of 61-9-85-3 cells (P < 0-05) and of the order of 67 % of those cells are obviously
giant trophoblast cells. This figure will be a slight underestimate for trophoblast
182
J. D. ANSELL AND M. H. L. SNOW
c
Fig. 5. Fixed preparations of trophoblast outgrowths. (A) Control, after 4 days in
culture, showing giant cells (GC) and inner cell mass (ICM). (B) Part of a control
enlarged to show giant nuclei (N) and a mitotic figure (M) in the ICM. (C) Outgrowth
from a blastocyst vesicle grown with 0-05 /tCi/ml tritiated thymidine, after 4 days in
culture. Note the absence of inner cell mass. (D) Part of (C) enlarged for comparison
with (B).
Table 2. Mean cell number in control blastocysts and tritiated thymidine
treated vesicles, and in outgrowths from similarly treated embryos
Treatment
Control
Mean
S.E.
005/tCi/ml Mean
S.E.
Blastocysts
(total cell no.)
Outgrowths
(total cell no.)
Outgrowths
(no. of giant cells)
61-9
±6-4
30-3
±2-5
85-3
±8-2
291
±2-5
56-6
±6-5
26-7
±2-3
cells present, since only giant trophoblast cells are counted, not those which
may be diploid or in the early stages of enlargement.
Control blastocysts transferred to kidney and testis both developed ectoplacental cone trophoblast and showed varying degrees of embryonic development (see Table 3), whilst approximately half of the vesicles treated at 0-01 and
0-025 /iCi/ml developed haemorrhagic nodules in the kidney composed only of
trophoblast tissue with no other embryonic development present. No growth
was noted on transfer of a group of vesicles treated at 0-05 /*Ci/ml. The testis
proved a harsher environment for vesicle development. Good embryonic and
trophoblast development was recorded for control blastocysts and four of the
five groups of 0-01 /*Ci/ml vesicles transferred developed some trophoblast
tissue in the testis. No trace of trophoblast or embryonic tissue developed from
any of the 0-025 and 0-05 /tCi/ml vesicles or from the 0-01 /tCi/ml vesicles transferred singly. Fig. 6 shows the various types of trophoblast development obtained.
Development of trophoblast in vitro
183
Table 3. The development oftritiated thymidine treated and control
blastocysts in the kidney and cryptorchidized testis
(T signifies trophoblast development only and E a range of embryonic development
from egg cylinder, to small patches of embryonic tissue enclosed within proliferating
trophoblast.)
No. of
single (S) or No. showing
Tritiated
multiple (M)
some
Type of
thymidine
development development
transfers
(/tCi/ml)
Ectopic site
Kidney
0
001
0025
Cryptorchidized testis
005
0
0
001
001
0025
0025
005
005
7M
3M
5M
1M
5M
2S
5M
3S
4M
3S
4M
4S
7
2
2
0
5
1
4
0
0
0
0
0
T,E
T
T
—
T,E
T, E
T
—
—
—
—
—
wjjk > \--fi"
Fig. 6. Blastocyst growth in ectopic sites. (A) The development shown by a single
control blastocyst 10 days after transfer beneath the capsule of a cryptorchidized testis.
Both embryonic and trophoblast development have occurred. (B) Part of a nodule of
pure trophoblast developed in the kidney from a group of [3H]thymidine treated
blastocysts. (C) Part of a nodule of pure trophoblast developed in the testis from a
group of [3H]thymidine treated blastocysts.
184
J. D. ANSELL AND M. H. L. SNOW
DISCUSSION
Snow (1973a, b) observed that trophoblast vesicles produced by treatment
with [3H]Tdr at a concentration of 0-01 /^Ci/ml still enclosed several inner cell
mass cells, whilst at 0-025 /*Ci/ml the inner cell mass was reduced to two or
three cells in 50 % of the treated embryos. In the remainder of these and in
approximately 80 % of embryos treated at 0-05 fiC\jm\ there was a complete
absence of inner cell mass cells.
Our experiments show that trophoblast vesicles produced by such treatments
do not lose the ability to hatch, nor are the trophoblast cells prevented from
becoming giant and outgrowing in culture. The presence of the inner cell mass
does not directly promote the hatching of embryos from their zona pellucidae
nor does it govern the initiation or development of the polyploidization of
primary trophoblast giant cells. This data also confirms that vesicles produced
by the higher concentrations of [3H]Tdr are composed entirely of trophoblast
cells.
Although significant differences are noted between the numbers hatching in
the control group and those treated with the higher doses of [3H]Tdr, this is
not thought to be an effect of the thymidine treatment per se. Since hatching is
the crucial stage in the outgrowth process and it is at this stage that differences
are manifested, the reduction in cell number, such that the cells present cannot
exert enough physical pressure on the zona pellucida to cause its rupture and
thereby hatch, is more likely to reduce the hatching rate.
The percentage of trophoblast cells in normal blastocysts has been estimated
to be approximately 78% or 48 cells in a blastocyst containing 62 cells by
Barlow et al. (1972), and approximately 71 % in blastocysts from mice of the
Q strain, treated at low doses of tritiated thymidine (0-001 /tCi/ml) but in all
other respects normal by Homer & McLaren (1974). Our data suggests that
during outgrowth from normal blastocysts there is some proliferation of trophoblast cells. No such proliferation was observed in vesicle outgrowths. A similar
situation arises from the ectopic transfer results. As the amount of inner cell
mass present in the ectopic site was reduced, so the degree of trophoblast proliferation went down. No haemorrhagic nodules or invasive giant cells were
found in any of the transfer sites of vesicles treated at 0-05 //Ci/ml, though both
embryonic and trophoblast development occurred on transfer of control blastocysts. Without proliferation, trophoblast cells would not be in evidence 10 days
after transfer since primary giant cells would be unlikely to survive that length
of time.
The proliferation of trophoblast into ectoplacental cone may then be dependent primarily on the presence of the inner cell mass. In its absence trophoblast
cells do not proliferate, but cease cell division and enlarge. A similar conclusion
was reached by Gardner & Johnson (1972), in an investigation of the independent growth of inner cell mass and trophoblast, derived from the mouse
Developmen t of trophoblast in vitro
185
blastocyst by surgery. The other possibility that the proliferating trophoblast of
the ectoplacental cone is actually derived from the inner cell mass has been
ruled out by the recent blastocyst reconstruction experiments of Gardner,
Papaioannou & Barton (1973).
The data presented for rates of outgrowth may also indicate another aspect
of the inner cell mass control over blastocyst development. In its absence the
outgrowth rate is faster. The presence of the inner cell mass may have then some
governing function over the time of attachment or implantation, possibly delaying that process until the optimum cell number for its successful completion
is attained.
We thank the Ministry of Agriculture, Fisheries and Food (J.D. A.) and the Ford Foundation (M.H.L.S.) for financial support.
REFERENCES
BARLOW, P., OWEN, D. A. J. & GRAHAM, C. (1972). DNA synthesis in the preimplantation
mouse embryos. /. Embryol. exp. Morph. 27, 431-445.
J. D. & BRINSTER, R. L. (1965). Biometrical problems in the study of early mammalian embryos in vitro. J. exp. Zool. 158, 39-48.
BOWMAN, P. & MCLAREN, A. (1970). Viability and growth of mouse embryos after in vitro
culture and fusion. /. Embryol. exp. Morph. 23, 693-704.
GARDNER, R. L. & JOHNSON, M. H. (1972). An investigation of inner cell mass and trophoblast tissues following their isolation from the mouse blastocyst. J. Embryol. exp. Morph.
28,279-312.
GARDNER, R. L., PAPAIOANNOU, V. E. & BARTON, S. C. (1973). Origin of the ectoplacental
cone and secondary giant cells in mouse blastocysts reconstituted from isolated trophoblast and inner cell mass. /. Embryol. exp. Morph. 30, 561-572.
GWATKIN, R. B. L. (1966). Amino acid requirements for attachment and outgrowth of
mouse blastocysts in vitro. J. cell. Physiol. 68, 335-345.
HILLMAN, N., SHERMAN, M. I. & GRAHAM, C. F. (1972). The effect of spatial arrangement on
cell determination during mouse development. /. Embryol. exp. Morph. 28, 263-278.
3
HORNER, D. & MCLAREN, A. (1974). The effect of low concentration of [ H]thymidine on
pre- and post-implantation mouse embryos. Biol. ofReprod. (In the Press.)
SNOW, M. H. L. (1973 a). Abnormal development of pre-implantation mouse embryos grown
in vitro with [3H]thymidine. /. Embryol. exp. Morph. 29, 601-215.
3
SNOW, M. H. L. (19736). The differential effect of H-thymidine upon two populations of
cells in pre-implantation mouse embryos. In The Cell Cycle in Development and Differentiation. Br. Soc. Devi Biol. Symp. no. 1 (ed. M. Balls and F. S. Billett), pp. 311-324.
Cambridge University Press.
SNOW, M. H. L. & ANSELL, J. D. (1974). The chromosomes of the giant cells in mouse trophoblast. Proc. R. Soc. Lond. B 187, 93-98.
BIGGERS,
{Received 20 May 1974)