Inner cell masses (ICMs) were dissected from and day blastocysts and cultured in contact with day morulae. Blastocysts and morulae were homozygous for different electrophoretic variants of the enzyme glucose phosphate isomerase (GPI). Aggregation of ICMs and morulae was observed, and such aggregates were able to form blastocysts in vitro and morphologically normal foetuses in utero. GPI analysis of these conceptuses revealed that most were chimaeric. However, donor ICM-type isozyme was only detected in the embryonic and extra-embryonic fractions of the chimaeras and never in the trophoblastic fraction.

Thus, ICM cells appear unable to form trophoblast derivatives even when exposed to ‘outside’ conditions as experienced by developing trophoblast cells. This is evidence that ICM cells, although not overtly differentiated, are determined by days.

There are two cell populations, morphologically and physiologically distinct, in the day mouse blastocyst: the inner cell mass (ICM) and the trophoblast cells. ICM cells give rise to the embryo and certain extra-embryonic tissues, while the trophoblast cells form the ectoplacental cone and trophoblastic giant cells of the later conceptus (Snell & Stevens, 1966; Gardner, Papaioannou & Barton, 1973). However, it is not yet clear whether the fate of both cell types is irreversibly fixed by the blastocyst stage. By days, trophoblast cells have acquired several specific properties (Gardner, 1972), but ICM cells still resemble those of earlier cleavage stages in several ways :

However, this lack of overt differentiation does not necessarily imply that ICM cells are still labile. When ICMs are isolated from mouse and rat blastocysts and injected into mouse blastocysts, they only colonize the presumptive ICM derivatives of the host embryos and never contribute to the trophoblast (Gardner & Johnson, 1973, 1975; Gardner, 1975). However, the failure of ICM cells to produce trophoblast in this situation does not prove that they are determined, since they were placed in their normal environment. The present experiments were designed to provide a more critical analysis of the determinative state of the ICM by studying its development after isolation or transplantation to a different embryonic site.

Conditions suitable for testing whether ICM cells can form trophoblast are suggested by blastomere aggregation experiments, which have shown that outside blastomeres in aggregates tend to contribute progeny to the trophoblast and trophoblast derivatives and inside blastomeres to the ICM and ICM derivatives (Hillman, Sherman & Graham, 1972). Thus, on this ‘inside-outside’ hypothesis, exposure of ICM cells to ‘outside’ conditions might be expected to induce them to form trophoblast, if they are still naïve. This problem was approached in two ways. One was to study the further development of ICMs whose surface cells had been exposed to the outside by simply isolating them from blastocysts. The results of these experiments will be discussed in a succeeding paper (Rossant, 1975). The other approach, described in this paper, was to attempt to aggregate ICMs with morulae, thus exposing donor ICM cells to ‘outside’ conditions prior to normal host trophoblast formation.

In the initial experiments, single day ICMs were paired with day morulae to test whether aggregation was possible. In later experiments, the number of ICM cells exposed to the outside was increased by aggregating either single day ICMs (30–50 cells) or paired day ICMs (26–32 cells) with each morula.

Recovery of embryos from donor females

Blastocysts and morulae were obtained, after superovulation or natural mating, from mice homozygous for different alleles of the gene for glucose phosphate isomerase (GPI). Gpi-lb/Gpi-lb ICM donor blastocysts were obtained from an inbred strain carrying the alleles Yellow (Ay) and extreme non-agouti (ae). Morulae were obtained from random-bred CFLP mice (Carworth Europe) of genetic constitution Gpi-1a/Gpi-lb. Donor blastocysts were recovered by flushing the uteri between 11.00 and 15.00 on the 4th day after mating (day blastocysts) and between 10.00 and 12.00 on the 5th day after mating (day blastocysts). Eight to sixteen cell morulae were obtained by flushing the oviducts between 14.00 and 19.00 on the 3rd day after mating (-day morulae).

PBI medium (Whittingham & Wales, 1969) containing 10 % foetal calf serum was used for recovery, storage, microsurgery and transfer of embryos, and Ml6 medium was used for culture (Whittingham, 1971).

Micromanipulation and culture in vitro

Donor ICMs were dissected from day and day blastocysts using a Leitz micromanipulator assembly, as described by Gardner (1972). The zonae pellucidae were removed from the host morulae by a 0-5 % solution of Pronase (Calbiochem, Grade B) in phosphate-buffered saline (Mintz, 1962a). Each morula was then placed in a drop of Ml6 under light liquid paraffin together with one or two -day ICMs or one day ICM. ICMs and morulae were brought into contact at 37 °C using a blunt, siliconized glass needle. The culture dishes were then transferred to an incubator at 37 °C and gassed with a mixture of 5 % CO2 in air. Contact of ICMs and morulae was checked after 30 min to 1 h.

Some ICM/morula pairs were cultured until the blastocyst stage (24–36 h) to observe and photograph the process of aggregation. However, when transfer to pseudopregnant recipients was to be performed, the time in culture was kept to the minimum required for aggregation (2–4 h), since it has been shown that the longer mouse eggs are kept in culture, the lower their viability after transfer (Bowman & McLaren, 1970). In only one series of experiments were the embryos cultured overnight before transfer.

Cell counts

Six blastocysts which had developed from ICM/morula aggregates were fixed and processed histologically. Three such blastocysts and two control blastocysts cultured from morulae were fixed in ants’ cocoons, embedded in wax and sectioned serially at 4 μm (Mintz, 1971). The other three experimental blastocysts were fixed in Dalton’s fluid (Dalton, 1955) and embedded in Araldite resin. Two control blastocysts were also fixed in this way. Thick sections (c. 2 μm) were cut from these, using a Huxley ultramicrotome and the sections were stained with a 50:50 mixture of 1 % methylene blue and 1 % Azur II in 1 % borax. Cell counts were made from drawings of the serial sections on transparent paper.

Transfer to pseudopregnant recipients

Recipient mice were used on the 3rd day of pseudopregnancy. ICM/morula aggregates were transferred to one uterine horn and, wherever possible, control morulae were transferred to the contralateral horn. In one experimental series, the control horn contained paired morulae, which had been aggregated in vitro under the same conditions used for ICM/morula aggregation.

Analysis of postimplantation development

In the initial experiments, recipients were killed at or days of pregnancy and any implants in the experimental horns were dissected into embryonic, extra-embryonic (amnion, chorion, allantois and yolk sac), and trophoblastic (ectoplacental cone and trophoblastic giant cells) fractions. Conceptuses of this age provided ample tissue for analysis, but contamination of the trophoblastic fraction with extra-embryonic membranes was apparent during dissection. Thus, in later experiments, recipients were killed at days, when separation of trophoblast and membranes seemed complete. The dissected samples were treated and analysed electrophoretically for GPI by the method of Chapman (Chapman, Whitten & Ruddle, 1971). A single recipient was killed near term ( days) enabling analysis of many different tissues of the conceptuses.

Observations on aggregation in vitro

Successful aggregation of single day ICMs with day morulae was observed in vitro (Fig. 1A-E). Aggregation was achieved within 2–4 h (Fig. 1C) and continued culture usually resulted in blastocysts with single, large ICMs (Fig. 1E). A high proportion of aggregates developed to the blastocyst stage if left in culture (Table 1). However, the proportion of day ICMs aggregating successfully with morulae was small. Altogether, 88 out of 124 (71 %) single or pairs of day ICMs aggregated with morulae, while only 40 out of 119 (33 %) day ICMs did so.

Table 1.

Blastocyst formation in ICM/morula aggregates cultured in vitro

Blastocyst formation in ICM/morula aggregates cultured in vitro
Blastocyst formation in ICM/morula aggregates cultured in vitro
Fig. 1.

Stages in aggregation of ICMs and morulae in vitro. (A) 312-day ICM (left) and morula (right) in culture; (B) 1 h after initial contact, aggregation beginning; (C) 312h after contact; (D) 18 h after contact, cavitation beginning; (E) 36 h after contact, blastocyst formed; (F) blastocyst formed from a control morula after 36 h in culture.

Fig. 1.

Stages in aggregation of ICMs and morulae in vitro. (A) 312-day ICM (left) and morula (right) in culture; (B) 1 h after initial contact, aggregation beginning; (C) 312h after contact; (D) 18 h after contact, cavitation beginning; (E) 36 h after contact, blastocyst formed; (F) blastocyst formed from a control morula after 36 h in culture.

Cell counts

The blastocysts from ICM/morula aggregates appeared to have larger ICMs than control blastocysts (Fig. 1E, F) and cell counts confirmed this visual impression (Table 2).

Table 2.

Results of cell counts on blastocysts derived from ICM/morula aggregates

Results of cell counts on blastocysts derived from ICM/morula aggregates
Results of cell counts on blastocysts derived from ICM/morula aggregates

Implantation rates

The overall rates of implantation and embryonic development of ICM/ morula aggregates were consistently lower than those of control morulae (Table 3). Also, the proportion of recipients which did not become pregnant was larger in all the experimental transfers than in the control series. The most valid comparison of implantation rates is probably made by considering only pregnant females with ICM/morula aggregates in one horn and control morulae in the contralateral horn (Table 4). The difference between experimental and control implantation rates is then less striking, although still apparent for two day ICMs/morula and one day ICM/morula aggregates. The implantation rate of day ICM/morula aggregates is also lower than that of morula/ morula aggregates (Table 5). However, the number of recipients is too small to test whether the differences between experimental and control implantation rates are statistically significant.

Table 3.

Implantation rates of ICM /morula aggregates

Implantation rates of ICM /morula aggregates
Implantation rates of ICM /morula aggregates
Table 4.

Implantation rates of ICM /morula aggregates

Implantation rates of ICM /morula aggregates
Implantation rates of ICM /morula aggregates
Table 5.

Implantation rates of 412day ICM/morula aggregates and morula /morula aggregates

Implantation rates of 412day ICM/morula aggregates and morula /morula aggregates
Implantation rates of 412day ICM/morula aggregates and morula /morula aggregates

GPI activity in conceptuses developed from ICM/morula aggregates

One day ICM/morula aggregates

Eight conceptuses were analysed for GPI and four showed GPI activity of both ICM donor type and host morula type (Table 6, Fig. 2, C1-4). In all four chimaeras, the contribution of ICM-type isozyme to the embryonic and extra-embryonic fractions was approximately equal to the contribution from the morula. However, in two out of four, no ICM-type isozyme could be detected in the trophoblastic fraction. In the other two (C2 and C4) a weak GPI-1B activity was detected in the trophoblastic fraction.

Table 6.

Distribution of GPI isozymes in chimaeric conceptuses derived from ICM/morula aggregates

Distribution of GPI isozymes in chimaeric conceptuses derived from ICM/morula aggregates
Distribution of GPI isozymes in chimaeric conceptuses derived from ICM/morula aggregates
Fig. 2.

Electrophoretograms of conceptuses derived from ICM/morula aggregates. E, Embryonic fraction; EX, extra-embryonic fraction; T, trophoblastic fraction. (A) Gpi-1a/Gpi-1a mouse standard; (B) Gpi-lb/Gpi-lb mouse standard. Numbers correspond to conceptus code numbers in Table 6. No photographs available for C11, C12, C13.

Fig. 2.

Electrophoretograms of conceptuses derived from ICM/morula aggregates. E, Embryonic fraction; EX, extra-embryonic fraction; T, trophoblastic fraction. (A) Gpi-1a/Gpi-1a mouse standard; (B) Gpi-lb/Gpi-lb mouse standard. Numbers correspond to conceptus code numbers in Table 6. No photographs available for C11, C12, C13.

Two day ICMsi morula aggregates

Ten. implanted embryos or resorbing embryos were analysed electrophoretically and nine were found to be chimaeric (Table 6, C5−13; Fig. 2, C5−10). A variety of patterns of GPI activity was observed in the embryonic and extra-embryonic fractions, ranging from C8, where all of the tissues of the embryo and its extra-embryonic membranes showed only ICM-type activity, to C5, where the only contribution of GPI-IB isozyme was a minor one to the extra-embryonic fraction. However, no ICM donor isozyme was detected in the trophoblastic fraction of any embryo analysed at days. ICM-type GPI activity was detected in the trophoblastic fraction of C8, which was analysed at days.

One day ICM/morula aggregates

The GPI activity of seven conceptuses was analysed and two were found to be chimaeric (Table 6, Fig. 2, C14_15). Both were too small to separate embryonic and extra-embryonic fractions but the contribution of ICM-type isozyme to the two egg-cylinders was quite large. There was no GPI-IB isozyme detectable in the trophoblastic fraction of either conceptus.

Isolated day and day mouse ICMs are able to aggregate with ;-day morulae to form blastocysts in vitro and morphologically normal foetuses in utero. Most (15/25) showed GPI activity of both donor ICM and host morula type. However, in all but three such chimaeras, GPI of ICM type could only be detected in the embryo and extra-embryonic membranes, and not in the ectoplacental cone and trophoblastic giant cells (Table 6, Fig. 2). The weak contribution of ICM-type isozyme to the trophoblastic fraction in three conceptuses (C2, C4, C8) was almost certainly due to contamination with extra-embryonic membranes. C2 and C4 were among three chimaeras analysed at days, when clean dissection of the trophoblast was difficult, since union of the chorion and allantois with the ectoplacental cone had already occurred (Snell & Stevens, 1966). C8 was analysed at days when the chorioallantoic placenta is fully formed, making contamination with extra-embryonic membranes inevitable. In all other chimaeras analysed, the separation of trophoblast and extra-embryonic membranes seemed complete.

Thus, it appears that placing ICM cells on the outside of morulae cannot induce them to form trophoblast, although in blastomere aggregates outside cells tend to give rise to trophoblast and trophoblastic derivatives (Hillman et al. 1972). Therefore, the present experiments suggest that the ICM cells of the mouse blastocyst, although not overtly differentiated, are determined by days.

At present, the possibility that some donor ICM cells contribute to the mural trophoblast surrounding the blastocoelic cavity cannot be excluded. Mural trophoblast cells cease dividing and start transforming into primary giant cells early on the 5th day of pregnancy (Dickson, 1966). Thus, donor ICM cell progeny would only be present in large enough numbers to be detectable by GPI analysis if they contributed to the proliferating polar trophoblast overlying the inner cell mass. However, if ICM cells do form trophoblast, the likelihood of 11 conceptuses showing no donor ICM isozyme in the trophoblastic fraction (Table 6, Fig. 2) is remote, unless the ICM cells preferentially form mural trophoblast. Confirmation that no trophoblast cells are formed thus requires a marker that can detect individual cells. Use of [3H]thymidine-labelled ICMs would be questionable because of its toxicity to ICM cells (Snow, 1973). Aggregation, of rat ICMs with mouse morulae may provide a suitable marker since the distribution of individual rat and mouse cells in interspecific chimaeras can be detected by immunofluorescence (Gardner & Johnson, 1973).

Formation of chimaeric foetuses after aggregation of ICMs and morulae is further evidence of the capacity for regulation possessed by pre-implantation mouse embryos, since the cells differ considerably in age. Regulation for asynchrony in cell age has been shown previously for cleavage stages, where embryos which are 10 or 12 h asynchronous can aggregate and form normal blastocysts (Mulnard, 1971; Stern & Wilson, 1972). At the blastocyst stage, chimaeras can be formed after injection of day ICMs into day blastocysts (Gardner, 1971). However, the formation of chimaeric embryos after aggregation of ,-day ICMs and day morulae is the first unequivocal example of regulation for 2 days asynchrony in embryonic cell age. ICM/morula aggregation also demonstrates for the first time that cells from the blastocyst stage can aggregate with pre-blastocyst cells and form normal embryos.

The low rates of implantation of ICM/morula aggregates, particularly day ICM/morula aggregates, also remain to be explained (Tables 3,4). Cell aggregation itself does not seem to reduce implantation since morula/morula aggregates implanted at a much higher rate than day ICM/morula aggregates (Table 5). Another possibility is that the increased size of the ICM in ICM/morula aggregates might reduce the ratio of mural to proliferating polar trophoblast. Since attachment is believed to begin in the abembryonic mural trophoblast (Boyd & Hamilton, 1952), such a reduction might affect the efficiency of implantation. It is also possible that the particularly low rate of implantation of day ICM/ morula aggregates is connected with the difficulty in achieving aggregation of these stages. Tight junctions occur between the endoderm cells of the day ICM (Enders, 1971) and these may interfere with cell aggregation. Thus, some cells might remain on the outside and adversely affect implantation in some way.

I wish to thank Dr R. L. Gardner, Dr V. E. Papaioannou and Mrs S. C. Barton. The author is in receipt of an M.R.C. Research Studentship and the work was supported by the Ford Foundation and the Medical Research Council.

Bowman
,
P.
&
Mclaren
,
A.
(
1970
).
Viability and growth of mouse embryos after in vitro culture and fusion
.
J. Embryol. exp. Morph
.
23
,
693
704
.
Boyd
,
J. D.
&
Hamilton
,
W. J.
(
1952
).
Cleavage, early development and implantation of the egg
.
In Marshairs Physiology of Reproduction
, 3rd ed. (ed.
A. S.
Parkes
), pp.
1
110
.
London
:
Longmans, Green
.
Chapman
,
V. M.
,
Whitten
,
W. K.
&
Ruddle
,
F. H.
(
1971
).
Expression of paternal glucose phosphate isomerase-1 (Gpi-1) in preimplantation stages of mouse embryos
.
Devi Biol
.
26
,
153
158
.
Dalton
,
A. J.
(
1955
).
A chrome-osmium fixative for electron microscopy
.
Anat. Rec
.
121
,
281
.
Dickson
,
A. D.
(
1966
).
The form of the mouse blastocyst
.
J. Anat
.
100
,
335
348
.
Enders
,
A. C.
(
1971
).
Fine structure of the blastocyst
.
In Biology of the Blastocyst
(ed.
R. J.
Blandau
), pp.
71
88
.
University of Chicago Press
.
Gardner
,
R. L.
(
1971
).
Manipulations on the blastocyst
.
In Advances in the Biosciences, no. 6: Schering Symposium on Intrinsic and Extrinsic Factors in Early Mammalian Development
(ed.
G.
Raspé
), pp.
279
296
.
Oxford
:
Pergamon
.
Gardner
,
R. L.
(
1972
).
An investigation of inner cell mass and trophoblast tissue following their isolation from the mouse blastocyst
.
J. Embryol. exp. Morph
.
28
,
279
312
.
Gardner
,
R. L.
(
1975
).
Analysis of determination and differentiation in the early mammalian embryo using intra- and inter-specific chimaeras
.
In The Developmental Biology of Reproduction: 33rd Symposium of the Society for Developmental Biology
(ed.
C. L.
Markert
&
J.
Papaconstantinou
).
Academic Press
:
New York
.
Gardner
,
R. L.
&
Johnson
,
M. H.
(
1973
).
Investigation of early mammalian development using interspecific chimaeras between rat and mouse
.
Nature, New Biol
.
246
,
86
89
.
Gardner
,
R. L.
&
Johnson
,
M. H.
(
1975
).
Investigation of cellular interaction and deployment in the early mammalian embryo using interspecific chimaeras between the rat and mouse
.
In Cell Patterning : Ciba Foundation Symposium
29
, pp.
183
200
.
ASP
:
Amsterdam
.
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
.
J. Embryol. exp. Morph
.
30
,
561
572
.
Hillman
,
N.
,
Sherman
,
M. I.
&
Graham
,
C. F.
(
1972
).
Effect of spatial arrangement on cell determination during mouse development
.
J. Embryol. exp. Morph
.
28
,
263
278
.
Kaufman
,
M. H.
&
Gardner
,
R. L.
(
1974
).
Diploid and haploid mouse parthenogenetic development following in vitro activation and embryo transfer
.
J. Embryol. exp. Morph
.
31
,
635
642
.
Mintz
,
B.
(
1962a
).
Experimental study of the developing mammalian egg: removal of the zona pellucida
.
Science, N. Y
.
138
,
594
595
.
Mintz
,
B.
(
1962b
).
Formation of genotypically mosaic mouse embryos
.
Am. Zool
.
2
,
432
.
Mintz
,
B.
(
1971
).
Allophenic mice of multi-embryo origin
.
In Methods in Mammalian Embryology (ed. J. C. Daniel)
, pp.
186
214
.
San Francisco
:
W. H. Freeman
.
Mulnard
,
J. G.
(
1971
).
Manipulation of cleaving mammalian embryos with special reference to a time-lapse cinematographic analysis of centrifuged and fused mouse eggs
.
In Advances in the Biosciences, no. 6: Schering Symposium on Intrinsic and Extrinsic Factors in Early Mammalian Development
(ed.
G.
Raspé
), pp.
255
277
.
Oxford
:
Pergamon
.
Rossant
,
J.
(
1975
).
Investigation of the determinative state of the mouse inner cell mass. II. The fate of isolated inner cell masses transferred to the oviduct
.
J. Embryol. exp. Morph
.
33
,
991
1001
.
Snell
,
G. D.
&
Stevens
,
L. C.
(
1966
).
Early embryology
.
In Biology of the Laboratory Mouse
, 2nd ed. (ed.
E. L.
Green
), pp.
205
245
.
New York
:
McGraw Hill
.
Snow
,
M. H. L.
(
1973
).
Abnormal development of preimplantation mouse embryos grown in vitro with [3H]thymidine
.
J. Embryol. exp. Morph
.
29
,
601
615
.
Stern
,
M. S.
&
Wilson
,
I. B.
(
1972
).
Experimental studies on the organization of the preimplantation mouse embryo. I. Fusion of asynchronously cleaving eggs
.
J. Embryol. exp. Morph
.
28
,
247
254
.
Tarkowski
,
A. K.
(
1961
).
Mouse chimaeras from fused eggs
.
Nature, Lond
.
190
,
857
860
.
Whittingham
,
D. G.
(
1971
).
Culture of mouse ova
.
J. Reprod. Fert. Suppl
.
14
,
7
21
.
Whittingham
,
D. G.
&
Wales
,
R. G.
(
1969
).
Storage of two-cell mouse embryos in vitro
.
Aust. J. Biol. Sci
.
22
,
1065
1068
.