/ . Embryol. exp. Morph. Vol. 36, 1, pp. 163-174, 1976 \ 63

Printed in Great Britain

Investigation of inner cell massdetermination by aggregation of isolated rat inner

cell masses with mouse morulae

By J. ROSSANT1

From the Department of Zoology, Oxford

SUMMARYInner cell masses (ICMs) dissected from 4-̂ -day rat blastocysts were aggregated with 2^-day

mouse morulae. Successful aggregates formed blastocysts in vitro and morphologically normal5i-day conceptuses in the mouse uterus. Immunofluorescent analysis of these conceptusesrevealed that rat cells were only present in the embryonic ectoderm and endoderm and neverin the trophectoderm derivatives, although rat trophoblast did develop in the mouse uterus invarious control experiments. The single-cell resolution of this technique extends the resultsobtained from aggregating mouse ICMs with mouse morulae and provides strong evidencethat ICM cells, although not overtly differentiated, are determined by the blastocyst stage.

INTRODUCTION

The inner cell mass (ICM) isolated from the mouse blastocyst, although notan overtly differentiated tissue (Gardner, 1972) is apparently determined by3% days post coitum, as judged by its inability to contribute to post-implantationtrophoblast after aggregation with morulae or after transfer to the oviduct andlater injection into blastocysts (Rossant, 1975 a, b). In these experiments, electro-phorettc variants of glucose phosphate isomerase (GPI) were used as geneticmarkers. However, the low sensitivity of this technique means that a minorcontribution of ICM cells to the proliferating trophoblast or a contribution tothe non-dividing mural trophoblast giant cells cannot be excluded (Gardner,Papaioannou & Barton, 1973). A marker that can detect individual cells is re-quired to eliminate this possibility. No such markers are yet available for themouse (Gardner & Johnson, 1975), but immunofluorescent analysis of rat/mouse chimaeras permits identification of single rat or mouse cells (Gardner& Johnson, 1973, 1975). Thus, an attempt was made to extend the results ofexperiments using mouse ICM/mouse morula aggregates by aggregating ratICMs with mouse morulae. Immunofluorescent analysis can then be used todiscover whether any rat ICM cells contribute to the trophoblast of resultingconceptuses. Analysis was performed at 5^ days of pregnancy in the mouse

1 Author's address: Department of Zoology, South Parks Rd., Oxford 0X1 3PS, U.K.II-2

164 J. ROSSANT

when all three presumptive derivatives of the trophectoderm - ectoplacentalcone, primary giant cells and extra-embryonic ectoderm - should be present(Gardner & Johnson, 1975; Gardner & Papaioannou, 1975).

MATERIALS AND METHODS

Recovery of embryos from donor females

Mouse morulae (8-16 cell) were flushed from the oviducts of CFLP females(Anglia Laboratory Animals Ltd) on the afternoon of the 3rd day after mating.Rat morulae at similar cleavage stages were flushed from the uterotubal junc-tions of CFHB random-bred females (Anglia Laboratory Animals Ltd) on theafternoon of the 4th day after mating, while rat blastocysts were obtained fromthe uterus on the afternoon of the 5th day. It is assumed throughout that ratearly development is 24 h behind that of the mouse.

PB1 medium +10 % foetal calf serum (FCS) was used for recovery, storage,microsurgery and transfer of embryos (Whittingham & Wales, 1969) and Ml6medium + 10 % FCS was used for culture (Whittingham, 1971).

Aggregation of culture and embryos

1. Rat ICM/mouse morula aggregates

ICMs were dissected from 4^-day rat blastocysts and aggregated with mousemorulae as described previously for mouse ICM/mouse morula aggregates(Rossant, 1975a). All aggregates were cultured in M16. A few pairs were cul-tured to the late blastocyst stage to observe and photograph the process ofaggregation. The remainder were cultured for either 2-4 h or 24 h (latemorula/early blastocyst) before transfer to pseudopregnant mice.

2. Rat blastocysts

Some 4^-day rat blastocysts were not dissected but were stored briefly in PB1+10 % FCS before transfer to mice.

3. Rat morulae

Rat morulae (3£ days) were cultured for 24 h in M16 or M16 +10 % FCSbefore transfer to recipient rats or mice. In most cases the zonae were removedby Pronase (Mintz, 1962) before culture.

4. Rat morula /mouse morula aggregates

The zonae were removed from both rat and mouse morulae by Pronase andrat/mouse pairs were brought into contact in drops of M16 or M16 +10 % FCS.The cultures were then transferred to a 37 °C incubator. Contact between themorulae was checked after 1 h. After 24 h in culture, aggregated embryos weretransferred to pseudopregnant mice.

Investigation of inner cell mass determination 165

Transfer to recipient animals

All four types of embryos listed above were transferred to recipient mice on the3rd day of pseudopregnancy. Rat recipients were used for uterine transfer on the4th day of pregnancy after the oviducts had been ligated on day 2. Rat embryoscultured for 24 h were transferred unilaterally so that the contralateral hornprovided a control for effective ligation of the oviduct.

Analysis of post-implantation developmentHistology

Recipient mice which had received non-cultured or cultured rat embryos, andone recipient which contained rat morula/mouse morula aggregates, were killed onthe 6th day after mating (5^-day embryos). Uteri containing implants were fixedin AFA, processed and embedded in wax (Orsini, 1962). Serial sections were cutat 6-7 /.im, stained with haemalum and eosin and examined for embryonicderivatives.

Recipient rats which had received cultured rat embryos were killed on the7th day after mating and uteri containing implants were processed as above.

Immunofluorescent analysis

Mice which had received rat ICM/mouse morula aggregates and rat morula/mouse morula aggregates were killed on the sixth day of pregnancy. Deciduawere processed and embedded in wax (Sainte-Marie, 1962) and sectioned at6 jam. The sections were treated with antisera specific for mouse and rat anti-gens as described previously (Gardner & Johnson, 1973) and examined byepifluorescent illumination in a Zeiss fluorescence microscope (Primary filter,barrier filter HBO 200). Camera lucida drawings were made of all sections,marking rat and mouse cells.

RESULTS

In vitro observations

1. Rat ICMjmouse morula aggregates

Successful aggregation of rat ICMs with mouse morulae was achieved in vitroin Ml6 (Table 1). The rate of aggregation was lower than for mouse ICM/morula combinations (Rossant, 1975 a) but a large proportion of successfulaggregates formed blastocysts after 24 h in culture (Table 1).

2. Rat morulae

Rat morulae were cultured to the blastocyst stage in both M16 and M16 +10 % FCS. However, success was rather variable and only 32/66 and 10/28morulae respectively underwent further cleavage.

166 J. ROSSANT

Table 1. Rates of in vitro aggregation and blastocyst formation of ratICMj mouse morula and rat morula I mouse morula combinations

Type of combination Culture medium No. pairs

No. aggregated(percentageaggregation)

No. blastocystsformed (percen-tage formation)

Rat ICM/mouse M16 48morula

Rat morula/mouse Ml6 45morula

Rat morula/mouse M16 +10 % FCS 143morula

23(48) 19(83)

31(69) 26(84)

65(45) 45(69)

Table 2. Implantation rates {only pregnant recipients considered)

Type of embryo transferredType of No. No. No. (%) No. (%)recipient recipients transferred implants embryos

Rat ICM/mouse morulaaggregates

Rat blastocystsCultured rat blastocystsCultured rat blastocystsRat morula/mousemorula aggregates

Mouse

MouseMouseRatMouse

5

2448

22

10162236

11(50)

9(90)10(62)8(36)

23C64)

7(32)

8(80)8(50)4(18)

14(39)

3. Rat morulajmouse morula aggregates

Successful aggregation of rat and mouse morulae was achieved in culture inboth M16 and M16 +10 % FCS (Table 1). The rather low rate of aggregationwas chiefly due to the fairly frequent failure of rat morulae to develop in culture,as reported above. A large proportion of the successful aggregates formed blas-tocysts after about 24 h in culture (Table 1), as reported previously by otherworkers (Mulnard, 1973; Stern, 1973; Zeilmaker, 1973).

Analysis of post-implantation development

1. Rat ICMjmouse morula aggregates

The implantation rate of these aggregates was rather low (Table 2; seeRossant, 1975 a). However, seven embryos were obtained and examined byimmunofluorescence. Five proved to be interspecific chimaeras; their patternsof chimaerism are summarized in Table 3. Conceptus 90a was obtained from anaggregate transferred after 2-4 h in culture, whereas the others were culturedfor 24 h before transfer. All five were morphologically normal although embryos90a and 95b were retarded slightly and lacked extra-embryonic ectoderm. Noneshowed contribution of rat cells to the mural trophoblast, ectoplacental cone orextra-embryonic ectoderm.

Investigation of inner cell mass determination 167

Table 3. Estimation by immunofluorescence of the proportion of rat cells in5\-day conceptuses derived from rat ICMjmouse morula aggregates

Mural Polar Extra-Conceptus tropho- tropho- Proximal Distal Embryonic embryoniccode no. blast blast endoderm endoderm ectoderm ectoderm

89a — — ++ ++ + + + —89c — — + + — —89e — — — + + — —90a — — + + + + + + + + + n.p.95b — — — — + + + + n.p.

Total + foreach tissue 0 0 7 8 9 0

Total no.embryos withrat cells ingiven tissue 0 0 3 4 3 0

Key to tables 3 and 4. + signs represent visual estimations of the proportion of rat cellsin each tissue. + + + + = all rat; + + + = mostly rat > 50%; + + = some rat < 50%;+ = very little rat; - = no rat; n.p. = tissue not present.

2. Rat blastocysts

As a control for survival of rat trophoblast in the mouse uterus, non-culturedrat blastocysts were transferred to recipient mice. A high proportion werecapable of inducing decidual formation (Table 2). Histological examination ofthe 5-|-day decidua revealed that most contained embryonic cells. Six out ofeight contained a definite egg-cylinder but this was never well organized (Fig. 1).Ectoderm and endoderm were generally delineated but no division into em-bryonic and extra-embryonic ectoderm was apparent and the orientation ofthe embryos was often abnormal. All conceptuses contained trophoblast cellsand in six out of eight definite mural giant cells could be identified (Fig. 1).Pycnotic cells were present to a greater or lesser extent in all conceptuses.

3. Rat morulae

To control for the ability of rat trophoblast to survive in the mouse afterdevelopment in culture, blastocysts developed from rat morulae cultured for24 h in M16 + 10%FCS were transferred to pseudopregnant mice. Decidualformation was induced (Table 2) and 50 % of the embryos transferred producedembryonic structures. Four such structures consisted solely of groups of tropho-blast cells with some giant cell formation (Fig. 2). The other four consisted ofwell-expanded trophoblastic vesicles, containing only a few dispersed cells ontheir inner surface (Fig. 3). No organized ICM was present and the inner cellswere presumably endoderm. The beginning of giant cell formation was alsoapparent in these structures.

168 J. ROSSANT

In all figures, the grid bar represents 50/<m.Fig. 1. Section of 5^-day conceptus developed from non-cultured rat blastocysttransferred to the mouse uterus. Arrows indicate definite trophoblast giant cells.

When rat blastocysts cultured in Ml6 alone were transferred to rats, the im-plantation rate was low (Table 2). However, considering the fairly high implan-tation rate in transfers to mice, this may be largely due to lack of practice inperforming rat uterine transfers. Of the eight decidua formed, four had normalrat egg-cylinders and two had trophoblast cells only.

Investigation of inner cell mass determination 169

Fig. 2. Group of trophoblast cells produced 5-j- days after transfer of culturedrat blastocyst to the mouse uterus. Arrows indicate definite trophoblast giant cells.

4. Rat morulajmouse morula aggregates

To control for the development of rat trophoblast in competition with mousecells, rat morula/mouse morula aggregates were transferred to mouse uteri. Thenumber of post-implantation embryos formed from these aggregates was rather

170 J. ROSSANT

Fig. 3. Section of 5^-day trophoblast vesicle containing only a few endoderm cellsafter transfer of cultured rat blastocyst to the mouse uterus.

low (Table 2). The first three embryos were examined histologically and werefound to be egg-cylinder-like but not well organized (Fig. 4). Trophoblast,ectoderm and endoderm were present in all three, but the expected divisioninto embryonic and extra-embryonic ectoderm was not found. However, sincetrophoblast cells were found, a further 11 conceptuses were analysed by immuno-fluorescence to see if any of the trophoblast cells were rat.

Investigation of inner cell mass determination 171

Fig. 4. Section of 5-i-day conceptus derived from rat morula/mouse morula aggregate.

The patterns of chimaerism in these embryos are summarized in Table 4.None was a completely normal 5-|-day embryo. Some resembled retarded blasto-cysts (e.g. R/M 19.3-4) while others were more advanced but more disorganized(e.g. R/M 19.3-3). The orientation of some was abnormal (e.g. R/M 28.3-2).However, distal and proximal endoderm, mural trophoblast, ectoplacental coneand embryonic ectoderm were present in all embryos. There was a definitetendency for the rat cells to colonize the endoderm in preference to other tissues(Table 4) but 7 out of 11 contained rat ectoderm cells. Also, 10 out of 11contained rat giant trophoblast cells, and 9 out of 11 rat polar trophoblast.

172 J. ROSSANT

Table 4. Estimation by immunoftuorescence of the proportion of rat cells in5\-day conceptuses derived from rat morulajmouse morula aggregates

Conceptus Mural Polar Proximal Distal Embryoniccode no. trophoblast trophoblast endoderm endoderm ectoderm

92a74 + + ++ + + + + +96b74 — — + + + + —R/M 19-3-1 + — + + + + + —R/M 19-3-2 + + + + + + + + + —R/M 19-3-3 + + + + + + + + + + +R/M 19-3-4 + + + + + + + + —R/M 28-3-1 + + + + + + + + + + + + + +R/M 28-3-2 + + + + + + + + + +R/M 28-3-3 + + + + + + + + + +R/M 9-5 + + + + + + + + + + + + +R/M 15-5 + + + + + + + + + + + +

Total + for each tissue 17 17 30 27 15Total no. embryoswith rat cells ingiven tissue 10 9 11 11 7

Conceptuses 92a74 and 96b74 and the three conceptuses examined histo-logically were derived from aggregates cultured in M16 while the rest had beencultured in M16 + 10%FCS. No obvious differences in morphology wereapparent between the two groups of embryos.

DISCUSSION

Aggregation of 4^-day rat ICMs with 2^-day mouse morulae has been achievedin vitro and morphologically normal blastocysts were produced. After transferto pseudopregnant mice, these blastocysts formed apparently normal 5^-dayconceptuses, of which a high proportion (5/7) were inter-specific chimaeras.However, immunofluorescent analysis revealed rat cells only in the presumptiveICM derivatives, i.e. embryonic ectoderm, distal and proximal endoderm andnever in the trophectoderm derivatives, i.e. ectoplacental cone, extra-embryonicectoderm and mural trophoblast (Gardner & Johnson, 1975; Gardner & Papaio-annou, 1975). Because of the single-cell resolution of the analysis these resultscompletely exclude the possibility of a contribution, however minor, of ICMcells to the trophoblast. Thus they extend the conclusions drawn from GPIanalysis of mouse ICM/mouse morula aggregates reported previously (Rossant,1975 a).

However, since interspecific chimaeras were used, the validity of the presentresults could be queried, unless it can be shown that rat trophoblast can developin the mouse until at least 5^ days of pregnancy. Non-cultured rat blasto-cysts produced both proliferating trophoblast and giant cells at 5£ days in themouse, although the embryonic regions were rather disorganized (Fig. 1).

Investigation of inner cell mass determination 173

Culture of rat morulae to the blastocyst stage before transfer to mice also didnot prevent trophoblast formation although no conceptus contained an organ-ized ICM (Figs. 2, 3). Finally, rat trophoblast, both proliferating and giant,could even be formed in competition with mouse cells in rat morula/mousemorula aggregates (Table 4, Fig. 4). Development of rat trophoblast in themouse under such a variety of conditions means that the absence of rattrophoblast in conceptuses derived from rat ICM/morula aggregates can beconsidered as a valid result. Thus, the present experiments provide strongevidence that ICM cells, although not overtly differentiated, are determined bythe blastocyst stage. This is supported by recent evidence of molecular differen-tiation between ICM and trophectoderm (Van Blerkom, Barton & Johnson,1976).

The present experiments also provide some new information on the morphol-ogy of rat/mouse chimaeras. Chimaeras derived from rat ICM/mouse morulaaggregates are morphologically normal at least in early stages, as are thoseproduced by injection of rat ICMs into mouse blastocysts (Gardner & Johnson,1973, 1975). However, none of the 14 5^-day conceptuses derived from ratmorula/mouse morula aggregates could be considered normal (Fig. 4). Thereasons for this difference are not clear, but at least two possible explanationscan be suggested.

Firstly, culture may produce deleterious effects on rat morulae and not on ratICMs, since consistent success in culturing rat morulae to the blastocyst stagewas not achieved. In particular, the predominance of endodermal chimaerism inrat morula/mouse morula aggregates (Table 4) might be related to the effects ofculture. Cultured rat blastocysts never produced egg-cylinder structures whentransferred to the mouse, but four out of eight produced trophoblast vesiclescontaining only a few dispersed endoderm cells. However, six out of eight non-cultured rat blastocysts did produce egg-cylinders containing ectoderm cells.Thus there seems to be a correlation between culture of rat embryos and lack ofectoderm formation, although Tarkowski has observed similar trophoblastvesicles containing only endoderm cells after transfer of rat eggs to the mouseoviduct (Tarkowski, 1962).

The second factor exerting a deleterious effect on the development of ratmorula/mouse morula chimaeras may be the formation and continued presenceof rat trophoblast, since neither rat ICM/mouse morula nor rat ICM/mouseblastocyst chimaeras contain this tissue. Tarkowski (1962) has suggested thatthe abnormal development of rat eggs after implantation in the mouse may bedue to defective interaction of the rat trophoblast with the mouse uterine epi-thelium. Similar effects could be occurring in some rat morula/mouse morulaaggregates. Mystkowska (1975) has reported similar poor postimplantationdevelopment from bank vole morula/mouse morula aggregates, and she sug-gested that the wide taxonomic gap and differences in the course of embryo-genesis between the two species were the main causes. However, the present

174 J. ROSSANT

experiments indicate that even morulae of more closely related species like themouse and the rat are rarely able to integrate and form normal conceptuses.

I should like to thank Dr R. L. Gardner, Dr M. H. Johnson and Mr A. Copp for valuablediscussion. Special thanks are due to Dr Johnson, who provided the specific antisera andlaboratory facilities for performing the immunofluorescent analysis. Mrs L. Ofer and MrsS. Clutterbuck-Jackson provided technical assistance. The author was supported by a MedicalResearch Council Research Studentship and a Beit Memorial Junior Research Fellowship.The work was supported by Medical Research Council project grants to Dr Gardner andDr Johnson.

REFERENCES

GARDNER, R. L. (1972). An investigation of inner cell mass and trophoblast tissue followingtheir isolation from the mouse blastocyst. / . Embryol. exp. Morph. 28, 279-312.

GARDNER, R. L. & JOHNSON, M. H. (1973). Investigation of early mammalian developmentusing interspecific chimaeras between rat and mouse. Nature (New BioL), Lond. 246, 86-89.

GARDNER, R. L. & JOHNSON, M. H. (1975). Investigation of cellular interaction and deploy-ment in the early mammalian embryo using interspecific chimaeras between the rat andmouse. In Cell Patterning (Ciba Found. Symp. 29), pp. 183-200. Amsterdam: AssociatedScientific Publishers.

GARDNER, R. L. & PAPAIOANNOU, V. E. (1975). Differentiation in the trophectoderm andinner cell mass. In The Early Development of Mammals {2nd Symposium of the BritishSociety for Developmental Biology) (ed. M. Balls & A. E. Wild), pp. 107-132. CambridgeUniversity Press.

GARDNER, R. L., PAPAIOANNOU, V. E. & BARTON, S. C. (1973). Origin of the ectoplacentalcone and secondary giant cells in mouse blastocysts reconstituted from isolated tropho-blast and inner cell mass. / . Embryol. exp. Morph. 30, 561-572.

MINTZ, B. (1962). Experimental study of the developing mammalian egg: removal of the zonapellucida. Science, N. Y. 138, 594-595.

MULNARD, J. G. (1973). Formation de blastocystes chimeriques par fusion d'embryons de ratet de souris au stade VIII. C. r. hebd. Seanc. Acad. Sci. Paris 276, 379.

MYSTKOWSKA, E. T. (1975). Development of mouse-bank vole interspecific chimaeric embryos./ . Embryol. exp. Morph. 33, 731-744.

ORSINI, M. W. (1962). Technique of preparation, study and photography of benzyl-benzoatecleared material for embryological studies. / . Reprod. Fert. 3, 283-287.

ROSSANT, J. (1975a). Investigation of the determinative state of the mouse inner cell mass.I. Aggregation of isolated inner cell masses with morulae. / . Embryol. exp. Morph. 33,979-990.

ROSSANT, J. (19756). Investigation of the determinative state of the mouse inner cell mass.II. The fate of isolated inner cell masses transferred to the oviduct. / . Embryol. exp. Morph,33,991-1001.

SAINTE-MARIE, G. (1962). A paraffin embedding technique for studies employing immuno-fluorescence. J. Histochem. Cytochem. 10, 250-256.

STERN, M. S. (1973). Chimaeras obtained by aggregation of mouse eggs with rat eggs. Nature,Lond. 243, 472-473.

TARKOWSKI, A. K. (1962). Interspecific transfers of eggs between rat and mouse. / . Embryol.exp. Morph. 10, 476^95.

VAN BLERKOM, J., BARTON, S. C. & JOHNSON, M. H. (1976). Molecular differentiation in thepreimplantation mouse embryo. Nature, Lond. 259, 319-321.

WHITTINGHAM, D. G. (1971). Culture of mouse ova. / . Reprod. Fert. Suppl. 14, 7-2.1.WHITTINGHAM, D. G. & WALES, R. G. (1969). Storage of two-cell mouse embryos in vitro.

Aust. J. BioL Sci. 22, 1065-1068.ZEILMAKER, G. H. (1973). Fusion of rat and mouse morulae and formation of chimaeric

blastocysts. Nature, Lond. 242, 115-116.

(Received 28 January 1976; revised 26 April 1976)