The in vitro growth and morphogenesis of mouse embryos from early primitive-streak stage to early-somite stage is described. The embryo culture method employs a static culture system, a conventional chemically defined medium (Dulbecco’s modified Eagle’s medium), supplemented with additional glucose, glutamine and suitably prepared serum. The method of serum preparation is important for successful culture. Both mouse serum and rat serum support good development of primitive-streak-stage mouse embryos. Over 60% of early-streak stage and about 90 % of late-streak stage grow and develop for 48 h in vitro. During the first 24 h in culture, total growth of the embryos, as reflected by protein content, size and morphology is the same as in vivo.

The development of mammalian embryos is conveniently divided into four phases: preimplantation, implantation and gastrulation, organogenesis, and growth.

Embryo size militates against the in vitro culture of whole embryos during the growth phase, but the other phases have been studied with some success. Pre-implantation stages of many species are routinely grown in vitro with minimal disturbance to either timing, form of development or subsequent viability (Whittingham, 1975), but successful culture of the later stages is largely confined to rats and mice (New, 1978). In the rat, using the techniques developed by D. A. T. New, it is possible to achieve normal development of late-streak and head-fold stages (0-5 somites) for nearly 4 days in culture and later stages, up to - days of gestation, for somewhat shorter periods (Buckley, Steele & New, 1979; Cockcroft, 1976). For the mouse, Clarkson, Doering & Runner (1969), Pienkowski, Solter & Koprowski (1974) and Kochhar (1975) reported limited success with late organogenesis stages, and Sadler (1979) describes good growth and development of early-somite embryos in rat serum for periods up to 48 h.

The intervening phase, from implantation to head-fold formation, includes the process of gastrulation which seems to represent a hiatus for the culture techniques. Many methods are available for the mouse which mimic the process of implantation and some of these allow the development to pre-primitive-streak egg cylinders, albeit rather slowly in comparison with development in vivo (Jenkinson & Wilson, 1970; Hsu, 1973; Hsu, Baskar, Stevens & Rash, 1974; Pienkowski et al 1974; McLaren & Hensleigh, 1975). But the major hurdle during this phase of development seems to be primitive-streak formation - the onset of gastrulation. At this transition point morphology of the cultured embryos usually becomes abnormal and only limited numbers of embryos proceed to somite stages (Hsu et al. 1974; Wiley & Pedersen, 1977). Recently a marked improvement in the yield of somite-stage embryos has been reported by Hsu (1979), using a system requiring careful choice of several different media formulations and a stringent schedule of media changing. We have focused attention on the egg cylinder to early-somite stages and below describe a simplified method for the culture of embryos at the primitive-streak stages.

Embryos were obtained from the random bred Q and MFI strains of mice. Early-primitive-streak stage (6·5 days p.c.) and late-primitive-streak stage (7·5 days p.c.) embryos were studied. The embryos were dissected from the decidua. Reichert’s membrane and ectoplacental cone trophectoderm were removed. It is important not to damage the primitive endoderm surface of the egg cylinder. Embryos were cultured in groups of four to six in 30 mm Falcon plastic Petri dishes containing 2·0-2·5 ml culture medium. These small dishes were then placed in groups in larger plastic containers (large Petri dishes or conveniently sized boxes) in which wet filter paper had been placed to ensure a humidified atmosphere. The whole assembly was placed in an incubator at 37 °C with a humidified atmosphere of 5 % CO2 in air.

Culture media

Four chemically defined media have been tested, supplemented with various sera. The media, NCTC-109, Waymouth’s, RPM1 1640, and Dulbecco’s modified Eagle’s medium (DMEM) foetal calf serum (FCS) and newborn calf serum (NCS) were obtained from Gibco Bio-Cult Ltd. Mouse serum (MS) was collected in the laboratory from MFI, Q, GW, C3H, C57BL and a number of hybrid strains. The serum was obtained by immediately centrifuging the blood collected from etherized mice, in a way similar to the collection of rat serum (Steele, 1972). Rat serum (RS) was collected from albino rats of unknown strain which were kindly supplied by Dr C. Hetherington, Clinical Research Centre, Harrow.

Benzyl penicillin (100 i.u./ml) and streptomycin sulphate (50/μg/ml) (Glaxo) were added to all media.

Assessment criteria

The development of the embryos in culture was assessed by comparison of morphology with the embryos of equivalent post-coital age in utero. Special emphasis was placed on the development of head fold, heart, neural tube, somites and gut formation. The size of the embryos was measured from camera-lucida drawings done at x 60 magnification. After macroscopic examination, some embryos were fixed in Sanfelice’s fluid, wax embedded and examined histologically. The remaining embryos were used for the determination of protein content by the colorimetric method of Lowry, Rosebrough, Farr & Randall (1951).

In pilot experiments late-streak-stage embryos were grown in each of the synthetic media supplemented with 20 % FCS (after McLaren & Hensleigh, 1975). After mixing but before gassing with 5 % CO2 in air these media have different osmolarities and pH values at 20 °C (Table 1). Development in all media was somewhat slower than in vivo but two important observations were made during these pilot studies: (1) the best yolk-sac and head-fold morphology seem to be correlated and were seen in Waymouth’s and DM EM, media with a high glucose concentration ; (2) after 24 h embryos in DMEM were growing more vigorously, estimated on the basis of yield of metaphase plates from a conventional air drying chromosome spreading technique. The second observation is supported by the evidence of further culture. Comparatively few embryos grew well in Waymouth’s or RPM1 1640 during the second 24 h period, whereas in DMEM 50 % proceeded through somitogenesis and developed beating hearts. Despite inferior head-fold morphology nearly 38 % of embryos in NCTC 109 progressed into somitogenesis. A more detailed comparison between DMEM and NCTC 109 confirmed that the high glucose medium is superior (Table 2). Further study therefore focused on DMEM and variations based on that medium.

Table 1.

A comparison between the four media used in pilot studies. Each was supplemented with 20 % FCS. pH and molarity were measured on complete medium but prior to gassing with 5 % CO2 in air

A comparison between the four media used in pilot studies. Each was supplemented with 20 % FCS. pH and molarity were measured on complete medium but prior to gassing with 5 % CO2 in air
A comparison between the four media used in pilot studies. Each was supplemented with 20 % FCS. pH and molarity were measured on complete medium but prior to gassing with 5 % CO2 in air
Table 2.

Development of late-streak-stage embryos in NCTC-109 or DM EM supplemented with 20 % FCS

Development of late-streak-stage embryos in NCTC-109 or DM EM supplemented with 20 % FCS
Development of late-streak-stage embryos in NCTC-109 or DM EM supplemented with 20 % FCS

A series of experiments was carried out to investigate optimal concentrations of serum (FCS, NCS, MS and RS) and the importance or otherwise of adding pyruvate, glutamine and bicarbonate to the basic medium just prior to use (to compensate for decomposition during storage in solution in the case of pyruvate and glutamine, or to correct for pH variations introduced with serum in the case of bicarbonate). These experiments produced the following results.

(1) For late-primitive-streak-stage embryos, 30 % FCS or NCS supports normal development of some embryos but 50-100 % mouse serum or rat serum is optimal for embryonic development;

(2) Additional L-glutamine is beneficial when added to the medium at a final concentration of 2 HIM;

(3) Sodium pyruvate should be omitted from the stock DMEM and added to complete the medium immediately prior to use (at a concentration of 0-1 mM);

(4) The preparation of serum is also important. Heat inactivation of serum should be done immediately before use. Embryotrophic properties are destroyed by freezing and storage after inactivation, but active serum can be stored at — 20 °C for several months with little loss of potency. Serum should not be filter-sterilized if this can be avoided, and certainly not following heat inactivation. However gamma-irradiated, non-filtered serum is not suitable for culture. In several replicate experiments using gamma-irradiated NCS not one embryo made normal head folds.

Table 3 shows the respective merits of FCS and NCS for late-streak-stage embryos. Embryos cultured in NCS-supplemented medium developed much better than in FCS-supplemented medium for the first 24 h. More embryos showed development of yolk sac, neural fold, beating heart, invaginating foregut and formed more somites. Development of embryos in FCS + DMEM was very poor in the second 24 h. Although development in NCS + DMEM was retarded, a high percentage of the embryo showed development of neural tube, initial closure of cephalic fold and elongation of trunk.

Table 3.

Development of late-streak embryos in DM EM supplemented with either 30 % FCS or 30 % NCS

Development of late-streak embryos in DM EM supplemented with either 30 % FCS or 30 % NCS
Development of late-streak embryos in DM EM supplemented with either 30 % FCS or 30 % NCS

With early-primitive-streak stages, FCS and NCS are inadequate, and embryonic development is abnormal. For these earlier stages, MS or RS is essential for good cultures and is also an improvement for late-primitive-streak stages. At the time of culture, 6·5-day embryos were either in the pre-primitive-streak or early-primitive-streak stage (Fig. 1). After 24 h in vitro, 65-88 % of the embryos cultured in mouse serum or rat serum (whole or diluted) reached the late-primitive-streak stage (Table 4) and showed normal development of embryonic and extraembryonic structure, principally the yolk sac (Figs. 2 and 3). There was significant but not excessive expansion of the yolk sac and amniotic cavity. The size and protein content of the embryos developed in vitro for 24 h was the same as those at 7· days in vivo (Table 5). In the embryos failing to develop normally, expansion of the yolk sac was usually excessive and a single large vesicle resulted. The abnormal embryos also contained less protein (Table 5). About 55-79 % of the embryos developed for another 24 h to the head-fold and early-somite stage (Table 4). Equivalent-aged (8·5 day) embryos in vivo have five to seven pairs of somites, well developed head fold, closing neural tube, invaginating foregut and early formation of heart tube. The cultured embryos showed a wide range of developmental stage: from presomite neural plate to early-somite stage with prominent head folds and beating heart (Figs. 4 and 5). The more advanced embryos had normal numbers of somites, but a shorter body axis and significantly less protein when compared to 8-day embryos in vivo (Table 5, Figs. 6 and 7). Histologically, the development of neural fold, heart, gut and mesenchyme were apparently normal (Figs. 12 and 13).

Table 4.

Development of primitive-streak-stage embryos in whole mouse serum, DM EM+50% mouse serum, whole rat serum and D MEM +50% rat serum

Development of primitive-streak-stage embryos in whole mouse serum, DM EM+50% mouse serum, whole rat serum and D MEM +50% rat serum
Development of primitive-streak-stage embryos in whole mouse serum, DM EM+50% mouse serum, whole rat serum and D MEM +50% rat serum
Table 5.

Growth of embryos in vitro, culture medium is DMEM+50 % mouse serum

Growth of embryos in vitro, culture medium is DMEM+50 % mouse serum
Growth of embryos in vitro, culture medium is DMEM+50 % mouse serum
Fig. 1.

Early-primitive-streak stage (6·5-days p.c.) embryos at the time of explanation. Bar = 500 μm.

Fig. 1.

Early-primitive-streak stage (6·5-days p.c.) embryos at the time of explanation. Bar = 500 μm.

Fig. 2.

Late-primitive-streak-stage (7·5-days p.c.) embryos in vivo. Bar = 500 μm.

Fig. 2.

Late-primitive-streak-stage (7·5-days p.c.) embryos in vivo. Bar = 500 μm.

Fig. 3.

Early-primitive-streak-stage embryos cultured for 24 h in vitro. Bar = 500 μm.

Fig. 3.

Early-primitive-streak-stage embryos cultured for 24 h in vitro. Bar = 500 μm.

Fig. 4.

Early-primitive-streak-stage embryos cultured for 48 h in vitro. Bar = 500 μm.

Fig. 4.

Early-primitive-streak-stage embryos cultured for 48 h in vitro. Bar = 500 μm.

Fig. 5.

One of the best embryos developed to early-somite stage at the end of the culture. Extraembryonic membranes were dissected away. Bar = 500 μm.

Fig. 5.

One of the best embryos developed to early-somite stage at the end of the culture. Extraembryonic membranes were dissected away. Bar = 500 μm.

Fig. 6.

Four embryos after culturing for 48 h in vitro, starting from early-primitivestreak stage. Yolk sacs were dissected away. Bar = 500 μm.

Fig. 6.

Four embryos after culturing for 48 h in vitro, starting from early-primitivestreak stage. Yolk sacs were dissected away. Bar = 500 μm.

Fig. 7.

Four 8-5-days early-somite-stage embryos in vivo. Yolk sacs and allantois were removed. Bar = 500 μm.

Fig. 7.

Four 8-5-days early-somite-stage embryos in vivo. Yolk sacs and allantois were removed. Bar = 500 μm.

Between 84 and 100 % of the late-primitive-streak-stage (7·5-day) embryos (Fig. 8) developed normally in culture during the first 24 h (Table 4). The embryos had six to seven pairs of somites, well-formed head folds, closing neural tube, invaginating foregut (Figs. 14-17) and some had a beating heart. Gross morphology was similar to the 8·5-day embryos in vivo (Fig. 9,10 and 11). Total growth as measured by total protein content and body axis length also appeared normal and no significant differences in protein content and size were observed between similar-staged embryos maintained in vitro or in vivo (Table 5). The few late-streak embryos failing to develop normally have invariably attached to the dish where the embryonic tissues grew out as a monolayer disrupting normal morphogenesis in the process. These embryos were most probably damaged during explantation from the decidua. During the second 24 h in culture, the growth of the embryos was depressed and overall development was retarded. However, there was further development of blood circulation (vigorous heart beat and yolk-sac circulation) and formation of craniofacial (cephalic fold, mandibular arch, optic vesicles) trunk (neural tube, somites) and tail-bud structures, The best embryos had about 15-20 somites and axial rotation had begun. Although some embryonic structures like the head and heart were of abnormal size, histologically the tissue appeared very healthy. No extensive necrosis was found and many cell divisions were seen especially in the neuro-epithelium.

Fig. 8.

Late-primitive-streak stage (7·5 days p.c.) embryos at the time of explanation. Bar = 500μm.

Fig. 8.

Late-primitive-streak stage (7·5 days p.c.) embryos at the time of explanation. Bar = 500μm.

Fig. 9.

Late-primitive-streak-stage embryos cultured for 24 h in vitro. Bar = 500 μm.

Fig. 9.

Late-primitive-streak-stage embryos cultured for 24 h in vitro. Bar = 500 μm.

Fig. 10.

Four embryos after culturing for 24 h in vitro. Yolk sacs and allantois were removed. Bar = 500 μm.

Fig. 10.

Four embryos after culturing for 24 h in vitro. Yolk sacs and allantois were removed. Bar = 500 μm.

Fig. 11.

Four 8·-day early-somite-stage embryos in vivo. Yolk sacs and allantois were removed. Bar = 500μm.

Fig. 11.

Four 8·-day early-somite-stage embryos in vivo. Yolk sacs and allantois were removed. Bar = 500μm.

Fig. 12.13.

Early-primitive-streak-stage embryos cultured for 48 h in vitro. NF, neural fold. HF, head fold. HT, heart. FG, invaginating fore gut. Bar = 200 μm.

Fig. 12.13.

Early-primitive-streak-stage embryos cultured for 48 h in vitro. NF, neural fold. HF, head fold. HT, heart. FG, invaginating fore gut. Bar = 200 μm.

Fig. 14—17.

Late-primitive-streak-stage embryos cultured for 24 h in vitro. Sections were through head region (Fig. 14), heart (Fig. 15), posterior trunk region (Fig. 16), and mid trunk region (Fig. 17). Bar = 100 μm. DA, dorsal aorta; HM, head mesendyme; HT, heart tube; NE, neuroepithelium; NF, neural fold; NT, neural tube; PC, pericardial cavity; SM, somite; TM, trunk mesoderm.

Fig. 14—17.

Late-primitive-streak-stage embryos cultured for 24 h in vitro. Sections were through head region (Fig. 14), heart (Fig. 15), posterior trunk region (Fig. 16), and mid trunk region (Fig. 17). Bar = 100 μm. DA, dorsal aorta; HM, head mesendyme; HT, heart tube; NE, neuroepithelium; NF, neural fold; NT, neural tube; PC, pericardial cavity; SM, somite; TM, trunk mesoderm.

Cultures in whole serum are not significantly superior to those using mouse serum or rat serum diluted 50 % with DMEM. There are differences between batches of rat serum which are not associated with degree of haemolysis in preparations. Since mouse serum is pooled from many mice, variation is not seen in these preparations. As observed by Steele (1972) the mode of serum preparation is important, immediately centrifuged serum being better than serum centrifuged after blood clotting. Table 6 describes the protocol of serum and medium preparation we found best.

Table 6.

Protocol for preparation of serum supplemented DMEM for mouse embryo culture

Protocol for preparation of serum supplemented DMEM for mouse embryo culture
Protocol for preparation of serum supplemented DMEM for mouse embryo culture

The results of this study indicate that primitive-streak-stage mouse embryos can be cultured through gastrulation to early-somite stage in a simple static culture system. The embryos develop best in whole mouse or rat serum, and in medium supplemented with 50 % mouse or rat serum. Over 60 % of the early-primitive-streak-stage embryos develop to the early-somite stage. Developmental events such as gastrulation, head-fold and neural-tube formation, gut invagination, heart growth and formation of the first six to eight somites all occurred on schedule and appeared normal. Total growth of embryos, as measured by protein content and size, is essentially the same as in vivo during the first 24 h in culture. Growth of embryos is depressed in the second 24 h and development of embryos beyond the early-somite stage (equivalent to 8-5 days in vivo) was retarded. It seems, therefore, that this culture system is most suitable for the observation and experimental manipulation of embryos between primitive-streak stages and early-somite stage for periods of morphogenesis of about 24 h. The inadequacy of the present method for culturing embryos beyond the somite stages could be twofold. First, the essential nutrient and energy substrate in the culture medium may be depleted during the initial culture period and would not support normal growth of older embryos. Rat embryos at and beyond the head-fold stage are known to have a stringent requirement of nutrients and energy substrates for normal development in vitro (Cockroft, 1979). Secondly, it has been shown that early stages of rat and mouse embryos require initially a low oxygen tension in the culture system and as development progresses a higher oxygen tension is necessary (New, Coppola & Cockroft, 1976; Buckley et al. 1978; Sadler, 1979; Morriss & New, 1979). The use of 5 % CO2 in air (i.e. about 20 % O2) may be beneficial for early embryonic development in this study, because in the static culture described, gaseous diffusion may be less efficient and a low oxygen tension is therefore maintained in the medium. The same gas phase would be insufficient for older embryos when there is a more demanding oxygen requirement. The transfer of cultured late-streak-stage embryos from a static culture to a rolling-bottle culture during the second 24 h of culture may perhaps improve the success rate and allow normal development of embryos beyond the early-somite stage (Sadler, 1979).

During the development of early-primitive-streak-stage embryos, the primitive endoderm occasionally becomes raised into fluid-filled blisters over the surface of the embryonic part of the egg cylinder. Such embryos do not develop normally. Similarly, overexpansion of the yolk sac and amniotic cavity are often accompanied by degeneration of embryonic structures. All these phenomena suggest that the normal osmoregulation between embryonic compartments and the environment has failed to operate. Whether this is due to intrinsic errors in embryonic physiology during culture or because media/culture conditions overwhelm the embryo’s regulatory capacities is not known. It may not be coincidence that the most successful culture medium also has the highest osmolarity.

The differences seen in sera prepared in different ways may suggest a way to identify the embryotrophic factor(s). Heat inactivation does not remove these factors but does alter them so that they can be destroyed at least partially by freezing and removed by Millipore filtration. The observation that filtration of fresh serum will lower its potency may indicate that the factors are already partly denatured before heat inactivation. Taken in conjunction with the fact that immediately centrifuged serum, i.e. serum collected before blood clotting occurs, is superior in embryo cultures, it is possible that the differences in serum types with respect to their embryotrophic properties are created entirely by artifacts of preparation. Commercially available serum is generally collected after whole blood has clotted and is sterilized before dispatch by membrane filtration. These various preparative steps may have depleted the serum of identifiable factors such as, but in addition to, the proteins found in immediately centrifuged serum, but which are apparently absent in serum collected after whole blood clotting (Klein, Minghetti, Jackson & Vogler, 1978).

Buckley
,
S. K. L.
,
Steele
,
C. E.
&
New
,
D. A. T.
(
1978
).
In vitro development of early postimplantation rat embryos. Devi Biol
.
65
,
396
403
.
Clarkson
,
S. G.
,
Doering
,
J. V.
&
Runner
,
M. N.
(
1969
).
Growth of postimplantation mouse embryos cultured in a serum supplemented chemically defined medium
.
Teratology
2
,
181
186
.
Cockroft
,
D. L.
(
1976
).
Comparison of in vitro and in vivo development of rat fetuses
.
Devi Biol
.
48
,
163
172
.
Cockroft
,
D. L.
(
1979
).
Nutrient requirements of rat embryos undergoing organogenesis in vitro
.
J. Reprod. Fert
.
57
,
505
510
.
Hsu
,
Y-C.
(
1973
).
Differentiation in vitro of mouse embryos to the stage of early somite
.
Devi Biol
.
33
,
403
11
.
Hsu
,
Y-C.
(
1979
).
In vitro development of individually cultured whole mouse embryos from blastocyst to early somite stage. Devi Biol
.
68
,
453
461
.
Hsu
,
Y-C.
,
Basker
,
J.
,
Stevens
,
L.
&
Rash
,
J.
(
1974
).
Development in vitro of mouse embryos from the two-cell stage to the early somite stage
.
J. Embryol. exp. Morph
.
31
,
235
245
.
Jenkinson
,
E. J.
&
Wilson
,
I. B.
(
1970
).
In vitro support system for the study of blastocyst differentiation in the mouse
.
Nature, Lond
.
228
,
776
778
.
Klein
,
N. W.
,
Minghetti
,
P. P.
,
Jackson
,
S. K.
&
Vogler
,
M. A.
(
1978
).
Serum protein depletion by cultured rat embryos
.
J. exp. Zool
.
203
,
313
319
.
Kochhar
,
D. M.
(
1975
).
The use of in vitro procedures in teratology
.
Teratology
11
,
274
287
.
Lowry
,
A. H.
Rosebrough
,
N. J.
,
Farr
,
A. L.
&
Randall
,
R. J.
(
1951
).
Protein measurement with the Folin reagent
.
J. biol. Chem
.
193
,
265
275
.
Mclaren
,
A.
&
Hensleigh
,
H.
(
1975
).
Culture of mammalian embryos over the implantation period
.
In The Early Development of Mammals
(ed.
M.
Balls
&
A. E.
Wild
), pp.
45
60
.
Cambridge University Press
.
Morriss
,
G. M.
&
New
,
D. A. T.
(
1979
).
Effect of oxygen concentration on morphogenesis of cranial neural folds and neural crest in cultured rat embryos
.
J. Embryol. exp. Morph
.
54
,
17
35
.
New
,
D. A. T.
(
1978
).
Whole-embryo culture and the study of mammalian embryos during organogenesis
.
Biol. Rev
.
53
,
81
122
.
New
.
D. A. T.
,
Coppola
,
P. T.
&
Cockroft
,
D. L.
(
1976
).
Improved development of head-fold rat embryos in culture resulting from low oxygen and modification of the culture serum
.
J. Reprod. Fert
.
48
,
219
222
.
Pienkowski
,
M.
,
Solter
,
D.
&
Koprowski
,
H.
(
1974
).
Early mouse embryos: growth and differentiation in vitro
.
Expl Cell Res
.
85
,
424
428
.
Sadler
,
T. W.
(
1979
).
Culture of early somite mouse embryos during organogenesis
.
J. Embryol. exp. Morph
.
49
,
17
25
.
Steele
,
C. E.
(
1972
).
Improved development of rat ‘egg-cylinders’ in vitro as a result of fusion of the heart primodia
.
Nature New Biology
237
,
150
151
.
Whittingham
,
D. G.
(
1975
).
Fertilisation, early development and storage of mammalian ova in vitro
.
In The Early Development of Mammals
, pp.
1
24
.
Cambridge University Press
.
Wiley
,
L. M.
&
Pedersen
,
R. A.
(
1977
).
Morphology of mouse egg cylinder development in vitro: A light and electron microscope study
.
J. exp. Zool
.
200
,
389
402
.