We have investigated the developmental potential of mouse blastocysts cultured under a variety of conditions. A number of parameters were used as criteria for development and differentiation, namely hatching of blastocysts from the zona pellucida and their adhesion to the substratum, outgrowth and polyploidization of trophoblast cells, increase in cell number, protein content, β-glucuronidase activity, appearance of lactate dehydrogenase A subunits, plasminogen activator production, and Δ5,3βhydroxysteroid dehydrogenase activity. Under optimal culture conditions, embryos grew relatively rapidly and expressed all the differentiative markers for which they were tested. Under less supportive conditions, the production of the markers was usually reduced quantitatively; the expression of some markers could also be considerably delayed or even totally prevented. In fact, embryos cultured in the least nutritive medium (one designed to support development only through pre-implantation stages) appeared to be in a state of metabolic quiescence closely resembling that of blastocysts in ovariectomy-induced delay. Overall, the results of our investigations lead us to propose that the expression of each of the aforementioned markers is probably under independent control and subject to intrinsic programming. Finally, the observation that some markers are produced by embryos in suboptimal media whereas others are not, suggests that the minimum metabolic level necessary for expression varies from one marker to another.

Mouse blastocysts cultured in appropriate serum-containing media acquire a number of properties analogous to those observed during the peri-implantation period (the time just prior to, during, and shortly after, implantation) in utero (for reviews, see Sherman & Wudl, 1976; Jenkinson, 1977). Among these are the ability of the blastocyst to hatch from the zona pellucida and the acquisition of an adhesive surface and subsequently of migratory properties by the outer, trophoblast layer. Biochemical analyses indicate that blastocysts also express several gene products for the first time during the peri-implantation period both in utero and in culture. These include lactate dehydrogenase (LDH) A subunits (Auerbach & Brinster, 1967, 1968; Monk & Ansell, 1976; Monk & Petzoldt, 1977; Spielmann, Eibs, Jacob-Miller & Bischoff, 1978), plasminogen activator (Strickland, Reich & Sherman, 1976; Sherman, 1980) and Δ5,3β-hydroxysteroid dehydrogenase (3β-HSD) (Chew & Sherman, 1975; Sherman & Atienza, 1975). The fact that blastocysts not only express these properties in vitro but do so in proper sequence and close to the temporal schedule observed in utero suggests that these events are either intrinsically programmed or triggered in response to activators present in serum-containing culture medium as well as in the uterine milieu. The recent development of a partially defined medium, EM2, which is serum-free but nevertheless supports blastocyst development and differentiation through peri-implantation stages (Rizzino & Sherman, 1979), tends to favour the former view.

Blastocysts cultured in EM2 medium hatch, attach and grow out along the culture dish with the same high frequencies observed in serum-containing media, albeit with some delay. However, in several other serum-free media tested, these events were delayed even further or prevented totally (Rizzino & Sherman, 1979; Sherman et al. 1979). We and others had found previously that hatching, attachment and outgrowth were not necessary for the expression of some differentiative markers by cultured blastocysts (Sherman, 1972 a, b; Barlow & Sherman, 1972; Spielmann et al. 1978). In the present study, we have attempted a more comprehensive comparison of the developmental properties of blastocysts in optimal medium and under culture conditions which are either suboptimal for, or preclusive of, the attachment and outgrowth of blastocysts. Accordingly, in addition to assaying blastocysts under the various culture conditions for the expression of the ‘new’ gene products mentioned above, we have also determined their cell numbers and mitotic indices, total protein contents, β-glucuronidase activities [since levels of this enzyme rise rapidly under optimal post-blastocyst culture conditions (Wudl & Sherman, 1976)] and extents of polyploidization. Furthermore, we have compared the properties of blastocysts cultured in media which do not permit attachment and outgrowth with blastocysts under similar constraints in the uteri of ovariectomized females.

Our results indicate that under conditions of metabolic insufficiency, certain differentiative gene products are expressed, whereas the appearance of others is either prevented or substantially delayed. Reasons for this apparent dichotomy are considered.

Embryos

Embryos were obtained from SWR/J (Jackson Laboratory, Bar Harbor, ME) females, superovulated (Runner & Palm, 1953) and mated with SJL/J (Jackson Laboratory) males. Blastocysts were collected on the afternoon of the 4th day of pregnancy (the day of observation of the sperm plug is considered the first day) by flushing the dissected uterine horns with phosphate-buffered saline (PBS - solution A of Dulbecco & Vogt). Prior to culture, all embryos were washed once with PBS. Delayed blastocysts were obtained by ovariectomising pregnant females early on the 3rd day of pregnancy (Yoshinaga & Adams, 1966). Progesterone therapy during delay was not used. Embryos were collected 7 days after ovariectomy in the same way as for normal blastocysts except that they were flushed into a culture dish containing a pad of 2% agarose (SeaKem, MCI Biomedical, Rockland, ME) to prevent attachment (Shalgi & Sherman, 1979). Time in culture is referred to in terms of equivalent gestation days (EGD), the age of the embryos from the time of conception. (When assays were carried out for 24 h periods the EGD is expressed in half days; for example, EGD 5·5 refers to an assay begun on EGD 5 and terminated on EGD 6.)

Media and culture conditions

Bovine serum albumin (BSA; Pentex/Miles Laboratories, Elkhart, IN) was made up as a 6% solution in distilled water, dialysed extensively at 4°C against distilled water, adjusted to a concentration of 3% and stored in aliquots at – 20°C. Fetuin (from bovine serum, Pedersen method, Calbiochem, La Jolla, CA) was solubilized as described by Rizzino & Sato (1978), dialysed extensively at 4°C against acidified distilled water (adjusted to pH 2·5 with 5 N-HCI), and stored in aliquots at –20°C. Aliquots were thawed only once and during use were stored for up to two weeks at 4°C. In all media the following antibiotics were used: penicillin (100u/ml), streptomycin (100μg/ml) and kanamycin (100u/ml) (Gibco, Grand Island, NY).

PCM, the preimplantation culture medium of Goldstein, Spindle & Pedersen (1975), was variously supplemented with 0·05% fetuin (PCM + FET), with amino acids and vitamins (MEM vitamins, Gibco) at the concentrations used by Spindle & Pedersen (1973) (PCM+AA+VIT), or with all of these components (PCM+AA+VIT + FET).

EM2 is a serum-free medium which supports post-implantation development (Rizzino & Sherman, 1979). It consists of equal parts of (i) NCTC 109 medium (Microbiological Associates, Bethesda, MD) supplemented with 0·3% BSA and antibiotics, and (ii) PCM + AA + VIT. This mixture is further supplemented with 0·05% fetuin. EM2 with fetuin omitted (EM2-FET) was also used in these experiments.

NCTC 109 medium supplemented with antibiotics and with 10% heat inactivated (56°C, 20 min) fetal calf serum (FCS; selected lots from Microbiological Associates) was the control medium for these experiments because it has been found to be optimal in this laboratory for blastocyst development (Sherman, 1975).1 This complete medium (abbreviated hereafter as cNCTC) was also used in culture dishes coated with a thin pad of 2% agarose (Sherman, 1978) (cNCT C/agarose).

Unless otherwise noted, all cultures were carried out in 3 ml medium in 35mm tissue-culture-grade plastic dishes (Falcon, Oxnard, CA). Twenty to thirty embryos were cultured per dish. Two milliliters of the medium were changed every second day except where indicated. Cultures were maintained at 37°C in an atmosphere of 5% CO2 in air and at high humidity.

Hatching, attachment and outgrowth

Embryos were inspected daily with a dissecting microscope to determine times of hatching from the zona pellucida, attachment to the substratum and outgrowth of trophoblast cells. T50 refers to the time at which 50% of the embryos had reached a given developmental stage.

Cell counts, mitotic indices and nuclear DNA measurements

The method of Tarkowski (1966), as modified by Pedersen & Cleaver (1975), was employed to produce cellular spreads from whole embryos. The procedures for determination of nuclear DNA content by microfluorometry of Feulgenstained spreads have been described elsewhere (Barlow & Sherman, 1972; Wudl & Sherman, 1978). Liver nuclei DNA values were used to standardize the microfluorometer. Cell numbers and mitotic indices (MI) were determined from the same cell spreads by observation with phase contrast optics.

Protein content

Embryos were washed through three changes of PBS. The zona pellucida, where present, was removed after the first wash by a brief exposure to PBS adjusted to pH 2·5 with 5 N-HCI and containing 0·4% (w/v) polyvinyl pyrrolidone (MW 40000) (Handyside & Barton, 1977). In this and other studies, embryos attached to, and/or outgrowing along, the substratum were carefully detached and collected with a drawn-out and flame-polished capillary tube. Thereafter these embryos were processed in the same way as non-adherent embryos. The washed embryos were transferred in groups of three to five into 10 μl of 0·1 M lithium borate buffer, pH 8·7, containing 0·05% sodium dodecyl sulphate. Samples were frozen and thawed three times and the protein separated from amino acids and peptides (MW ⩽ 6000 daltons) on a Bio-Gel P6 (Bio-Rad Laboratories, Richmond, CA) column. Protein content was measured using fluorescamine reagent (Böhlen, Stein, Dairman & Udenfriend, 1973) and a preparative analytical peptide analyzer (Hoffman-LaRoche Inc., Nutley, NJ) standardized with BSA (manuscript in preparation).

Enzyme analyses

β-Glucuronidase assays were carried out by microflurometric determinations using 4-methylumbelliferyl-β-D-glucuronide (Sigma, St Louis, MO; 3·10−4 M in 0·1 M acetate buffer, pH 5·6) as substrate (Wudl & Sherman, 1976). Each assay sample contained a sufficient number of embryos (from 2 to 30, depending upon EGD and culture medium used) to give an aggregate enzyme activity between 10−7 and 10−6 moles 4-methylumbelliferone produced per hour at 37°C. The production of plasminogen activator was assayed by the fibrinagar overlay procedure (Beers, Strickland & Reich, 1975; Sherman, 1980). The production of plasminogen activator was measured semi-quantitatively by determining the size of the lysis zones, after overnight incubation in the reaction mixture, under a dissecting microscope. Scoring was in arbitrary units as follows: no zone of lysis, 0; faint zone of lysis, 0·5; distinct lysis zone, 0·1–1·0 mm diameter, 1·0; 1·1 to 2·0 mm diameter, 2·0; greater than 2 mm diameter, 3·0. Δ5,3β-Hydroxysteroid dehydrogenase (3β-HSD) activity was determined by radioimmunoassay measuring the conversion of pregnenolone to progesterone by groups of 10–15 embryos (Marcal, Chew, Salomon & Sherman, 1975; Salomon & Sherman, 1975). For LDH analyses, embryos were washed three times with PBS, placed in groups of 50 to 100 into distilled water, and frozen and thawed three times. The samples were then mixed with ethylene glycol (17% final concentration) and bromphenol blue (Sigma), electrophoresed on 0·5 mm thick 8% polyacrylamide slab gels (Dietz & Lubrano, 1967) until the bromphenol blue marker reached the bottom of the gel (approx. 4 h), and stained for LDH activity by the procedure of Shaw & Koen (1968).

Hatching, attachment and outgrowth

Hatching occurred under all culture conditions and in most cases attained levels which were similar to those observed in optimal (cNCTC) medium (Fig. 1). However, both the extent and rate of hatching were substantially reduced in PCM + AA+VIT (Fig. 1D). Although hatching was normal when this medium was further supplemented with fetuin (Fig. 1C), it was noted that during the initial several hours of culture, blastocysts in both of these media did not appear to be as fully expanded as those under other culture conditions. Hatching was delayed slightly, if at all, in EM2 media. Although all blastocysts hatched in PCM media lacking amino acids (Figs. 1A and B), there was evidence of a delay.

Fig. 1

Hatching, attachment and outgrowth of blastocysts in different culture media. Embryos were collected on the fourth day of pregnancy and cultured in groups of 20–30 in 35 mm dishes containing 3 ml of the following media (for full description of culture media see Materials and Methods): (A) PCM+FET; (B) PCM; (C) PCM + AA+VIT+FET; (D) PCM+AA+VIT; (E) EM2; (F) EM2-FET; (G) cNCTC; (H) cNCTC/agarose. Each point represents observations on a total of 50 to 300 embryos from three separate experiments. Embryos were scored by examination under a dissecting microscope for hatching from the zona pellucida (○), attachment to the substratum (▵), and trophoblast outgrowth (▫).

Fig. 1

Hatching, attachment and outgrowth of blastocysts in different culture media. Embryos were collected on the fourth day of pregnancy and cultured in groups of 20–30 in 35 mm dishes containing 3 ml of the following media (for full description of culture media see Materials and Methods): (A) PCM+FET; (B) PCM; (C) PCM + AA+VIT+FET; (D) PCM+AA+VIT; (E) EM2; (F) EM2-FET; (G) cNCTC; (H) cNCTC/agarose. Each point represents observations on a total of 50 to 300 embryos from three separate experiments. Embryos were scored by examination under a dissecting microscope for hatching from the zona pellucida (○), attachment to the substratum (▵), and trophoblast outgrowth (▫).

Attachment of blastocysts occurred soon after hatching in cNCTC medium (Fig. 1G). Embryos attached to the substratum, although after longer periods of culture, in the serum-free media, provided that fetuin was present (Figs. 1 A, C and E). Attachment was completely prevented by coating the dishes with an agarose pad (Fig. 1H) and was only transient in serum-free media lacking fetuin (Figs. 1 B, D and F).

Outgrowths of trophoblast cells were produced by all blastocysts in cNCTC medium. Outgrowths were also observed in EM2 medium and in PCM+AA + VIT + FET but their appearance was delayed and the extent of outgrowth was smaller, particularly in the latter medium. Blastocysts cultured in PCM + FET rarely gave rise to trophoblast outgrowths (Fig. 1 A) and on such occasions only a few outgrowing cells were in evidence. Outgrowths were never observed under the other culture conditions.

Cell number and mitotic index

Under optimal conditions (Fig. 2D), cell number increased almost exponentially throughout the culture period. Embryos prevented from attachment and outgrowth by the presence of an agarose pad contained considerably fewer cells (Fig. 2D). By EGD 11, many of these embryos appeared disorganized and showed signs of necrosis. Cell numbers observed in embryos cultured in EM2 medium initially increased exponentially, although with a prolonged doubling time compared with embryos in cNCTC medium (Fig. 2C). Omission of fetuin from EM2 medium did not affect the initial rise in cell number, but did influence the final numbers reached (Fig. 2C). The gross morphology of the embryos was similar to those cultured in cNCTC medium on agarose pads. In PCM + AA+VJT + FET, embryos averaged two cell doublings by EGD 9 with no subsequent increase in cell number (Fig. 2B). In the other suboptimal media, maximum cell numbers reached only about 120 to 130 and this number declined after EGD 8 (Figs. 2 A, B).

Fig. 2

Cell number and mitotic index of cultured blastocysts. Culture conditions were as follows: (A) PCM ( ○), PCM+FET ( ▵); (B) PCM + AA+VIT (○), PCM + AA + VIT + FET (▵); (C) EM2-FET (○), EM2 (▵); (D) cNCTC/agarose (○), cNCTC (▵). Fourth day values (●) are from blastocysts collected without culture. Embryos were collected from culture at the times indicated and cell spreads were made on microscope slides as described in Materials and Methods. The cells were counted and examined for mitotic figures by phase contrast microscopy. Between 5 and 12 embryos were examined and a mean cell number calculated. Mitotic index (MI) values reflect the percent of cells in mitosis (each telophase pair was counted as a single mitotic figure).

Fig. 2

Cell number and mitotic index of cultured blastocysts. Culture conditions were as follows: (A) PCM ( ○), PCM+FET ( ▵); (B) PCM + AA+VIT (○), PCM + AA + VIT + FET (▵); (C) EM2-FET (○), EM2 (▵); (D) cNCTC/agarose (○), cNCTC (▵). Fourth day values (●) are from blastocysts collected without culture. Embryos were collected from culture at the times indicated and cell spreads were made on microscope slides as described in Materials and Methods. The cells were counted and examined for mitotic figures by phase contrast microscopy. Between 5 and 12 embryos were examined and a mean cell number calculated. Mitotic index (MI) values reflect the percent of cells in mitosis (each telophase pair was counted as a single mitotic figure).

Whereas the MI in cNCTC medium declined gradually over the culture period, consistent with the observed tailing off of the rate of increase of cell number, the MI of embryos in most of the suboptimal media fell more precipitously (Fig. 2). The observation that the MI of embryos were similar in cNCTC medium in the absence or presence of an agarose pad was interesting in view of the differences in cell number (Fig. 2D). In the serum-free media, the MI values generally correlated with the observed cell numbers. There was often a transient increase in the MI of embryos on EGD 7 and again on EGD 9, in each case on the day following renewal of the culture medium.

Protein content

The total protein content of embryos cultured in cNCTC medium increased approximately 100-fold over the period of culture (Fig. 3 D), reflecting the increase in cell number. Embryos cultured on agarose had a proportionately greater increase in protein content than in cell number (see Table 1). Similar relationships were observed for embryos cultured in EM2 media (Fig. 3C and Table 1), that is, the absence of fetuin resulted in an increase in the ratio of protein content to cell number. The increase in protein content in these media over the period of culture was approximately 25-fold. Embryos cultured in PCM supplemented with amino acids and vitamins showed an increase in protein content of only three-to fourfold either in the presence or absence of fetuin (Fig. 3 B). Once again, although the increase in protein content was small compared with that of embryos in optimal medium, it was continuous, unlike the increase in cell number. Finally, embryos cultured in PCM showed a pattern of protein content mirroring the cell number data: there was an increase until EGD 7 followed by a gradual decrease over the remaining culture period (Fig. 3 A). However, in PCM + FET, this decrease was not observed.

Table 1

Protein contents and β-glucuronidase activities of uncultured fourth day or delayed blastocysts and of blastocysts cultured in various media

Protein contents and β-glucuronidase activities of uncultured fourth day or delayed blastocysts and of blastocysts cultured in various media
Protein contents and β-glucuronidase activities of uncultured fourth day or delayed blastocysts and of blastocysts cultured in various media
Fig. 3

Protein content of cultured blastocysts. Culture conditions and symbols are the same as for Fig. 2. Embryos were prepared for protein analysis as described in Materials and Methods. Samples contained three to five embryos and values were averaged from three separate experiments.

Fig. 3

Protein content of cultured blastocysts. Culture conditions and symbols are the same as for Fig. 2. Embryos were prepared for protein analysis as described in Materials and Methods. Samples contained three to five embryos and values were averaged from three separate experiments.

β-Glucuronidase activity

Under all culture conditions, the β-glucuronidase activity of the embryos rose continuously through EGD 10 (Fig. 4). There were, however, variations in the ratios of enzyme activity per cell and per unit protein (Table 1). As with total protein content, the enzyme activity per cell was generally less in media permitting outgrowth than in the equivalent media, lacking fetuin (or with an agarose pad), which precluded outgrowth. The data in Table 1 also indicate that there is a large increase in the amount of β-glucuronidase activity relative to protein content during the period of culture under all conditions used.

Fig. 4

β-Glucuronidase activity of cultured blastocysts. Culture conditions and symbols were the same as for Fig. 2 and enzyme activity was determined as described in Materials and Methods. Between four and six samples from two separate experiments, each sample containing several embryos, were evaluated and mean values calculated.

Fig. 4

β-Glucuronidase activity of cultured blastocysts. Culture conditions and symbols were the same as for Fig. 2 and enzyme activity was determined as described in Materials and Methods. Between four and six samples from two separate experiments, each sample containing several embryos, were evaluated and mean values calculated.

DNA content

The DNA content of the largest trophoblast cells of blastocysts cultured in cNCTC medium underwent five to six cycles of endoreduplication by EGD 11 (Fig. 5D), about one cycle more than the largest trophoblast cells of embryos cultured in cNCTC medium on an agarose pad. The largest trophoblast cells from embryos cultured in EM2 media (Fig. 5C) or in supplemented PCM (Fig. 5 B) underwent fewer cycles and the time of onset of polyploidization was delayed. The presence or absence of fetuin did not markedly affect the ploidy levels reached in these media. Fetuin was, however, essential for the onset of polyploidization of trophoblast cells in otherwise unsupplemented PCM (Fig. 5 A): in PCM alone there was no evidence of polyploidization during seven days of cultnre, whereas polyploidization was observed in trophoblast cells cultured in PCM + FET, although only a single endoreduplicative cycle had taken place by EGD 11.

Fig. 5

Nuclear DNA content of cultured blastocysts. Culture conditions and symbols were the same as for Fig. 2. Cell spreads of embryos collected in two separate experiments were made as for Fig. 2 and the nuclei were stained with Schiff’s reagent. DNA measurements were made by microfluorometry and expressed as multiples of the haploid amount (C) by reference to values obtained for liver nuclei. Each point represents the average of measurements on a total of 18 to 30 nuclei, taking the largest three to five nuclei from each of at least five separate embryos.

Fig. 5

Nuclear DNA content of cultured blastocysts. Culture conditions and symbols were the same as for Fig. 2. Cell spreads of embryos collected in two separate experiments were made as for Fig. 2 and the nuclei were stained with Schiff’s reagent. DNA measurements were made by microfluorometry and expressed as multiples of the haploid amount (C) by reference to values obtained for liver nuclei. Each point represents the average of measurements on a total of 18 to 30 nuclei, taking the largest three to five nuclei from each of at least five separate embryos.

Plasminogen activator production

Under all culture conditions, lysis zones indicating secretion of plasminogen activator first became detectable in embryos assayed between EGD 5 and 6 (Fig. 6). In fact, there was little difference in the size the lysis zones prior to EGD 7. Although embryos continued to secrete plasminogen activator for the duration of the culture period, the average size of the lysis zone differed among culture conditions as would be expected from the other parameters already considered. It should be stressed that lysis zone size is only a semi-quantitative estimation of plasminogen activator production and secretion. For example, outgrowing blastocysts might have larger lysis zones than those failing to outgrow partly because lysis zone size includes the extent of outgrowth in the former case. Another consideration is that the arbitrary units used to represent activity are based on linear measurements, i.e. the diameter of the lysis zones, whereas a more realistic estimate of the amount of enzyme secreted might be in terms of the volume of the lysis zone (which is difficult to measure because of variability in the thickness of the agar overlay). It is, nevertheless, clear from these experiments that embryos under all culture conditions secreted plasminogen activator continuously and without an initial delay and that the amount of plasminogen activator produced per embryo was similar in EM2 and cNCTC medium, substantially less in PCM + AA + VIT, and even less in unsupplemented PCM.

Fig. 6

Plasminogen activator activity in cultured blastocysts. Culture conditions and symbols were the same as for Fig. 2. Enzyme activity was determined by the fibrin-agar overlay procedure and lysis zones were inspected after overnight incubation. The sizes of the lysis zones are expressed in terms of arbitrary units as explained in Materials and Methods. Each point represents the analysis of at least 20 embryos.

Fig. 6

Plasminogen activator activity in cultured blastocysts. Culture conditions and symbols were the same as for Fig. 2. Enzyme activity was determined by the fibrin-agar overlay procedure and lysis zones were inspected after overnight incubation. The sizes of the lysis zones are expressed in terms of arbitrary units as explained in Materials and Methods. Each point represents the analysis of at least 20 embryos.

Δ5,3β-Hydroxysteroid dehydrogenase activity

There were dramatic effects of culture medium on the ability of embryos to convert pregnenolone to progesterone via 3β-HSD activity (Fig. 7). Three points are noteworthy: (a) embryos in PCM (with or without fetuin) failed to produce detectable amounts of progesterone (Fig. 7A); (b) in PCM + AA + VIT small amounts of 3β-HSD activity were observed but, as with polyploidization, the time of appearance of measurable levels of progesterone was delayed substantially; and (c) embryos in EM2 medium (with or without fetuin) appeared to produce much more progesterone than those in cNCTC medium. The last observation is expected on the basis of previous findings and is related to the fact that the inner cell mass (ICM) of the blastocyst develops more extensively in cNCTC medium than in EM2 medium (see Sherman, Atienza, Salomon & Wudl, 1977; Rizzino & Sherman, 1979).

Fig. 7

Conversion of pregnenolone to progesterone by blastocysts cultured in various media. Culture conditions and symbols were the same as for Fig. 2 except that all media were supplemented with 1 μg/ml pregnenolone and that FCS, where used, was dextran-norit treated. 3β-HSD activity was measured for groups of .10 to 15 embryos by determining the amount of progesterone secreted into the medium [Chew & Sherman (1975) have demonstrated that 95% or more of the progesterone formed by blastocysts can be detected in the culture medium at any time during the assay period]. Each point represents the average of determinations on samples of medium from at least six blastocyst cultures from three separate experiments.

Fig. 7

Conversion of pregnenolone to progesterone by blastocysts cultured in various media. Culture conditions and symbols were the same as for Fig. 2 except that all media were supplemented with 1 μg/ml pregnenolone and that FCS, where used, was dextran-norit treated. 3β-HSD activity was measured for groups of .10 to 15 embryos by determining the amount of progesterone secreted into the medium [Chew & Sherman (1975) have demonstrated that 95% or more of the progesterone formed by blastocysts can be detected in the culture medium at any time during the assay period]. Each point represents the average of determinations on samples of medium from at least six blastocyst cultures from three separate experiments.

Lactate dehydrogenase activity

Prior to implantation, mouse embryos contain LDH-1 (four B subunits) as the predominant isozyme, whereas following implantation electrophoretic analysis reveals the presence of enzyme containing A subunits, namely LDH-5 (4A), LDH-4 (3A, IB) and LDH-3 (2A, 2B) (Auerbach & Brinster, 1967, 1968; Monk & Ansell, 1976; Monk & Petzoldt, 1977; Spielmann et al. 1978). Consistent with these reports, blastocysts cultured in cNCTC medium produced A subunits within four days. By EGD 10, A subunits were predominant, since only LDH-4 and LDH-5 bands were observed (Table 2). Similar patterns were observed in EM2 media and in PCM+AA+VIT, both with and without fetuin. However, in PCM with or without fetuin, no LDH activity was detectable by EGD 10 under our conditions. Conversely, blastocysts cultured in cNCTC on an agarose pad showed the normal appearance of A subunits, but the persistence of a strong LDH-1 band on EGD 10 suggested that there had been no obvious diminution of B subunits.

Table 2

Lactate dehydrogenase electrophoretic profiles from embryos cultured in various media

Lactate dehydrogenase electrophoretic profiles from embryos cultured in various media
Lactate dehydrogenase electrophoretic profiles from embryos cultured in various media

In cNCTC/agarose, some blastocysts form aggregates whereas others do not. Accordingly, we analyzed groups of aggregated and nonaggregated blastocysts for LDH activity. We observed that on EGD 6 the aggregated group of embryos contained bands of LDH-2 and LDH-3 as well as LDH-1, whereas non-aggregated embryos contained only LDH-1. However, by EGD 10, the profiles for the two groups contained all five LDH bands.

Blastocysts delayed from implanting in vivo

In order to determine whether embryos prevented from attachment and/or outgrowth in vitro resembled those prevented from implanting in vivo in other respects, we compared the properties of blastocysts maintained in ovariectomised mothers for one week or cultured for a similar period of time, either in PCM, PCM + FET, PCM+AA+VIT or EM2-FET (Table 3).

Table 3

Comparison of properties of blastocysts delayed from implantation in utero with those prevented from attachment and/ or outgrowth in culture for an equivalent period of time

Comparison of properties of blastocysts delayed from implantation in utero with those prevented from attachment and/ or outgrowth in culture for an equivalent period of time
Comparison of properties of blastocysts delayed from implantation in utero with those prevented from attachment and/ or outgrowth in culture for an equivalent period of time

In general, the results indicated that the properties of implantation-delayed blastocysts most closely resembled those of embryos cultured in PCM, with or without fetuin, and suggested that the development of blastocysts cultured in PCM+AA+VIT or EM2-FET was more advanced than in blastocysts delayed in vivo. It should be noted in particular that in vivo-delayed blastocysts resemble embryos cultured in PCM in that they fail to polyploidize, secrete relatively small amounts of plasminogen activator and do not possess detectable 3β-HSD and LDH-5 activities.

In this study, we have for the first time attempted to quantitate a spectrum of developmental parameters as well as markers of differentiation for blastocysts developing under a variety of culture conditions. In cNCTC, a medium which has been demonstrated previously to support substantial post-blastocyst development in vitro (Sherman, 1975), embryos resemble qualitatively those developing in vivo with respect to the timing and sequence of appearance of the markers under investigation. However, even these in vitro conditions, which are the most favorable that we have used, do not support embryonic growth at a rate similar to that which occurs in vivo. For example, an 8th day conceptus in utero contains approximately 20 μg total protein (Sherman, unpublished observations) and the egg cylinder alone contains about 16000 cells (Snow, 1976); in cNCTC on EGD 8, after the most rapid period of growth in vitro, embryos contain only 600 ng protein and about 900 cells, including trophoblast.

On the basis of the protein values, we can estimate that the approximate protein doubling time in cNCTC is about 19 h from the 4th day to EGD 8 (approximately five doublings in 96 h), whereas it is twice as fast (about ten doublings in 96 h, giving a doubling time of 9·6 h) in utero. The calculation of the difference in doubling time with respect to cell number is complicated by the fact that all cells of the blastocyst appear capable of division from the 4th to the 5th day, both in vitro (Fig. 2) and in utero (Barlow, Owen & Graham, 1972), but thereafter some of the cells become non-dividing as they polyploidize to form giant trophoblast cells. If we use 42 cells as an estimate of the average number of the latter cells on the 5th day (see Table 2 of Sherman & Atienza, 1975), then the doubling time of the remainder of the population is approximately 18 h in cNCTC (56 cells to 920 cells in 72 h) and approximately 9 h in utero (56 cells to 16 000 cells in 72 h). Although these estimates do not take into account other possible influential factors such as cell death, they nevertheless suggest that (a) protein and cell doubling times are about twice as rapid in vivo as in vitro and (b) since the protein doubling time (of both ICM and trophectoderm derivatives) is very similar to the cell doubling time (ICM derivatives only), protein content most likely increases at the same rate in the total ICM-derived cell population as in the total trophectoderm-derived cell population.

Several previous studies have focussed upon the molecular requirements for hatching, attachment and outgrowth in vitro. Contrary to the observations of Spindle & Pedersen (1973), we have not noted that hatching is inhibited by the absence of amino acids (Fig. 1). This discrepancy might be due to the different protein concentrations (and thus the potential source of exogenous amino acids) in the media: in the study by Spindle & Pedersen, the medium contained 1% FCS, that is, approximately 0·5 mg/ml total protein, whereas the concentration of BSA in PCM amounts to 3 mg/ml. Nevertheless, aside from a requirement for glucose (Wordinger & Brinster, 1976), presumably as an energy source for blastocyst expansion and contraction during hatching, the blastocyst appears to be undemanding in terms of nutrient requirements for shedding of the zona pellucida.

Blastocysts will attach permanently to the culture dish and trophoblast cells will outgrow along it only in the presence of an appropriate substratum, an adequate source of energy, and an exogenous supply of amino acids (Gwatkin, 1966a, b; Spindle & Pedersen, 1973; Wordinger & Brinster, 1976; Van Blerkom, Chavez & Bell, 1979; Rizzino & Sherman, 1979). Both serum and fetuin preparations contain a component which promotes blastocyst adherence, though not to an agarose pad (Fig. 1; Sherman, 1978). Fetuin does not appear to be required for the onset of blastocyst adhesiveness; however, attachment is only transient in the absence of fetuin preparations or serum (Fig. 1 B, D, F). We should stress that we have not yet established fetuin itself to be the agent which facilitates adhesion since our previous studies indicate that the fetuin preparations used to contain contaminating proteins, albeit at relatively low levels (Rizzino & Sherman, 1979).

Unlike Gwatkin (1966a, b) and Spindle & Pedersen (1973), we have not found that arginine and leucine are required for blastocyst attachment in our media (e.g. Fig. 1 A). We can confirm the requirement for amino acids for trophoblast outgrowth (Gwatkin, 1966a, b; Spindle & Pedersen, 1975) in our serum-free media provided that albumin and fetuin preparations are dialyzed prior to use. However, the observed delay of trophoblast outgrowth in serum-free media containing fetuin and amino acids (Figs. 1 C, E) suggests that other factors, as yet undefined, are also involved.

In general, blastocysts cultured under conditions which prohibit trophoblast outgrowth (that is, in serum-free media lacking fetuin or in serum-containing medium with an agarose pad) develop almost as well as those growing in a permissive medium. There are, however, some notable exceptions. Embryos prevented from outgrowth contain fewer cells beyond three days of culture than those capable of outgrowth. We believe this difference to be due primarily to an increase in the death rate of ICM-derived cells which become internalized in greater numbers by the lack of outgrowth and which, therefore, have impaired access to nutrients. This view is supported by the observed differences in protein content relative to cell number in outgrowing vs. non-outgrowing embryos: the former differences are generally much smaller than the latter (cf. Figs. 2, 3), suggesting that giant trophoblast cells, which undoubtedly contain more protein than individual ICM cells, constitute a greater proportion of the cell population in the absence of outgrowth. Figure 5 confirms that trophoblast cells polyploidize to similar extents in the presence or absence of outgrowth.

During development in utero and under optimal culture conditions, periimplantation embryos show a progressive decrease in LDH 1 activity and the appearance for the first time of LDH A subunits (Auerbach & Brinster, 1967, 1968; Monk & Ansell, 1976; Monk & Petzoldt, 1977; Spielmann et al. 1978). Monk & Petzoldt (1977) proposed from studies on embryos cultured in hanging drops that cell adhesion (either to the substratum or to other cells) was the stimulus for the production of A subunits of LDH. Spielmann et al. (1978), however, failed to observe suppression of A-subunit production in cultured blastocysts which remained viable but failed to hatch from their zona pellucidae. Our results are intermediate between those of Monk & Petzoldt and of Spielmann and his colleagues: individual embryos cultured on agarose do produce detectable levels of A subunits, but at a later stage than aggregated embryos under the same culture conditions. It is, therefore, likely, as considered by Monk & Petzoldt (1977), that even contact of blastomeres with each other following collapse or occlusion of the blastocoel, as eventually occurs when individual blastocysts are cultured on agarose or when they fail to hatch under other culture conditions, is adequate to stimulate production of LDH A subunits. Our LDH studies indicate that as well as producing LDH A subunits, blastocysts cultured on agarose uniquely maintain high levels of LDH B subunits. Whether this reflects the continuing synthesis of these subunits or a decrease in their degradation rate is not clear. Nevertheless, the pattern of enzyme activity (LDH 5 and LDH 1 greater than LDH 3) indicates that the A and B subunits do not interact freely, presumably because they are produced at different times and/or in different cells.

The observation that blastocysts initiate polyploidization in PCM supplemented with fetuin, but not in PCM alone (Fig. 5 A), was unexpected and is not readily explained. We have previously proposed that the role of the fetuin preparation in blastocyst development in vitro is to provide a substratum suitable for attachment and outgrowth (Rizzino & Sherman, 1979). However, outgrowth does not occur in PCM + FET due to the lack of added amino acids; therefore, fetuin, or a contaminating factor in the fetuin preparation, presumably facilitates the onset of polyploidization in this medium at some other level.

Because the various parameters under study do not necessarily respond in the same way to alteration of the culture conditions, we propose that there are variable thresholds for gene expression during the peri-implantation period. We have separated the parameters measured into three classes based upon the nutrient requirements for their expression (Table 4). We should stress, however, that this compartmentalization is not meant to be rigid; indeed, there might be a contiguous spectrum of thresholds for expression of genes during the periimplantation period, and this might have been apparent if we measured many more parameters under a larger array of culture conditions. Nevertheless, it is clear that by alteration of culture conditions, we have been able to suppress the appearance of some gene products but not of others.

Table 4

Threshold levels for expression of developmental and differentiative markers during peri-implantation stages of mouse embryogenesis

Threshold levels for expression of developmental and differentiative markers during peri-implantation stages of mouse embryogenesis
Threshold levels for expression of developmental and differentiative markers during peri-implantation stages of mouse embryogenesis

A variety of factors might account for the differential effects that we have observed. For instance, differences in the complexity of the events might explain why there is a higher threshold for division of cells in the ICM (involving DNA replication and mitosis) than for polyploidization of trophoblast cells (which involves only replication). The same might apply to trophoblast outgrowth vs adhesion. Also, exceptional enzyme stability might result in increasing β-glucuronidase activities, even under the least favourable culture conditions [Wudl (1974) has found enzyme activity to be linear for at least 168 h at 37°C and at least 60 h at 56°C]. Differential gene expression during the peri-implantation period might also reflect differential times of mRNA transcription, mRNA stability and/or preferential mRNA translation. It appears that all of these factors are involved in gene expression during preimplantation stages of mammalian embryogenesis (reviewed by Sherman, 1979; see also Van Blerkom et al. 1979, for a consideration of the control of gene expression during implantation delay and during release from delay). Experiments in progress are consistent with the view that mRNAs for some of the low threshold events in Table 5 are produced earlier than those for some of the high threshold events (J. Schindler, unpublished observations). It is anticipated that with further study, evidence will also be found for translational control of gene expression during the periimplantation period.

We are grateful to Ms Jill Lunn for computation of DNA values and to Drs J. Schindler, J. Monahan and H. Ennis for comments on the manuscript.

Auerbach
,
S.
&
Brinster
,
R. L.
(
1967
).
Lactate dehydrogenase isozymes in the early mouse embryo
.
Expl Cell Res
.
46
,
89
92
.
Auerbach
,
S.
&
Brinster
,
R. L.
(
1968
).
Lactate dehydrogenase isozymes in mouse blastocyst cultures
.
Expl Cell Res
.
53
,
313
315
.
Barlow
,
P. W.
,
Owen
,
D. A. J.
&
Graham
,
C. F.
(
1972
).
DNA synthesis in the preimplantation mouse embryo
.
J. Embryol. exp. Morph
.
27
,
431
445
.
Barlow
,
P. W.
&
Sherman
,
M. I.
(
1972
).
The biochemistry of differentiation of mouse trophoblast: studies on polyploidy
.
J. Embryol. exp. Morph
.
27
,
447
465
.
Beers
,
W. H.
,
Strickland
,
S.
&
Reich
,
E.
(
1975
).
Ovarian plasminogen activator: relationship to ovulation and hormonal regulation
.
Cell
6
,
387
394
.
Bohlen
,
P.
,
Stein
,
S.
,
Dairman
,
W.
&
Udenfriend
,
S.
(
1973
).
Fluorometric assay of proteins in the nanogram range
.
Archs Biochem. Biophys
.
155
,
213
220
.
Chew
,
N. J.
&
Sherman
,
M. I.
(
1975
).
Biochemistry of differentiation of mouse trophoblast: A5,3β-Hydroxysteroid dehydrogenase
.
Biol. Reprod
.
12
,
351
359
.
Dietz
,
A. A.
&
Lubrano
,
T.
(
1967
).
Separation and quantitation of lactate dehydrogenase isoenzymes by disc electrophoresis
.
Analyt. Biochem
.
220
,
246
257
.
Goldstein
,
L. S.
,
Spindle
,
A. I.
&
Pedersen
,
R. A.
(
1975
).
X-ray sensitivity of the preimplantation mouse embryo in vitro
.
Radiat. Res
.
62
,
276
287
.
Gwatkin
,
R. B. L.
(
1966a
).
Amino acid requirements for attachment and outgrowth of the mouse blastocyst in vitro
.
J. cell. Physiol
.
68
,
335
344
.
Gwatkin
,
R. B. L.
(
1966b
).
Defined media and development of mammalian eggs in vitro
.
Ann. N. Y. Acad. Sci
.
139
,
79
90
.
Handyside
,
A. H.
&
Barton
,
S. C.
(
1977
).
Evaluation of the technique of immunosurgery for the isolation of inner cell masses from mouse blastocysts
.
J. Embryol. exp. Morph
.
37
,
217
226
.
Jenkinson
,
E. J.
(
1977
).
The in vitro blastocyst outgrowth system as a model for the analysis of peri-implantation development
.
In Development in Mammals
, vol.
II
(ed.
M. H.
Johnson
), pp.
151
172
.
Amsterdam
:
North-Holland
.
Marcal
,
J. M.
,
Chew
,
N. J.
,
Salomon
,
D. S.
&
Sherman
,
M. I.
(
1975
).
Δ6,3β-Hydroxy-steroid dehydrogenase activities in rat trophoblast and ovary during pregnancy
.
Endocrinology
96
,
1270
1279
.
McLaren
,
A.
&
Hensleigh
,
H. C.
(
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
.
Monk
,
M.
&
Ansell
,
J.
(
1976
).
Patterns of lactic dehydrogenase isozymes in mouse embryo over the implantation period in vivo and in vitro
.
J. Embryol. exp. Morph
.
36
,
653
662
.
Monk
,
M.
&
Petzoldt
,
I.
(
1977
).
Control of inner cell mass development in cultured mouse blastocysts
.
Nature, Lond
.
265
,
338
339
.
Pedersen
,
R. A.
&
Cleaver
,
J. E.
(
1975
).
Repair of UV damage to DNA of implantationstage mouse embryos in vitro
.
Expl Cell Res
.
95
,
247
253
.
Rizzino
,
A.
&
Sato
,
G.
(
1978
).
Growth of embryonal carcinoma cells in serum-free-medium
.
Proc. natn. Acad. Sci. U.S.A
.
75
,
1844
1848
.
Rizzino
,
A.
&
Sherman
,
M. I.
(
1979
).
Development and differentiation of mouse blastocysts in serum-free medium
.
Expl Cell Res
.
121
,
221
233
.
Runner
,
M. N.
&
Palm
,
J.
(
1953
).
Transplantation and survival of unfertilized ova of the mouse in relation to postovulatory age
.
J. exp. Zool
.
124
,
303
316
.
Salomon
,
D. S.
&
Sherman
,
M. I.
(
1975
).
Implantation and invasiveness of mouse blastocysts on uterine monolayers
.
Expl Cell Res
.
90
,
261
268
.
Shalgi
,
R.
&
Sherman
,
M. I.
(
1979
).
Scanning electron microscopy of the surface of normal and implantation-delayed mouse blastocysts during development in vitro
.
J.exp.Zool.21ti
,
69
80
.
Shaw
,
C. R.
&
Koen
,
A. L.
(
1968
).
Starch zone electrophoresis of enzymes
.
In Chromatographic and Electrophoretic Techniques
, vol.
n
(ed.
I.
Smith
), pp.
325
364
.
New York
:
Interscience
.
Sherman
,
M. I.
(
1972a
).
Biochemistry of differentiation of mouse trophoblast: esterase
.
Expl Cell Res
.
75
,
449
459
.
Sherman
,
M. I.
(
1972b
).
The biochemistry of differentiation of mouse trophoblast: alkaline phosphatase
.
Devi Biol
.
27
,
337
350
.
Sherman
,
M. I.
(
1975
).
The culture of cells derived from mouse blastocysts
.
Cell
5
,
343349
.
Sherman
,
M. I.
(
1978
).
Implantation of mouse blastocysts in vitro
.
In Methods in Mammalian Reproduction
(ed.
J. C.
Daniel
, Jr.
), pp.
247
257
.
New York
:
Academic Press
.
Sherman
,
M. I.
(
1980
).
Studies on the temporal correlation between secretion of plasminogen activator and stages of early mouse embryogenesis
.
Oncodevl Biol. Med
.
1
. (In the press.)
Sherman
,
M. I.
&
Atienza
,
S. B.
(
1975
).
Effects of bromodeoxyuridine, cytosine arabinoside and Colcemid upon in vitro development of mouse blastocysts
.
J. Embryol. exp. Morph
.
34
,
467
484
.
Sherman
,
M. I.
,
Atienza
,
S. B.
,
Salomon
,
D. S.
&
Wudl
,
L. R.
(
1977
).
Progesterone formation and metabolism by blastocysts and trophoblast cells in vitro
.
In Development in Mammals
, vol.
n
(ed.
M. H.
Johnson
), pp.
209
233
.
Amsterdam
:
North Holland
.
Sherman
,
M. I.
,
Shalgi
,
R.
,
Rizzino
,
A.
,
Sellens
,
M. H.
,
Gay
,
S.
&
Gay
,
R.
(
1979
).
Changes in the surface of the mouse blastocyst at implantation
.
Ciba Foundation Symp
.
64
,
33
52
.
Sherman
,
M. I.
&
Wudl
,
L. R.
(
1976
).
The implanting mouse blastocyst
.
In The Cell Surface in Animal Embryogenesis and Development
(ed.
G.
Poste
&
G. L.
Nicholson
), pp.
81
125
.
Amsterdam
:
Elsevier North Holland
.
Snow
,
M. H. L.
(
1976
).
Embryo growth during the immediate post-implantation period
.
Ciba Foundation Symp
.
40
,
53
70
.
Spielmann
,
H.
,
Eibs
,
H.-G.
,
Jacob-Müller
,
U.
&
Bischoff
,
R.
(
1978
).
Expression of lactate dehydrogenase isozyme 5 (LDH-5) in cultured mouse blastocysts in the absence of implantation and outgrowth
.
Biochem. Genet
.
16
,
191
202
.
Spindle
,
A. I.
&
Pedersen
,
R. A.
(
1973
).
Hatching, attachment and outgrowth of mouse blastocysts in vitro: fixed nitrogen requirements
.
J. exp. Zool
.
186
,
305
318
.
Strickland
,
S.
,
Reich
,
E.
&
Sherman
,
M. I.
(
1976
).
Plasminogen activator in early embryogenesis: enzyme production by trophoblast and parietal endoderm
.
Cell
9
,
231
240
.
Tarkowski
,
A. K.
(
1966
).
An air-drying method for chromosome preparations from mouse eggs
.
Cytogenetics
5
,
394
400
.
Van Blerkom
,
J.
,
Chavez
,
D. J.
&
Bell
,
H.
(
1979
).
Molecular and cellular aspects of facultative delayed implantation in the mouse
.
Ciba Found. Symp
.
64
,
141
172
.
Wordinger
,
R. J.
&
Brinster
,
R. L.
(
1976
).
Influence of reduced glucose levels on the in vitro hatching, attachment, and trophoblast outgrowth of the mouse blastocyst
.
Devi Biol
.
53
,
294
296
.
Wudl
,
L. R.
(
1974
).
Ph.D. thesis, SUNY, Buffalo, N.Y
.
Wudl
,
L. R.
&
Sherman
,
M. I.
(
1976
).
In vitro studies of mouse embryos bearing mutations at the T locus: tw5 and t12
.
Cell
9
,
523
531
.
Wudl
,
L. R.
&
Sherman
,
M. I.
(
1978
).
In vitro studies of mouse embryos bearing mutations in the T complex: t6
.
J. Embryol. exp. Morph
.
48
,
127
151
.
Yoshinaga
,
K.
&
Adams
,
C. E.
(
1966
).
Delayed implantation in the spayed, progesterone treated adult mouse
.
J. Reprod. Fert
.
12
,
593
595
.

The fluorescamine assay as used by us to measure protein contents involves reaction primarily with lysine residues (Bohlen et al. 1973). We had assumed that proteins in cultured blastocysts contained, on average, an equal amount of all amino acids, so that the lysine content would be 5%. From actual amino acid analyses of proteins from cultured blastocysts (Sellens, M. H., Stein, S., and Sherman, M. I., manuscript in preparation), we have now learned that the average lysine content is approximately 7%. Therefore, protein values given in Fig. 3 are uniformly overestimated by approximately 30%.

1

It has been reported (McLaren & Hensleigh, 1975) that postblastocyst development was very poor in cNCTC. We and others (H. Spielmann, personal communication; C. Hammerberg, personal communication) have observed enormous variation in the effectiveness of NCTC-109 medium from different suppliers. We have found postblastocyst development in supplemented NCTC-109 medium to be highly reproducible provided that we use pretested lots of serum and medium which is less than six months old and which has been purchased from Microbiological Associates.